Accepted Manuscript

Efficacy of a hybrid nanofluid in a new microchannel heat sink

equipped with both secondary channels and ribs

Mehdi Bahiraei, Mohammad Jamshidmofid, Marjan Goodarzi

PII: S0167-7322(18)34289-2
DOI: doi:10.1016/j.molliq.2018.10.003
Reference: MOLLIQ 9748
To appear in: Journal of Molecular Liquids
Received date: 19 August 2018
Revised date: 28 September 2018
Accepted date: 2 October 2018

Please cite this article as: Mehdi Bahiraei, Mohammad Jamshidmofid, Marjan Goodarzi
, Efficacy of a hybrid nanofluid in a new microchannel heat sink equipped with both
secondary channels and ribs. Molliq (2018), doi:10.1016/j.molliq.2018.10.003

This is a PDF file of an unedited manuscript that has been accepted for publication. As
a service to our customers we are providing this early version of the manuscript. The
manuscript will undergo copyediting, typesetting, and review of the resulting proof before
it is published in its final form. Please note that during the production process errors may
be discovered which could affect the content, and all legal disclaimers that apply to the
journal pertain.
 ACCEPTED MANUSCRIPT

Efficacy of a hybrid nanofluid in a new microchannel heat sink

equipped with both secondary channels and ribs

Mehdi Bahiraei1,*, Mohammad Jamshidmofid1, Marjan Goodarzi2

1
Department of Mechanical Engineering, Kermanshah University of Technology, Kermanshah, Iran
2
Sustainable Management of Natural Resources and Environment Research Group, Faculty of Environment and

T
IP
Labour Safety, Ton Duc Thang University, Ho Chi Minh City, Vietnam

CR
* Corresponding author: Mehdi Bahiraei (E-mail: m.bahiraei@kut.ac.ir)

Tel: +988337259980

US
Email of “Marjan Goodarzi”: marjan.goodarzi@tdtu.edu.vn
AN
Abstract
M

This paper aims to evaluate the thermohydraulic attributes of a hybrid nanofluid containing
ED

graphene–silver nanoparticles in a microchannel heat sink equipped with the ribs and secondary

channels. In addition to the heat transfer surface increment, the ribs direct the flow towards the
PT

secondary channels, and intensify the flow mixing. Meanwhile, the secondary channels increase
CE

the flow area, which reduces the pressure drop due to the presence of the ribs. The results show

that combining the three approaches, namely employing the nanofluid, ribs and secondary
AC

channels in the microchannel improves the heat sink performance significantly. With increasing

either concentration or Reynolds number, the temperature decreases, the temperature uniformity

enhances, and the regions with highest temperatures become smaller. Additionally, the average

convective heat transfer coefficient enhances with increasing the concentration and Reynolds

number such that with increase of the concentration from 0 to 0.1% at Re=100, a 17%

1
 ACCEPTED MANUSCRIPT

enhancement happens in the convective heat transfer coefficient. Moreover, the bottom surface

temperature decreases with increment of the concentration such that a 3.42 K reduction occurs

with increasing the concentration from 0 to 0.1% at Re=100. Meanwhile, the flow experiences a

greater pumping power at higher Reynolds numbers and concentrations.

T
Keywords: Hybrid nanofluid, microchannel heat sink, rectangular ribs, secondary channels,

IP
graphene nanoplatelets, thermal performance

CR
1. Introduction

US
After inventing the integrated circuits (ICs), electronics has witnessed a crucial switch towards
AN
miniaturizing electronic devices. Due to the significant increase in heat generation rates from the

new electronic chips, essential engineering challenges are noticed in thermal management and
M

cooling of these electronic components. In fact, a miniature silicon chip might include billions of
ED

transistors which act with great frequencies. In addition, novel development in laser industry has

caused the utilizing devices with great heat fluxes like high energy mirrors or laser diode
PT

arrangements having large powers. This leads to considerable heat production rates from such
CE

equipment. As a result, the common techniques such as employing air cooling methods have

become insufficient to convey these great heat fluxes. Recently, several cooling approaches
AC

including micro-heat pipe, micro-jet impingement, microelectro-hydrodynamic and

microchannel heat sink have been experimented to meet the requests of great heat dissipation

rates and keeping a small junction temperature. Among these methods, microchannel heat sinks

can be the best selection owing to their promising characteristics like compactness, small weight,

and high ratio of area to volume [1, 2]. The initial concept of microchannel heat sink was

2
 ACCEPTED MANUSCRIPT

introduced by Tuckerman and Pease [3]. The authors argued that their suggested outline can

dissipate a heat flux by 105 W/m2 with highest temperature difference of 71°C between the inlet

liquid and substrate.

After the work of Tuckerman and Pease [3], many investigators have studied thermohydraulic

efficiency of the rectangular microchannels. Lee et al. [4] experimentally investigated single-

T
phase heat transfer inside a rectangular microchannel for different hydraulic diameters. They

IP
certified the conventional correlations for the prediction of thermal attributes in the rectangular

CR
microchannel. Peng [5] studied the friction and heat transfer characteristics of a microchannel

heat sink at various hydraulic diameters. The authors claimed that both flow and heat transfer in

US
this device are considerably affected by the channel aspect ratio. Toh et al. [6] numerically
AN
evaluated the friction factor in a microchannel heat sink. It was shown that the Poiseuille number

reduces by the Reynolds number decrement. This observation was attributed to the decrease in
M

the viscosity because of greater temperatures at small Reynolds number.

