Accepted Manuscript

Numerical study of nanofluids thermal and hydraulic

characteristics considering Brownian motion effect in micro fin
heat sink

Wenwen Guo, Guoneng Li, Youqu Zheng, Cong Dong

PII: S0167-7322(18)31071-7
DOI: doi:10.1016/j.molliq.2018.04.152
Reference: MOLLIQ 9055
To appear in: Journal of Molecular Liquids
Received date: 2 March 2018
Revised date: 27 April 2018
Accepted date: 30 April 2018

Please cite this article as: Wenwen Guo, Guoneng Li, Youqu Zheng, Cong Dong ,
Numerical study of nanofluids thermal and hydraulic characteristics considering Brownian
motion effect in micro fin heat sink. The address for the corresponding author was
captured as affiliation for all authors. Please check if appropriate. Molliq(2017),
doi:10.1016/j.molliq.2018.04.152

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

Numerical study of nanofluids thermal and hydraulic characteristics

considering Brownian motion effect in micro fin heat sink

Wenwen Guo,a Guoneng Li,a Youqu Zheng,*,a Cong Donga

a
Department of Energy and Environment System Engineering, Zhejiang University of
Science and Technology, Hangzhou 310023, China

PT
* Corresponding author: zyq888@zust.edu.cn

RI
Abstract
The thermal and hydraulic performance in a micro fin heat sink (MFHS) with

SC
ZnO-water nanofluids is investigated numerically using the static and dynamic single
phase model (i.e. SSPM and DSPM). For DSPM, both the Brownian motion effect on
NU
thermal conductivity and viscosity was taken into consideration. The model is
established and validated by comparing with literatures. It is found that the DSPM
MA

results are higher than SSPM in heat transfer performance and pressure drop by taking
the Brownian motion effect into account. With Re ranging from 85 to 595 the heat
D

transfer coefficient of DPSM and SSPM at ZnO volume concentration () of 3.0% are
E

enhanced by 25.6-38.3% and 16.6-23.8% respectively. Particle size and volume

PT

concentration are both critical to the nanofluids heat transfer performance of MFHS.
DSPM considering Brownian motion by utilize proper thermal conductivity and
CE

viscosity model, which is probably more appropriate for investigating the influence of
nanofluids particle size. Nanofluids of higher volume concentration and lower particle
AC

size are in favor of higher efficiency of heat sink in certain condition. The overall heat
transfer efficiency () was also evaluated and the maximum of 34.9 was obtained at
=3.0% and dP=30nm using DSPM (water=29.97).

Keywords: micro fin heat sink; nanofluids; Brownian motion; numerical study;
particle size.

1
 ACCEPTED MANUSCRIPT

Nomenclature

average temperature of bottom surface

MFHS micro fin heat sink Ts (K)
SSPM static single phase model Tf average fluid bulk temperature (K)
DSPM dynamic single phase model q heat flux (W/m2)
TPM Two phase model Amin minimum cross area (m2)
T temperature (K) C constant
width of micro fin heat sink

PT
W Eu Euler number
(mm)
length of micro fin heat sink
L Qf volume flow rate (m3/s)

RI
(mm)
longitudinal pitch of fin centers
SL Pf wetted perimeter (mm)

SC
(mm)
transverse pitch of fin centers
ST Af cross section area of fin (mm2)
(mm)
NU
Hf height of pin fin (mm) vmax maximum velocity (m/s)
Df diameter of pin fin (mm) k thermal conductivity (W/(mK))
NL longitudinal pin fin number
MA

Greek symbols
NT transverse pin fin number  viscosity (kg/(ms))
dc clearance of pin fin (mm)  overall heat transfer efficiency
D

Cp specific heat (J/(kgK))  density (kg/m3)

P 
E

pressure drop (Pa) distance between nanoparticles (m)


PT

dp average particle diameter (mm) volume concentration (%)

M molecular weight Subscripts
NA Avogadro constant bf base fluid
CE

vol.% volume percent nf nanofluid

VB Brownian velocity p particle
AC

Re Reynolds number 0 initial state

Nu Nusselt number s surface
dh f hydraulic diameter of fin (m)
heat transfer coefficient
h
(W/(m2K))

kB Boltzmann constant

2
 ACCEPTED MANUSCRIPT

1. Introduction
With the development of semiconductor and integrated chips, electronic devices in
modern times become faster as well as hotter. Better computing capability is
accompanied by much higher heat flux. Generated by electronic chips, the dissipated
heat flux can reach a high level [1]. For purpose of protecting electronic equipment
and promoting hardware reliability, much attention has increased in the enhancement

PT
of heat dissipation techniques of electronic chips. As one of the effective approaches,
heat sinks are capable of cooling computer chips and decreasing the surface

RI
temperature of electronic devices. As a result, the hydraulic and thermal properties of

SC
heat sinks with different structures were concerned and investigated by many
researchers [2, 3]. Wu et al. [4] discussed the impact of surface roughness as well as
NU
hydrophilcity on heat transfer performance of heat sinks, and observed that in large
Reynolds numbers the Nusselt number and friction loss have got an obvious increase.
MA

To further increase the thermal dissipation property, millimeter-level channels and fins
with different shapes, arrangement types and scales were adopted in heat sink surface.
Those are respectively called micro channel heat sink (MCHS) and micro fin heat
D

sink (MFHS). Khan et al. [5] and Moores et al. [6] studied the fin shape influence on
E

flow and heat transfer behavior of heat sink with micro fins and developed a series of
PT

correlations applying to several kinds of geometry. Tullius et al. [7] found that among
CE

the commonly used fin shapes, square and ellipse micro fins have better thermal
performance than diamond and triangle fins, at a price of higher pressure loss.
AC

Traditional heat transfer fluid of heat sinks is air, followed by water, ethylene
glycol and some other single phase heat transfer media. To satisfy the ever-increasing
needs in cooling performance, nanofluids are thought to be a promising heat transfer
media of heat sinks. First introduced by Choi in 1995 [8], the concept of nanofluids
has caused much attention. Nanofluids are nanoparticle suspensions preparing by
adding nanoparticles into the base fluid (two-step method) or directly chemical
synthesis in the base fluid (one-step method). Researches indicated that a slight
quantity of nanoparticles can increase the base fluid thermal conductivity dramatically

3
 ACCEPTED MANUSCRIPT

and improve the thermal performance [9, 10]. So far, the widely used nanofluids
include Al2O3 nanofluids [11, 12], CuO nanofluids [13, 14], Fe2O3 nanofluids [15, 16]
and ZnO nanofluids [17], etc.
Nanofluids have been employed to improve the cooling ability of heat sinks both in
experimental and numerical study. Duangthongsuk et al. [17] conducted the
experiments of ZnO and SiO2 nanofluids with different concentration in a micro

PT
circular pin fin heat sink. The measured data elucidated that nanofluids could enhance
heat transfer performance of fluids, along with an increase in pumping power. Zhou et

RI
al. [18] used self-prepared silver nanofluids and compared their thermal and hydraulic
performance with the base fluid in a novel drop-shaped micro fin heat sink. Prepared

SC
by one-step method, the silver nanoparticles were all in 3-7 nm without aggregation
and showed a great heat transfer improvement. Ali and Arshad [19] studied the
NU
influence of angle of square fins with 9.5 vol. % graphene nanoplatelets on heat
transfer performance. Their experimental results displays that the heat sink structure
MA

did matter to the nanofluids performance. As the angle of square fins ranged from 90
to 22.5, the heat transfer enhancement varied from 19.68 to 23.86%, indicating the
D

significance of heat sink structure.

