International Journal of Heat and Mass Transfer 107 (2017) 712–722

Contents lists available at ScienceDirect

International Journal of Heat and Mass Transfer

journal homepage: www.elsevier.com/locate/ijhmt

Experimental study on the heat transfer and flow characteristics

of nanofluids in the built-in twisted belt external thread tubes
Bin Sun ⇑, Amin Yang, Di Yang
School of Energy and Power Engineering, Northeast Dianli University, Jilin City 132012, China

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

Article history: In order to study the heat transfer and flow characteristics of fluid flowing through the built-in twisted
Received 19 July 2016 belt external thread tubes, in the Reynolds number range of 2000–12,000, the heat transfer and flow
Received in revised form 21 November 2016 characteristics were studied respectively for the Cu, Al, Al2O3, Fe2O3, multiwalled carbon, and Graphite
Accepted 26 November 2016
nanofluids with the mass fraction of 0.1–0.6%. The results show that the Cu nanofluids has the best con-
Available online 3 December 2016
vective heat transfer effect among numerous nanofluids, and the heat transfer characteristics reach the
optimum when the mass fraction of Cu nanofluids is 0.5%. The heat transfer performance of built-in
Keywords:
twisted belt external thread tubes increased 50.32% compared to the horizontal tube, indicating that
Nanofluids
Twisted-tape inserts in outside thread Tube
built-in twisted belt external thread tubes has the ability to strengthen heat transfer. With the decrease
Heat transfer characteristics of the pitch, the heat transfer coefficient and flow resistance were both increased. According to the exper-
Resistance characteristics imental data, the comprehensive analysis of hot performance coefficient was carried out and the corre-
lation of convective heat transfer and flow resistance in the built-in twisted belt external thread tubes
was obtained.
Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction in the review articles [7–10]. Most of them observed an improve-

ment in heat transfer with nanofluids even though a few reported
Since 1990s, the researchers began to explore the application of negative result. Only a few researchers have considered nanofluids
nano materials technology in the field of heat transfer enhance- as working fluid in loop heat pipes. Duangthongsuk [11] study the
ment, and a new generation of efficient heat transfer and cooling effect of heat transfer and pressure drop of TiO2-water nanofluids
technology was studied. In 1995, the Argonne National Laboratory in double countercurrent heat transfer, the results show that the
of the United States, Choi et al. [1] put forward a new concept – convective heat transfer coefficient of the nanofluids is improved
nanofluids for the first time. 6–11% compared with the base fluid, and the convective heat
In recent years, scholars at home and abroad have conducted a transfer coefficient increases with the increase of mass flow rate
lot of research on nanofluids technology. Thermal physical proper- and the decrease of the temperature of the nanofluids. Xuan
ties of nanoparticles, Xuan [2] discuss the relevant laws of et al. [12] studied the flow and convective heat transfer character-
nanoparticle properties, particle size, particle fraction, fluid sus- istics of Cu-water nanofluids in conventional and small channel flat
pension stability and fluid temperature impact on the thermal con- tubes. It is found that the convective heat transfer coefficient can
ductivity of nanofluids. The results show that at the rated Reynolds be increased by adding nanoparticles to the liquid and the flow
number, the nanoparticles suspended in the fluid can effectively resistance is not increased significantly, which shows that adding
enhance the heat transfer process, the nanofluids has a higher heat a small amount of nanoparticles in water will not cause additional
transfer coefficient than the original base fluid. Noreen Sher Akbar resistance loss. Some scholars [13–16] examined the Magnetic
et al. [3–6] intended for investigating the effects of heat flux and fluid dynamics flow of non-Newtonian nanofluid in a pipe by
induced magnetic field for the peristaltic flow of different mathematical models. The results revealed that Nusselt number
nanoparticles. increases with an increase of nanoparticle volume fraction, Ray-
Many researchers have used nanofluids in different types of leigh numbers and inclination angle. Darzi [17] study heat transfer
heat pipes to get better heat transfer performance as mentioned and flow characteristics of Al2O3-water nanofluids in the casing
tube heat exchanger, the results show that the nanofluids have
⇑ Corresponding author. great application prospects in the field of heat transfer enhance-
E-mail address: sunbin@nedu.edu.cn (B. Sun). ment, and in the experiment, the friction resistance is little

http://dx.doi.org/10.1016/j.ijheatmasstransfer.2016.11.084
0017-9310/Ó 2016 Elsevier Ltd. All rights reserved.
 B. Sun et al. / International Journal of Heat and Mass Transfer 107 (2017) 712–722 713

Nomenclature

A effective heat transfer area, m2 m kinematic viscosity, m2s

C specific heat capacity, J kg
1°/C
1 q density, kgm
3
D diameter of the inner wall, m u nanoparticles volume fraction, %
f flow friction factor x nanoparticles mass fraction, %
h convective heat transfer coefficient, Wm
2K
1
l length, m Greek symbols
Pr Prandtl number bf base fluid
Pe Berkeley number Exp experiment
DP pressure drop, Pa in inlet
Q average heat flow, W m weight
Q average heat flux, Wm
2 nf nanofluids
q flow, kgs
1 out outlet
Re Reynolds p nanoparticles
DT temperature difference, °C Reg regression equation
T temperature, °C v volume
u flow rate, m s
1 w tube wall
k thermal conductivity, Wm
1K
1 water water
l dynamic viscosity, Pas

