Journal Pre-proof

The effect of flow pulsation on Al2 O3 nanofluids heat transfer behavior in

a helical coil: a numerical analysis

Wenwen Guo, Guoneng Li, Youqu Zheng, Cong Dong

PII: S0263-8762(20)30026-5
DOI: https://doi.org/10.1016/j.cherd.2020.01.016
Reference: CHERD 3967

To appear in: Chemical Engineering Research and Design

Received Date: 14 October 2019

Revised Date: 7 January 2020
Accepted Date: 13 January 2020

Please cite this article as: Guo W, Li G, Zheng Y, Dong C, The effect of flow pulsation on
Al2 O3 nanofluids heat transfer behavior in a helical coil: a numerical analysis, Chemical
Engineering Research and Design (2020), doi: https://doi.org/10.1016/j.cherd.2020.01.016

This is a PDF file of an article that has undergone enhancements after acceptance, such as
the addition of a cover page and metadata, and formatting for readability, but it is not yet the
definitive version of record. This version will undergo additional copyediting, typesetting and
review before it is published in its final form, but we are providing this version to give early
visibility of the article. 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.

© 2020 Published by Elsevier.

The effect of flow pulsation on Al2O3 nanofluids heat transfer
behavior in a helical coil: a numerical analysis

Wenwen guoa, Guoneng Lia, Youqu Zhenga,*, Cong Donga

a Department of Energy and Environment System Engineering, Zhejiang University of

Science and Technology, Hangzhou 310023, China

of
* Corresponding author: zyq888@zust.edu.cn Tel: +86-571-85070019

ro
Graphical abstract

Z=6 Z=5 Z=4 Z=3 Z=2

-p
Z=1
re
90°
Inlet
lP

0°
na
ur

Nanofluids heat transfer

in a helical coil under
pulsation
Jo

Highlights
 Nanofluids heat transfer in a helical coil under pulsation condition was
simulated
 Impact of various pulsation parameters on the heat transfer performance was
1
 studied

  h( pul ) / h(bf ) got its maximum value of 1.35 at the optimum pulsation

frequency
 The enhancing mechanisms of nanofluids heat transfer under pulsation were
explored

Abstract
Nanofluids heat transfer in a helical coil under pulsation condition was numerically
simulated. The dynamic single phase model which considered Brownian motion effect

of
was utilized to study the Al2O3 nanofluids heat transfer in helical coils. The reliability
test results were consistent with experimental results. Then, pulsation was further

ro
imposed. The influence of the pulsation frequency on the heat transfer performance
was explored. And the changes of instantaneous h with frequency were further studied
-p
to interpret the results. The axial velocity contours, axial velocity distributions, and
cross-section velocity vectors at various planes along the coil were examined. The
re
results revealed that there was an optimum pulsation frequency for heat transfer
augmentation within the scope of Re. The secondary flow generated in cross region
lP

and the counter-rotating vortex formed in the axial direction all devoted to the helical
coil heat transfer augmentation under pulsation.
na

Keywords: nanofluids; pulsation flow; nanoparticle; helical coil; secondary flow.

ur

Nomenclature

A the heat transfer area (m2) Greek symbols

Jo

Am non-dimensional amplitude  the Womersley parameter

interaction between fluid particles and
Cp specific heat (J/(kgK)) 
nanoparticles (m)
De De number  distance between nanoparticles (m)

DSPM dynamic single phase model  viscosity (kg/(ms))

f frequency (Hz)  density (kg/m3)
f(-) friction factor  pulsation period (s)

2
 convection heat transfer
h  volume concentration (%)
coefficient (W/(m2K))
time-average convection heat
h  angular frequency (rad/s)
transfer coefficient (W/(m2K))

H coil pitch (mm) Subscripts

He helical coil number a axial

HTP heat transfer performance bf base fluid

the surface average of the

n
J ABS absolute value of vorticity in the C coil
cross section (s-1)
k thermal conductivity (W/(mK)) hmax maximum heat transfer coefficient

of
m fluid mass flow rate (kg/s) nf nanofluid

Nu Nusselt number p particle

ro
Nu time-average Nusselt number pul pulsation
r helical coil tube radius (mm) S straight pipe
Rc coil radius (mm) st steady
t
T
time (s)
temperature (K)
t
w
-p
time
wall
re
U X-axis velocity (m/s)

V Y-axis velocity (m/s)

lP

VB Brownian velocity

vol.% volume percent

na

W Z-axis velocity (m/s)

Z spiral number

P pressure drop (Pa)

ur

1 Introduction
Jo

Curved tubes are used extensively in various industry departments. Compared with
straight tubes, curved tubes such as spiral and helically coiled tubes have more
compact structure, larger heat transfer area, and higher heat transfer rate (Naphon and
Wongwises, 2006). To promote the energy efficiency of exchanger with certain
geometry and configuration, many heat transfer enhancement techniques have been
explored by researchers, among which developing new heat transfer medium with
3
appropriate thermal properties is a typical approach. Nanofluids are promising
alternatives to traditional heat transfer media due to their high thermal conductivity
and well heat transfer performance (HTP). They are mixtures of fluid and suspended
nanoparticles (metals, metal oxides, carbon nanotubes, etc.). Since the idea of
nanofluids was put forward in 1995, they have aroused interest of researchers and
been applied in heat transfer enhancement of various types of heat exchangers
(Mahian et al.,2018; Guo et al., 2018a; Sheremet et al. 2017; Bondareva et al. 2018).