ED

Many other scholars have concentrated on the influence of cross sectional shape on

thermohydraulic efficiency of microchannel heat sinks via examining different shapes like
PT

triangular, circular, trapezoidal, and so forth. Gunnasegaran et al. [7] numerically investigated
CE

the influence of three shapes, namely rectangular, trapezoidal and triangular microchannel heat

sinks. It was found that maximum heat transfer enhancement is achieved in the rectangular heat
AC

sink followed by the trapezoidal and triangular ones. Alfaryjat et al. [8] assessed the effect of

three different cross-sectional shapes such as circular, hexagonal, and rhombus configurations.

They reported that the hexagonal heat sink presents the maximum pressure loss and heat transfer

amount. It was also revealed that the rhombus microchannel demonstrates the minimum thermal

resistance and friction coefficient.

3
 ACCEPTED MANUSCRIPT

Although great heat transfer rates can be obtained through microchannel heat sinks, more

improvements are still required to meet cooling demands of electronic components. Because of

very small sizes of these devices, flow in ordinary straight microchannels is laminar, which

decreases the ability of microchannel heat sinks to remove high heat fluxes. Thus, several

scholars have tried to employ various approaches for heat transfer enhancement in these systems.

T
Modifications in the geometry as well as employing superior coolants are the most efficient

IP
methods for this purpose.

CR
Ribs are recently used as one of the heat transfer enhancement methods owing to their ability in

interrupting thermal boundary layer, which can intensify flow mixing and consequently heat

US
transfer amount. However, rather intense flow disturbances and blocking-flow phenomenon due
AN
to applying ribs can significantly increase pressure loss and pumping power as well. Therefore,

for decreasing the pressure losses due to ribs, many scholars have tried to increase the flow area
M

through employing different grooves in heat sinks. Li et al. [9] simulated fluid flow and heat
ED

transfer in a microchannel heat sink with ribs in central part of the channel and triangular

grooves arranged on the walls. The results indicated that the suggested outline contributes in
PT

increasing flow mixing by promoting the jetting, throttling and chaotic advection. Moreover, it
CE

caused larger flow area which decreased the pressure loss. Xia et al. [10] compared the

performance of four ribs with different shapes including trapezoidal, rectangular, triangular and
AC

circular with fan-shaped reentrant cavities. It was found that at Reynolds numbers lower than

300, the microchannel having trapezoidal ribs and fan-shaped reentrant cavities shows the

maximum performance.

Utilizing better coolants is also one of the methods for enhancing performance of microchannel

heat sinks. Nanofluids can be excellent candidates for this goal. In fact, the results reported from

4
 ACCEPTED MANUSCRIPT

many papers prove that nanofluids possess much better thermal characteristics compared with

conventional liquids [11-20]. Due to the appropriate attributes of nanofluids, some researchers

have employed them as coolant in microchannel heat sinks [21-23]. Rahimi-Gorji et al. [24]

evaluated heat transfer attributes for a microchannel heat sink operated with various nanofluids

including Cu, Al2O3, Ag and TiO2 nanoparticles in water and ethylene glycol. In order to reach

T
an optimum outline for the heat sink geometry, the Response Surface Methodology (RSM) was

IP
adopted. It was shown that the Brownian motion of nanoparticles intensifies with the

CR
concentration increment and thus, the temperature difference between the coolant and wall

reduces. Bahiraei and Heshmatian [25] assessed the irreversibility and thermohydraulic features

US
of a hybrid nanofluid within two microchannel heat sinks. Increase of the concentration or
AN
velocity intensified the pumping power in both heat sinks. In addition, the heat transfer had a

greater contribution in irreversibility compared with the friction. They also reported that utilizing
M

nanofluids is a desirable method for applications of electronics cooling.

ED

Up to now, different nanofluids have been introduced with various kinds of nanoparticles and

base liquids. Nanoparticles having great thermal conductivity are the appropriate candidates for
PT

utilization in preparing nanofluids, because even at the small concentrations, result in high heat
CE

transfer amounts. Graphene has recently attracted the attention of many scholars because of its

excellent thermal characteristics, and can be a suitable choice to reach high heat transfer rates.
AC

During recent years, some papers have been published with considering graphene sheets in

nanofluids [26]. Zhou et al. [27] evaluated the heat transfer features of oscillating heat pipes by

use of a nanofluid containing graphene nanoplatelets. They reported that the thermal resistance

reduces 83.6% in the case of using 2 vol.% nanofluid in comparison with the pure water.

Ranjbarzadeh et al. [28] experimentally studied heat transfer and flow attributes of a water–

5
 ACCEPTED MANUSCRIPT

graphene oxide nanofluid in isothermal heat pipe. It was found that the thermal conductivity of

nanofluid at concentration of 0.1% was 28% higher than that of the base liquid, which led to a

40% enhancement in heat transfer rate.

According to the studies carried out in this area, it is evident that the methods of disrupting flows

such as employing ribs can augment the heat transfer of microchannel heat sinks with penalty of

T
rather great pressure loss. On the other hand, the secondary flow approaches are able to improve

IP
thermal attributes with a minor pressure loss. Moreover, utilizing nanofluids can enhance heat

CR
exchange process in microchannel heat sinks. To the best knowledge of the authors, no study has

been carried out with combining the three methods of utilizing ribs, secondary channels and

US
nanofluids in microchannel heat sinks. Accordingly, the purpose of the current research is to
AN
evaluate hydrothermal characteristics of a hybrid nanofluid containing graphene–silver

nanoparticles inside a microchannel heat sink proposed by Ghani et al. [2]. It is noteworthy that
M

this heat sink benefits from the advantages of secondary channels which cause greater flow area
ED

to decrease the pressure loss due to the ribs. Additionally, this combination intensifies the flow

mixing noticeably owing to capability of the ribs in directing higher flow towards secondary
PT

channels, which can enhance the flow passage between neighbor channels.
CE

2. Definition of the microchannel heat sink and nanofluid

AC

Fig. 1 illustrates the microchannel heat sink under study in which the ribs and secondary

channels have been applied to enhance cooling performance (also see Fig. 2). The heat flux is

introduced from its bottom surface, and the material of the heat sink is cooper. The ribs disturb

thermal boundary layer and can enhance heat transfer rate. The size and pitch of the ribs is set

such that they result in a rational pressure drop along with significant heat transfer amount.