changed compared with the base fluid. Utomo [18] studied the exchanger, twisted belt with twisted ratio Y is 4.149, 4.95, 5.882,
convective heat transfer characteristics of nanofluids in a horizon- 8.54, 15.649, the tube is used in the heat exchanger, which has
tal tube under the condition of laminar flow, the results show that obvious effect on the industrial application, it is found that the
the addition of nanoparticles has a great influence on the thermal enhancement amplitude is inversely proportional to the Reynolds
properties of the fluid, so the heat transfer efficiency of the fluid is number Re, when the twisted belt with twisted ratio Y is 4.149,
further enhanced. The development of nanotechnology has become the enhanced amplitude is relative to the ordinary heat exchanger
a widely used, interdisciplinary, and development potential of the enhanced 44.7%. Kumar [25]applied the helical twisted belt
technology and has also become one of the main research direc- methodology to solar water heaters, and determined the twisted
tions to solve the energy problem. ratio range to be 3–12. It turns out: solar water heaters inserted
Smithberg and Landis [19] studied the theory of twisted belt for with a twisted belt can improve the heat transfer effect about
the first time. A twisted belt was inserted into a tube to enhance 18–70%, but also increase the resistance 87–132%.
heat transfer and the manufacturing process was simplified at a The first paper on the study of heat transfer enhancement of the
low cost. Tubes with a twisted belt inside make fluid rotate. The external screw thread is proposed by the American scholar Lawson
swirling fluid enhances fluid flow velocity and the flow path near [26] in 1966. In recent years, many scholars at home and abroad
the wall, which strengthens the fluid turbulence of boundary layers have carried out a lot of research on the methods of heat transfer
as well as the mixture of boundary layer flow and main flow, enhancement, such as rough surface and extended surface, these
thereby increasing the heat transfer effect. Many scholars have car- enhanced heat transfer methods have improved the heat transfer
ried out a more detailed and comprehensive study on the twisted effect to different extent, when the fluid flows in the threaded tube,
belt. The diameter of tube, the twisted belt twist rate, fluid Prandtl the inner surface of the spiral shape is raised to causes the laminar
number effects of different aspects of twisted belt tube perfor- flow in the region of the transition zone is formed in a vertical vor-
mance to do a detailed overview. tex, the vortex causes periodic disturbance of the raised part of the
Maddah [20] studied the heat transfer characteristics of the Laminar flow medium, cause fluid particle in boundary layer dis-
nano fluids in a twisted tape tube, the experiment was carried turbance into disorganized turbulent state, when fluid flows
out in a twisted tape tube with different concentration of nanoflu- through the raised parts and produce new vortex and the vortex
ids, the results show that the nanofluids can effectively enhance flow direction is always perpendicular to the flow direction of
the heat transfer effect, twist ratio have a significant effect on medium laminar, which makes the flow state of the transition zone
the change of heat exchange and frictional resistance in the can change, thermal resistance becomes small, the heat transfer
twisted. Bhadouriya [21] experimental studied inner tube for rate increase [27]. At present, there is a certain market in the
reversing square tube, the outer tube is an annular casing device thread tubes heat exchanger or heat exchange element, it has been
of the circular tube, characteristics of heat transfer and friction widely used in the petroleum industry of our country and used in
resistance of air flow, the experimental results indicate that the heat exchangers, and some are used in the boiler flue tube. The test
ring parameter is small, the friction coefficient and the Nusselt shows that when the thermal resistance of the tube external fluid
number is greater. Liu [22] has a Numerical simulation study on is controlled by the thermal resistance, the appearance of the
the formation and development of spiral vortex in short twisted threaded tube is about 2.5 times larger than the same horizontal
tape, it is found that the radial velocity is the main factor for the tube heat exchanger, to make up for the deficiency of heat transfer
formation of eddy current, the size of the twisted ratio directly coefficient of the tube outer membrane, improve the heat transfer
influences the formation of the spiral vortex. Azmi [23] numerical effect. Reduce the size and weight of the heat exchanger. In addi-
simulation studied convective heat transfer characteristics of SiO2 tion, the threaded tube heat exchanger has a Strong antifouling
nanofluids in a twisted tape tube, and the analysis is carried out, ability and corrosion resistance ability, has a Long use cycle.
compared with water in the built in twisted tape, the heat transfer The thermal efficiency of a high-heat-exchange tube is known
coefficient of the SiO2 nanofluids with a volume fraction 3% is to be significantly higher than that of a common copper tube
increased 94.1%, friction coefficient is increased 160%. Sivashar- [28–32]. Thus, the heat transfer characteristics of nanofluids in
mugam et al. [24], the spiral twisted belt is applied to the heat an the built-in twisted belt external thread tubes were studied in
714 B. Sun et al. / International Journal of Heat and Mass Transfer 107 (2017) 712–722

this paper and the best comprehensive effects of the ratio and per-
formance are achieved. So this experiment chose the pitch of
2.5 mm 3 mm 2 mm external thread for heat transfer experiments.
In order to further improve the heat transfer effect, in the tube is
embedded with a twisted ratio of 2.5 built. The Reynolds number
is selected in the range of 12,000–2000, the convective heat trans-
fer characteristics of Cu-water nanofluids with different mass frac-
tion in three different pitch of the built-in twisted belt external
thread tubes were compared and studied, the results can provide
reference for future research on heat transfer characteristics of
nanofluids surface.

2. Experimental system, fluid preparation, and protocol

Fig. 2. The built-in twisted belt external thread tubes.
2.1. Experimental system

In this paper, the experimental device is a double tube, as

shown in Figs. 1 and 2. The length of the horizontal copper tube of Pt100 thermal resistance and thermal formula. Heating capacity
and built-in twisted belt external thread tubes are 1000 mm; the was derived from the added energy value of nanofluids in the inner
inner diameter and external diameter for built-in twisted belt tube, and was verified through the energy loss value of hot water in
external thread tubes is 12 mm and 16 mm (the thickness is the outer tube. Outer tube diameter is 40 mm; wall thickness is
1 mm, the pitch is 2 mm, 2.5 mm, and 3 mm, the reverse ratio is 2 mm and tube length is 1100 mm. Importer and exporter of the
2.5). The entrance steady segment length is 35 cm, and the length outer tube are arranged in 2 Pt100 thermal resistance to measure
of exit steady flow section is 35 cm. Pt100 temperature meters the incoming and outgoing water temperature. The hot water tank
were installed for continuous logging of the inlet and outlet fluid requires a rod heated to 2 kW to heat water, and in order to reach
temperature, and to measure the inlet and outlet temperature of the temperature requirement of 15–45 °C, a temperature control
the fluid. A Rosemount 3051S capacitive differential pressure device is installed. Meanwhile, the preheated section also requires
transmitter was also used to measure the pressure difference. a temperature control device to make the temperature of the drag
The outer tube uniformly supported four Pt100 thermal resistance nanofluids reach the set temperature in the experiment. Finally,
sensors to measure the wall temperature. Finally, in order to use Advantech USB-4716 data acquisition to collect all the mea-
achieve the insulation effect, the entire device was wrapped in sured data.
glass wool and aluminum foil tape. In order to maintain physical A hot water tank provides a stable heat source in the experi-
stability, flow resistance was measured when no heat transfer ment, and the circulated cool water in the cooling water tank
was applied. During the forced convection heat transfer experi- serves to cool the nanofluids and guarantee that the inlet temper-
ments, the qualitative temperature of the fluid’s inlet and outlet ature is kept constant. The related equipment parameters of exper-
arithmetic mean temperature was measured. The average temper- imental device system are shown in Table 1. Concrete experiment
ature of the outer wall was calculated from average measurement steps can be found in Ref. [33] (see Fig. 3).