In recent studies, nanofluids were applied in coil heat exchangers to enhance the heat
transfer. (Pan et al., 2014; Khoshvaght-Aliabadi et al., 2018; Soto-Meca et al., 2016;

of
Narrein and Mohammed, 2014). Khoshvaght-Aliabadi et al. (2018) designed a new
type of spirally-coiled twisted-duct in which the performance index of 1.88 was

ro
obtained by using 1.0% Cu-water nanofluids. Akhavan-Behabadi et al. (2012) found
that in helical coils the nanofluids exhibited better HTP than the water and oil based
-p
fluid and the dimensions of the coils played an important role. Kahani et al. (2013)
compared the HTP of two metal oxide nanofluids (volume concentration of 0.25-1.0%)
re
in helical coils. The experimental results revealed Al2O3 nanofluids had better
performance than TiO2 nanofluids with the same volume concentration. Kannadasan
lP

et al. (2012) made a comparison of nanofluids heat transfer in turbulence with

different arrangements of helical coils and obtained similar heat transfer coefficient
and friction factors for the horizontal and vertical helical coils. They also found higher
na

concentration of nanofluids were in favor of Nusselt number (Nu) increment. Ghobadi

and Muzychka (2013) emphasized the importance of the radius of curvature of coils
on heat transfer augmentation and discovered the heat transfer decreased with
ur

increasing radius of curvature. Mohammed and Narrein (2012) discovered the helix
radius and inner tube diameter had an influence on the HTP of the coiled tube. In
Jo

curved tubes, the fluid particle motion is not always parallel to the coil tube axis
because of the sub-movement generated by secondary flow (Sudo et al., 2006;
Mondal et al., 2007) which is beneficial for heat transfer enhancements (Dravid et al.,
1971).

At present, one of the technical bottlenecks for nanofluids used as heat transfer media
is their instability, restricting the wide application of nanofluids. Attracting the
4
extensive attentions of researchers, pulsation flow has merits of preventing
nanoparticle sedimentation and also shows potential to enhance HTP in certain heat
exchangers by altering the thickness of boundary layer. At present, pulsating flow has
a wide use in modern power-producing facilities, industrial equipment and chemical
technology. According to the experiment data published by Jin et al. (2007), the heat
transfer improvement in a triangular grooved tube was up to 350% with flow
pulsation and mainly derived from strong mixing effect caused by vortex generated by
pulsation flow. Moschandreou and Zamir (1997) found the pulsation had a positive
effect on the fluid bulk temperature and Nu increment over a range of pulsation
frequency (f). But when the frequency increased or decreased to a value out of this

of
range, the effect would be reversed and the pulsation could restrain the heat transfer in

ro
turn. Wang and Zhang (2005) made a numerical calculation and concluded that for a
circular pipe there was an optimum Womersley number at which the heat transfer can
be maximally improved. On the other hand, however, there are publications asserting
-p
that the enhancement effect of flow pulsation was indistinctive for some heat
exchangers. Some researchers (Yu et al., 2004) found that pulsation flow didn’t have
re
impact on the time-average Nusselt number for heat transfer in a circular tube with
constant wall heat flux. Mehta and Khandekar (2015) studied laminar pulsating flow
lP

in a square mini-tube adopting the non-intrusive infra-rea thermography at three flow

frequencies (0.05, 1.0 and 3 Hz) and concluded that the enhancement was unattractive
na

with respect to laminar pulsating flow and mainly observed in the developing region
of the tube. Hence, whether pulsation exerts a positive influence on heat transfer
improvement is determined by heat exchanger structure, pulsation frequency,
ur

Reynolds number (Re) etc. The study was designed to examine the possibility of heat
transfer enhancement by the pulsating flow in a coiled tube.
Jo

Although pioneers have done some experimental or numerical work on pulsatile flow
in curved pipes (Timité et al., 2010; Vester et al., 2015; Smith, 1975). There are few
works concerning nanofluid heat transfer in coil tubes with pulsation, especially in
helical coiled tubes. Kharvani et al. (2016) used a rotating ball valve as a pulse
generator and studied the effect of the pulsation on the HTP of water in a spiral-coil.
They discovered that the overall thermal performance was enhanced up to 26% at
5
Re=6220-16300 and f=0-20Hz. Doshmanziari et al. (2016) investigated the HTP of
Al2O3-DW nanofluids in spiral-coils under pulsation and found the overall heat
transfer coefficient increased 23 % by using nanofluids heat transfer media coupled
with flow pulsation. Introducing pulsation flow in the nanofluids helical coils heat
transfer system could improve the stability of nanofluids and promote the thermal
performance of the heat exchangers. Thus, integrating these various heat transfer
enhancement techniques has a promising prospect in engineering application and the
HTP of nanofluids in a helical coil with pulsation flow are worth to be explored.

In the study, nanofluids heat transfer in a helical coil with pulsation has been

of
numerically simulated. In the established model, Al2O3 nanofluids with different
volume concentrations were used as the heat transfer media. The nanofluids heat

ro
transfer result was compared with deionized water and reported literature results.
Then, the impact of pulsation parameters (including f and Re etc.) on the flow and
-p
HTP were studied in detail. And the enhancing mechanisms of nanofluids heat
transfer in helical coils under pulsation were also discussed.
re
2 Numerical studies
lP

2.1. Physical model introduction

A horizontal helical coil was built based on the geometry displayed in figure 1. The
na

radius of the helical coil tube (r) is 3.5 mm. The coil radius (Rc) and coil pitch (H) is
35 and 42 mm respectively. And the spiral number (Z) is 6. In order to avoid the
ur

impact of laminar entrance region, an adiabatic straight tube with 210 mm long is
constructed before the helical coil (Kahani et al., 2013).
Jo

Al2O3-water nanofluids were utilized as the working fluid, taking the heat away from
helical coil wall. The volume concentrations were 0.5, 1.0 and 1.5%. Table 1 lists the
physical properties of water and Al2O3 nanoparticles.