6
 ACCEPTED MANUSCRIPT

According to Figs. 1 and 2, the secondary channels are the oblique ones located in the middle

walls. Indeed, the flow is deviated towards these channels after impingement to the ribs and

hence, a proper flow mixing with the neighbor part occurs. It can be mentioned that in addition

to the role of directing the flow towards the secondary channels, the ribs also increase the heat

transfer surface. Moreover, the secondary channels increase the flow area and therefore, decrease

T
the pressure drop due to the ribs. The flow enters the heat sink from the inlet, shown in Fig. 1,

IP
and after passing the relevant routes and gaining the introduced heat, exits from the outlet. The

CR
length, width and height of the heat sink are 10 mm, 0.6 mm and 0.4 mm, respectively. Also, the

area of the bottom surface is 6 mm2. All the dimensions of the heat sink are listed in Table 1

based on the symbols shown in Fig. 1.

US
AN
M
ED
PT
CE
AC

Fig. 1. Schematic diagram and geometrical parameters of microchannel.

7
 ACCEPTED MANUSCRIPT

T
IP
CR
Fig. 2. Concept of ribs and secondary channels in the microchannel heat sink.

US
Table 1. Dimensions of the microchannel heat sink.
AN
Geometrical parameters X1 X2 X3 X4 X5 X6 X7 X8 L θ
Quantity (µm) 500 75 37.5 250 100 300 100 600 10000 45°
M

The nanofluid used in this research is a hybrid nanofluid containing a nanopowder synthetized
ED

with decorating silver nanoparticles on graphene nanoplatelets. A complete description related to

the method of preparing this new nanofluid has been stated in Ref. [29]. In that work,
PT

functionalization of the Graphene Nanoplatelets (GNPs) was executed through acid treatment in
CE

the first step and then, the functionalized GNPs were decorated by the silver nanoparticles

through Ag (NH3)2OH solution.

AC

3. Governing equations

In order to examine the hydrothermal characteristics of the hybrid nanofluid in the microchannel

heat sink, the continuity, momentum and energy equations are solved by applying variable

thermophysical properties. The flow regime is laminar, and the nanofluid is considered

incompressible and Newtonian. In fact, based on the relevant literature, nanofluids in such low
8
 ACCEPTED MANUSCRIPT

concentrations behave similar to Newtonian fluids. In addition, it is noteworthy that, as per Table

2, the thermophysical properties of the nanofluid are temperature dependent and hence space

dependent, and as a result, they cannot be moved outside the derivative operator. Moreover, the

steady state assumption is considered for this problem.

Conservation of mass:

T
.(ρv)  0 (1)

IP
Conservation of momentum:

CR
.(  vv )  P  .(v) (2)

US
Conservation of energy:

.(  vc pT)  .(k T ) (3)

AN
where k is the thermal conductivity, ρ indicates the density, μ is the viscosity, and cp represents
M

the specific heat. Besides, v , T and P indicate velocity, temperature and pressure, respectively.

The governing equation for conduction heat transfer in the solid walls is also written as Eq. (4).
ED

It should be noted that there is no heat source in the solid parts and also, the thermophysical
PT

properties of the solid walls are considered as invariant.

2T  0 (4)
CE
AC

3.1. Nanofluid properties

A uniform nanoparticle concentration is considered for the nanofluids under study because based

on the published experimental results, the nanofluids prepared by Yarmand et al. [29] were stable

and no sedimentation of particles was seen for up to 60 days.

Eq. (5) is used for evaluating the specific heat of the nanofluid.

c p ,nf  (1   )c p , f   c p , p (5)
9
 ACCEPTED MANUSCRIPT

where subscripts f, p and nf refer to the base fluid, particles and nanofluid, respectively. Also, φ

indicates the volume concentration.

With the aid of the experimental data presented by Yarmand et al. [29], the viscosity, thermal

conductivity and density of the hybrid nanofluid are modeled that are temperature-dependent.

Table 2 summarizes the developed models for concentrations of 0, 0.02%, 0.06% and 0.1%. Note

T
IP
that the temperature is in Celsius in these models. Moreover, it is noteworthy that Yarmand et al.

[29] carried out the experiments for the temperatures between 20°C and 40°C.

CR
US
Table 2. Models of thermophysical properties developed from experimental data.

φ=0 φ=0.02%
  0.001 (0.0003631T  0.04121T  1.777)
2
  0.001(0.0002476T 2  0.03364T  1.711)
AN
k  0.002 T  0.5488 k  1.78106 (T ) 2.869  0.5898
  0.2968T  1004   0.304T  1004.3
M

φ=0.06% φ=0.1%
  0.001(5.396T 0.4884
 0.0484)   0.001 (19.19T 1.13  0.6251)
ED

k  1.855105 (T )3  0.001365(T )2  0.03402(T )  0.3567 k  2.928106 T 2.815  0.6752

  0.3T  1004.8   0.302T  1005.2
PT
CE

3.2. Boundary conditions

Uniform velocity and temperature are considered at the inlet of the heat sink. Zero relative
AC

pressure is used at the outlet of the heat sink. Besides, the no slip condition is employed on the

walls, and a constant heat flux is assumed on the bottom surface.