Fig. 1. Experimental device system.

 B. Sun et al. / International Journal of Heat and Mass Transfer 107 (2017) 712–722 715

Table 1
Experimental equipments.

Experimental equipments Model Parameter

Flowmeter LWGYA-6 0.1–0.6 m3/h
Differential manometer Rosemount3051S 0–6.5 kPa
Temperature sensor PT100 0–100 °C
Data acquisition unit USB-4716
Hot resistance Pt100 0–150 °C
Magnetic force driving Pump 125 W 30 m
Heating water tank 2 kW

Fig. 4. TEM image of Cu nanofluids.

Table 3
Thermal parameters of Cu-water nanofluids.

T x With dispersant Without dispersant

lnf knf lnf knf

Fig. 3. Installation drawing of experiment test section. 25 0.1 90.74 0.6002 89.39 0.6009
0.2 92.01 0.6004 89.42 0.6012
2.2. Preparation of nanofluids 0.3 93.82 0.6006 89.45 0.6013
0.4 95.58 0.6008 89.67 0.6016
0.5 98.21 0.6011 89.84 0.6019
Deionized water is used as base fluid, while adding sodium
30 0.1 81.69 0.6182 80.02 0.6191
dodecyl benzene sulfonate (SDBS) as a dispersion. The specifica- 0.2 82.06 0.6184 80.04 0.6192
tions for the purchased nanoparticles are shown in Table 2. The 0.3 84.30 0.6186 80.07 0.6193
preparation of nanofluids is ‘‘two step method” [34–36] and prepa- 0.4 85.54 0.6188 80.09 0.6195
ration methods are as follows: 0.5 88.22 0.6191 80.11 0.6198
35 0.1 73.87 0.6272 72.27 0.6279
(1) According to a preferred molar ratio of 1:2, add an amount of 0.2 74.08 0.6274 72.29 0.6281
0.3 76.40 0.6276 72.31 0.6283
nanofluids and SDBS in the deionized water, stir 30 min in
0.4 77.82 0.6278 72.33 0.6286
the electromagnetic stirrer. 0.5 80.42 0.6281 72.35 0.6289
(2) Configure nanofluids and shock for 60 min in the ultrasonic
vibration analyzer to fully disperse particles to form a stable Unit: T(°C), x (%), lnf (10
5 Pas), knf (W/mK).
nanofluid.

Table 3, the addition of dispersion has effect on the thermal con-

2.3. Stability analysis of nanofluids
ductivity, while the viscosity of nanofluids with dispersion
increases slightly, and the differences become bigger as the disper-
Transmission electron microscopy (TEM) was used to observe
sion concentration increases.
the dispersion and stability of nanofluids, and the results are
shown in Fig. 4.
To confirm that the addition of the dispersant has a great influ- 2.4. Evaluation of uncertainty
ence on the thermal properties of the nanofluids, especially on the
viscosity, the SNB-3 type rotary viscosity meter and DRE-type III In order to obtain more accurate results, the technician must
thermal conductivity coefficient measuring instrument were used first evaluate system uncertainty [37]. In this research’s experi-
to measure the viscosity lnf and thermal conductivity knf of the ment, thermal conductivity and dynamic viscosity coefficient were
Cu-water nanofluids with or without dispersion at different measured by the DRE-III-type thermal conductivity detector and
temperatures, and the results are listed in Table 3. As shown in SNB-3 digital rotary viscometer, so the relative error is 3% and

Table 2
Parameters of various particles.

Parameter Cu Al A12O3 Fe2O3 Graphite Mutiwalled carbon

Average grain diameter/nm 40 40 20 40 40 10
Purity/% 99.9% 99.9% 99.99% 99.9% 99.9% 98%
Color Brown Black White Bronzing Black Black
Shape Spherical Spherical Spherical Spherical Spherical Cylindrical
Density/gcm
3 8.9 2.7 3.9 5.24 2.18 1.3
Specific heat/Jg
1K
1 0.39 0.88 1.3 0.72 0.71 0.51
716 B. Sun et al. / International Journal of Heat and Mass Transfer 107 (2017) 712–722

Table 4 In order to create more accurate experimental data, when calcu-

Laboratory instruments and uncertainty. lating wall temperature and fluid temperature, multiple measure-
Instruments name Uncertainty (%) ments must be made to get the average value. Due to the high
Thermal resistance 0.48 thermal conductivity of copper and the wall thickness of copper
Differential pressure transmitter 0.48 tubing (approximately 1 mm), the temperature gradient of a cop-
Rotameter 0.48 per tube wall in this experiment is negligible, so T w is the mea-
sured temperature for thermal resistance.
X
4
Table 5 T¼ T wi =4 ð10Þ
Experimental variables and Uncertainty. i¼1

Variable name Uncertainty (%)