2.2. Governing equations

6
Single phase and two phase models are usually utilized for the simulation of nanofluid
convective heat transfer. Among the studies, few papers have reported the simulation
considering particle agglomeration. Nanoparticle agglomeration models are generally
applied in nanofluid thermal conductivity, nanofluid vaporizing droplet and nanofluid
optical properties studies (Du and Tang, 2015; Rashidi et al., 2018). In the study,
single phase model was adopted due to its much higher computational efficiency and
better convergence property. The particle agglomeration or clogging studies will be
conducted in our follow-up study.

Several assumptions were made to conduct the simulation: (1) the nanofluid is

of
incompressible and remains to be a single phase flow; (2) the radiation heat transfer is
negligible; (3) the heat flux is kept constant along the coil; (4) the wall thickness is

ro
negligible and the properties of the coil material are temperature-independent. The
governing equations for nanofluids in unsteady condition are as follows:
nf 
-p
Continuity:  ·( nf V )  0 (1)
t

( nf V )  
re
Momentum:  ·( nf V V )  P  nf  2 V (2)
t
( nf T )  

Energy equation:  · (  Cp)nf V T      knf T  (3)
lP

t  

 nf and Cpnf are calculated by the following equations:

nf   p  (1   ) bf (4)

na

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

ur

In terms of modeling nanofluids thermal conductivity and viscosity, Brownian motion

effect was considered. In our previous study (Guo et al., 2018b), nanofluid HTP was
Jo

simulated using dynamic single phase model which was also utilized in the study.
k nf and  nf are expressed as:

(6)

For Al2O3 nanofluids, f (T ,  ) and  can be calculated by the following equations

(Vajjha and Das, 2009).

7
 T
f (T ,  )  (2.8217 102   3.917 103 )( )
T0 (7)
 (3.0669 102   3.91123 103 )

  8.4407(100 )1.07304 (8)

 pVB d p2
nf  bf  (9)
72C

Where  and VB denote the distance between nanoparticles and the Brownian

velocity respectively:


3 dp (10)
6

of
1 18k BT (11)
VB 
dp  p d p

ro
1
C [c1d p c2 )  (c3d p  c4 )] (12)
bf

-p
Where KB is Boltzmann constant. c1, c2 , c3 and c4 are constant and the values of
them are −0.000001133, −0.000002771, 0.00000009, −0.000000393 respectively
(Vajjha and Das, 2009). dp an p are nanoparticle size and density. They are set in
re
accordance with the properties of nanofluids used in the simulation.
lP

2.3. Data reduction

The instantaneous convective heat transfer coefficient ( h(t ) ), instantaneous Nusselt
na

number ( Nu(t ) ), time-average convection heat transfer coefficient in a period ( h ) and

time-average Nusselt number ( Nu ) are obtained as:
m  Cp  (Tout (t )  Tin (t ))
h(t ) 
ur

(13)
A  (Tw (t )  Tb (t ))

1 


h h(t )dt (14)
Jo

Nu(t )  2h(t )  r / k (15)

Nu  2h  r / k (16)

Where Tout (t ) , Tin (t ) , Tw (t ) and Tb (t ) denote fluid outlet temperature, fluid inlet

temperature, helical wall temperature and the average fluid temperature

8
( Tb  (Tout  Tin ) / 2 ).

Re, De, He, and Pr are important dimension parameters in helical coil heat transfer
and flow characteristic study. They are expressed as:

Re  2 V  r /  (17)

De  Re (r / Rc)1/2 (18)

H 2 1/2
He  De  [1  ( ) ] (19)
2 Rc
Pr    Cp / k (20)

of
In a coil, the transition Re for the fluid from laminar to turbulent regime can be
calculated by Recri  12730 / ( Rc / r )0.32 (Kubair and Varrier, 1961). In the present study,

ro
the critical Re was 6094. And Re was kept lower than 4000 in the study. And the main
stream maintained laminar flow. As pulsation was imposed, some vortexes might
-p
generate and the flow tends to be a locally turbulent state.

2.4 Boundary conditions

re
At the entrance, the fluid flows along the Y-axis with the temperature of 293 K. A
uniform velocity profile with a sinusoidal pulsation is input by a user defined function.
lP

The inlet velocity is expressed as:

V(t)=Vm [1+Am sin(t)]; U (t )  W (t )  0 ;
Where Vm, t, Am and  (=2f) denote the mean velocity, time, the non-dimensional
na

amplitude, and the angular frequency respectively.

For all solid walls, the no-slip boundary condition is defined. And at the outlet, the
ur

pressure is set to be the atmosphere pressure.

The working fluid flows through the helical coil and is heated up by its wall with
Jo

constant heat flux. The change in heat flux will affect fluid temperature distribution,
while its influence on Nu is marginal (Jayakumar et al. 2008). Based on the proper
flow temperature rise (1 ~ 10 C), the heat flux value was set at 8000 W/m2, which
was determined by the heat-exchange area and the flow mass rate.

3 Numerical method and model validation

9
The fluent CFD solver was adopted to solve the simulation. The governing equations
as listed above. All simulations were performed by segregated method using
SIMPLEC algorithm. The second-order upwind schemes were used to discretize the
governing equations. A non-uniform structured grid was utilized to mesh the helical
coil volume. Refined grids were adopted in the boundary layer.

For all mesh configurations, Nu and friction factor ( f () ) were computed and
compared. Six cases were conducted. Based on the results (table 2), little difference
(deviations were below 0.3%) is found between mesh configuration cases 5 (252
thousand) and 6 (286 thousand) and the smaller mesh configuration (252 thousand)

of
was selected to perform following calculations.