It should be noted that the continuum assumption is valid only when mean free path of fluid

molecules is very smaller than flow length scale. Otherwise, non-continuum effects such as slip

condition become important. Knudsen number presents ratio of mean free path of fluid

10
 ACCEPTED MANUSCRIPT

molecules to channel length scale. For Knudsen number values of smaller than 0.001, the

continuum assumption is established, and no slip condition can be applied. In the current study,

the fluid is water with mean free path of about 0.3 nanometers, while the lowest flow dimension

here is 37.5 micrometers. As a result, Knudsen number is much less than 0.001 and therefore, the

no slip condition can be employed.

T
IP
4. Data processing

CR
The average convective heat transfer coefficient is evaluated by the following equation [2]:

qAs

US
h (6)
Acon. TW  Tm  AN
where q represents the wall heat flux, As indicates the heated area, Acon. is the convection heat

transfer area, Tw denotes the average temperature of microchannel wall, and Tm is average fluid
M

temperature.
ED

To assess the performance of microchannel heat sinks, the uniformity of temperature distribution

on the heated surface is an essential parameter. In reality, the difference between the highest and
PT

lowest temperatures on the heating surface determines the cooling quality. The ratio of this
CE

temperature difference to the heat flux applied is utilized here as a norm for uniformity of the

cooling [30]:
AC

Tmax  Tmin
 (7)
q

where Tmax and Tmin denote highest temperature and lowest temperature of heating surface,

respectively.

It should be noted that a lower θ indicates a more uniform temperature distribution.

11
 ACCEPTED MANUSCRIPT

For investigating the overall performance of heat sinks, pumping power required for fluid motion

should also be evaluated. Pumping power ( W ) can be obtained from Eq. (8).

W  V P (8)

where ΔP is the pressure drop, while V denotes the volumetric flow rate.

T
5. Numerical method and validation

IP
To perform the numerical simulation, the finite volume method is utilized. The second order

CR
upwind method is applied for solving the mass, momentum and energy equations, whereas the

US
approach employed for pressure-velocity coupling is the SIMPLE method. Convergence

criterion to stop the numerical solution is considered 10-6 for all variables.
AN
With investigating different grids to ensure the mesh independency, the best grid is selected by
M

evaluating two parameters of the pressure drop and mean surface temperature. As shown in

Table 3, there is no significant change in pressure drop and surface temperature when the number
ED

of elements becomes greater than 3919494. Therefore, this grid is chosen for the further
PT

simulations.

In order to validate the numerical simulation, the results of this contribution are compared with
CE

the results reported by Ghani et al. [2]. For this evaluation, the pressure drop of a rectangular
AC

microchannel heat sink, and the mean surface temperature of a microchannel heat sink having

the secondary oblique channels and rectangular ribs are taken into account. The results of

pressure drop and mean surface temperature at different Reynolds numbers are presented in

Table 4 and Fig. 3, respectively. It is noticed that the results are consistent and therefore, the

numerical method is valid.

12
 ACCEPTED MANUSCRIPT

Table 3. Results of grid independency.

No. Elements Twall (K) ΔP (Pa)
1 1015588 310.95 65484
2 2004324 310.66 70497
3 3039566 310.46 72167
4 3594126 310.41 72772
5 3919494 310.29 73966
6 4598589 310.29 73285

Table 4. Results of pressure drop obtained from current study in comparison with those reported by Ghani et al. [2].

T
ΔP (Pa)

IP
Re Present study Ghani et al. [2] Error (%)
100 4223.1 4107.8 2.8
200 8590.0 8110.0 5.9

CR
300 13098.7 12746.2 2.8
400 17747.7 17699.3 0.3
500 22535.5 22863.5 1.4

US
AN
M
ED
PT
CE
AC

Fig. 3. Results of mean surface temperature obtained from current study in comparison with the work conducted by

Ghani et al. [2].

13
 ACCEPTED MANUSCRIPT

6. Results and discussion

The hydrothermal characteristics of the hybrid nanofluid including surface temperature,

convective heat transfer coefficient, temperature uniformity, pressure drop and pumping power

are examined in the microchannel heat sink having the ribs and secondary channels. The

simulations are performed at Reynolds numbers of 100 to 500, and nanoparticle concentrations

T
of 0 to 0.1%. The fluid motion in the microchannel is steady, and its direction is from the left to

IP
the right. Fig. 4 shows the flow pathlines at Re=500 and φ=0.1% in which the color of the

CR
pathlines is based on the velocity magnitude.

US
AN
M
ED
PT
CE

Fig. 4. Pathlines for nanofluid flow at Re=500 and φ=0.1%.

AC

Fig. 5 illustrates the velocity vectors in one part of the microchannel heat sink at Re=500 and

φ=0.1%. This figure clearly shows dividing and mixing the flow among the channels as well as

the role of the ribs in directing the flow towards the secondary channels. As per the figure, the

velocity magnitude is smaller when the fluid passes the secondary channels. Intensification of the

14
 ACCEPTED MANUSCRIPT

flow mixing due to the presence of the ribs and secondary flows disturbs the thermal boundary

layers and therefore, can augment the heat transfer rate.

T
IP
CR
US
AN
M

Fig. 5. Vectors of velocity for Re=500 and φ=0.1%.

ED

Fig. 6 displays the velocity contours at the central, longitudinal and vertical cross sections
PT

through the center of the heat sink for Re=500 and φ=0.1%. Obviously, the high velocities are

observed upstream of the ribs and also in the narrower conduits. In contrast, the low velocities
CE

are seen downstream the ribs and also in the secondary channels. The longitudinal and vertical
AC

cross sections evidently show the small dead zones behind the ribs. Moreover, the minor dead

zones formed in the secondary channels are clear in the central cross section.