T nf ¼ ðT out þ T in Þ=2 ð11Þ
Friction factor, f 1.5
Heat transfer factor, h 8.5 The tube Reynolds number can be expressed as:
Reynolds number, Re 0.55
Nusselt number, Nu 11.4 Re ¼ ud=mnf ð12Þ
Tube diameter, d 0.09
Tube length, l 0.09 The tube flow velocity u can be expressed as:
Thermal conductivity K 3
Dynamic viscosity v u ¼ 4qv =pd
2
1 ð13Þ
The kinematic viscosity of nanofluids and deionized water are
measured by the SNB-3 digital rotary viscometer.
1%, respectively, depending on the accuracy of the instrument.
f ¼ 2DPd=lqnf u2 ð14Þ
Table 4 shows uncertainties of apparatus performance. Table 5
shows variables uncertainty show in the adequacy of variables.
All the acceptable ranges of experimental error were less than 3.2. Experimental verification theory
15%, which indicates strong reliability and accuracy for both
instruments and variables. Prior to the start of an experiment, one must check the experi-
mental device and verify the accuracy. The method of device veri-
3. Analysis of experimental data fication is to measure the convective heat transfer coefficient h and
flow resistance coefficient f of deionized water under turbulent
3.1. Data processing flow conditions at 25 °C ambient temperature in the tube. Verifica-
tion can be obtained from calculating the Nusselt number via the
The average heat flow is calculated as follows: convective heat transfer coefficient, according to the formula:

Q ¼ Q=A ð1Þ Nu ¼ h  D=k ð15Þ

The fluid flow in the built-in twisted belt tubes is in accordance
Q ¼ C nf  qm  DT ð2Þ
with the law of spiral motion. The calculated Nusselt number and
Among them: flow resistance coefficient of deionized water under turbulent flow
conditions compare with the rotating fluid heat transfer coefficient
DT ¼ T waterin
T waterout ð3Þ
correlations (15) and resistance coefficient correlation (17) that
reference [38] gives.
qm ¼ qv  qnf ð4Þ
Among them qm is volume flow: Nu ¼ 0:023  Re0:8  Pr0:4 ð16Þ
The average heat flux Q applying Newton cooling formula can
be expressed as: f ¼ 0:0791Re
0:25 ð17Þ

Q ¼ h  ðT w
T nf Þ ð5Þ As shown in Figs. 5 and 6, heat transfer characteristics based on

experimental results are in good agreement with the calculated
Among them h is average heat transfer coefficient, T w is average data generated by the theoretical formula. The overall error rate
temperature of tube wall T nf is average temperature of working is less than 4.7%. The flow characteristic results of the experiment
medium. agree well with the theoretical formula, and the overall error rate
From Eq. (1) the average convective heat transfer coefficient of was about 6.5%. So the experimental system meets the required
nanofluids can be obtained by the following: accuracy.

¼ Q
h ð6Þ
T w
T nf 4. Results and discussion

uqp C p þ ð1
uÞqbf C bf Cu nanofluids were used to conduct research using traditional

C nf ¼ ð7Þ
uqp þ ð1
uÞqbf horizontal copper tube and built-in twisted belt external thread
tubes resulting in a reverse ratio of 2.5 and the pitch of 2 mm,
The density of the nanofluids as follows: 2.5 mm and3 mm. The research objective was to explore the
changes in heat transfer and flow characteristics in the two types
qnf ¼ ð1
uÞqbf þ uqp ð8Þ
of heat exchange tubes. Studies using traditional and twisted tubes
Nanofluid converted mass fraction into volume fraction: showed advantages of each, which allowed to be recommended as
a model for best practices.
1
u¼ ð9Þ In this study, the data represent the average value of the mea-
ð1=xÞðqp =qbf Þ
sured data based on experiments conducted under the same condi-
 B. Sun et al. / International Journal of Heat and Mass Transfer 107 (2017) 712–722 717

0.0095 16000
DI water
experimental data 14000 0.5% Al2O3
0.0090 calculated data
0.5% Al
12000 0.5% Fe2O3
0.0085 0.5% Multiwalled carbon
10000
0.5% Graphite
0.5% Cu
8000
0.0080

h
f

6000
0.0075
4000

0.0070 2000

0
0.0065 2000 4000 6000 8000 10000 12000
4000 6000 8000 10000 12000 14000
Re Re

Fig. 5. Deionized water flow characteristics compared with Blasius approximate Fig. 7. Effect of nanoparticles on convective heat transfer coefficient at different Re
formula. in the built-in twisted belt external thread tubes.

120 DI water
10000
0.5% Al2O3
110 experimental data
calculated data 0.5% Al
8000 0.5% Fe2O3
100
0.5% Multiwalled carbon
90 6000 0.5% Graphite
0.5% Cu
h

80
4000
Nu

70
2000
60

50 0

40 2000 4000 6000 8000 10000 12000

Re
30
4000 6000 8000 10000 12000 14000 16000
Fig. 8. Effect of nanoparticles on convective heat transfer coefficient at different Re
Re in the horizontal tube.

Fig. 6. Deionized water heat transfer characteristics compared with Dittus-Boelter

experimental correlation.
about 1.5 times the heat transfer coefficient of the DI water. Anal-
ysis of the reasons; multiwalled carbon, Al2O3 and Al nanofluids
viscosity is too low, affecting the performance of its heat transfer
tions for each type of tubing. Each parallel experiment was
effect and deteriorating its heat transfer capacity. By using the
repeated 3–5 times.
SNB-3 Digital Viscometer on multiwalled carbon, Al2O3, Al nanoflu-
ids viscosity measurement, its value is significantly less than Cu,
4.1. Influence of nanoparticles on nanofluids heat transfer Graphite nanofluids, the average of about 70–80%, influencing
characteristics and flow characteristics the performance of its heat transfer effect. Under the usual condi-
tions, different nanofluids in the same mass fraction, the large den-
Experiments were carried out to measure the temperature, flow sity of thermal conductivity, because of the small size of the
and other parameters at different Reynolds number for the Cu, Al, density, and the intermolecular arrangement closely, so it will
A12O3, Fe2O3, multiwalled carbon, and Graphite nanofluids with accelerate the heat transfer. Cu has excellent electrical conductiv-
different mass fraction. Then the convective heat transfer coeffi- ity and thermal conductivity, the nanofluids formed by this sub-
cient and the flow resistance coefficient of each condition were cal- stance has better heat transfer performance than multiwalled
culated, and the comparative analysis was carried out. As shown in carbon, Al2O3, Al nanofluids.
Figs. 7 and 8, the convective heat transfer coefficient of Cu nanoflu- Furthermore, it is also found that the convection heat transfer
ids is bigger than any other nanofluids at the same condition, and is performance of the built-in twisted belt external thread tubes is
about 2.5 times the heat transfer coefficient of the DI water. The better than the horizontal tube at the same condition by compar-
Graphite nanofluids and the multiwalled carbon nanofluids are ing the results in Figs. 7 and 8, which indicates that the built-in
about 2.1 times and 1.8 times the convective heat transfer coeffi- twisted belt external thread tubes can be used for heat transfer
cient of the DI water separately. The convective heat transfer coef- enhancement. we can find that the convective heat transfer coeffi-
ficient of Fe2O3, Al2O3 and Al nanofluids are almost the same, and is cient of nanofluids in the built-in twisted belt external thread
718 B. Sun et al. / International Journal of Heat and Mass Transfer 107 (2017) 712–722