The simulation results of water heat transfer and flow characteristics were compared

ro
with empirical formula, and model reliability was verified. For helical coils under
constant heat flux condition, Nu proposed by Manlapaz and Churchill (1981) is:

Nu   4.364 

4.636 
x1 
3
 De  
  1.816 
3/2

 
 x2  
1/3 -p (21)
re
1342 2 1.15
Where x1  (1.0  2
) , x2  1.0 
De Pr Pr
lP

The pressure drop of straight pipe at laminar flow can be calculated according to the
equations below (White, 1991):

4 f s ( dL ) V 2
Ps  (22)
na

f s ()  16 / Re (23)
For fully developed flow of helical pipes (7<Rc/r<104), Srinivasan et al. (1970)
ur

proposed a friction factor calculation formula. It is expressed as:

1 for De  30 
Jo

f C ( )  
 0.419 De0.275 for 30  De  300 (24)
f S ( )  
0.1125De
0.5
for De  300 
Figure 2 depicts the comparison between theoretical and simulation results. As shown
in the figure, the simulation results and the theoretical one have some difference with
the deviation within 12%. It confirmed the reliability and accuracy of the simulation
model and results.

10
4 Results and discussion
4.1 Nanofluid HTP in a helical coil

Al2O3 nanoparticle is commonly used in nanofluid preparation and heat transfer test
due to its high thermal conductivity and easy accessibility. In the paper, the HTP of
Al2O3 nanofluids in helical coil was investigated and the results were presented in
figure 3(a). It was evident that the convection heat transfer coefficient of the three
nanofluids was higher than that of water. The increase in heat transfer coefficient
exceeded the increase in thermal conductivity, which might be ascribed to the

of
dynamic micro-scale heat and energy transfer caused by Brownian motion of
nanoparticles (Azizian et al., 2011).

ro
At different nanoparticle concentrations, h( nf ) / h(bf ) saw a similar trend with Re
-p
change. With addition of 0.5 vol. % Al2O3, the convection heat transfer coefficient
increased about 8% in the range of Re. With further increasing in Al2O3 volume
re
concentration, h( nf ) / h(bf ) further increased to 1.11 for 1.0 % Al2O3-water and 1.14
lP

for 1.5 % Al2O3-water. It is noted that nanofluid viscosity is usually higher than the
base fluid. And to obtain a certain Re, the nanofluid velocity is usually higher than the

base fluid. And h(nf ) / h(bf ) at a fix Re would be higher than that at a fix velocity. Since a
na

fixed-Re comparison base may be a more common choice, a fixed Re base was
adopted in the following study.
ur

To validate the model of nanofluids heat transfer, the simulation results were also
compared with reported literatures. The structure dimension of the coil was
Jo

constructed based on coil II in the paper reported by Kahani et al. (2013). As

illustrated in Figure 3(b), high matching of simulation and experimental results were
seen at both higher and lower Re, and the highest deviation (~16.8%) was found at
Re=1880. Not fully considering the influence of Pr in the literature might be
attributed to the deviation (Kahani et al., 2013). Overall, the deviation was in an
acceptable range, which indicated the reliability of the model for nanofluid heat
11
transfer simulation.

4. 2. Nanofluid heat transfer with pulsation

As previously mentioned, Al2O3 nanofluids with higher nanoparticle concentration

could enhance the convection heat transfer coefficient to a higher degree. In the
following discussion, 1.5% Al2O3-water was adopted. For the pulsation flow, the
Womersley parameter (nondimensional frequency,   ro (2 f /  )1/2 ) is used to
describe the ratio between oscillatory and viscous effect. When <1, a lager viscous

of
layer appears near the wall, which is not beneficial for heat transfer. In the study the
Womersley parameter was set in the range of 1 to 30.

ro
As demonstrated in figure 4(a), h( pul ) / h( nf ) firstly increased as pulsation frequency
increased and then decreased at Re=1327~3114. Within the scope of Re, there was an
-p
optimum pulsation frequency. The lower Re was, the lower optimum pulsation
frequency could be. For Re=1327, h( pul ) / h( nf ) got its maximum value 1.11 at the
re
optimum pulsation frequency of 4 Hz. When Re increased to 3114, the maximum
value increased to 1.18 at the optimum pulsation frequency of 9 Hz. Lee et al. (1999)
lP

and Jin et al. (2007) also found that there existed an optimal frequency for each Re. A
maximum heat transfer enhancement (1.3) in an axisymmetric wavy channel was
obtained at Re=100 and St=3 in the study of Lee et al. (1999). In the steady flow, the
na

heat transfer is promoted by self induced fluctuation as Re increases. While in the

pulsation flow, the heat transfer is enhanced by the repeating sequence of vortex
ur

generation, growth, and expansion (Jin et al. 2007) for the flow through the curved
tubes (such as triangular grooved channel, wavy channel). If the pulsation frequency
Jo

is too high, the vortex growth would not be completed and vertex expansion not
enough for an effective mixing. If the pulsation frequency is too low, the vertex
generation rate would also be too low. Thus, in order to obtain a sufficient fluid
mixing, there is an optimum pulsation frequency for the vortex growth and vortex
expansion to the mainstream.

An empirical correlation was established to predict the overall performance of the

12
coiled tube under different frequencies. It was expressed as:

Nu( pul ) / Nu( nf )  0.317 0.133 Re0.0564 Pr 0.664 (f=1~12 Hz, Re=1889~3127, Standard
Deviation =0.024)
Where  denotes pulsation frequency number (   ro (2 f /  )1/2 ).