15
 ACCEPTED MANUSCRIPT

T
IP
CR
US
AN
Fig. 6. Contours of velocity for different cross sections at Re=500 and φ=0.1%.
M

According to Table 2, the thermal conductivity, viscosity and density of the hybrid nanofluid are
ED

temperature dependent. Fig. 7 presents the contours of these properties as well as the temperature

contour along the microchannel at the cross section through the heat sink middle for Re=500 and
PT

φ=0.1%. It is clear that because the nanofluid flows from the left to the right, the temperature
CE

increases towards the right side. As a result, the thermal conductivity enhances while the

viscosity and density reduce from the left to the right. The temperature dependency of the
AC

thermal conductivity is a positive factor since it improves the uniformity of the temperature

distribution. In fact, in the regions with higher temperatures, the thermal conductivity is also

greater which results in an enhanced heat transfer there. In addition, the viscosity has an inverse

relationship to the temperature and thus, the viscosity decreases in the hotter areas, which

decreases the hydrodynamic resistance against the fluid motion.

16
 ACCEPTED MANUSCRIPT

T
IP
CR
US
AN
M
ED

Fig. 7. Contours of temperature, thermal conductivity, viscosity and density at Re=500 and φ=0.1%.
PT

Fig. 8 illustrates the temperature distribution in both fluid part and solid part as well as the
CE

streamlines related to the areas that possess the velocities lower than 1 m/s at different Reynolds

numbers and concentrations. As can be observed, the nanofluid demonstrates a superior cooling
AC

compared with the base fluid, which is due to its greater thermal conductivity. Moreover, with

increasing the Reynolds number, the temperature reduces in both fluid and solid parts. As

mentioned before, the flow velocity is rather low in the secondary channels and behind the ribs

and hence, the fluid temperature intensifies in these regions. Another point that can be seen in the

figure is that the dead zones are smaller for the nanofluid in comparison with the base fluid. It

17
 ACCEPTED MANUSCRIPT

should be noted that, as mentioned, the streamlines in this figure have been illustrated only for

low velocity regions and as can be noticed, the temperatures are higher in these areas because the

residence time of flow particles is greater.

T
IP
CR
US
AN
M
ED
PT

Fig. 8. Dead zones and temperature contours at central cross section for different Reynolds numbers and
CE

concentrations.

Figs. 9 and 10 show the temperature contours respectively for the vertical and longitudinal cross
AC

sections at different Reynolds numbers and concentrations. It is noticed in both figures that with

increasing either concentration or Reynolds number, the temperature decreases, the temperature

uniformity improves, and the regions with highest temperature become smaller. Meanwhile,

because of the dead zone formation in the corners of the microchannel (also see Fig. 6) and low

velocities in those regions, the temperature is greater there.

18
 ACCEPTED MANUSCRIPT

T
IP
CR
US
AN
M

Fig. 9. Temperature contours at vertical cross section for (a) Re=300 and φ=0, (b) Re=500 and φ=0, (c) Re=300 and
φ=0.1% and (d) Re=500 and φ=0.1%.
ED
PT
CE
AC

Fig. 10. Temperature contours at longitudinal cross section for pure water and nanofluid at different Reynolds
numbers and concentrations.

19
 ACCEPTED MANUSCRIPT

Fig. 11 displays the temperature contours in both fluid and solid parts as well as the streamlines

at three different cross sections along the microchannel for Re=500 and φ=0.1%. It should be

noted that background of these contours shows the temperature distribution whereas the

streamlines have been drawn on them. Moreover, the color of the streamlines is based on the

velocity magnitude and therefore, the amounts of both temperature and velocity are shown on the

T
color bar. It is noteworthy that these cross sections are related to the last half of the heat sink. It

IP
is perceived that the temperatures of the fluid and solid walls increase along the heat sink. The

CR
first cross section is related to behind one of the ribs. In the middle part of this cross section, due

to the flow rotation and formation of vortices, the fluid experiences the higher temperatures.

US
Moreover, the greater velocities are clearly observed at the two border conduits of this cross
AN
section, which is owing to their smaller flow area. The second cross section is drawn for

upstream one rib and the entrance of secondary channels. As can be noticed, in the regions that
M

the fluid enters the secondary channels, the vortices are formed, and the temperature is greater
ED

there. The third cross section is related to the region having the narrow channels, and because the

velocity is great in the narrow channels, the residence time of the fluid is small and hence, the
PT

fluid temperature in these border channels is lower than that in the main channel.
CE
AC

20
 ACCEPTED MANUSCRIPT

T
IP
CR
US
AN
Fig. 11. Temperature contours as well as streamlines at different vertical cross sections for Re=500 and φ=0.1%.
M

The average convective heat transfer coefficient of the microchannel heat sink is depicted in Fig.
ED

12 in terms of Reynolds number at different concentrations. Evidently, the average convective

PT

heat transfer coefficient enhances with the concentration increment. This is because the

graphene–silver nanoparticles have an excellent thermal conductivity, which improves the heat
CE

transport in the fluid. Thereby, with increase of the concentration from 0 to 0.1% at Re=100, a
AC

17% enhancement happens in the convective heat transfer coefficient. In addition, increment of

Reynolds number increases the convective heat transfer coefficient because the boundary layers

become thinner at greater Reynolds numbers.

21
 ACCEPTED MANUSCRIPT

T
IP
CR
US
Fig. 12. Convective heat transfer coefficient in terms of Reynolds number at different concentrations.
AN
Because the bottom surface of the heat sink locates on electronic processors, the amount of

temperature on this surface is of substantial importance as well. As a result, the average

M

temperature on the bottom surface is presented in Table 5 for different Reynolds numbers and
ED

concentrations. According to this table, the bottom surface temperature decreases with increment

of the concentration such that a 3.42 K reduction occurs with increasing the concentration from 0
PT

to 0.1% at Re=100. It is interesting to note that the temperature reduction with the concentration
CE

increment is more significant at lower Reynolds numbers. This is due to the fact that at higher

Reynolds numbers, the inertia term is stronger and consequently, adding the nanoparticles will
AC

have smaller effects on heat transfer process. On the contrary, at lower Reynolds numbers, the

diffusion term is more dominant and thus, the effect of concentration increase is more profound.