tubes is higher 50.34% than that in horizontal circular tube, the is bigger than any other kinds of measured nanofluids. What is
results show that the built-in twisted belt external thread tubes more, it is easy to find that the flow resistance coefficient of the
has good development prospects for improving the effect of con- built-in twisted belt external thread tubes is larger than that of
vective heat transfer. Analysis of the reasons for: On the one hand, the horizontal tube at the same condition because of the existing
the addition of nanoparticles increases the thermal conductivity of of built-in twisted belt.
the medium, thermal physical properties were enhanced, and the
particles, the liquid and the wall collide each other, so as to 4.2. Influence of pitch size on Cu nanofluids heat transfer
strengthen the heat transfer effect. Different nanofluids under the characteristics and flow characteristics in the built-in twisted belt
same mass fraction, large density has large thermal conductivity, external thread tubes
because large density has small volume and the molecules are
arranged closely, so it will accelerate the heat transfer. On the In Fig. 11(a), the convective heat transfer coefficient increases
other hand, built-in twisted belt has a spiral motion and a two with the increase of the Reynolds number and the increasing range
flow, fluid and built-in twisted belt has collision and heat transfer, is more and more. The reason is that with the increase of Re, the
the effective flow of the fluid is improved in the near wall region, increase of the movement speed of the nano powder and distur-
so as to improve the convective heat transfer. The external thread bance enhancement of nano powders, on the one hand, the tem-
is to expand the surface area of the tube, the increase of heat trans- perature distribution in the pipeline is more uniform, on the
fer surface can be seen as a decrease of heat transfer resistance. other hand, the thickness of the boundary layer decreases, which
As shown in Figs. 9 and 10, the flow resistance coefficient of all leads to the enhancement of convective heat transfer effect. When
kinds of nanofluids decreases with the increase of Re in both the the pitch of the built-in twisted belt external thread tubes is 2 mm,
built-in twisted belt external thread tubes and the horizontal tube, the convective heat transfer coefficient increase from 0.1% to 0.6%,
and it is almost inversely proportional to the Re. The addition of have a increasing effect with the mass fraction of Cu nanofluids. In
nanoparticles will obviously increase the flow resistance of DI the Reynolds number range of 2000–8000, the convective heat
water, and the influence of Fe2O3 nanoparticles on flow resistance transfer coefficient of the Cu nanofluids with a mass fraction of
0.5% reaches the maximum value. Analysis of the Cu nanofluids
test results are as follows: In the nominal range of concentrations,
as thermal motion of nanoparticles with increasing concentration
becomes more intense, the heat exchange of particles interacting
with particles as well as particles interacting with substrate solu-
tion will become more frequent. When it exceeds the nominal con-
centration range, the increase of fluid viscosity becomes the main
factor affecting heat exchange. And as the viscosity increased,
rod-like micelle structure also generated the adherence phe-
nomenon causing the thermal motion of nanoparticles and heat
transfer to cause the particle-to-particle and particles-
to-substrate interaction to slow down making the solution more
viscous thereby weakening the effect of heat transfer, the heat
transfer coefficient will decrease with the increase of the concen-
tration of the nanofluids, so there is the best concentration of the
nanofluids.
In the Reynolds number range of 8000–12,000, the convective
heat transfer coefficient of the Cu nanofluids with a mass fraction
of 0.6% reaches the maximum value. This is because in the low Rey-
Fig. 9. Effect of nanoparticles on flow resistance coefficient at different Re in the nolds number range, the influence of nanofluids viscosity on heat
built-in twisted belt external thread tubes. transfer becomes more important than the influence of nanoparti-
cles concentration for Cu nanofluids with mass fraction of 0.5% and
0.6%. As a result, compared with the Cu nanofluids with mass frac-
tion of 0.5%, the increase of viscosity resulted in the weaken of
effective collisions between nanoparticles as well as nanoparticles
and substrate solution for Cu nanofluids with mass fraction of 0.6%,
so the heat transfer between nanoparticles was also reduced. As
the Reynolds number increases continuously, the helical move-
ment of fluid flow formed in the built-in twisted belt tubes at a
high flow rate, which strengthened the effective collisions between
nanoparticles as well as nanoparticles and substrate solution, and
compensated for the negative effect of the deterioration of heat
transfer due to the fluid viscosity increasing.
From Fig. 11(b) and (c), under the conditions of 2.5 mm and
3 mm pitch, the convective heat transfer coefficient of Cu nanoflu-
ids increases with the increase of mass fraction, and the amplitude
is gradually increasing. When the mass fraction is reached 0.6%, the
convective heat transfer coefficient reaches the maximum value
and there is a certain difference with the phenomenon of Fig. 11
(a). Analysis of the reason is that if the pitch is large there is only
rotational action and has no longitudinal perturbation effect, so it
Fig. 10. Effect of nanoparticles on flow resistance coefficient at different Re in the is not conducive to heat transfer. The heat transfer coefficient of
horizontal tube. built-in twisted belt external thread tubes at the pitch of 2 mm
 B. Sun et al. / International Journal of Heat and Mass Transfer 107 (2017) 712–722 719

