Renewable Energy 94 (2016) 213e222

Contents lists available at ScienceDirect

Renewable Energy
journal homepage: [Link]/locate/renene

Thermal enhancement of solar parabolic trough collectors by using

nanofluids and converging-diverging absorber tube
E. Bellos*, C. Tzivanidis, K.A. Antonopoulos, G. Gkinis
Department of Thermal Engineering, National Technical University of Athens, Zografou, Heroon Polytechniou 9, 15780 Athens, Greece

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

Article history: Parabolic trough collectors are the most mature technology for utilizing the solar energy in high tem-
Received 7 December 2015 perature applications. The objective of this study is the thermal efficiency enhancement of the com-
Received in revised form mercial parabolic collector IST-PTC by increasing the convective heat transfer coefficient between the
8 March 2016
working fluid and the absorber. There are two main factors which influence on this parameter, the
Accepted 18 March 2016
working fluid type and the absorber geometry. For this reason three working fluids are investigated,
thermal oil, thermal oil with nanoparticles and pressurized water. Moreover, a dimpled absorber tube
with sine geometry is tested because this shape increases the heat transfer surface and increases the
Keywords:
PTC
turbulence in the flow. The final results show that these two techniques improve the heat transfer co-
Nanofluids efficient and the thermal efficiency of the collector. More specifically, the use of nanofluids increases the
Thermal enhancements collector efficiency by 4.25% while the geometry improvement increases the efficiency by 4.55%.
Dimpled absorber Furthermore, collector parameters such as the heat loss coefficient, the exergetic efficiency, the pressure
Solidworks losses and the absorber temperature are presented for all the examined cases. The model is designed
with Solidworks and is simulated by its flow simulation studio.
© 2016 Elsevier Ltd. All rights reserved.

1. Introduction research has been focused on the improvements of PTCs in order to

increase their efficiency on high operating temperature levels. The
Fossil fuel and electricity costs have increasing trends in the main idea is based on increasing the heat transfer coefficient inside
recent years, a fact that renders the alternative and renewable the absorber, something that reduces the mean absorber temper-
energy sources to be more and more attractive solutions. Solar ature and as a consequence the thermal losses. The flow inside the
energy utilization is vital for our society because of its abundance tubular absorber depends on the working fluid and geometry of the
and its zero CO2 footprints [1e3]. Solar energy is able to cover a tube.
great variety of energy needs, as domestic hot water production [4], In the recent years, a great deal of research has focused on
house-heating [5], industrial heat demand [6] and electricity pro- working fluid selection which has a significant role in the col-
duction in solar power plants [7e9]. Parabolic trough collector lector performance. The typical working fluids are water, anti-
(PTC) is a kind of concentrated collector with use in applications freeze fluid (water and approximately 40% glycol by volume) [11],
with high temperature demand [10]; electricity production and thermal oil and air. The use of nanoparticles in water or in oil is
industrial processes are the main applications where PTCs are used an idea for improving the heat conductivity of the working fluid.
for solar energy utilization. A lot of studies have focused on this area by testing different
The main parts of this collector are the evacuated tube and the nanofluids on usual solar collectors as FPC and PTC. The most
reflector, while a tracking system is necessary for efficient opera- used nanofluids contain the following nanoparticles: Al2O3, Cu,
tion. Their thermal efficiency is high enough because the thermal TiO2, Al, Fe, CuO and SiO2 [12] with Al2O3 and Al to be used more
losses of the evacuated tube are lower than other conventional in studies and applications. Zadeh et al. [13] studied the use of
technologies, as flat plate collectors (FPC) [10]. However, a lot of synthetic oil/Al2O3 as working fluid in the commercial collector
LS-2 PTC and concluded that nanofluids improve the efficiency
about 10%. In this study the simulation tool was fluent which also
has been used for similar simulations by Sokhansefat et al. [14]
* Corresponding author. and Mwesigye et al. [15]. The use of Al2O3/water nanofluid as
E-mail address: bellose@[Link] (E. Bellos).

[Link]
0960-1481/© 2016 Elsevier Ltd. All rights reserved.
214 E. Bellos et al. / Renewable Energy 94 (2016) 213e222

base fluid in flat plate collectors was investigated by Said et al. 2. Theoretical analysis and methodology
[16] and Shojaeizadeh et al. [17] energetically and exergetically
and both studies concluded that nanofluids improve the collector 2.1. Energetic and exergetic performance
performance. Khullar et al. [18] analyzed the use of aluminum
nanoparticles in the thermal oil (Therminol VP-1) and proved This paragraph presents the main parameters of the analysis.
enhancements in efficiency about 5e10%. Other types of nano- Thermal efficiency, solar available radiation and useful energy are
fluids have been investigated in many studies. Kasaeian et al. [19] necessary in every solar system calculation.
tested nanotube/oil based nanofluids and Risi et al. [20] gas- The useful energy that working fluid carries can be determined
based nanofluid. by the energy balance in its control volume, as equation (1)
On the other hand, many studies have focused on geometry presents:
improvements of the absorber in order to increase the heat transfer
rate. More specifically, the idea is to create higher turbulence de- _ p ,ðTout  Tin Þ;
Qu ¼ m,c (1)
gree inside the flow, something that increases the heat transfer
The available solar energy in the collector area can be calculated
coefficient. Many techniques have been applied experimentally and
as the product of the reflector aperture and of the solar beam
numerically; in the first category, objects are located inside the tube
radiation.
in order to mix the flow better while in the second the absorber
geometry is designed with dimples which act as passive vortexes Qs ¼ Aa ,Gb ; (2)
inside the flow. Too and Benito [21] studied a dimpling absorber for
air tubular absorber and found improvement in efficiency and Thermal efficiency of the collector is the ratio of these two
simultaneously stated that the pressure drop in their case is lower quantities, as equation (3) presents:
than in cases with coil or tape inserts.
Qu
Chen et al. [22] compared dimpled and smooth tubes, hth ¼ ; (3)
concluding that the heat transfer coefficient inside the dimpled Qs
ones is greater. Kalinin et al. [23] used annular turbulizers in order The IST-PTC collector which is investigated in this study has an
to enhance the heat transfer coefficient inside the tubes. By this efficiency curve according to equation (4). This equation is taken by
way, the heat transfer area decrease is ranged from 30% to 50%. SANDIA Reports [10,31].
Kareem et al. [24] analyzed numerically and experimentally spirally