E

Numerical studies about nanofluids thermal and hydraulic characteristics were also
PT

reported in literatures [20-28]. Generally speaking, there are two approaches in

dealing with nanofluids computational problems. One method is single phase model
CE

(SPM), in which the nanoparticles have no relative velocity and thermally equilibrate
with the base fluid. Another one is two phase model (TPM), in which the particles are
AC

regarded as an individual phase, and the interaction with the fluid phase was
calculated. Although in some cases the two phase model has higher precision, lots of
studies still adopted single phase model in the computational analysis of heat sinks
due to its much higher computational efficiency and better convergence property,
bringing much convenience to engineering application. And studies show that with
proper modeling of thermal conductivity and viscosity, single phase model could also
be suitably utilized in the nanofluids problem [23, 14]. Seyf and Feizbakhshi [25]
made a numerical study on MFHS heat transfer properties with single phase model,
4
 ACCEPTED MANUSCRIPT

and found that the Nusselt number could increase 10.35% for 4 vol.% CuO nanofluids.
And increasing volume concentration can lead to an enhancement of heat transfer rate
and pressure loss. Hasan et al. [26] presented the effects of fin shape on Al2O3-water
nanofluid heat transfer properties in a MFHS in laminar regime. They found that
compared with square and triangular fins, the MFHS with circular fins shows better
performance. Snoussi et al. [27] compared the simulation results of heat transfer and

PT
flow properties in a micro channel heat sink and presented the velocity and
temperature distribution. Lelea et al. [28] simulated the working process of a MCHS

RI
with 1.0 -5.0 vol.% Al2O3 nanofluids and concluded that the thermal performance of
nanofluids did improve based on Reynolds number. But they also observed that based

SC
on constant power or velocity, the enhancement effect was uncertain.
For single phase models, various models were established for the description of
NU
thermal conductivity and viscosity of nanofluids [29-41]. The most classical model
was founded by Maxwell [30], in which the thermal conductivity of the two-phase
MA

mixture was treated as a function of volume concentration of nanoparticles. Base on

that, renovated models considering the shape factors of nanoparticle were proposed,
D

such as Hamilton & Crosser [31] and Yu & Choi [32] model. It is worth to note that in
E

some circumstance considering the Brownian motion effect can improve the thermal
PT

conductivity accuracy. Xuan and Li [33] built a model considering the volume
concentration (), size and properties of nanoparticles. Koo et al. [34] and Vajjha et al.
CE

[35] modified Maxwell model and added the dynamic term considering the Brownian
motion of nanoparticles, which was reported to coincide with nanofluid heat transfer
AC

experiment in some cases and was applied in the study for nanofluid thermal
conductivity description [36, 37].
Another important physical property for nanofluids is viscosity, of which the most
classic models were proposed by Einstein [38] and Brinkman [39]. Einstein [38]
investigated the motion state of a single sphere particle in linearly viscous fluid and
calculated energy needed for the motion of the particle. The obtained correlation of
nanofluids viscosity to the volume concentration of nanoparticles was reported
applicable to low concentration cases. Brinkman [39] ameliorated the Einstein model
5
 ACCEPTED MANUSCRIPT

and enlarged the range of application to higher volume concentration of nanoparticles.

Considering the Brownian motion effect, Masoumi et al. [40] deduced a correlation of
nanofluids viscosity based on the relative velocity of nanoparticle to base fluid in a
volumetric element. The Masoumi model was used and thought to have satisfactory
precision in literatures [41] and was adopted in the study.
To our knowledge, there are few reports concerning nanofluid Brownian motion

PT
effect in numerical study of MFHS. In the present work, we use two models: the static
single phase model (SSPM) without considering the Brownian motion, and the

RI
dynamic single phase model (DSPM) in which the Brownian motion effect is taken
into account. The influences of Reynolds number, nanoparticle size and volume

SC
concentration on heat transfer and flow characteristics and the overall heat transfer
efficiency evaluation for the two models are investigated in detail for ZnO nanofluids
NU
in the laminar regime of a MFHS.
MA

2. Mathematical modelling and governing equations

As illustrated in Figure 1, a MFHS with circular fins is established and the fins with
D

1mm diameter and 0.5 mm height are stagger-arranged in the heat sink channel. The
E

bottom is heated with a constant heat flux while other outside surfaces are regarded as
PT

adiabatic. Flowing from the inlet of heat sink channel, the ZnO-water nanofluids can
CE

pass through the micro fins and cool the bottom surface. The dimensions of the
MFHS are described in detail in Table 1. And the physical property of base fluid and
AC

nanoparticles are listed in Table 2.

The applicability of single phase model for nanofluids was accepted in literatures
[42]. Therefore, the nanofluids are regarded as single phase flow and the N-S
equations for nanofluids in steady condition are as follows:

·( nf V )  0 (1)

 
·( nf V V )  P  nf 2 V (2)

 

· (  Cp )nf V T      knf T  (3)
 
6
 ACCEPTED MANUSCRIPT

Where  nf , Cpnf , knf and nf are density, specific heat capacity, thermal

conductivity and viscosity of nanofluids.  nf and C p are obtained by the equations:

nf   p  (1   ) bf (4)

( Cp)nf   ( Cp) p  (1   )( Cp)bf (5)

With respect to the thermal conductivity and viscosity of ZnO nanofluids, two

PT
models are introduced for comparative study: static single phase model (SSPM) and
dynamic single phase model (DSPM) with a Brownian motion term.

RI
In the static single phase model, nanofluids thermal conductivity was expressed by

SC
Maxwell model while nanofluids viscosity was from Corcione correlation [43]:

k p  2kbf  2(k p  kbf )

knf  kbf (6)
k p  2kbf  (k p  kbf )
NU
bf
nf  (7)
 d p 1.03 
MA

1  34.87( ) 
 dbf 
Where dp means the average diameter of nanoparticles and dbf was defined as
D

follows:
E

1
 6M 3
dbf  0.1  (8)
PT

 N 
 A bf 

Where M denotes the molecular weight and NA is the Avogadro constant.