16000 16000
DI water DI water
0.1% Cu 14000 0.1% Cu
14000 0.2% Cu
0.2% Cu
0.3% Cu 0.3% Cu
12000 12000
0.4% Cu 0.4% Cu
0.5% Cu 0.5% Cu
10000 10000 0.6% Cu
0.6% Cu

8000 8000

h
h

6000 6000

4000 4000

2000 2000

0 0
2000 4000 6000 8000 10000 12000 2000 4000 6000 8000 10000 12000
Re Re

(a) pitch 2mm (b) pitch 2.5mm

16000 16000
DI water
DI water p=2
14000 0.1% Cu 14000 DI water p=2.5
0.2% Cu
DI water p=3
0.3% Cu 0.5% Cu p=2
12000 12000
0.4% Cu 0.5% Cu p=2.5
0.5% Cu 0.5% Cu p=3
10000 10000
0.6% Cu
8000
8000
h
h

6000
6000
4000
4000
2000
2000
0
0 2000 4000 6000 8000 10000 12000
2000 4000 6000 8000 10000 12000
Re
Re
(c) pitch 3mm (d) three different pitch
Fig. 11. Effect of pitch size and Cu nanoparticles concentration on fluid convective heat transfer coefficient.

more than at the pitch of 2.5 mm, the increase is 12%, the heat the surface area of the tube, and the increase of heat transfer sur-
transfer coefficient of built-in twisted belt external thread tubes face can be seen as a decrease of heat transfer resistance.
at the pitch of 2.5 mm more than at the pitch of 3 mm, the increase As shown in Fig. 12, the flow resistance coefficient of nanofluids
is 9%. The coefficient of heat transfer increases with the decrease of decreases with the increase of the Reynolds number in the built-in
pitch. twisted belt external thread tubes. As shown in Fig. 12(d), the flow
By comparing Fig. 11(b) with Fig. 8, it can be found that the con- resistance coefficient increases with decreasing of pitch at the
vective heat transfer coefficient of Cu nanofluid in the built-in same condition. The flow resistance coefficient of built-in twisted
twisted belt external thread tubes is 50.34% higher than that in belt external thread tubes with pitch of 2 mm is 17% higher than
the horizontal circular tube, which shows that the built-in twisted that of 2.5 mm, and the flow resistance coefficient of tubes with
belt external thread tubes has good development prospects for pitch of 2.5 mm is 12% higher than that of 3 mm. The flow resis-
improving the effect of convective heat transfer. Analysis of the tance coefficient decreases with the increase of pitch, the larger
reasons for: On the one hand, the addition of nanoparticles the pitch, the smaller the resistance, the more close to the horizon-
increases the thermal conductivity of the medium, thermal physi- tal tube, when the pitch is small, the laminar flow of the bottom
cal properties were enhanced, and the particles, the liquid and the will be disturbed, but also lead to the mainstream rotation, thereby
wall collide each other, so as to strengthen the heat transfer effect. increasing the flow resistance.
Different nanofluids under the same mass fraction, large density
has large thermal conductivity, because large density has small
volume and the molecules are arranged closely, so it will accelerate 4.3. The coefficient of thermal performance
the heat transfer. On the other hand, the existing of built-in twisted
belt may cause the spiral motion and a two flow of the fluid, and As mentioned above, the nanofluids shows the characteristics
the collision and heat transfer may also happen between fluid for enhancing convection heat transfer compared with DI water,
and built-in twisted belt. In conclusion, the effective flow of the and the convective heat transfer effect of nanofluids in the built-
fluid was improved in the near wall region, and the convective heat in twisted belt external thread tubes is very good. Therefore, the
transfer was strengthened. The external thread was used to expand coefficient of thermal performance [39] which is an important
720 B. Sun et al. / International Journal of Heat and Mass Transfer 107 (2017) 712–722

0.095 0.080
DI water
0.090
DI water
0.1% Cu 0.075 0.1% Cu
0.2% Cu 0.2% Cu
0.085 0.3% Cu 0.070 0.3% Cu
0.4% Cu 0.4% Cu
0.080
0.5% Cu 0.5% Cu
0.065
0.6% Cu 0.6% Cu
0.075
0.060
0.070

f
f

0.065 0.055

0.060 0.050

0.055 0.045

0.050
0.040
2000 4000 6000 8000 10000 12000
2000 4000 6000 8000 10000 12000
Re
Re

(a)pitch 2 mm (b) pitch 2.5mm

0.090
0.070 DI water p=2
DI water
0.085 DI water p=2.5
0.1% Cu
DI water p=3
0.065 0.2% Cu 0.080 0.5% Cu p=2
0.3% Cu
0.075 0.5% Cu p=2.5
0.4% Cu
0.060
0.5% Cu
0.5% Cu p=3
0.070
0.6% Cu
0.055 0.065
f
f

0.060
0.050
0.055

0.045 0.050

0.045
0.040
0.040

0.035 0.035
2000 4000 6000 8000 10000 12000 2000 4000 6000 8000 10000 12000
Re Re

(c) pitch 3 mm (d)three different pitch

Fig. 12. Effect of pitch size and Cu nanoparticles concentration on the fluid flow resistance coefficient.

evaluation for the overall effect of the built-in twisted belt external
thread tubes and the nanofluids was introduced in this paper:
3.5
0.5% Al2O3
ðNunf =Nubf Þ
g¼ 1=3
ð18Þ 0.5% Al
ðf nf =f bf Þ 0.5% Fe2O3
the coefficient of thermal performance

0.5% Multiwalled carbon

3.0
The variation of coefficient of thermal performance for different 0.5% Graphite
0.5% Cu
nanofluids in the built-in twisted belt external thread tubes with
pitch of 2 mm is shown in Fig. 13. The coefficient of thermal perfor-
mance for all kinds of measured nanofluids is greater than 1, and 2.5
there is a peak value for the thermal performance coefficient of
each kind of nanofluids. The thermal performance coefficient of
Cu nanofluids is the largest, which shows the comprehensive char- 2.0
acteristics of convective heat transfer and flow resistance of Cu
nanofluids are the best among all kinds of nanofluids studied in
this paper. 1.5