corrugated tubes and finally proved that the increase in heat Tin  Tam T  Tam 2
transfer coefficient is accompanied by increase in pressure drop hth ¼ 0:762  0:2125,  0:001672,Gb , in ;
Gb Gb
along the tube. Hong et al. [25] studied a converging-diverging tube
(4)
with spaced twisted-tapes for various geometric combinations.
They ended up that a careful design is needed because other geo- The heat loss coefficient of the collector expresses the specific
metric combinations increase the efficiency and others lead to heat losses of the absorber and can be calculated according to
lower efficiency. The use of objects inside the flow was studied by equation (5), if the heat losses (Qloss) are known:
many researchers and the results have shown improvements in
heat transfer. Cheng et al. [26] used vortex generators inside the Qloss
UL ¼ ; (5)
absorber of a PTC and concluded that the enhancement is varied Aro ,ðTr  Tam Þ
from 2% to 13% according to the water inlet temperature. The use of
other inserts as coils [27], twisted tapes [28,29] and perforated Where Aro is the outer surface of the absorber and it is calculated
plate inserts [30] have been analyzed and heat transfer enhance- as:
ments have been noticed. Other inserts that can be used are conical
Aro ¼ p,Dro ,L; (6)
nozzles, V-nozzles, screw-tapes and conical-rings which act as
swirl generators [27]. Moreover, the use of porous matric foam The optical efficiency modifier is the real optically efficiency for
inside the tube is an innovative idea because by this way the con- incident angle q to the maximum possible optical efficiency, for
tact surface is increasing and also the mixing conditions are more q ¼ 0 .
intense.
The goal of this study is to compare the PTC efficiency en- hopt ðqÞ
KðqÞ ¼ ; (7)
hancements came from using different working fluids and from hopt ðq ¼ 0ο Þ
geometry improvements. For this reason a commercial collector,
the IST-PTC (Industrial Solar Technology Corporation product) is For IST-PTC the manufacturer gives the next approximation for
analyzed. In the first part of this analysis, a validation of the model this parameter [10]:
is presented. The validation data are taken by SANDIA Reports
2
[10,31] where the collector's efficiency has been measured. The KðqÞ ¼ cosðqÞ þ 0:0003178,q  0:0000143,q ; (8)
use of nanoparticles inside the base fluid, which is thermal oil, is The exergetic output from the collector is given from the next
investigated in order to determine the impact on the collector equation [35,36]:
performance. Moreover, pressurized water is also tested, because


this is the basic working medium in conventional applications. In T
the next part of this analysis, a converging-diverging absorber
_ p ,Tam ,ln out ;
Eu ¼ Qu  m,c (9)
Tin
tube is used in the same collector in order to determine the en-
hancements in the efficiency. Finally, all the cases are compared The available solar exergy is usually calculated by Petela [37]
energetically and exegetically in order find the best way for formula in the following equation:
improving the IST-PTC efficiency. The model designed with Sol- "

 #
idworks and simulated in its flow simulation studio. This simu- 4 Tam 1 Tam 4
Es ¼ Ac ,Gb , 1  , þ , ; (10)
lation tool has been used in many other studies of solar collectors 3 Tsun 3 Tsun
[32e34].
 E. Bellos et al. / Renewable Energy 94 (2016) 213e222 215

The sun temperature (Tsun) in equation (10) is equal to 4350 K. greater density and thermal conductivity, while its specific heat
This value is the 3/4 of the real temperature of the sun (5800 K) in capacity is lower compared to thermal oil. The dynamic viscosity
its outer layer [38]. The exergetic efficiency is the ratio of useful has similar values in both cases, with the nanofluid viscosity to be
exergy output to exergy input, as the next equation shows [36,37]: a bit greater.

Eu
hex ¼ ; (11) 2.3. Heat transfer analysis
Es
The exergetic efficiency is related to the maximum possible The heat transfer from the absorber to working fluid is the core
produced work and usually is used in systems with electricity of the analysis and depends on the heat convection coefficient.
production. PTC is the most suitable technology for solar power Greater values of this parameter lead to lower absorber tempera-
plants. For this reason, the exergetic analysis is essential for ture which means lower thermal heat losses, according to equation
comparing the examined cases. (5). More specifically, equation (16) presents how the useful energy
depends on the heat convection coefficient h, which is referred in
many cases as heat transfer coefficient because the conduction in
2.2. Nanofluid properties
the absorber can be neglected.
 