Furthermore, the temperature of the bottom surface reduces significantly by increasing the

Reynolds number such that it decreases 13.88 K at φ=0.1% with increase of Reynolds number

from 100 to 500.

22
 ACCEPTED MANUSCRIPT

Table 5. Average temperature of bottom surface at different Reynolds numbers and concentrations.

Mean surface temperature (K)

Re
φ=0 φ=0.02% φ=0.06% φ=0.1%
100 324.11 322.84 321.88 320.69
200 314.00 313.44 312.88 312.15
300 310.59 310.20 309.78 309.21
400 308.85 308.55 308.21 307.73

T
500 307.79 307.54 307.24 306.81

IP
In addition to the average surface temperature, the maximum temperature of the bottom surface

CR
is also very important in the heat sinks. Indeed, a low maximum surface temperature shows a low

possibility of hot spot formation on the surface of electronic chips. It can be mentioned that hot

US
spots decrease the performance and life of electronic components, and have destructive effects.
AN
Fig. 13 illustrates the maximum temperature of the bottom surface in terms of the Reynolds

number at different concentrations. As can be perceived, the maximum temperature decreases by

M

increasing either Reynolds number or concentration. Moreover, the maximum temperature

ED

experiences a more intense reduction with the concentration increment at lower Reynolds

numbers. Thereby, with increase of the concentration from 0 to 0.1%, the maximum temperature
PT

reduces 5.34 K at Re=100, and 1.34 K at Re=500.

CE
AC

23
 ACCEPTED MANUSCRIPT

T
IP
CR
US
Fig. 13. Maximum temperature of bottom surface in terms of Reynolds number at different concentrations.
AN
Along with average and maximum temperatures of bottom surface, uniformity of temperature

distribution is also crucial in examining thermal performance of heat sinks. This factor is
M

investigated here with the aid of parameter θ (Eq. (7)), such that a smaller θ means more even
ED

temperature distribution. Fig. 14 depicts the parameter θ at different Reynolds numbers and

concentrations. It is seen that the temperature uniformity improves with increment of the
PT

concentration and Reynolds number such that the best temperature uniformity occurs at Re=500
CE

and φ=0.1%. This finding is also certified by Fig. 15 in which the temperature contours of

bottom surface are displayed at two Reynolds numbers and two concentrations. The
AC

improvement of temperature uniformity at the higher concentration and Reynolds number is

clearly observed in this figure.

24
 ACCEPTED MANUSCRIPT

T
IP
CR
Fig. 14. Parameter θ in terms of Reynolds number at different concentrations.

US
AN
M
ED
PT
CE
AC

Fig. 15. Contours of bottom surface temperature at different Reynolds numbers and concentrations.

In order to better clarify the effect of employing the nanofluid on the cooling process in the heat

sink, Fig. 16 shows the difference between the bottom surface temperature and average

temperature of the fluid at different concentrations and Reynolds numbers. Clearly, at all

25
 ACCEPTED MANUSCRIPT

Reynolds numbers, the temperature difference decreases with the concentration increment, which

proves the promising efficacy of the nanofluid compared with the base fluid.

T
IP
CR
US
AN
M

Fig. 16. Temperature difference between channel bottom and fluid at different Reynolds numbers and
ED

concentrations.
PT

In heat sinks, in addition to evaluating thermal characteristics, investigation of hydrodynamic

features such as pressure drop and pumping power should also be carried out for a thorough
CE

assessment of them. The pumping power is a crucial parameter because demonstrates rate of
AC

energy consumption by the system. This subject becomes even more important in the case of

using nanofluids because addition of the nanoparticles to a base fluid may intensify pressure drop

in addition to heat transfer enhancement. It should be noted that the viscosity increases with

adding the nanoparticles. Meanwhile, the characteristic lengths in microchannels are small,

which can lead to significant pressure drops. Figs 17 and 18 display respectively the pressure

contours and pumping power values at different Reynolds numbers and concentrations. It can be
26
 ACCEPTED MANUSCRIPT

observed that both pressure drop and pumping power intensify with increasing either Reynolds

number or concentration. At higher Reynolds numbers, the velocity gradients on the walls are

more intense due to the greater velocities, which augment the wall shear stress. Consequently,

the pressure drop and pumping power are higher at greater Reynolds numbers. In addition, there

are two main reasons for increase of the pumping power and pressure drop with the

T
concentration increment. The first reason is that the viscosity increases by rising the

IP
concentration, and the second one is that the flow velocity must be increased with the viscosity

CR
increment at a constant Reynolds number (note that the viscosity is in the denominator of

Reynolds number). As a result, the pumping power and pressure drop will have the greater

values at higher concentrations.

US
AN
M
ED
PT
CE
AC

27
 ACCEPTED MANUSCRIPT

T
IP
CR
US
AN
M

Fig. 17. Pressure contours for (a) Re=300 and φ=0, (b) Re=300 and φ=0.1%, (c) Re=500 and φ=0 and (d) Re=500
and φ=0.1%.
ED
PT
CE
AC

28
 ACCEPTED MANUSCRIPT

T
IP
CR
US
Fig. 18. Pumping power in terms of Reynolds number at different concentrations.
AN
According the fact that adding the nanoparticles increases the pressure drop in addition to the
M

heat transfer enhancement, the Performance Evaluation Criterion (PEC) is employed here to
ED

examine the amount of heat transfer enhancement in comparison with pumping power increment

(Eq. (9)). Fig. 19 shows this parameter in terms of the Reynolds number at different nanoparticle
PT

concentrations. It is found that the PEC value enhances with increasing the concentration, while
CE

it reduces by the Reynolds number increment. This reveals that adding the nanoparticles

improves the heat transfer more than the pumping power increment. It is known that due to small
AC

length scales in microchannels, pressure drop intensifies significantly with increasing Reynolds

number. This fact has caused that the microchannel under study demonstrates greater PEC values

at lower Reynolds numbers.