CE

In the dynamic single phase model, both the Brownian motion effect on thermal
conductivity and viscosity was taken into consideration. Vajjha and Das [35] modified
AC

the classic static model of Maxwell on the basis of literatures [44, 45], in which the
static part is from the Maxwell theory and the dynamic term is caused by Brownian
motion of nanoparticles. The dynamic part was introduced from kinetic theory and the
influences of particle size, temperature, and volume concentration were counted.

(9)

7
 ACCEPTED MANUSCRIPT

Where f (T ,  ) and  denote the interaction of fluid parcels and nano particles. The

coefficients in f (T ,  ) and  for ZnO nanofluids are determined by experimental

results.
T
f (T ,  )  (2.8217 102   3.917 103 )( )
T0 (10)
 (3.0669 10   3.91123 10 )
2 3

PT
  8.4407(100 )1.07304 (11)

As for viscosity in DSPM, Masoumi model [40] was used. The Masoumi model is

RI
introduced from shear stress analysis of nanoparticle surface and the relative velocity

SC
between particles and fluid is considered:
 pVB d p2
nf  bf  (12)
72C
NU
Where VB and  are the Brownian velocity and the distance between nanoparticles
respectively:
MA

1 18k BT (13)
VB 
dp  p d p
D


 3 dp (14)
6
E
PT

1 (15)
C [c1d p c2 )  (c3d p  c4 )]
bf

In sum, the thermal conductivity and viscosity model for numerical analysis of the
CE

MFHS is listed in Table 3.

The MFHS domain was meshed according to the finite volume method. Hybrid of
AC

structured and unstructured element was used in meshing the MFHS. At inlet, uniform
nanofluids entered into the micro fin heat sink section with a temperature of T0=298K
and a velocity based on certain Reynolds number. The bottom surface of the MFHS
was uniformly heated by a heat flux of 10000W/m2. The other side wall surfaces were
assumed to be adiabatic. As for discretization, second-order upwind is employed in
solving the conservation equations.
For the MFHS, the Reynolds number was defined as:

8
 ACCEPTED MANUSCRIPT

 vmax d h
Re  f
(16)


Where vmax is the maximum velocity in the fluid and d h f is the hydraulic

diameter of the fins:

Qf
vmax  (17)
Amin

PT
4 Af
dhf 
Pf (18)

RI
Nusselt number is defined as:

SC
hd h f
Nu  (19)
k
NU
Where h denotes:
MA

q
h=
Ts - T f (20)
D

Where Ts and T f is the average temperature of bottom surface and the average
E

fluid temperature respectively.

PT

Euler number is defined as the dimensionless momentum loss for the MFHS:
2p
CE

Eu  (21)
 vmax
2
NL
AC

3. Grid independency and model validation

Four schemes were carried out in the grid independency study. As shown in Table 4,
the intermediate and fine mesh was used and grid adaption was made. The grid in the
near wall section with relative high gradient of velocity and temperature fields was
refined in Scheme 2 and Scheme 4 in order to improve the accuracy at lower
computational cost. It is observed that the largest difference is between Scheme 1 and
Scheme 2. As the Grid element number increases from 369837 to 581800 or higher,
the Nusselt number changed no more than 0.3%. Therefore, the intermediate mesh
9
 ACCEPTED MANUSCRIPT

size with gird refined in the near wall area was used in the following numerical
calculation.
The model validation of the present work can be divided into two sections: one is
the validation of the computational fluid dynamic model used for thermal and
hydraulic description in the MFHS domain, and the other one is the validation of
physical property model. For the first part, the obtained simulation results are

PT
compared with the empirical correlation of micro fin heat sinks by Tullius et al. [7].
The Tullius correlation concerning Nusselt number and pressure drop is as follows:

RI
S L 0.2 St 0.2 H f 0.25 dh 0.4 0.6 0.36 Pr 0.25
Nu f  0.0937( ) ( ) ( ) (1  ) Re f Pr ( )
dh f dh f dhf dhf Prs (22)

SC
 vmax
2
p  N L f (23)
2
NU
Where f is defined as the friction factor of MFHS:
S L 0.2 St 0.2 H f 0.18 dh 0.2 0.435
f  2.963( ) ( ) ( ) (1  ) Re f
MA

dhf dhf dhf d hf (24)

dh  H c  H f
(25)
D

The comparison of present study results for pure water with literature data is
E

displayed in Figure 2. Both Nu and P coincide with the correlation in the present
PT

Reynolds number range. The differences are within 10%. The models concerning
thermal conductivity and viscosity of nanofluids in the study have been validated and
CE

employed by previous researchers in experimental or numerical study on other

devices [41, 46, 47], meaning that they are able to analyze the heat transfer and
AC

hydraulic performance of the MFHS based on the precondition of the models.

4. Results and discussions

A computational simulation was performed to explore the heat transfer and
hydraulic properties of the MFHS with ZnO nanofluids. The results of SSPM and
DSPM that considering the Brownian motion were compared and the effects of
Reynolds number, particle size and the volume concentration on the nanofluids heat

10
 ACCEPTED MANUSCRIPT

transfer of MFHS were investigated in detail. Also the overall heat transfer efficiency
was evaluated and the Nu correlations were built.
4.1 Effect of Reynolds number
Figure 3 displays the comparison of heat transfer performance of ZnO-water
nanofluids at =1.5% and =3.0% as the function of Reynolds number. In Figure 3,
the heat transfer coefficients both increase with Re. With Re ranging 85 - 595, the heat
transfer coefficients of DPSM at =1.5% increased from 3735 to 10507 W/(m2K),

PT
15.0-20.2% higher than those of pure water respectively. The heat transfer coefficients

RI
of DPSM at = 3% increased from 4300 to 11470 W/(m2K), 25.6-38.3% higher than
those of pure water respectively. This heat transfer enhancement result is in the range

SC
of enhancement data reported by Ali and Arshad [48], in which a heat transfer
enhancement of 16.11-37.78% was obtained in a micro square fin heat sink with
NU
3.99-4.31 vol.% TiO2 nanofluids at Re=250-600. With the addition of nanoparticles,
the Nusselt numbers of both SSPM and DSPM also see an enhancement compared
MA

with the results of pure water. At Re=425 and =1.5%, for instance, the Nusselt
number of SSPM and DSPM are respectively 2.7 and 4.9% higher than pure water. In
D

addition, DSPM generally can obtain better heat transfer results than SSPM,
E

indicating that taking the Brownian motion effect into account can obtain much higher
PT

heat transfer performance.