The use of nanoparticles (np) in the base fluid (bf) leads to the
Qu ¼ h,Ari , Tr  Tfm ; (16)
improvement of working fluid properties. The nanofluid (nf)
properties are depended on the volumetric fraction of the nano- Where Ari is the internal surface of the tube and Tfm the mean
particles in the nanofluid and is symbolized with 4 (vol %). Usually temperature of the working fluid. In many cases, this temperature is
this fraction takes low values close to 1% or 2% depended on the calculated as the average of the inlet and outlet temperature.
kind of the nanoparticle. The properties of the nanofluid are given In theoretical basis, the coefficient h is calculated by the Nusselt
from the next set of equation. The density of the mixture is pre- number which is depended on the flow conditions and the geom-
sented by equation (12) [39]: etry of the problem. Reynolds number and Prandtl number are
conjugated in Nusselt calculation. For the general case, the hy-
rnf ¼ rbf ,ð1  4Þ þ rnp ,4; (12)
draulic diameter Dh is proposed. This geometry parameter can be
The specific heat capacity of the nanofluid is given by the next calculated as:
equation [40]:
4,Acs
Dh ¼ ; (17)
rbf ,ð1  4Þ rnp ,4 Pcs
cp;nf ¼ ,cp;bf þ ,c ; (13)
rnf rnp p;np P and At symbolize the perimeter of the cross section and the
area of this.
For the calculation of thermal conductivity, many correlations
Nusselt number is determined as equation (18) shows:
have been used. The Maxwell equation is the more prevailed
[41,42]: h,Dri
  Num ¼ ; (18)
k
knp þ 2,kbf  2,4, kbf  knp
knf ¼ knp kbf knp
; (14) Reynolds number is able to be written as:
kbf
þ2þ kbf
u,Dh 4,m_
Re ¼ ¼ ; (19)
The mixture viscosity is calculated as equation (15) shows [43]: n Pcs ,m
 
Prandtl number is calculated as.
mnf ¼ mbf , 1 þ 2:5,4 þ 6:5,42 ; (15)
m,cp
In this study, the nanofluid is a mixture of thermal oil and Al2O3, Pr ¼ ; (20)
k
with the volumetric fraction 4 to be 2%. The usual values for this
parameter range from 0.1% to 4% [17], so an intermediate value was In turbulent flow, with Reynolds number greater than 2300, the
selected to be analyzed. The nanoparticles are spherical with a Nusselt number is given by Colburn correlation [44] according to
diameter of about 20 nm. Table 1 includes the properties of the equation (21):
nanoparticles, of the thermal oil and of the nanofluid for three
typical temperature levels. Nu ¼ 0:023,Re0:8 ,Pr0:4 ; (21)
Fig. 1 illustrates the properties comparison between the ther- By combining equations (17e21), the heat convection coefficient
mal fluid and the nanofluid. It is obvious that the nanofluid has can be written as:
! !
  c0:4 0:6
0:8 p ,k P 0:2
Table 1 h ¼ 0:1743,m , , cs ¼ Cmass ,Cprop ,Cgeom ;
Properties of nanoparticles and of the nanofluid-base fluid in three temperature m0;4 Acs
levels (300 K e 400 K e 500 K).
(22)
T(K) Material r (kg/m3) cp (J/kgK) k (W/mK) m (Pas)
The above equation shows that the heat transfer coefficient
e Al2O3 4000 773 40 e
depends on the mass flow rate (Cmass), the working fluid properties
300 Thermal oil 1058 1570 0.13570 0.003570
Nanofluid 1117 1513 0.14392 0.003758 (Cprop) and the geometry configuration of the absorber (Cgeom). The
400 Thermal oil 977 1850 0.12425 0.000739 enhancement of this coefficient depends on these factors sepa-
Nanofluid 1037 1767 0.13179 0.000778 rately as equation (22) suggests. For this reason, in this study the
500 Thermal oil 889 2120 0.10900 0.000331 improvement of h, based on the geometry and on the working fluid,
Nanofluid 951 2007 0.11615 0.000348
will be presented as different examined cases.
216 E. Bellos et al. / Renewable Energy 94 (2016) 213e222

Fig. 1. Comparison of properties between thermal oil and thermal oil/Al2O3 (oil nanofluid); a) density b) specific thermal capacity c) thermal conductivity d) dynamic viscosity.

In literature, Xuan and Li [45] suggested the equation (23) for - Static pressure in the tube outlet.
Nusselt calculation in the nanofluid. This correlation is experi- - Heat convection coefficient between the outer cover surface
mental and takes into consideration the fraction 4 of nanoparticles and the ambient air.
and is similar to Colburn equation (21). b) Materials:
  - The cover is made from glass.
Nu ¼ 0:0059, 1 þ 7:628,40:6886 ,Pe0:001 ,Re0:9238 ,Pr0:4 ; - The absorber is selected to be made of cooper.
- The reflector has a mirror surface.
(23) c) Radiation surfaces
This equation is applied in turbulent flows, from Reynolds 2300 - Absorber outer surface is selective.
to 25000 and for volumetric fraction 4 from 0.3% to 2%. - Inner and outer cover surfaces have values from the selected
material.
- The reflector surface was set “symmetrical in radiation” in
2.4. Methodology order to reflect the radiation.
d Outputs
In the first part of this study, the IST PTC performance is vali- - Working fluid bulk temperature in the outlet.
dated by experimental data in literature [10,31]. After this, the ef- - Mean receiver temperature in its outer surface.
ficiency enhancement is investigated by two ways. In the first test, - Mean volumetric cover temperature in its volume.
the use of different working fluids is examined by testing Al2O3 - Receiver thermal losses.
nanoparticles in the thermal oil and by using pressurized water. In - The total enthalpy difference between the outlet and the inlet
the second test, the internal surface of the absorber becomes wavy surfaces of the tube.
in order to increase the heat transfer between the thermal oil and e) Mesh
the tube. With these two techniques, the factors Cprop and Cgeom, - A basic mesh in the computational volume is used.
which determined in equation (22) are examined separately. The - Local mesh in the evacuated tube is applied.
final results determine which way is more efficient and the per- - Extra refinement in the fluid and in the partial cells is added.
formance increase in every case.
The model was designed in Solidworks and the simulation was 3. Examined model
made by its Flow simulation studio. Different values of inlet tem-
perature are tested and in every case the outlet temperature is The first part of the analysis was the design of the IST-PTC in
obtained, as the most important result. Moreover, the mean Solidworks. This collector is a commercial collector with smaller
receiver temperature, the mean cover temperature and the thermal module than other commercial collectors, as LS-2. For reducing the
heat losses are extra results which are useful for the calculations. In computational domain and for simulating a real collector, the IST-
order to make a suitable simulation, the user has to determine the PTC was selected as the ideal one. The designed model is shown
proper boundary conditions of the model, the radiation surface, the in Fig. 2.
material of every part, the simulations goals and of course the The main dimensions of this model and the simulation param-
mesh. eters are included in Table 2. It is essential to state that the receiver
emittance depends on the receiver temperature [31] and a linearly
a) Boundary conditions: correlation was used in the simulation program. According to
- Inlet mass flow rate and temperature of the working fluid in Table 2, the emittance is equal to 0.18 for receiver temperature at
the tube inlet. 80  C and 0.27 at 400  C. The heat convection coefficient between
 E. Bellos et al. / Renewable Energy 94 (2016) 213e222 217