Nu Nu0
PEC 
( f f 0 )1 3 (9)

29
 ACCEPTED MANUSCRIPT

where Nu and f represent the Nusselt number and friction factor, respectively, while subscript 0

refers to the base fluid.

T
IP
CR
US
AN
M

Fig. 19. PEC values in terms of Reynolds number at different concentrations.

ED

7. Conclusion
In this study, the hydrothermal features of a hybrid nanofluid containing graphene–silver
PT

nanoparticles in a microchannel heat sink equipped with the ribs and secondary channels are
CE

examined. Along with increase of the heat transfer surface, the ribs have the role of directing the

flow towards the secondary channels. Meanwhile, the secondary channels increase the flow area,
AC

which reduces the pressure drop due to the presence of the ribs. The results show that employing

the nanofluid in the microchannel heat sink enhanced with the ribs and secondary channels

improves the heat sink performance significantly. The main conclusions of this work are as

follows:

30
 ACCEPTED MANUSCRIPT

- With increasing either concentration or Reynolds number, the temperature decreases, the

temperature uniformity enhances, and the regions with highest temperatures become smaller.

- The average convective heat transfer coefficient enhances with increasing the concentration and

Reynolds number such that with increase of the concentration from 0 to 0.1% at Re=100, a 17%

enhancement happens in the convective heat transfer coefficient.

T
- The bottom surface temperature decreases with increment of the concentration such that a 3.42

IP
K reduction occurs with increasing the concentration from 0 to 0.1% at Re=100.

CR
- The temperature of the bottom surface reduces significantly by increasing the Reynolds number

such that it decreases 13.88 K at φ=0.1% with increase of Reynolds number from 100 to 500.

US
- The dead zones are smaller for the nanofluid in comparison with the base fluid.
AN
- Because of the dead zone formation in the corners of the microchannel and low velocities in

those regions, the temperature is greater there.

M

- The temperature reduction with the concentration increment is more significant at lower
ED

Reynolds numbers.

- The flow experiences a greater pumping power at higher Reynolds numbers and concentrations.
PT
CE
AC

31
 ACCEPTED MANUSCRIPT

References

[1] S. Abubakar, C.S. Nor Azwadi, A. Ahmad, The use of Fe 3O4-H2O4 nanofluid for heat transfer enhancement in

rectangular microchannel heat sink, J. Adv. Res. Mater. Sci. 23 (2016) 15–24.

[2] I.A. Ghani, N.A.C. Sidik, R. Mamat, G. Najafi, T.L. Ken, Y. Asako, W.M.A.A. Japar, Heat transfer

enhancement in microchannel heat sink using hybrid technique of ribs and secondary channels, International Journal

of Heat and Mass Transfer 114 (2017) 640–655.

T
[3] D.B. Tuckerman, R. Pease, High-performance heat sinking for VLSI, IEEE Electron Dev. Lett. 2 (1981) 126–

IP
129.

CR
[4] P.-S. Lee, S.V. Garimella, D. Liu, Investigation of heat transfer in rectangular microchannels, Int. J. Heat Mass

Transf. 48 (2005) 1688–1704.

US
[5] X.F. Peng, G.P. Peterson, Convective heat transfer and flow friction for water flow in microchannel structures,

Int. J. Heat Mass Transf. 39 (1996) 2599–2608.

AN
[6] K.C. Toh, X.Y. Chen, J.C. Chai, Numerical computation of fluid flow and heat transfer in microchannels, Int. J.

Heat Mass Transf. 45 (2002) 5133–5141.

M

[7] P. Gunnasegaran, H. Mohammed, N. Shuaib, R. Saidur, The effect of geometrical parameters on heat transfer
ED

characteristics of microchannels heat sink with different shapes, Int. Commun. Heat Mass Transf. 37 (2010) 1078–

1086.
PT

[8] A. Alfaryjat, H. Mohammed, N.M. Adam, M. Ariffin, M. Najafabadi, Influence of geometrical parameters of

hexagonal, circular, and rhombus microchannel heat sinks on the thermohydraulic characteristics, Int. Commun.
CE

Heat Mass Transf. 52 (2014) 121–131.

[9] Y. Li, G. Xia, D. Ma, Y. Jia, J. Wang, Characteristics of laminar flow and heat transfer in microchannel heat sink
AC

with triangular cavities and rectangular ribs, Int. J. Heat Mass Transf. 98 (2016) 17–28.

[10] G. Xia, Y. Zhai, Z. Cui, Numerical investigation of thermal enhancement in a micro heat sink with fan-

shaped reentrant cavities and internal ribs, Appl. Therm. Eng. 58 (2013) 52–60.

[11] M. Bahiraei, K. Gharagozloo, M. Alighardashi, N. Mazaheri, CFD simulation of irreversibilities for laminar

flow of a power-law nanofluid within a minichannel with chaotic perturbations: An innovative energy-efficient

approach, Energy Conversion and Management, 144 (2017) 374–387.

32
 ACCEPTED MANUSCRIPT

[12] S.M. Hosseini, M.R. Safaei, M. Goodarzi, A.A.A.A. Alrashed, T.K. Nguyen, New temperature, interfacial shell

dependent dimensionless model for thermal conductivity of nanofluids, International Journal of Heat and Mass

Transfer, 114 (2017) 207–210.