In the single phase modeling of nanofluids heat transfer problem, the numerical
CE

results are closely related to the description of Brownian motion effect on the physical
properties of nanofluids. Brownian motion is a random movement of nanoparticles
AC

and has an impact on both thermal conductivity and viscosity of nanofluids. Based on
DSPM, the influence factors of Brownian motion on thermal conductivity and
viscosity of nanofluids mainly consist of average particle size, thermodynamic
temperature and volume concentration of nanoparticles. To further improve its
applicability more parameters such as particle shape distribution and particle material
are expected to be included. Figure 4 depicts the effect of Reynolds number on
pressure drop and Euler number of the MFHS with ZnO-water nanofluids at =1.5%
and =3.0%. It is observed that as Re increases the pressure drop increased to as high
11
 ACCEPTED MANUSCRIPT

as 2224 Pa while the value of Eu decreased from 0.75 to 0.25 for DSPM at =1.5%.
The present data show that with addition of nanofluids the pressure loss would
inevitably increase to some extent but the friction loss proportion in certain Re range
would decline.
To ascertain the heat transfer and flow characteristics of the MFHS, Figure 5 shows
the stream lines of the MFHS with ZnO-water nanofluids (=1.5%, dp=30nm, DSPM).

PT
It is worth to note that the stream lines near the bottom surface tend to completely
flow around the fins at lower Re. While at higher Re, the stream lines near the bottom

RI
surface are more likely to pass through the channel between the micro pin fins and the
wake length is relatively longer, which could also affect the heat transfer performance

SC
and pressure drop of MFHS.
4.2 Effect of particle size
NU
Particle size is another significant factor in the nanofluids heat transfer
performance of MFHS. As shown in Figure 6, the heat transfer coefficient and the
MA

Nusselt number both show a decreasing trend as the particle size increases from 30 to
77nm. For 1.5% ZnO-Water nanofluids at Re=425, the heat transfer coefficient of
D

SSPM reduces gently from 8803 to 8723 W/(m2K), while the heat transfer coefficient
E

of DSPM reduces from 9438 to 8992 W/(m2K). For 1.5% ZnO-water nanofluids at
PT

Re=425, h/hw decreases from 16.0 to 10.5%, and from 8.2 to 7.2 % as particle size
increases from 30 to 77 nm by DSPM and SSPM method respectively. The particle
CE

sizes of nanofluids have great influence on heat transfer characteristics as discovered

by many researchers [49]. In consideration of the study of particle size influence on
AC

heat transfer, DSPM considering Brownian motion by utilize proper thermal

conductivity and viscosity model, which is probably more appropriate for
investigating the influence of nanofluids particle size.
In SSPM, the nanoparticle is assumed to be in a steady and static state, and heat
conduction occurs through the uniform mixture of base fluid and nanoparticles. The
heat conduction of DSPM, however, was induced not only by the static thermal wave
propagation, but also by the dynamic micro-scale heat and energy transfer caused by
Brownian motion of nanoparticles. Resulting from their collision with the fast-moving
12
 ACCEPTED MANUSCRIPT

molecules in the fluid, Brownian motion is the random motion of particles suspended
in a fluid. On the one hand, for a single particle the smaller particle size means less
amount of liquid molecules could impact on the particle surface, leading to higher
unbalancedness and acceleration of the particle by the force of the liquid molecules.
Thus lower particle size can contribute to a relatively strong Brownian motion effect.
On the other hand, at certain volume concentration lower particle diameter can lead to

PT
higher specific surface area, which means the particles can take along more
surrounding fluid in the random movement. As a result, when the particle size of

RI
nanofluids declines, the micro-scale interaction is more remarkable and the heat
transfer performance in DSPM is better. When the particle size increased, the heat

SC
transfer coefficient of DSPM decreased drastically and the Nusselt number of the two
models becomes relatively close to each other. In literatures some experimental result
NU
discrepancies were reported in the nanofluid heat transfer experiments with the same
type of nanoparticles and similar flow conditions [50, 51]. Part of the reason may be
MA

that the nanofluids usually have different particle sizes and the particle size
distribution tends to change with time due to particle interaction and aggregation.
D

As the particle size of ZnO increases, the pressure drops of SSPM and DSPM fall
E

down at Re=425 and =1.5%. The pressure drop of SSPM reduces from 1004 to 936
PT

Pa, while the pressure drop of DSPM decreases from 1306 to 995 Pa, as shown in
Figure 7. And the decreasing rate of DSPM reduces as the particle size increases. The
CE

Euler number is kept almost unchanged as the particle size increases.

4.3 Effect of volume concentration
AC

Figure 8 depicts the relation of nanoparticle volume concentration with the heat
transfer coefficient of the heat sink. It is observed that the heat transfer coefficient of
the MFHS goes up steadily with the volume concentration of nanoparticle increasing
from 0.6 to 3.0%. At Re =425 and dP=30nm, the heat transfer coefficient of both
SSPM and DSPM experienced an evidently increase. Compared with pure water, the
heat transfer coefficient of SSPM enhanced 8.2% and 17.9% for 1.5% vol. and 3.0
vol.% ZnO-water nanofluids respectively. The DSPM considering the Brownian
motion effect generally has higher heat transfer performance than the SSPM at all the
13
 ACCEPTED MANUSCRIPT

simulated volume concentrations. As expected, the heat transfer coefficient of DSPM

enhanced 16.0% and 27.8% for 1.5% vol. % and 3.0 vol.% ZnO-water nanofluids
respectively.
The effect of volume concentration on the pressure drop and the overall heat sink
heat transfer efficiency also revealed the similar tendency. As shown in Figure 9, the
pressure drop of the MFHS increased with volume concentration in both SSPM and

PT
DSPM, although in DSPM the pressure drop increases with a higher slope. It should
be stressed that the influences of volume concentration on the thermal and hydraulic

RI
property are usually two-sided: one is enhancing the thermal conductivity of the heat
transfer fluid, and the other is increasing the fluid viscosity, contributing to a thicker

SC
boundary layer and higher pressure drop.
4.4 Overall heat transfer efficiency
NU
To look into the heat transfer performance of the MFHS, the temperature
distributions of the MFHS bottom surface at two volume concentration with
MA

ZnO-water nanofluids as the coolant (Re=85, dp=30nm, DSPM) are presented in

Figure 10. The surface temperature on the wake side of the pin fins is relatively lower
D

than the upstream side and the surface temperature rises along the flow direction.
E

After nanoparticle addition, the surface temperature of the MFHS decreases obviously.
PT

And the surface temperature of nanofluids at higher volume concentration is lower

than that at lower volume concentration. ZnO-Water nanofluid is effective in
CE

improving the cooling ability of heat sinks.