used in these concentrating collectors in order to minimize the

transversal incident angles and the reflected rays to be delivered
correctly to the receiver. Fig. 4 shows that the calculated optical
modifier K is close to the theoretical one of equation (8). This result
also enforces the reliability of the developed model.

3.2. Converging-diverging absorber tube model

The influence of the tube geometry on the collector efficiency

was investigated in this section. According to equation (22), the
optimum tube is the one with the higher perimeter Pcs. In order to
increase this quantity locally and the total inner tube surface, the
inside surface of the tube was designed to be wavy as a converging-
Fig. 2. Examined model of IST-PTC in Solidworks
diverging tube. The variation of the inside tube diameter makes the
flow more turbulent and the heat transfer conditions are improved.
Moreover, the greater inside surface allows more useful energy to
Table 2 be delivered to the fluid and simultaneously the mean receiver
Model dimensions [31] and simulation parameters. temperature to be reduced. Fig. 5b depicts the wavy surface and the
Simulation parameters Values Model dimensions Values equation of the internal diameter across the “x” direction. It is
important to state that the selection of this function was not
εr (Tr ¼ 80  C) 0.18 W 2300 mm
εr (Tr ¼ 400  C) 0.27 L 6100 mm random and it was the result of a small evaluation between
εc 0.86 f 800 mm different geometry configurations. Equation (24) shows the equa-
hopt (q ¼ 0ο) 0.792 Aa 14.03 m2 tion of the receiver internal diameter in the case of the wavy tube
Gb 1000 W/m2 Dri 47 mm (Fig. 5b).
m 0.21 kg/sec Dro 51 mm
Tam 10  C Dci 72 mm
Dri ¼ 47 þ 2,sinð0:2,xÞ; (24)
hca 10 W/m2K Dco 75 mm
It is noticeable that Dri and x are measured in mm in equation
(24).

ambient and cover was selected to be 10 W/m2K, a typical value for

4. Results
low air velocity conditions, as the prototype experimental
conditions.
4.1. Working fluid investigation

3.1. Validation of the IST-PTC The first study case examines the influence of the working fluid
in the collector performance. More specifically, the thermal and the
In this section, the designed modeled is simulated for different exergetic efficiencies are calculated in order to determine the most
operating condition in order to predict its efficiency curve and to suitable solution. Moreover the receiver temperature, the heat
make a comparison with the literature results. More specifically, transfer coefficient and the thermal loss coefficient are presented in
different values for the thermal oil temperature were examined in order to explain the variations in the efficiencies between the
order to predict the thermal efficiency curve. The results of this examined cases. The typical working fluid of the IST-PTC is thermal
simulation are given in Fig. 3 and it is obvious that the efficiency oil and the model validation (Figs. 3 and 4) uses this working me-
curves are close to each other. The theoretical efficiency is a bit dium. The utilization of Al2O3 nanoparticles (4 ¼ 2%) inside the
lower for higher fluid temperature level because the correlation of thermal oil is investigated in this paragraph. Moreover, the most
equation (4) responds to temperature levels up to 300  C. usual working fluid in thermal devices, the pressurized water, is
The next examined parameter in the validation analysis is the investigated and the reason for using a greater pressure is for
optical efficiency modifier for different incident angles, in the
longitude direction. It is important to state that a tracking system is
1

0.80
0.8
0.75
0.6
0.70
K

calculated
ηth

0.65 calculated 0.4

theoreƟcal
0.60 theoreƟcal
0.2
0.55

0.50 0
0 0.05 0.1 0.15 0.2 0.25 0.3 0 20 40 60 80
(Tin-Tam)/Gb θ (o)

Fig. 3. Efficiency comparison between developed model (red line) and measured ef- Fig. 4. Optical efficiency modifier validation. With blue line the calculated values from
ficiency by SANDIA (blue line). The small figure inside presents the absorber emittance. the model and with the red one the measured by SANDIA. (For interpretation of the
(For interpretation of the references to colour in this figure legend, the reader is references to colour in this figure legend, the reader is referred to the web version of
referred to the web version of this article.) this article.)
218 E. Bellos et al. / Renewable Energy 94 (2016) 213e222

Fig. 7. Exergetic efficiency comparison for operation with three different working
fluids.