As the pulsation frequency increased, P( pul ) / P( nf ) increased at first and then
decreased gradually after reaching the peak value (figure 4(b)). And the higher Re was,
the higher P( pul ) / P( nf ) could be. At the optimum pulsation frequency, it is evident
from the figure 4(b) that ( h( pul ) / h( nf ) )/ ( P( pul ) / P( nf ) ) was always higher than 1,

of
indicating higher heat transfer efficiency could be obtained by using pulsation
technique.

ro
In order to get clear about the influence of pulsation frequency on h , the changes of
instantaneous convective heat transfer coefficient (h(t)) with pulsation frequency at
-p
Re=1327 was further studied. As explicated in figure 5, the extent of variation of
h(t ) at lower frequency is higher than that of higher frequency. At the lower
re
frequency (f=1), h(t ) fluctuated in the range of 2054 and 3054 W/(m2K) with h
slightly higher than that in the steady flow ( h st=2443 W/(m2K)). At the higher
lP

frequency (foptimum=4), h(t ) fluctuated in the range of 2655 and 2771 W/(m2K),
which was always higher than that in the steady flow. Under this condition,
h( pul ) / h( st ) reached 1.11.
na

To explain the phenomenon, the change trends of some other parameters (including

V (t ) , Tw (t ) and Tb (t ) ) in a period were also studied. Under constant heat flux,

ur

h(t ) reached its maximum value at the minimum ( Tw (t ) - Tb (t ) ). As displayed in

Jo

figure 6(a), at the pulsation frequency of 1, V (t ) and Tw (t ) showed an obviously

opposite trend over time, with some phase lag. h(t ) got the maximum value at a

relatively high flow velocity. When pulsation frequency increased to 4 Hz, the phase

lag increased to /2. It was surprised to find that h(t ) got the maximum value when

V (t ) <Vm, while h(t ) got the minimum value when V (t ) >Vm. The heat transfer

13
enhancement mechanism under pulsation condition is discussed in the following
section.

Figure 7 displays the axial velocity contours at various planes along the coil at
Re=1327. At the inlet (0), all the velocity contours exhibit symmetrical ring-like
structure and the maximum velocity is at the circle center. As the fluid flowed past the
curved tube, a centrifugal force was generated in the presence of curvature in ducts.
By the centrifugal force, the fluid particles flowed outward. After 90, periodically,
the maximum velocity centers in the flow moved toward left and right at every 180
bend. It improved fluid particle movement in radial direction which promoted fluid

of
particles mixing and heat transfer enhancement (Tohidi et al., 2015). At 90 and 180,
there were some differences in the velocity contour of the flow at different conditions.

ro
While after 270, they tended to exhibited similar shapes.

-p
As presented in figure 7 (a)~(c), at f=1 Hz, h(t ) got the maximum value at a

relatively high flow velocity. It was obvious that the higher flow rate was beneficial
re
for heat transfer enhancement. However, h(t ) reached the maximum value at a
lP

relatively low flow rate at f=4 Hz. Axial velocity distributions for steady flows and
pulsatile flows at f=4 Hz and t=thmax were further studied in order to explain it. It was
consistent with figure 7. The fluid particles of maximum velocity moved to the left
na

and right periodically. As investigated in figure 8, for steady and pulsation flows, the
fluid particles reach the maximum velocity at the same axial position. While at the
ur

each side of the maximum velocity, the axial velocity decreasing rate was faster for
pulsation flows at all cross sections. And in the areas near the tube inner wall, the
Jo

axial velocity decreased to below zero, which indicated the existence of a

counter-rotating vortex. Kiml et al. (2004) employed angled ribs to induce secondary
flow. He found the presence of the vertex near the inner wall reduced the thickness of
thermal and velocity boundary layers and enhanced the heat transfer rate (Kiml et al.,
2004).

The velocity distribution at the cross section can also affect the heat transfer
14
performance. The velocity vectors at various planes along the coil are plotted in figure
9. In all conditions, the fluid particles had a rotation trend in the cross area, but the
rotating vortexes only appeared in the pulsation flows at f=4 Hz, t=thmax (figure 9 (c)).
In the curved tubes, the fluid particles moved toward the external wall due to
centrifugal force. And because of flow continuity, the fluid near the tube wall was
forced to flow inside, forming a rotation trend. In the deceleration phase of a certain
pulsation frequency, the rotating vortexes can be formed. As seen in figure 9(c), the
pulsation induced the secondary flow near the inner wall, which conveyed the hot
fluid from the heated wall to the cold water and greatly increased the heat transfer

of
performance (Song et al., 2008). The comparison of the temperature contours of
different cross sections (figures 9 (a) and (c)) found that higher temperature was seen

ro
at the inner tube. After pulsation was imposed, it was obvious to find that the high
temperature region decreased, the thermal boundary layers tended to thin and a
-p
sufficient mix of hot and cold liquid was formed, as discovered in figure 9 (c).
re
Secondary flow is key to heat transfer enhancement. And the absolute vorticity flux
n
( J ABS ) can be used to evaluate the intensity of secondary flow.
lP

n
J ABS  (   n dA) / A
A

n
na

Where J ABS is the surface average of the absolute value of vorticity in the cross

n
section (Song and Wang, 2008). The value of J ABS can be calculated by integrating
ur

 n with dA and dividing the result by area (A). It was obtained with the help of

n
CFD post software. Figure 10 illustrates the distribution of J ABS at various planes
Jo

n
along the coil. As working fluid flows along the channel, the value of J ABS showed a

n
decrease trend. As predicted, the value of J ABS under the pulsation flow (f=4 Hz,

t=thmax) at the cross section was higher than that under steady flow (Re=1327). In
summary, the secondary flow generated in the cross section and the counter-rotating
vortex formed in axial direction all devoted to the helical coils heat transfer
15
augmentation under pulsation.