[13] A. Karimipour, S.A. Bagherzadeh, M. Goodarzi, A.A. Alnaqi, M. Bahiraei, M.R. Safaei, M. Safdari Shadloo,

Synthesized CuFe2O4/SiO2 nanocomposites added to water/EG: Evaluation of the thermophysical properties beside

sensitivity analysis & EANN, International Journal of Heat and Mass Transfer, 127 (2018) 1169–1179.

T
[14] M. Sheikholeslami, H.R. Kataria, A.S. Mittal, Effect of thermal diffusion and heat-generation on

IP
MHD nanofluid flow past an oscillating vertical plate through porous medium, Journal of Molecular Liquids, 257

CR
(2018) 12–25.

[15] J.A. Esfahani, M.R. Safaei, M. Goharimanesh, L.R. Oliveira, M. Goodarzi, S. Shamshirband, E.P.B. Filho,

US
Comparison of experimental data, modelling and non-linear regression on transport properties of mineral oil based
AN
nanofluids, Powder Technol. 317 (2017) 458–470.

[16] M. Bahiraei, A. Rezaei Gorjaei, A. Shahidian, Investigating heat transfer and entropy generation for mixed
M

convection of CuO–water nanofluid in an inclined annulus, Journal of Molecular Liquids 248 (2017) 36–47.
ED

[17] Anu K., Hemalatha J., Ultrasonic and magnetic investigations of the molecular interactions in zinc doped

magnetite Nanofluids, Journal of Molecular Liquids, 256 (2018) 213–223.

PT

[18] M. Bahiraei, M. Alighardashi, Investigating non-Newtonian nanofluid flow in a narrow annulus based on

second law of thermodynamics, Journal of Molecular Liquids 219 (2016) 117–127.

CE

[19] Goshayeshi, H.R., Goodarzi, M., Safaei, M.R., Dahari, M. Experimental study on the effect of inclination angle

on heat transfer enhancement of a ferrofluid in a closed loop oscillating heat pipe under magnetic field,
AC

Experimental Thermal and Fluid Science, 74 (2016) 265–270.

[20] M. Bahiraei, R. Khosravi, S. Heshmatian, Assessment and optimization of hydrothermal characteristics

for a non-Newtonian nanofluid flow within miniaturized concentric-tube heat exchanger considering designer’s

viewpoint Applied Thermal Engineering 123 (2017) 266–276.

33
 ACCEPTED MANUSCRIPT

[21] M. Bahiraei, S. Heshmatian, M. Keshavarzi, Multi-attribute optimization of a novel micro liquid block working

with green graphene nanofluid regarding preferences of decision maker, Applied Thermal Engineering 143 (2018)

11–21.

[22] M.R. Sohel, R. Saidur, S.S. Khaleduzzaman, T.A. Ibrahim, Cooling performance investigation of electronics

cooling system using Al2O3–H2O nanofluid, International Communications in Heat and Mass Transfer 65 (2015)

89–93.

T
[23] M. Bahiraei, S. Heshmatian, Application of a novel biological nanofluid in a liquid block heat sink for cooling

IP
of an electronic processor: Thermal performance and irreversibility considerations, Energy Conversion and

CR
Management 149 (2017) 155–167.

[24] M. Rahimi-Gorji, O. Pourmehran, M. Hatami, D.D. Ganji, Eur. Phys. J. Plus 130 (2015) 22.

US
[25] M. Bahiraei, S. Heshmatian, Thermal performance and second law characteristics of two new
AN
microchannel heat sinks operated with hybrid nanofluid containing graphene–silver nanoparticles, Energy

Conversion and Management 168 (2018) 357–370.

M

[26] M. Bahiraei, N. Mazaheri, Application of a novel hybrid nanofluid containing graphene–platinum nanoparticles
ED

in a chaotic twisted geometry for utilization in miniature devices: Thermal and energy efficiency considerations,

International Journal of Mechanical Sciences, 138–139 (2018) 337–349.

PT

[27] Y. Zhou, X. Cui, J. Weng, S. Shi, H. Han, C. Chen, Experimental investigation of the heat transfer performance

of an oscillating heat pipe with graphene nanofluids, Powder Technology, 332 (2018) 371–380.
CE

[28] R. Ranjbarzadeh, A. Karimipour, M. Afrand, A. Homayoon, M. Isfahani, A. Shirneshan, Empirical analysis of

heat transfer and friction factor of water/graphene oxide nanofluid flow in turbulent regime through an isothermal
AC

pipe, Applied Thermal Engineering, 126 (2017) 538–547.

[29] H. Yarmand, S. Gharehkhani, G. Ahmadi, S.F. Seyed Shirazi, S. Baradaran, E. Montazer, M.N.M. Zubir, M.S.
Alehashem, S.N. Kazi, M. Dahari, Graphene nanoplatelets–silver hybrid nanofluids for enhanced heat transfer,
Energy Conversion and Management 100 (2015) 419–428.
[30] M. Bahiraei, S. Heshmatian, Efficacy of a novel liquid block working with a nanofluid containing graphene
nanoplatelets decorated with silver nanoparticles compared with conventional CPU coolers, Applied Thermal
Engineering 127 (2017) 1233–1245.

34
 ACCEPTED MANUSCRIPT

 Hydrothermal features of hybrid nanofluid in a microchannel heat sink are assessed

 The heat sink is equipped with the rectangular ribs and secondary channels

 Employing nanofluid, ribs and secondary channels enhances heat sink performance

 Cooling efficiency improves with rising either concentration or Reynolds number

 Flow experiences higher pumping power at higher Reynolds numbers and concentrations

T
IP
CR
US
AN
M
ED
PT
CE
AC

35