quantities. Higher receiver temperature increases the heat losses

and reduces the thermal efficiency, results that are presented in
Figs. 6, 8 and 9 respectively. The last examined parameter is the
heat transfer coefficient between tube and fluid (Fig. 10). Greater
values of this quantity lead to lower receiver temperature for pro-
ducing a constant useful energy. By this way, the heat loss coeffi-
cient is getting lower and the efficiency is improved. Figs. 6 and 10
Fig. 5. Examined absorber tubes. a) Cylindrical tube and b) Converging-diverging tube show that the greater efficiency is followed by higher heat transfer
with sine shape. coefficient. As a conclusion, the properties of every working me-
dium determine the heat convection coefficient which is the most
important parameter for the collector efficiency.
keeping it in liquid phase. For this reason, the use of water in high
temperature applications is followed by operating in high pressure
levels.
4.2. Tube geometry investigation
Fig. 6 shows the thermal efficiency comparison between the
examined fluids. Nanofluid performs better than thermal oil, while
In this paragraph the impact of tube geometry on the collector
the pressurized water is the best solution. The difference is getting
performance is investigated. The cylindrical tube (initial geometry)
greater while the fluid inlet temperature increases, a result that
compared to a wavy geometry (new geometry) for operation with
indicates the gain of using of nanofluid and pressurized water in
thermal oil. According to Fig. 11, the wavy geometry or converging-
greater temperature levels. It is essential to focus in nanofluid
diverging tube is more efficient than the usual tube. It can be
because the use of water is complicated due to high pressure levels.
explained by equation (22), where the heat transfer coefficient
Fig. 7 illustrates the exergetic efficiency of these working fluids. The
increases with higher inner tube surface. The difference between
results are very interesting, because nanofluid and water are more
the curves is getting greater for higher temperature levels, some-
efficient than thermal oil, with the first to be the suitable one for
thing that indicates the use of a wavy surface in application with
the higher temperature levels. It is remarkable that the exergetic
high temperature demand. Fig. 12 shows the exergetic efficiency
efficiency of the pressurized water is not greater than the one of
between the examined cases and the results again show
nanofluid, although the thermal efficiency of water is greater. This
improvement with the new geometry. Fig. 13 illustrates that the
result can be explained by the greater water specific heat capacity,
new geometry receiver has lower temperature levels and conse-
as equation (9) shows. The greater heat capacity leads to lower
quently lower thermal loss coefficient (Fig. 14).
temperature increase in the fluid and to lower mean fluid tem-
The heat transfer coefficient is greater in new wavy geometry, a
perature, a factor with negative impact on the exergetic output of
fact that explains the greater thermal and exergetic efficiency (see
the collector.
Fig. 15). The reason for the higher heat transfer coefficient is the
Figs. 8 and 9 can be evaluated together, because the receiver
more turbulent conditions inside the tube. The diverging-
temperature and the thermal loss coefficient are conjugated

Fig. 6. Thermal efficiency comparison for operation with three different working Fig. 8. Receiver temperature ¼ comparison for operation with three different working
fluids. fluids.
 E. Bellos et al. / Renewable Energy 94 (2016) 213e222 219

470
420
370

Tr (oC)
320
270
new geometry
220
iniƟal geometry
170
120
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35
(Tin-Tam)/Gb
Fig. 9. Thermal loss coefficient comparison for operation with three different working Fig. 13. Receiver temperature comparison for the two examined geometry with
fluids. thermal oil as working fluid.

10
9
8
7

UL (W/m2K)
6
5
4
3
new geometry
2
iniƟal geometry
1
0
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35
Fig. 10. Heat transfer coefficient for operation with three different working fluids. (Tin-Tam)/Gb

Fig. 14. Thermal loss coefficient comparison for the two examined geometry with
0.80 thermal oil as working fluid.

0.75 new geometry

iniƟal geometry converging tube forces the flow to be mixed better and to create a
0.70 more uniform temperature profile inside the tube. Moreover, the
sine geometry creates dimples in the tubes which create vortexes
ηth

0.65
inside the flow, something that enhances the heat transfer from the
0.60 hot tube to the fluid. Furthermore, the internal surface of this wavy
tube has greater area than the typical tube, a fact that leads to lower
0.55
absorber temperature and as a consequence to lower thermal
0.50 losses.
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35
(Tin-Tam)/Gb
4.3. Discussion of the results and deeper analysis
Fig. 11. Thermal efficiency comparison for the two examined geometry with thermal
oil as working fluid. In sections 4.1 and 4.2 two ways for increasing the efficiency of
IST-PTC were presented. The results show that a higher heat

0.40
400
0.35
350
0.30
300
0.25
h (W/m2K)

250
ηex

0.20
200
0.15
150 new geometry
0.10 new geometry
100 iniƟal geometry
0.05 iniƟal geometry
50
0.00
0
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35
(Tin-Tam)/Gb
(Tin-Tam)/Gb
Fig. 12. Exergetic efficiency comparison for the two examined geometry with thermal
Fig. 15. Heat transfer coefficient comparison for the two examined geometry with
oil as working fluid.
thermal oil as working fluid.
220 E. Bellos et al. / Renewable Energy 94 (2016) 213e222

Table 3
Heat transfer coefficient enhancement for all examined cases.

Tin 10  C 50  C 100  C 150  C 200  C 250  C 300  C 350  C Mean

Nanofluid 1.75% 2.70% 3.77% 5.40% 7.02% 8.89% 10.27% 10.92% 6.34%
Wavy tube 0.34% 0.71% 1.60% 3.79% 5.06% 6.66% 7.77% 8.10% 4.25%
Press. water 0.16% 1.00% 2.33% 4.11% 5.40% 6.95% 8.03% 8.42% 4.55%

Table 4
Thermal efficiency enhancement for all examined cases.