Conclusions

A numerical study on nanofluid HTP in a helical coil under pulsation was performed
in the paper. The effect of nanofluid concentration, pulsation frequency and Re on the
HTP was studied in detail. The following conclusions can be made:

(1) High nanoparticle concentration was beneficial for the increase of h( nf ) / h(bf ) . For

1.5 % Al2O3-water, h( nf ) / h(bf ) reached 1.14 in the range of Re (1327~3114).

of
(2) Within the scope of Re, there was an optimum pulsation frequency for heat
transfer augmentation. The lower Re was, the lower optimum pulsation frequency

ro
could be. For Re=3114, h( pul ) / h( nf ) and h( pul ) / h(bf ) got its maximum value of

1.18 and 1.35 at the optimum pulsation frequency of 9 Hz respectively.

n
-p
(3) The secondary flow intensity ( J ABS ) of the pulsation flow (Re=1327, f=4 Hz,
re
t=thmax) at the cross section was higher than that of steady flow.
(4) The secondary flow generated in the cross section and the counter-rotating vortex
lP

formed in the axial direction both devoted to the helical coils heat transfer
augmentation under pulsation.
na

Declaration of interests

☒ The authors declare that they have no known competing financial interests or personal
ur

relationships that could have appeared to influence the work reported in this paper.
Jo

Acknowledgments
This work was supported by the National Natural Science Foundation of China (No.
51906220, 51606169), Natural Science Foundation of Zhejiang province
(LQ19E060002); the Fundamental Research Funds for the Provincial Universities
(2019QN06).

16
References
Azizian R, Doroodchi E, Moghtaderi B. Effect of nanoconvection caused by Brownian motion on

the enhancement of thermal conductivity in nanofluids. Industrial & Engineering Chemistry

Research, 2011, 51(4): 1782-1789.

Akhavan-Behabadi M A, Pakdaman M F, Ghazvini M. Experimental investigation on the

convective heat transfer of nanofluid flow inside vertical helically coiled tubes under uniform

wall temperature condition. International Communications in Heat & Mass Transfer, 2012,

39(4): 556-564.

of
Bondareva N S, Sheremet M A, Oztop H F, et al. Free convection in an open triangular cavity

filled with a nanofluid under the effects of Brownian diffusion, thermophoresis and local

ro
heater. Journal of Heat Transfer, 2018, 140(4): 042502.

-p
Dravid A N, Smith K A, Merrill E W, et al. Effect of secondary fluid motion on laminar flow heat

transfer in helically coiled tubes. AIChE Journal, 1971, 17(5): 1114-1122.

re
Du M, Tang G H. Optical property of nanofluids with particle agglomeration. solar energy, 2015,

122: 864-872.
lP

Doshmanziari F I, Zohir A E, Kharvani H R, et al. Characteristics of heat transfer and flow of

Al2O3/water nanofluid in a spiral-coil tube for turbulent pulsating flow. Heat & Mass Transfer,

2016, 52(7): 1-16.

na

Ghobadi M, Muzychka Y S. Effect of entrance region and curvature on heat transfer in mini scale

curved tubing at constant wall temperature. International Journal of Heat & Mass Transfer,
ur

2013, 65(5): 357-365.

Guo W, Li G, Zheng Y, et al. Laminar convection heat transfer and flow performance of Al2O3–
Jo

water nanofluids in a multichannel-flat aluminum tube. Chemical Engineering Research and

Design, 2018a, 133: 255-263.

Guo W, Li G, Zheng Y, et al. Numerical study of nanofluids thermal and hydraulic characteristics

considering Brownian motion effect in micro fin heat sink. Journal of Molecular Liquids,

2018b, 264: 38-47.

IAPWS, Revised Release on the IAPS Formulation 1985 for the Thermal Conductivity of

17
 Ordinary Water Substance, International Association for the Properties of Water and Steam,

London, 1998, 23.

Jin D X, Lee Y P, Lee D Y. Effects of the pulsating flow agitation on the heat transfer in a

triangular grooved channel. International Journal of Heat & Mass Transfer, 2007, 50(15):

3062-3071.

Jayakumar J S , Mahajani S M , Mandal J C , et al. Experimental CFD estimation of heat transfer

in helically coiled heat exchanger. Chemical Engineering Research and Design, 2008, 86(3):

221-232.

Kubair V, Varrier C B S, Pressure drop for liquid flow in helical coils, Transactions of the Indian

of
Institute of Chemical Engineers , 1961, 14: 93-97.

Kiml R, Magda A, Mochizuki S, et al. Rib-induced secondary flow effects on local circumferential

ro
heat transfer distribution inside a circular rib-roughened tube. International Journal of Heat &

Mass Transfer, 2004, 47(6): 1403-1412.

-p
Kannadasan N, Ramanathan K, Suresh S. Comparison of heat transfer and pressure drop in

horizontal and vertical helically coiled heat exchanger with CuO/water based nanofluids.
re
Experimental Thermal & Fluid Science, 2012, 42(5): 64-70.

Kahani M, Heris S Z, Mousavi S M. Comparative study between metal oxide nanopowders on

lP

thermal characteristics of nanofluid flow through helical coils. Powder Technology, 2013,

246(9): 82-92.
na

Kharvani H R, Doshmanziari F I, Zohir A E, et al. An experimental investigation of heat transfer

in a spiral-coil tube with pulsating turbulent water flow. Heat & Mass Transfer, 2016, 52(9):

1779-1789.
ur

Khoshvaght-Aliabadi M, Khaligh S F, Tavassoli Z. An investigation of heat transfer in heat

exchange devices with spirally-coiled twisted-ducts using nanofluid. Applied Thermal

Jo

Engineering, 2018, 143: 358-375.