Due to the heat transfer performance improvement is usually accompanied by
AC

higher pressure loss, the overall heat sink heat transfer efficiency to evaluate the heat
transfer enhancement quality is defined as:

  Nu Eu (26)
It means the overall heat transfer efficiency of the MFHS considering the pressure
loss caused by increasing the Reynolds number. As shown in Figure 11 (a), the overall
heat sink heat transfer efficiency of the MFHS continuously increase with Reynolds
number for both two models at =1.5% and =3.0%. It means that higher Re can

14
 ACCEPTED MANUSCRIPT

contribute to better thermal and hydraulic performance of the MFHS if permitted.

This was coincident with Seyf and Feizbakhshi [25], in which a numerical study was
done on the heat transfer enhancement of MFHS with CuO and Al2O3 nanofluids as
the coolants and similar conclusion was obtained. The overall heat sink heat transfer
efficiency of the MFHS with ZnO-water nanofluids at Re=425 and =1.5% is
calculated and displayed in Figure 11 (b). The highest overall heat transfer efficiency

PT
of the MFHS is 32.4, with a particle size of 30 nm. Compared with pure water, the
overall heat sink heat transfer efficiency of ZnO-water nanofluids is much higher. It

RI
can be seen that the smaller particle size of nanofluids could obtain higher overall heat
transfer efficiency, implying that nanofluids with smaller particle size could promote

SC
the heat transfer performance without extra cost in momentum loss. As the particle
size increases, the differences in the overall heat sink heat transfer efficiency of the
NU
two models become unobvious.
As displayed in Figures 11 (b) and (c), it’s obvious that lower particle size and
MA

higher volume concentration can facilitate the overall efficiency of the micro fin heat
sink in the simulated range (=0.6-3%, d=30-77 nm), which means the higher volume
D

concentration and lower particle size is in favor of higher efficiency of heat sink in
E

certain condition (Re =425). The maximum overall efficiency of 34.9 was obtained at
PT

=3.0% and dP=30nm using DSPM (water=29.6). However, it is difficult to prepare

nanofluids with high concentration and small particle size simultaneously.
CE

Nanoparticles tend to agglomerate by van der Waals force as soon as they are added to
the solution especially in high concentration. So far, many methods have been applied
AC

to prevent particle agglomeration, such as ultrasonic, dispersant addition. And in

many experimental studies, real particle size rather than primary particle size of
nanofluids used should be given for a better understanding of nanofluids heat transfer
enhancement mechanism.
4.5 Mathematical Correlations
The simulation results were applied to build Nusselt number (Nu) correlations for the
prediction of ZnO-water nanofluids heat transfer performance in MFHS. The
regression equation coefficients were obtained by the classical least square method
15
 ACCEPTED MANUSCRIPT

with the help of Origin software. The obtained regression equations were as follows:

Nu(SSPM)  0.201Re 0.534 Pr 0.293 0.0027 (27)

Nu(DSPM)  0.243Re 0.521 Pr 0.238 0.012 (28)

The comparison of predicted and simulation values of Nusselt number by SSPM and
DSPM models are displayed in Figure 12. The predicted values are in good agreement
with the simulation results with a deviation of approximately 5%. The proposed

PT
correlations can be used to predict heat transfer performance of ZnO-water nanofluids
in MFHS.

RI
5. Conclusions

SC
The numerical investigation is carried out in a MFHS with ZnO-water nanofluids.
The results of SSPM and DSPM models were compared and the effects of Reynolds
NU
number, particle size and volume concentration on the nanofluids heat transfer and
hydraulic properties of MFHS were investigated. The following conclusions can be
MA

arrived at:
(1) The heat transfer coefficient and the pressure loss continuously increase with
D

Reynolds number for both two models at =1.5% and =3.0%. The heat transfer
E

coefficients of DPSM at = 3% increased from 4300 to 11470 W/(m2K), 25.6-38.3%

PT

higher than those of pure water respectively. And DSPM has higher heat transfer
performance than SSPM, indicating that taking the Brownian motion effect into
CE

account can obtain much higher heat transfer performance.

(2) Particle size is an important factor in the nanofluids heat transfer and hydraulic
AC

performance of MFHS. The heat transfer coefficient and the Nusselt number both
show a decreasing trend with particle size increasing from 30 to 77nm. DSPM
considering Brownian motion by utilize proper thermal conductivity and viscosity
model, which is probably more appropriate for investigating the influence of
nanofluids particle size.
(3) Higher overall efficiency of the micro fin heat sink can be obtained at lower
particle size and higher volume concentration. The maximum of 34.9 was obtained at
=3.0% and dP=30 nm using DSPM (water=29.97). In experimental study, real particle
16
 ACCEPTED MANUSCRIPT

size rather than primary particle size of nanofluids used should be given.

Acknowledgement
We acknowledge the financial support from the National Natural Science Foundation
of China (No. 51476146, 51476145, 51606169); and the Education Department of
Zhejiang Province (No. Y201738887).

PT
References

RI
[1] X. L. Xie, W. Q. Tao, Y. L. He, Numerical study of turbulent heat transfer and

SC
pressure drop characteristics in a water-cooled minichannel heat sink. J. Electron.
Packag. 129 (3) (2007) 247–255.
[2] S. E. Ghasemi, A. A. Ranjbar, M. J. Hosseini, Forced convective heat transfer of
NU
nanofluid as a coolant flowing through a heat sink: Experimental and numerical
study. J. Mol. Liq. 248 (2017) 264–270.
MA

[3] P. S. Lee, C. J. Teo, Heat transfer enhancement in microchannels incorporating

slanted grooves. ASME 2008 First International Conference on Micro/Nanoscale
Heat Transfer (2008) 819–823.
D

[4] H. Y. Wu, C. Ping, An experimental study of convective heat transfer in silicon

E

microchannels with different surface conditions. Int. J. Heat Mass Transfer 46

PT

(14) (2003) 2547–2556.

[5] W. A. Khan, J. R. Culham, M. M. Yovanovich, Optimization of pin-fin heat sinks
CE

using entropy generation minimization. IEEE Trans. Compon. Packag. Technol.