Tin 10  C 50  C 100  C 150  C 200  C 250  C 300  C 350  C Mean

Nanofluid 0.34% 0.71% 1.60% 3.79% 5.06% 6.66% 7.77% 8.10% 4.25%
Wavy tube 0.16% 1.00% 2.33% 4.11% 5.40% 6.95% 8.03% 8.42% 4.55%
Press. water 1.75% 2.70% 3.77% 5.40% 7.02% 8.89% 10.27% 10.92% 6.34%

transfer coefficient leads to greater thermal efficiency. Table 3 The increase in inlet temperature makes the dynamic viscosity
presents the enhancement of heat transfer coefficient in all to be reduced and the Reynolds number to take higher values. As a
examined cases. The mean increase by the use of nanoparticles, result, the friction factor is reduced and the pressure drop is getting
pressurized water and wavy tube are 6.34%, 4.55% and 4.25% lower when the operating temperature in greater. Thus the curves
respectively. Table 4 presents the increase in thermal efficiency in in the Fig. 17 have a decreasing rate for greater values of the
these cases. The improvement in the water case is 6.34%, while in parameter [Tin-Tam]/Gb. Furthermore, Fig. 17 proves that the use of a
the cases of nanofluid and wavy tube is 4.25% and 4.55% respec- wavy surface creates higher pressure losses in the system
tively. It is obvious that the improvement is important in all cases compared to the other cases. The working fluid with the lower
with the use of water to be the most attractive solution. However, pressure drop is pressurized water and this result gives an extra
the demand for high pressure is a constraint, especially in high point to this fluid. The pressure drop with the wavy surface varies
temperature levels. For this reason, in applications that operate in
low and medium temperature levels, pressurized water is the
suitable solution, while in higher temperature levels the other so- 210
lutions are the most feasible.
It is remarkable that the greater enhancement is observed in 180
nanofluid case, while the maximum thermal efficiency improve-
ment in water case. This result indicates that except from heat 150
transfer coefficient, other factors influence on the collector per-
120
Nusselt

formance. For example, the specific heat capacity is a very impor-

tant parameter and the heat transfer conductivity, as well. These
90 SimulaƟon results
properties in water are greater than the respective in thermal oil
theoreƟcal (Xuan and Li)
and in nanofluid. 60
Another important notice is that the enhancement between the
examined cases is more intense for higher fluid temperature levels. 30
For this reason, the use of these technologies in newer parabolic
collectors is vital in order to make them a competitive technology. 0
More specifically, in low fluid temperature levels, the enhancement 0 5000 10000 15000 20000 25000

is about 1% while in high it reaches about 9%. Reynolds

For a deeper analysis of heat transfer coefficient in the
Fig. 16. Comparison of calculated Nusselt number (blue line) and theoretical (red line)
nanofluid case, the calculated Nusselt number is compared with according to Xuan and Li model. (For interpretation of the references to colour in this
the Xuan and Li model [45]. Equation (23) is the correlation for figure legend, the reader is referred to the web version of this article.)
the Nusselt number as a function of Reynolds Number, Prandtl
Number and volumetric fraction of nanoparticles 4. In our case,
this fraction is equal to 2% for all the examined cases. Fig. 16 350
Wavy geometry
depicts this comparison between the calculated Nusselt from Oil nanofluid
300
simulation results and the theoretical Nusselt for respective Thermal oil
operating conditions. More specifically, every couple of points 250 Pressurized water
respects to a different operation temperature for the collector.
ΔP (Pa)

The comparison shows that the theoretical values are about 21% 200
greater than the calculated. This difference is accepted and 150
Fig. 16 validates the heat transfer calculation for the case of
nanofluids. 100
Moreover it is essential to estimate the pressure losses in the
50
parabolic collector for all the examined cases. Pressure drop is an
important issue in great parabolic collector fields and for this 0
reason a specific analysis is presented. For the examined module 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35
with 6.1 m length pressure drop between inlet and outlet was (Tin-Tam)/Gb
calculated for the four examined cases. Fig. 17 shows the pressure
losses for various operation conditions. Fig. 17. Pressure drop for all the examined cases.
 E. Bellos et al. / Renewable Energy 94 (2016) 213e222 221