Lee B S , Kang I S , Lim H C . Chaotic mixing and mass transfer enhancement by pulsatile

laminar flow in an axisymmetric wavy channel. International Journal of Heat and Mass

Transfer, 1999, 42(14):2571-2581.

Manlapaz R L, Churchill S W. Fully developed laminar convection from a helical coil. Chemical

Engineering Communications, 1981, 9(1-6): 185-200.

18
Moschandreou T, Zamir M. Heat transfer in a tube with pulsating flow and constant heat flux.

International Journal of Heat & Mass Transfer, 1997, 40(10): 2461-2466.

Mondal R N, Huda M A, Ahmmed S F, et al. Effects of secondary flow on convective heat transfer

through a curved rectangular duct. Journal of Applied Sciences Research, 2007, 3(12):

1842-1850.

Mohammed H A and Narrein K. Thermal and Hydraulic Characteristics of Nanofluid Flow in a

Helically Coiled Tube Heat Exchanger, International Communications in Heat and Mass

Transfer,2012, 39: 1375-1383.

Mehta B, Khandekar S. Local experimental heat transfer of single-phase pulsating laminar flow in

of
a square mini-channel. International Journal of Thermal Sciences, 2015, 91: 157-166.

Mahian O, Kolsi L, Amani M, et al. Recent advances in modeling and simulation of nanofluid

ro
flows-part II: applications. Physics reports, 2018, 791: 1-59.

Naphon P, Wongwises S. A review of flow and heat transfer characteristics in curved tubes.
-p
Renewable & Sustainable Energy Reviews, 2006, 10(5):463-490.

Narrein K and Mohammed H A. Influence of Nanofluids and Rotation on Helically Coiled Tube
re
Heat Exchanger Performance, Thermochimica Acta, 2013, 564: 13-23.

Narrein K and Mohammed H A. Heat Transfer and Fluid Flow Characteristics in Helically Coiled
lP

Tube Heat Exchanger (HCTHE) Using Nanofluids: A review, Journal of Computational and

Theoretical Nanoscience, 2014, 11:970-984.

na

Pan C , Zhou Y , Wang J . CFD study of heat transfer for oscillating flow in helically coiled tube

heat-exchanger. Computers & Chemical Engineering, 2014, 69:59-65.

Rashidi S, Mahian O, Languri E M. Applications of nanofluids in condensing and evaporating systems.

ur

Journal of Thermal Analysis and Calorimetry, 2018, 131(3): 2027-2039.

Srinivasan P S, Nandapurkar S S, Holland F A. Friction factors for coils. Transactions of the

Jo

Institution of Chemical Engineers and the Chemical Engineer, 1970, 48(4-6): T156-T161.

Smith F T. Pulsatile flow in curved pipes. Journal of Fluid Mechanics, 1975, 71(1): 15-42.

Sudo K, Sumida M, Yamane R. Secondary motion of fully developed oscillatory flow in a curved

pipe. Journal of Fluid Mechanics, 2006, 237(237): 189-208.

Song K W, Wang L B. Relationship between heat transfer intensity and absolute vorticity flux

intensity in flat tube bank fin channels with Vortex Generators. Progress in Computational
19
 Fluid Dynamics An International Journal, 2008, 8(7): 496-502(7).

Soto-Meca A , Serna J , Velasco F J S . Heat and mass transfer enhancement in a double diffusive

mixed convection lid cavity under pulsating flow. Computers & Chemical Engineering, 2016,

94:128-140.

Sheremet M A, Cimpean D S, Pop I. Free convection in a partially heated wavy porous cavity

filled with a nanofluid under the effects of Brownian diffusion and thermophoresis. Applied

Thermal Engineering, 2017, 113: 413-418.

Timité B, Castelain C, Peerhossaini H. Pulsatile viscous flow in a curved pipe: Effects of pulsation

on the development of secondary flow. International Journal of Heat & Fluid Flow, 2010,

of
31(5): 879-896.

Tohidi A, Ghaffari H, Nasibi H, et al. Heat transfer enhancement by combination of chaotic

ro
advection and nanofluids flow in helically coiled tube. Applied Thermal Engineering, 2015,

86: 91-105.
-p
Vajjha R S, Das D K. Experimental determination of thermal conductivity of three nanofluids and

development of new correlations. International Journal of Heat and Mass Transfer, 2009,
re
52(21-22): 4675-4682.

Vester A K, Örlü R, Alfredsson P H. Pulsatile Turbulent Flow in Straight and Curved Pipes –
lP

Interpretation and Decomposition of Hot-Wire Signals. Flow Turbulence & Combustion,

2015, 94(2): 305-321.

na

White F M, Viscous fluid flow, McGraw Hill, New York, 1991.

Wang X, Zhang N. Numerical analysis of heat transfer in pulsating turbulent flow in a pipe.

International Journal of Heat & Mass Transfer, 2005, 48(19): 3957-3970.

ur

Yu J C, Li Z X, Zhao T S. An analytical study of pulsating laminar heat convection in a circular

tube with constant heat flux. International Journal of Heat & Mass Transfer, 2004, 47(24):
Jo

5297-5301.