28(2) (2005) 247–254.
[6] K. A. Moores, Y. K. Joshi, Effect of tip clearance on the thermal and
AC

hydrodynamic performance of a shrouded pin fin array. J. Heat Transfer 125 (6)
(2003) 999–1006.
[7] J. F. Tullius, T. K. Tullius, Y. Bayazitoglu, Optimization of short micro pin fins in
minichannels. Int. J. Heat Mass Transfer 55 (15–16) (2012) 3921–3932.
[8] S. U. S. Choi, Z. G. Zhang, W. Yu, F. E. Lockwood, E. A. Grulke, Anomalous
thermal conductivity enhancement in nanotube suspensions. Appl. Phys. Lett. 79
(14) (2001) 2252–2254.
[9] K. Nemade, S. Waghuley, A novel approach for enhancement of thermal
conductivity of CuO/H2O based nanofluids. Appl. Therm. Eng. 95 (2016) 271–
17
 ACCEPTED MANUSCRIPT

274.
[10] S. Akilu, A. T. Baheta, K. V. Sharma, Experimental measurements of thermal
conductivity and viscosity of ethylene glycol-based hybrid nanofluid with
TiO2-CuO/C inclusions. J. Mol. Liq. 246 (2017) 396–405.
[11] M. Sheikholeslami, Magnetohydrodynamic nanofluid forced convection in a
porous lid driven cubic cavity using Lattice Boltzmann method. J. Mol. Liq. 231
(2017) 555–565.
[12] W. Guo, G. Li, Y. Zheng, C. Dong, Laminar convection heat transfer and flow

PT
performance of Al2O3–water nanofluids in a multichannel-flat aluminum tube.
Chem. Eng. Res. Des. 133 (2018) 255–263.

RI
[13] M. Sheikholeslami, Influence of Lorentz forces on nanofluid flow in a porous
cylinder considering Darcy model. J. Mol. Liq. 225 (2017) 903–912.

SC
[14] M. Sheikholeslami, CuO-water nanofluid flow due to magnetic field inside a
porous media considering Brownian motion. J. Mol. Liq. 249 (2018) 921–929.
NU
[15] M. Sheikholeslami, Magnetic field influence on nanofluid thermal radiation in a
cavity with tilted elliptic inner cylinder. J. Mol. Liq. 229 (2017) 137–147.
MA

[16] M. Sheikholeslami, Numerical investigation of nanofluid free convection under

the influence of electric field in a porous enclosure. J. Mol. Liq. 249 (2018)
1212–1221.
D

[17] W. Duangthongsuk, S. Wongwises, An experimental study on the thermal and

hydraulic performances of nanofluids flow in a miniature circular pin fin heat
E

sink. Exp. Therm. Fluid Sci. 66(2015) 28–35.

PT

[18] M. Z. Zhou, G. Xia, L. Chai, J. Li, L. Zhou, Analysis of flow and heat transfer
characteristics of micro-pin fin heat sink using silver nanofluids. Sci. China:
CE

Technol. Sci. 55 (1) (2012) 155–162.

[19] H. M. Ali, W. Arshad, Effect of channel angle of pin-fin heat sink on heat transfer
AC

performance using water based graphene nanoplatelets nanofluids. Int. J. Heat

Mass Transfer 106 (2017) 465–472.
[20] M. Sheikholeslami, H. B. Rokni, CVFEM for effect of Lorentz forces on
nanofluid flow in a porous complex shaped enclosure by means of
non-equilibrium model. J. Mol. Liq. 254 (2018) 446–462.
[21] M. Sheikholeslami, M. Seyednezhad, Simulation of nanofluid flow and natural
convection in a porous media under the influence of electric field using CVFEM.
Int. J. Heat Mass Transfer 120 (2018) 772–781.
[22] M. Sheikholeslami, Numerical investigation for CuO-H2O nanofluid flow in a

18
 ACCEPTED MANUSCRIPT

porous channel with magnetic field using mesoscopic method. J. Mol. Liq. 249
(2018) 739–746.
[23] S. E. Ghasemi, A. A. Ranjbar, M. J. Hosseini, Forced convective heat transfer of
nanofluid as a coolant flowing through a heat sink: Experimental and numerical
study. J. Mol. Liq. 248 (2017) 264–270.
[24] F. Garoosi, F. Hoseininejad, M. M. Rashidi, Numerical study of heat transfer
performance of nanofluids in a heat exchanger. Appl. Therm. Eng. 105 (2016)
436-455.

PT
[25] H. R. Seyf, M. Feizbakhshi, Computational analysis of nanofluid effects on
convective heat transfer enhancement of micro-pin-fin heat sinks. Int. J. Therm.

RI
Sci. 58 (4) (2012) 168–179.
[26] M. I. Hasan, Investigation of flow and heat transfer characteristics in micro pin

SC
fin heat sink with nanofluid. Appl. Therm. Eng. 63(2) (2014) 598–607.
[27] Snoussi, L., N. Ouerfelli, K. V. Sharma, N. Vrinceanu, A. J. Chamkha, A. Guizani,
NU
Numerical simulation of nanofluids for improved cooling efficiency in a 3D
copper microchannel heat sink (MCHS) Phys. Chem. Liq. (2017) 1–21.
MA

[28] D. Lelea, I. Laza, The water based Al2O3 nanofluid flow and heat transfer in
tangential microtube heat sink with multiple inlets. Int. J. Heat Mass Transfer 69
(1) (2014) 264–275.
D

[29] W. W. Guo, G. N. Li, Y. Q. Zheng, C. Dong, Measurement of the thermal

conductivity of SiO2 nanofluids with an optimized transient hot wire method.
E

Thermochim. Acta 661 (2018) 84–97.

PT

[30] J. C. Maxwell, A Treatise on Electricity and Magnetism, Clarendon, Oxford,

1873.
CE

[31] R. L. Hamilton, O. K. Crosser, Thermal conductivity of heterogeneous

two-component systems. Ind. Eng. Chem. Fundam. 1(3) (1962) 187–191.
AC

[32] W. Yu, S. U. S. Choi, The Role of Interfacial Layers in the Enhanced Thermal
Conductivity of Nanofluids: A Renovated Maxwell Model. J. Nanopart. Res. 6 (4)
(2004) 355–361.
[33] Y. M. Xuan, Q. Li, Heat transfer enhancement of nanofluids. Int. J. Heat Fluid
Flow 21(1) (2000) 58–64.
[34] J. Koo, C. Kleinstreuer, A new thermal conductivity model for nanofluids. J.
Nanopart. Res. 6 (6) (2004) 577–588.
[35] R. S. Vajjha, D. K. Das, Experimental determination of thermal conductivity of
three nanofluids and development of new correlations. Int. J. Heat Mass Transfer

19
 ACCEPTED MANUSCRIPT

52 (21–22) (2009) 4675–4682.

[36] M. Mahmoodi, S. Kandelousi, Effects of thermophoresis and Brownian motion
on nanofluid heat transfer and entropy generation. J. Mol. Liq. 211 (2015) 15–24.
[37] H. R. Seyf, B. Nikaaein, Analysis of Brownian motion and particle size effects
on the thermal behavior and cooling performance of microchannel heat sinks. Int.
J. Therm. Sci. 58 (2012) 36–44.
[38] A. Einstein, Investigation on the Theory of Brownian Motion, Dover, New York,
1956.