4.4. Verify the regression equation

1.0
According to the correlative formula [40] and the experimental 2000 4000 6000 8000 10000 12000
data obtained the correlation formula between the Nu number and Re
the friction drag coefficient of Cu-water nanofluids in the built-in
twisted belt external thread tubes. Fig. 13. Variation of nanofluids thermal performance coefficient.
 B. Sun et al. / International Journal of Heat and Mass Transfer 107 (2017) 712–722 721

350 Cu nanofluid has a more outstanding comprehensive perfor-

mance in the enhanced convection heat transfer.
300 0.1% Cu (2) The convective heat transfer coefficient of built-in twisted
0.2% CU belt external thread tubes is strengthen than the horizontal
10%
0.3% Cu tube, but the Flow resistance coefficient of the built-in
250
0.4% Cu twisted belt external thread tubes is larger than the horizon-
Nusselt from equation

0.5% Cu tal tube, indicates that the built-in twisted belt external
200 0.6% Cu
thread tubes has the ability to enhance heat transfer, but it
will increase the flow resistance.
10%
150 (3) The Cu nanofluids is studied in the built-in twisted belt
external thread tubes with different pitch sizes, respectively.
100 It is concluded that the smaller the pitch the heat transfer
1.1171 -2.4461 0.8009 effect is stronger, but the flow resistance coefficient is
Nu=0.6981Re Pr (1+w)
bigger.
50
(4) Obtained the best mass fraction of Cu nanofluids in the built-
in twisted belt external thread tubes is 0.5, at this time it has
0
0 50 100 150 200 250 300 350 the best heat transfer effect, obtained the heat transfer and
flow resistance correlations the pitch of 2 when the Cu
Nusselt from experiment
nanofluids heat transfer effect is the best within the built-
Fig. 14. Compare Cu nanofluids number Nu fitted values and experimental values. in twisted belt external thread tubes.

0.095
Acknowledgements
0.090 0.1% Cu
0.2% Cu This study was supported by the Program for Science and tech-
0.085 0.3% Cu 10% nology development projects of Jilin province (20160101282JC).
0.4% Cu
0.080 0.5% Cu
friction from equation

0.6% Cu References
0.075
[1] S.U.S. Choi, J.A. Eastman, Enhancing thermal conductivity of fluids with
0.070 nanoparticles, Tech. Inform. Tech. Rep. 231 (1) (1995) 99–105.
10% [2] Xuan. Yimin, Theory and application of nano fluid energy transfer, Chin. Sci.
Tech. Sci. (2014) 269–279.
0.065
[3] N.S. Akbar, S. Nadeem, S.U. Rahman, et al., Nano fluid flow in tapering stenosed
-0.2445 0.4694 arteries with permeable walls, Int. J. Therm. Sci. 85 (6) (2014) 54–61.
0.060 f=0.5433Re (1+w) [4] N.S. Akbar, M. Raza, R. Ellahi, Influence of heat generation and heat flux on
peristaltic flow with interacting nanoparticles, Eur Physi J Plus 129 (8) (2014)
0.055 1–15.
[5] N.S. Akbar, M. Raza, R. Ellahi, Influence of induced magnetic field and heat flux
0.050 with the suspension of carbon nanotubes for the peristaltic flow in a
0.050 0.055 0.060 0.065 0.070 0.075 0.080 0.085 0.090 0.095 permeable channel, J Magn Magn Mater 381 (2015) 405–415.
friction from experiment [6] N.S. Akbar, M. Raza, R. Ellahi, Interaction of nanoparticles for the peristaltic
flow in an asymmetric channel with the induced magnetic field, Eur Physi J
Plus 129 (7) (2014) 1–12.
Fig. 15. Compare Cu nanofluids number friction fitted values and experimental
[7] Z.-H. Liu, Y.-Y. Li, A new frontier of nanofluid research – application of
values. nanofluids in heat pipes, Heat Mass Transfer 55 (2012) 6786–6797.
[8] O.A. Alawi, N.A.C. Sidik, H.A. Mohammed, S. Syahrullail, Fluid flow and heat
Nu ¼ 0:6981Re1:1171 Pr
2:4461 ð1 þ xÞ0:8009 ð19Þ transfer characteristics of nanofluids in heat pipes, Heat Mass Transfer 56
(2014) 50–62.
[9] R. Sureshkumar, S.T. Mohideen, N. Nethaji, Heat transfer characteristics of
f ¼ 0:5433Re
0:2445 ð1 þ xÞ0:4694 ð20Þ nanofluids in heat pipes: A review, Renew Sust Energ Rev 20 (4) (2013) 397–
410.
The fitting value and experimental value is shown in Figs. 14 [10] M.H. Buschmann, Nanofluids in thermosyphons and heat pipes: Overview of
and 15. In Fig. 14, the mean deviation is 4.37%, the maximum devi- recent experiments and modelling approaches, Int. J. Therm. Sci. 72 (10)
(2013) 1–17.
ation is 12.67%, the standard deviation is 3.05%, in Fig. 15, the mean [11] Weerapun Duangthongsuk, Somchai Wongwises, Heat transfer enhancement
deviation is 3.6%, the maximum deviation is 11.16%, the standard and pressure drop characteristics of TiO2–water nanofluid in a double-tube
deviation is 3.01%, so the two correlation can predict the experi- counter flow heat exchanger, Int. J. Heat Mass Transf. (2009) 2059–2067.
[12] Xuan. Yi-min, Li. Qiang, Investigation on convective heat transfer and flow
mental results well. Among the Re number range 2000–12,000, features of nanofluids, Heat Trans. 125 (1) (2003) 151–155.
Pr range 7.06–7.09, Concentration is less than 0.6%, for other [13] A. Zeeshan, R. Ellahi, M. Hassan, Magnetohydrodynamic flow of water/
ranges of applicability there is yet to be further verified. ethylene glycol based nanofluids with natural convection through a porous
medium, Eur Physi J Plus 129 (12) (2014) 1–10.
[14] S. Rashidi, M. Dehghan, R. Ellahi, et al., Study of stream wise transverse
5. Conclusions magnetic fluid flow with heat transfer around an obstacle embedded in a
porous medium, J Magn Magn Mater 378 (378) (2015) 128–137.
[15] R. Ellahi, M. Hassan, A. Zeeshan, Shape effects of nanosize particles in Cu–H2O
Based on the mass fraction of 0.1–0.6%, Cu nanofluids were nanofluid on entropy generation, Heat Mass Transfer 81 (2015) 449–456.
applied in the horizontal tube and built-in twisted belt external [16] R. Ellahi, The effects of MHD and temperature dependent viscosity on the flow
thread tubes that the pitch of 2 mm, 2.5 mm and 3 mm, respec- of non-Newtonian nanofluid in a pipe: Analytical solutions, Appl Math Model
37 (3) (2013) 1451–1467.
tively. Results show that: [17] A.A. Rabienataj Darzi, Mousa Farhadi, Kurosh Sedighi, Heat transfer and flow
characteristics of Al2O3–water nanofluid in a double tube heat exchanger, Int.
(1) A variety of nano fluids (Graphite, multiwalled carbon, Commun. Heat Mass Transfer 47 (2013) 105–112.
[18] Adi T. Utomo, Ehsan B. Haghighi, Ashkan I.T. Zavareh, The effect of
A12O3, Cu, Al, Fe2O3) were used into a horizontal tube and nanoparticles on laminar heat transfer in a horizontal tube, Int. J. Heat Mass
the built-in twisted belt external thread tubes, among them, Transf. 69 (2014) 69–77.
722 B. Sun et al. / International Journal of Heat and Mass Transfer 107 (2017) 712–722