from 160 Pa to 330 Pa while for the pressurized water case it varies x Direction x, mm
from 100 Pa to 158 Pa. Nanofluid oil and thermal oil creates similar
pressure losses, from 125 Pa to 202 Pa and 118 Pa to 192 Pa Greek symbols
respectively. DP Pressure drop, Pa
ε emittance
5. Conclusions h efficiency
q angular displacement in longitudinal direction,o
This study presents efficiency enhancements on parabolic m dynamic viscosity, Pa s
trough collectors. The commercial IST-PTC is designed and simu- r Density. kg/m3
lated in Solidworks for various operation conditions. In the first 4 Volumetric fraction of nanoparticles,%
part of the analysis, the model validation is carried out by taking
experimental correlations from the literature. After this part, two Subscripts and superscripts
kinds of enhancements are presented; the first is the working fluid a aperture
investigation and the second is a test of a new absorber geometry am ambient
configuration with a wavy internal surface. bf base fluid
The use of Al2O3 nanoparticles inside the thermal oil improved c cover
the mean efficiency by 4.25% while the use of pressurized water by ca Cover-ambient
6.34%. However, the demand of high pressure level in the case of ci inner cover
water, in order to keep it in the liquid phase makes the use of co outer cover
nanofluids as more attractive solution. On the other hand, the cs Cross-section
absorber tube design with a wavy inner geometry leads the tube to ex exergetic
be converging-diverging, something that creates more turbulent fm mean fluid
conditions in the flow. The results showed 4.55% mean efficiency geom geometry
improvement compared to usual tube geometry. h Hydraulic
It is interesting to state that in all study cases, the increase in the in inlet
efficiency is greater for higher fluid temperature levels. This in- loss losses
dicates that these enhancements are suitable for high temperature mass mass flow rate
applications. More specifically, the main reason for the efficiency nf nanofluid
improvement in these cases is the increase in heat transfer coeffi- np nanoparticle
cient inside the flow, while other parameters as fluid specific heat opt optical
capacity have positive impact on the collector efficiency. Addi- out outlet
tionally, it is essential to mention that the use of a wavy surface prop properties
increases the pressure losses in the collector, a factor that have to be r receiver
taken into consideration for this design. ri inner receiver
ro outer receiver
Acknowledgments s Solar
th thermal
The first author would like to thank Onassis Foundation for its u useful
financial support.
References
Nomenclature
[1] A. Fernandez-García, E. Zarza, L. Valenzuela, M. Pe rez, Parabolic-trough solar
A area, m2 collectors and their applications, Renew. Sustain. Energy Rev. 14 (2010)
1695e1721.
C coefficient
[2] G. Tsalikis, G. Martinopoulos, Solar energy systems potential for nearly net
CL local concentration ratio zero energy residential buildings, Sol. Energy 115 (2015) 743e756.
cp specific heat capacity, kJ/kg K [3] K. Nithyanandam, R. Pitchumani, Optimization of an encapsulated phase
D diameter, mm change material thermal energy storage system, Sol. Energy 107 (2014)
770e788.
E Exergy, W [4] N. Aste, M. Beccali, L.C. Tagliabue, Nomograph for rapid technical and eco-
f Focal length, mm nomic assessment of solar thermal systems for DHW production, Sol. Energy
Gb solar beam radiation, W/m2 86 (9) (2012) 2472e2485.
[5] E. Bellos, C. Tzivanidis, A. Delis, K.A. Antonopoulos, Comparison of a solar
h convection coefficient, W/m2K driven heat pump heating system with other typical heating systems with
k thermal conductivity, W/m K TRNSYS ECOS, Conference (2015) (Pau, France).
K Angle efficiency modifier [6] S. Kalogirou, The potential of solar industrial process heat applications, Appl.
Energy 76 (4) (2003) 337e361.
L tube length, mm [7] G. Pei, J. Li, J. Ji, Analysis of low temperature solar thermal electric generation
m mass flow rate, kg/s using regenerative organic rankine Cycle, Appl. Therm. Eng. 30 (2010)
Nu mean Nusselt number 998e1004.
[8] M. Wang, J. Wang, Y. Zhao, P. Zhao, Y. Dai, Thermodynamic analysis and
P Perimeter, m optimization of a solar-driven regenerative organic Rankine cycle (ORC) based
Pe Peclet number (¼Re ∙ Pr) on flat-plate solar collectors, Appl. Therm. Eng. 50 (2013) 816e825.
Pr Prandtl number [9] B.F. Tchanche, G. Papadakis, G. Lambrinos, A. Frangoudakis, Fluid selection for
a low-temperature solar organic rankine cycle, Appl. Therm. Eng. 29 (2009)
Q Heat flux, W
2468e2476.
Re Reynolds number [10] S.A. Kalogirou, Solar thermal collectors and applications, Prog. Energy
T Temperature, oC Combust. Sci. 30 (3) (2004) 231e295.
u Working fluid velocity, m/s [11] Y. Gao, Q. Zhang, R. Fan, X. Lin, Y. Yu, Effects of thermal mass and flow rate on
forced-circulation solar hot-water system: comparison of water-in-glass and
UL Losses coefficient, W/m2K U-pipe evacuated-tube solar collectors, Sol. Energy 98 (2013) 290e301.
W Width, mm [12] S. Hassani, R. Saidur, S. Mekhilef, A. Hepbasli, A new correlation for predicting
222 E. Bellos et al. / Renewable Energy 94 (2016) 213e222