PT
[39] H. Brinkman, The viscosity of concentrated suspensions and solutions. J. Chem.
Phys. 20 (1952) 571-571.

RI
[40] N. Masoumi, N. Sohrabi, A. Behzadmehr, A new model for calculating the
effective viscosity of nanofluids. J. Phys. D: Appl. Phys. 42(5) (2009) 055501–

SC
055506.
[41] H. Safikhani, A. Abbassi, A new dispersion model for thermal properties of
NU
nanofluids in flat tubes. Int. J. Therm. Sci. 109 (2016) 114–122.
[42] Y. M. Xuan, W. Roetzel, Conceptions for heat transfer correlation of nanofluids.
MA

Int. J. Heat Mass Transfer 43(19) (2000) 3701–3707.

[43] M. Corcione, Empirical correlating equations for predicting the effective thermal
conductivity and dynamic viscosity of nanofluids. Energy Convers. Manage. 52
D

(1) (2011) 789–793.

[44] Y. M. Xuan, Q. Li, W. Hu, Aggregation structure and thermal conductivity of
E

nanofluids. AIChE J. 49 (4) (2003) 1038–1043.

PT

[45] J. Koo, C. Kleinstreuer, Laminar nanofluid flow in microheat-sinks. Int. J. Heat

Mass Transfer 48 (13) (2005) 2652–2661.
CE

[46] T. Perarasu, M. Arivazhagan, P. Sivashanmugam, Experimental and CFD Heat

Transfer Studies of Al2O3-Water Nanofluid in a Coiled Agitated Vessel Equipped
AC

with Propeller. Chin. J. Chem. Eng. 21(11) (2013) 1232–1243.

[47] M. H. Jian, H. W. Tang, Y. T. Yang, Numerical simulation and optimization of
nanofluids in a complex micro heat sink. Numer. Heat Transfer, Part A 71 (3)
(2017) 341–359.
[48] H. M. Ali, W. Arshad, Thermal performance investigation of staggered and inline
pin fin heat sinks using water based rutile and anatase TiO2 nanofluids. Energy
Convers. Manage. 106 (2015) 793–803.
[49] E. V. Timofeeva, D. S. Smith, W. Yu, D. M. France, D. Singh, J. L.Routbort,
Particle size and interfacial effects on thermo-physical and heat transfer

20
 ACCEPTED MANUSCRIPT

characteristics of water-based α-SiC nanofluids. Nanotechnology 21(21) (2010)

215703.
[50] D. S. Wen, Y. Ding, Experimental investigation into convective heat transfer of
nanofluids at the entrance region under laminar flow conditions. Int. J. Heat
Mass Transfer 47 (24) (2004) 5181–5188.
[51] U. Rea, T. McKrell, L. W. Hu, J. Buongiorno, Laminar convective heat transfer
and viscous pressure loss of alumina–water and zirconia–water nanofluids. Int. J.
Heat Mass Transfer 52(7-8) (2009) 2042–2048.

PT
RI
SC
NU
MA
E D
PT
CE
AC

21
 ACCEPTED MANUSCRIPT

Figure captions:
Figure 1 A 3D Schematic representation of the micro fin heat sink
Figure 2 Comparison of (a) Nu and (b) P simulation results for pure water with
literature
Figure 3 Effect of Reynolds number on heat transfer characteristics of the MFHS
with ZnO-water nanofluids (=1.5%, =3.0%, dP=30nm)

PT
Figure 4 Effect of Reynolds number on pressure drop and Euler number of the MFHS
with ZnO-water nanofluids (=1.5%, =3.0%, dP=30nm)

RI
Figure 5 The stream lines of MFHS with ZnO-water nanofluids (=1.5%, dP=30nm,

SC
DSPM)
Figure 6 Effect of particle size on heat transfer characteristics of the MFHS with
NU
ZnO-water nanofluids (Re =425, =1.5%)
Figure 7 Effect of particle size on pressure drop and Euler number of the MFHS with
MA

ZnO-water nanofluids (Re =425, =1.5%)

Figure 8 Effect of volume concentration on heat transfer characteristics of the MFHS
with ZnO-water nanofluids (Re =425, dP =30nm)
D

Figure 9 Effect of volume concentration on pressure drop of the MFHS with

E

ZnO-water nanofluids (Re =425, dP=30nm)

PT

Figure 10 Temperature contours of MFHS bottom surface at different volume

concentration with ZnO-water nanofluids (Re=85, dp=30nm, DSPM)
CE

Figure 11 Effect of (a) Reynolds number (b) particle diameter (c) volume
concentration on the overall heat sink heat transfer efficiency of the MFHS with
AC

ZnO-water nanofluids
Figure 12 Comparison of predicted and experimental values of Nusselt number using
SSPM (a) and DSPM (b) models

22
 ACCEPTED MANUSCRIPT

PT
RI
Figure 1 A 3D Schematic representation of the micro fin heat sink

SC
16
(a)
14 Tullius et al. [7]
present study
NU
12
10%
10
MA
Nuf

4
D

2
E

0
100 200 300 400 500 600
Re
PT

(b)
CE

Tullius et al. [7]

1200 Present study
10%
AC
P (Pa)

800

400

0
100 200 300 400 500 600
Re

Figure 2 Comparison of (a) Nu and (b) P simulation results for pure water with
literature

23
 ACCEPTED MANUSCRIPT

12000 20
Water
=1.5% SSPM 18
=1.5% DSPM
10000 =3.0% SSPM 16
=3.0% DSPM
14
h (W/m2/K) 8000
12

Nu
10
6000
8
Water
= 1.5% SSPM

PT
4000 6
= 1.5% DSPM
= 3.0% SSPM 4
=3.0% DSPM

RI
2000 2
0 100 200 300 400 500 600
Re

SC
Figure 3 Effect of Reynolds number on heat transfer characteristics of the MFHS
with ZnO-water nanofluids (=1.5%, =3.0%, dP=30nm)
NU
4000
Water Water
=1.5% SSPM =1.5% SSPM
MA

3500 0.8
=1.5% DSPM =1.5% DSPM
=3.0% SSPM =3.0% SSPM
3000
=3.0% DSPM =3.0% DSPM

2500 0.6
P (Pa)

u
2000
E

0.4
1500
PT

1000
0.2
500
CE

0
0.0
0 100 200 300 400 500 600
Re
AC

Figure 4 Effect of Reynolds number on pressure drop and Euler number of the
MFHS with ZnO-water nanofluids (=1.5%, =3.0%, dP=30nm)