[19] E. Smithberg, F. Landis, Friction and forced convection heat-transfer [30] K. Bilen, M. Cetin, H. Gul, The investigation of groove geometry effect on heat
characteristics in tubes with twisted tape swirl generators, Heat Transfer 86 transfer for internally grooved tubes, Appl. Therm. Eng. 29 (2009) 753–761.
(1) (1964) 39–49. [31] S. De Schampheleire, P. De Jaeger, H. Huisseune, Influence of the louver and
[20] Heydar. Maddah, Mostafa. Alizadehc, Nahid. Ghasemi, Experimental study of delta winglet geometry on the thermal hydraulic performance of a compound
Al2O3/water nanofluid turbulent heat transfer enhancement in the horizontal heat exchanger, Int. J. Heat Mass Transf. 57 (1) (1999) 58–72.
double pipes fitted with modified twisted tapes, Int. J. Heat Mass Transf. 78 [32] Y. Zhu, H. Hu, G. Ding, Influence of metal foam on heat transfer characteristics
(2014) 1042–1054. of refrigerant–oil mixture flow boiling inside circular tubes, Appl. Therm. Eng.
[21] Rambir Bhadouriya, Amit Agrawal, S.V. Prabhu, Experimental and numerical 50 (2013) 1246–1256.
study of fluid flow and heat transfer in an annulus of inner twisted square duct [33] Bin. Sun, Zhimin. Zhang, Di. Yang, Improved heat transfer and flow resistance
and outer circular pipe, Int. J. Therm. Sci. 94 (2015) 96–109. achieved with drag reducing Cu nanofluids in the horizontal tube and built-in
[22] Wen. Liu, Bofeng. Bai, A numerical study on helical vortices induced by a short twisted belt tubes, Int. J. Heat Mass Transf. 95 (2016) 69–82.
twisted tape in a circular pipe, Case Stud. Therm. Eng. 5 (2015) 134–142. [34] S.K. Das, S.U.S. Choi, W.H. Yu, Science and Technology, John Wiley Sons Inc,
[23] W.H. Azmi, K.V. Sharma, P.K. Sarma, Numerical validation of experimental 2007.
heat transfer coefficient with SiO2 nanofluid flowing in a tube with twisted [35] E.J. Swanson, J. Tavares, S. Coulombe, Improved Dual-Plasma Process for the
tape inserts, Appl. Therm. Eng. 73 (2014) 294–304. Synthesis of Coated or Functionalized Metal Nanoparticles, IEEE T Plasma Sci
[24] P. Sivasharmugam, S. Sundaram, Improvement in performance of heat 36 (4) (2008) 886–887.
exchanger fitted with twisted tape, J. Energy Eng. 125 (1) (1999) 35–40. [36] I. Madni, C.Y. Hwang, S.D. Park, Mixed surfactant system for stable suspension
[25] A. Kumar, B.N. Prasad, Investigation of twisted tape inserted solar water of multiwalled carbon nanotubes, Colloid & Surface A Physicochem & Eng 358
heaters heat transfer, Renewable Energy 19 (3) (2000) 379–398. (1–3) (2010) 101–107.
[26] Lawson, C.G. Nua, R.J. Modonald, Transaction of the ANS 565(2) (1966) 38–42. [37] R.J. Moffat, Describing the uncertainties in experimental results, Exp. Fluid
[27] Qin Fuyou, San Bao Wang, Light tube and an outer screw thread tube for heat Therm. Sci. 1 (1) (1988) 3–17.
exchanger for thermal performance comparison, Chem. Eng. Equip. (2008) 28– [38] F. Gou, Investigation of heat transfer enhancement for twisted-tape inserts in
30. tubes, Chin. Inst. Atom. Energy (2002) 20–29.
[28] P. Bharadwaj, A.D. Khondge, A.W. Date, Heat transfer and pressure drop in a [39] L. Qiu, H.W. Deng, J.N. Sun, Pressure drop and heat transfer in rotating smooth
spirally grooved tube with twisted tape insert, Heat Mass Transfer 52 (2009) square U-duct under high rotation numbers, Heat Mass Transfer 66 (2013)
1938–1944. 543–552.
[29] Heat transfer characteristics for evaporation of R417A flowing inside [40] Wu Huiying, Wu Xinyu, Flow and heat transfer characteristics of a nanofluid in
horizontal smooth and internally grooved tubes, Energy Convers 49 (2008) a chip microchannel, J. Chem. Eng. 59 (9) (2008) 2181–2187.
1731–1739.