the thermal conductivity of nanofluids; using dimensional analysis, Int. J. Heat parabolic trough systems, Int. J. Hydrogen Energy 40 (44) (2015)
Mass Transf. 90 (2015) 121e130. 15254e15266.
[13] P.M. Zadeh, T. Sokhansefat, A.B. Kasaeian, F. Kowsary, A. Akbarzadeh, Hybrid [28] Aggrey Mwesigye, Tunde Bello-Ochende, Josua P. Meyer, Heat transfer and
optimization algorithm for thermal analysis in a solar parabolic trough col- entropy generation in a parabolic trough receiver with wall-detached twisted
lector based on nanofluid, Energy 82 (2015) 857e864. tape inserts, Int. J. Therm. Sci. 99 (2016) 238e257.
[14] T. Sokhansefat, A.B. Kasaeian, F. Kowsary, Heat transfer enhancement in [29] C. Chang, C. Xu, Z.Y. Wu, X. Li, Q.Q. Zhang, Z.F. Wang, Heat Transfer
parabolic trough collector tube using Al2O3/synthetic oil nanofluid, Renew. Enhancement and Performance of Solar Thermal Absorber Tubes with Cir-
Sustain. Energy Rev. 33 (2014) 636e644. cumferentially Non-uniform Heat Flux, Energy Procedia 69 (2015) 320e327.
[15] A. Mwesigye, Z. Huan, J.P. Meyer, Thermodynamic optimisation of the per- [30] A. Mwesigye, T. Bello-Ochende, J.P. Meyer, Multi-objective and thermody-
formance of a parabolic trough receiver using synthetic oileAl2O3 nanofluid, namic optimisation of a parabolic trough receiver with perforated plate in-
Appl. Energy 156 (2015) 398e412. serts, Appl. Therm. Eng. 77 (2015) 42e56.
[16] Z. Said, R. Saidur, M.A. Sabiha, A. Hepbasli, N.A. Rahim, Energy and exergy [31] V. Dudley, SANDIA, Report Test Results for Industrial Solar Technology Para-
efficiency of a flat plate solar collector using pH treated Al2O3 nanofluid, bolic Trough Solar Collector. SAND94e1117, Sandia National Laboratory,
Available online, J. Clean. Prod. 29 (July 2015). ISSN 0959e6526. Albuquerque, USA, 1995.
[17] E. Shojaeizadeh, F. Veysi, A. Kamandi, Exergy efficiency investigation and [32] C. Tzivanidis, E. Bellos, D. Korres, K.A. Antonopoulos, G. Mitsopoulos, Thermal
optimization of an Al2O3ewater nanofluid based flat-plate solar collector, and optical efficiency investigation of a parabolic trough collector, Case Stud.
Energy Build. 101 (2015) 12e23. Therm. Eng. 6 (2015) 226e237.
[18] V. Khullar, H. Tyagi, P.E. Phelan, T.P. Otanicar, H. Singh, R.A. Taylor, Solar en- [33] E. Bellos, C. Tzivanidis, D. Korres, K.A. Antonopoulos, Thermal analysis of a flat
ergy harvesting using a nanofluids-based concentrating solar collector, in: plate collector with solidworks and determination of convection heat coeffi-
Proceedings of MNHMT2012 3rd Micro/Nanoscale Heat & Mass Transfer In- cient between water and absorber, ECOS Conf. (2015) (Pau, France).
ternational Conference on March 3e6, Atlanta, USA, 2012. [34] E. Bellos, C. Tzivanidis, K.A. Antonopoulos, Thermal performance of a direct-
[19] A. Kasaeian, S. Daviran, R.D. Azarian, A. Rashidi, Performance evaluation and flow coaxial evacuated tube with solidworks flow simulation, in: 6th Inter-
nanofluid using capability study of a solar parabolic trough collector, Energy national Conference on “Experiments/Process/System Modelling/Simulation/
Convers. Manag. 89 (2015) 368e375. Optimization, Athens, July, 2015, pp. 8e11.
[20] A. Risi, M. Milanese, D. Laforgia, Modelling and optimization of transparent [35] T.J. Kotas, The Exergy Method of Thermal Plant Analysis, Krieger Publish
parabolic trough collector based on gas-phase nanofluids, Renew. Energy 58 Company, Malabar, FL, USA, 1995.
(2013) 134e139. [36] A. Bejan, Advanced Engineering Thermodynamics, Wiley Interscience, New
[21] Y.C. Soo Too, R. Benito, Enhancing heat transfer in air tubular absorbers for York, NY, USA, 1988.
concentrated solar thermal applications, Appl. Therm. Eng. 50 (10) (2013) [37] R. Petela, Exergy analysis of solar radiation, (Chapter 2), in: N. Enteria,
1076e1083. A. Akbarzadeh (Eds.), Solar Thermal Sciences and Engineering Applications,
[22] J. Chen, H. Müller-Steinhagen, G.G. Duffy, Heat transfer enhancement in CRC Press, Taylor & Francis Group, 2013.
dimpled tubes, Appl. Therm. Eng. 21 (5) (2001) 535e547. [38] S. Kalogirou, in: Soteris A. Kalogirou (Ed.), Solar Energy Engineering, Academic
[23] E.K. Kalinin, G.A. Dreitser, N.V. Paramonov, A.S. Myakochin, A.I. Tikhonov, Press, Boston, 2009.
S.G. Zakirov, E.S. Levin, L.S. Yanovsky, Comprehensive study of heat transfer [39] M. Ayatollahi, S.H. Nasiri, A.B. Kasaeian, Convection heat transfer modeling of
enhancement in tubular heat exchangers, Exp. Therm. Fluid Sci. 4 (6) (1991) Ag nanofluid using different viscosity theories, IIUM Eng. J. 13 (1) (2012).
656e666. [40] K. Khanafer, K. Vafai, A critical synthesis of thermophysical characteristics of
[24] Zaid S. Kareem, Shahrir Abdullah, Tholudin M. Lazim, [Link] Jaafar, nanofluids, Int. J. Heat. Mass Transf. 54 (2011) 4410e4428.
[Link] Wahid, Heat transfer enhancement in three-start spirally [41] W. Yu, S.U.S. Choi, The role of interfacial layers in the enhanced thermal
corrugated tube: experimental and numerical study, Chem. Eng. Sci. 134 (29) conductivity of nanofluids: a renovated Hamilton-Crosser model,
(2015) 746e757. J. Nanoparticle Res. 6 (4) (2004) 355e361.
[25] H.O.N.G. Mengna, D.E.N.G. Xianhe, H.U.A.N.G. Kuo, L.I. Zhiwu, Compound heat [42] J.C. Maxwell, A Treatise on Electricity and Magnetism, second ed., Clar-
transfer enhancement of a converging-diverging tube with evenly spaced endonPress, Oxford, UK, 1881.
twisted-tapes, Chin. J. Chem. Eng. 15 (6) (2007) 814e820. [43] G.K. Batchelor, The effect of Brownian motion on the bulk stress in a sus-
[26] Z.D. Cheng, Y.L. He, F.Q. Cui, Numerical study of heat transfer enhancement by pension of spherical particles, J. Fluid Mech. 83 (1977) 97e117.
unilateral longitudinal vortex generators inside parabolic trough solar re- [44] J. Leinhard IV, J. Leinhard V, A Heat Transfer Textbook, fourth ed., Philogiston
ceivers, Int. J. Heat Mass Transf. 55 (21e22) (2012) 5631e5641. Press, USA, 2012, pp. 354e360.
[27] Hacı Mehmet Şahin, Eşref Baysal, Ali Rıza Dal, Necmettin Şahin, Investigation [45] Y. Xuan, Q. Li, Investigation of convective heat transfer and flow features of
of heat transfer enhancement in a new type heat exchanger using solar nanofluids, J. Heat. Transf. 125 (2003) 151e155.