20
 Z=6 Z=5 Z=4 Z=3 Z=2 Z=1
H z

2Rc
x
Inlet 0

Outlet
y
2r 0°
Wall (constant heat flux)

of
ro
Figure 1 Schematic diagram of the computational domain of the helical coil

-p
re
lP
na
ur
Jo

21
 (a)
45
40
35 12%

30
Nu

25
20
15 Manlapaz and Churchill Eq. [28]
10 Present study

5
0
250 500 750 1000 1250
De

of
0.055
(b)

0.050

0.045

ro
0.040
fc (-)

0.035
Srinivasan et al. Eq. [29]
0.030 Present study
0.025

0.020

0.015
12%
-p
re
0.010
200 400 600 800 1000 1200
De

Figure 2 Comparison of (a) Nu and (b) f c (-) simulation results for water with
lP

theoretical equations
na
ur
Jo

22
 1.24
1.5% Al2O3-water (a)

1.20 1.0% Al2O3-water

0.5% Al2O3-water
1.16
h(nf)/h(w)

1.12

1.08

1.04

1.00
500 1000 1500 2000 2500 3000 3500 4000
Re

5000
(b)

of
4000

3000

ro
h W/(m K)
2

0.5%Al2O3-water
2000

1000

0
0 1000
Results reported by Kahani et al. []
Simulation results

2000 3000 4000

-p
re
Re

Figure 3 (a) Nanofluids heat transfer in a helical tube (b) Comparison of h simulation
results with experimental results for 0.5%Al2O3-water
lP
na
ur
Jo

23
 1.20
(a)

1.15

1.10
h(pul)\(h)(nf)

1.05 Re=1327
Re=1889
1.00 Re=2600
Re=3114
0.95
0 2 4 6 8 10 12
f(Hz)

of
1.06
(b)

Re=1327
Re=1889

ro
1.05
Re=2600
Re=3114
P(pul)/P(nf)

1.04

1.03
-p
re
1.02
0 2 4 6 8 10 12
lP

f (Hz)
Figure 4 Nanofluids (a) heat transfer and (b) flow performance under pulsation
(Am=0.5)
na
ur
Jo

24
 Time s (f=4Hz)
3200 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8

3000

2800
h W/(m /K)

2600
2

2400

2200

2000

1800
0.0 0.5 1.0 1.5 2.0 2.5 3.0

of
Time s (f=1Hz)

Figure 5 Instantaneous h(t) at Re=1327 f  1 and f  4

ro
-p
re
lP
na
ur
Jo

25
 3200
hmax (a)
h(t) W/(m K) 0.3
2

velcocity m/s
2800
V(t) 0.2
2400
0.1
h(t)
2000 hmin
301 0.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
300 TW(t)

299
T (K)

296.4 Tb(t)

296.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
t (s)

of
2800
hmax (b)
0.3

velcocity m/s
2760

ro
h(t) W/(m K)

V(t)
2

0.2
2720

2680 0.1

2640
h(t)
0.0
299.5 T (t)
W
0.2
hmin
0.4 0.6
-p
0.8 1.0
0.0
re
T (K)

299.0
296.5 T (t)
b
lP

296.0
0.0 0.2 0.4 0.6 0.8 1.0
t (s)

Figure 6 Instantaneous parameters at Re=1327 (a) f  1 and (b) f  4

na
ur
Jo

26
 0 90 180 270 360 450
(a)

(b)

(c)

of
(d)

ro
(e)
-p
re
Figure 7 Axial velocity contours at various cross-sections at Re=1327 (a) steady; (b)
f  1Hz , t=thmax; (c) f  1Hz , t=thmin; (d) f  4Hz , t=thmax; (e) f  4Hz , t=thmin.
lP
na
ur
Jo

27
 1.2

0 90
1.0

0.8
Va/Vmax

0.6
180
0.4 270

0.2

of
0.0

ro
-0.2
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
r/R

-p
Figure 8 Axial velocity distributions for steady flows (solid lines) and pulsatile flows
at f=4 Hz, t=thmax (corresponding dot lines) (Re=1327).
re
lP
na
ur
Jo

28
 Z=1 Z=2 Z=3 Z=4 Z=5
(a)

(b)

(c)

of
ro
Figure 9 Temperature contours and velocity vectors at various planes (90+360Z)°
along the coil at Re=1327 (a) steady; (b) f=4 Hz, t=thmin; (c) f=4 Hz, t=thmax

-p
re
lP
na
ur
Jo

29
 25

24

23
-1
JABS s

22
n

Steady flow
21 Pulsation flow (f=4 Hz, t=thmax)

20

of
1 2 3 4 5
Z
n
Figure 10 Distribution of J ABS at various cross-sections (90+360Z)° along the coil

ro
(Re=1327)

-p
re
lP
na
ur
Jo

30
Table 1 Physical properties of base fluid and nanoparticles
Particle Density Thermal conductivity Heat capacity
kg/m3 W/(mK) J/(KgK)
Al2O3 3890 36 880
Distilled =1.5221810-5 T31.8287710-2 T2+6.59684 T+2.5237310-2
water  = -4.7097210-9 T35.0329210-6 T2 + 5.9238210-3 T5.8653610-1
(IAPWS, = 3.1629010-11 T4  4.3346410-8 T3 + 2.2319110-5 T2  5.1233110-3
1998) T + 4.4328110-1
Cp = -3.3314310-8 T5 + 5.6596310-5 T4  3.8410210-2 T3 + 1.3025410

of
T2  2.20797103 T + 1.53885105

ro
Table 2 Grid independence test for water
Mesh configuration / Nu f (-)

No. thousand
-p
Case 1 114 (Very coarse) 26.445 0.03143
re
Case 2 123 (Coarse) 26.360 0.03150

Case 3 156 (Intermediate) 25.228 0.03125

lP

Case 4 175 (Intermediate) 25.182 0.03103

Case 5 252 (fine) 24.676 0.03116

na

Case 6 286 (fine) 24.673 0.03116

ur
Jo

31