24
 ACCEPTED MANUSCRIPT

PT
RI
SC
NU
MA
D

Figure 5 The stream lines of MFHS with ZnO-water nanofluids (=1.5%,

E

dP=30nm, DSPM)
PT
CE
AC

25
 ACCEPTED MANUSCRIPT

9800 1.36
h =1.5% SSPM
= 1.5% DSPM 1.32
9600
h/hw =1.5% SSPM 1.28
9400 =1.5% DSPM
h (W(m2K)) 1.24

h/hw
9200 1.20

9000 1.16

1.12
8800

PT
1.08
8600
1.04

RI
8400 1.00
30 40 50 60 70 80
Particle size (nm)

SC
9.9
9.8
NU
9.7 =1.5% SSPM
=1.5% DSPM
9.6
MA

9.5
9.4
Nu

9.3
9.2
D

Enhancement
9.1
E

9.0
Nu=9.04 (Water)
PT

8.9
8.8
20 30 40 50 60 70 80 90
CE

Particle size (nm)

Figure 6 Effect of particle size on heat transfer characteristics of the MFHS with
AC

ZnO-water nanofluids (Re =425, =1.5%)

26
 ACCEPTED MANUSCRIPT

1600 3.0
P =1.5% SSPM
=1.5% DSPM 2.5
1400
u =1.5% SSPM 2.0
=1.5% DSPM
1.5
P (Pa)
1200

1.0

Eu
1000
0.5

PT
0.0
800
-0.5

RI
600 -1.0
30 40 50 60 70 80

SC
Particle size (nm)

Figure 7 Effect of particle size on pressure drop and Euler number of the MFHS
NU
with ZnO-water nanofluids (Re =425, =1.5%)
MA

1.5
10500 d=30nm SSPM
d=30nm DSPM
10000 d=30nm SSPM 1.4
D

d=30nm DSPM
9500
h (W/m2/K)

1.3
h/hw

9000
PT

1.2
8500
CE

8000 1.1

7500
1.0
AC

0.5 1.0 1.5 2.0 2.5 3.0

Volume concentration (%)

Figure 8 Effect of volume concentration on heat transfer characteristics of the

MFHS with ZnO-water nanofluids (Re =425, dP =30nm)

27
 ACCEPTED MANUSCRIPT

2.50
d=30nm SSPM
2000 d=30nm DSPM
2.25
d=30nm SSPM
1500 d=30nm DSPM
P (Pa) 2.00

P/Pw
1000 1.75

1.50
500

PT
1.25
0

RI
1.00
0.5 1.0 1.5 2.0 2.5 3.0
Volume concentration (%)

SC
Figure 9 Effect of volume concentration on pressure drop of the MFHS with
ZnO-water nanofluids (Re =425, dP=30nm)
NU
MA
E D
PT
CE

Figure 10 Temperature contours of MFHS bottom surface at different volume

AC

concentration with ZnO-water nanofluids (Re=85, dp=30nm, DSPM)

28
 ACCEPTED MANUSCRIPT

50
(a)
45
40
35
30
 25
20
Water
15 1.5% SSPM (dP=30nm)
1.5% DSPM (dP=30nm)

PT
10
3.0% SSPM (dP=30nm)
5
3.0% DSPM (dP=30nm)
0

RI
0 100 200 300 400 500 600
Re

SC
34
=1.5% SSPM (Re=425)
(b)
=1.5% DSPM (Re=425)
33
NU
32

31


MA

Enhancement
30

=29.97 (water Re=425)

29
E D

28
20 30 40 50 60 70 80 90
PT

Particle size (nm)

36
d=30nm SSPM (Re=425)
CE

35 d=30nm DSPM (Re=425)

34
33
AC

32


Enhancement
31
30
=29.97 (water Re=425)
29
28
0.5 1.0 1.5 2.0 2.5 3.0
Volume concentration (%)
Figure 11 Effect of (a) Reynolds number (b) particle diameter (c) volume
concentration on the overall heat sink heat transfer efficiency of the MFHS with
ZnO-water nanofluids

29
 ACCEPTED MANUSCRIPT

14 14
(a) (b)
Predicted Nusselt number(SSPM)

12

Predicted Nusselt number(DSPM)

12 5% 5%

10 10

8 8

PT
6 6

4 4

RI
2 2

2 4 6 8 10 12 14 2 4 6 8 10 12 14

SC
Simulation Nusselt number (SSPM) Simulation Nusselt number (DSPM)

Figure 12 Comparison of predicted and experimental values of Nusselt number using

NU
SSPM (a) and DSPM (b) models
MA
E D
PT
CE
AC

30
 ACCEPTED MANUSCRIPT

Table 1 Dimensions of the MFHS in the study

Dimension size
Width of the MFHS. W (mm) 16
Length of the MFHS. L (mm) 25
Height of the MFHS channel. Hc (mm) 1
Longitudinal pitch of fin centers. SL(mm) 2
Transverse pitch of fin centers. ST (mm) 2
Height of pin fin. Hf (mm) 0.5

PT
Diameter of pin fin. Df (mm) 1
Longitudinal pin fin number. NL 12
Transverse pin fin number. NT 7

RI
Clearance of pin fin. dc (mm) 1

SC
Table 2 Physical property of base fluid and nanoparticles
Property Water ZnO nanoparticles
NU
 (kg/m )
3
998.2 2200
k (W/(mK)) 0.6 1.4
Cp (J/(kgK)) 4182 742
MA

μ (kg/(ms)) 0.001003 -
D

Table 3 Thermal conductivity and viscosity model

ID Thermal conductivity Viscosity
E

bf
k p  2kbf  2(k p  kbf ) nf 
PT

SSPM knf  kbf  d p 1.03 

k p  2kbf  (k p  kbf ) 1  34.87( ) 
 dbf 
CE

k p  2kbf  2(kbf  k p ) kT  pVB d p2

DSPM knf  kbf  5 104 bf C pbf f (T ,  ) nf  bf 
k p  2kbf  (kbf  k p )  pd p 72C
AC

Table 4 Grid independency study (pure water, Re=425)

Grid %Diff of
No. Meshing scheme Nu
elements Nu
1 195579 intermediate 8.50 -
2 369837 intermediate (refined in the near wall area) 9.04 6.4
3 581800 fine 9.07 0.3
4 768987 fine (refined in the near wall area) 9.10 0.2

31
 ACCEPTED MANUSCRIPT

Highlights

 A model of MFHS with ZnO nanofluids considering Brownian motion is

numerically built

 Heat transfer and pressure drop using DSPM are higher than using SSPM

 Higher volume concentration and lower particle size are in favor of heat transfer

 Overall heat transfer efficiency was evaluated

PT
RI
SC
NU
MA
E D
PT
CE
AC

32