Energy Conversion and Management 136 (2017) 283293

Contents lists available at ScienceDirect

Energy Conversion and Management

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

Power generation enhancement in a salinity-gradient solar pond power

plant using thermoelectric generator
Behrooz M. Ziapour , Mohammad Saadat, Vahid Palideh, Sadegh Afzal
Department of Mechanical Engineering, University of Mohaghegh Ardabili, Ardabil, Iran

a r t i c l e

i n f o

Article history:
Received 8 November 2016
Received in revised form 10 January 2017
Accepted 11 January 2017

Keywords:
Organic
Power generation
Rankine
Salinity
Solar pond
Thermoelectric generator

a b s t r a c t
Salinity-gradient solar pond (SGSP) has been a reliable supply of heat source for power generation when
it has been integrated with low temperature thermodynamics cycles like organic Rankine cycle (ORC).
Also, thermoelectric generator (TEG) plays a critical role in the production of electricity from renewable
energy sources. This paper investigates the potential of thermoelectric generator as a power generation
system using heat from SGSP. In this work, thermoelectric generator was used instead of condenser of
ORC with the purpose of improving the performance of system. Two new models of SGSP have been presented as: (1) SGSP using TEG in condenser of ORC without heat exchanger and (2) SGSP using TEG in
condenser of ORC with heat exchanger. These proposed systems was evaluated through computer
simulations. The ambient conditions were collected from beach of Urmia lake in IRAN. Simulation results
indicated that, for identical conditions, the model 1 has higher performance than other model 2. For models 1 and 2 in TLCZ = 90 C, the overall thermal efficiency of the solar pond power plant, were obtained
0.21% and 0.2% more than ORC without TEG, respectively.
2017 Elsevier Ltd. All rights reserved.

1. Introduction
The air pollution from production and utilization of fossil fuel
are the primary causes of global warming [1]. Fossil fuel leads to
long term environmental issues such as acid rain and greenhouse effect [2]. Under these conditions, the sustainable and
environment-friendly energy resources such as solar energy has
been identified as one of the promising sources of energy to replace
the non-renewable energy resources. Solar energy, directly or indirectly is the major source of renewable energy available to humankind. In solar engineering of thermal processes, a set of
enhancement techniques is widely used to improve the performance of heat exchangers. A salinity-gradient solar pond (SGSP)
is a stratified body of water that collects and stores solar energy
as thermal energy for long periods of time [3,4]. It normally consists of three layers: the upper convective zone (UCZ), the nonconvective zone (NCZ), and the lower convective zone (LCZ). The
upper convective zone is a layer of cooler, less salty water. The
non-convective zone is a layer where salinity increases with
increasing depth. This is the most important layer in a solar pond
because the salt gradient suppresses global circulation within the
pond. This layer acts as a transparent insulator that permits solar

Corresponding author.
E-mail address: behrooz_m_ziapour@yahoo.com (B.M. Ziapour).
http://dx.doi.org/10.1016/j.enconman.2017.01.031
0196-8904/ 2017 Elsevier Ltd. All rights reserved.

radiation to penetrate to the bottom of the pond. The lower convective zone is a layer of high-salinity brine, which even when
heated, remains so dense that it cannot rise to the surface of the
pond. This maintained stratification allows the radiation that
reaches the bottom of the pond to be stored as heat in the lower
convective zone. While is not as efficient as photovoltaic solar collectors, the costs of constructing and operating a solar pond are a
fraction of the costs of photovoltaic cells [5]. Solar ponds can provide reliable heat at temperatures between 50 and 90 C and have a
low capital cost since they are based on locally available low- cost
materials, and can be incorporated into salinity mitigation schemes
[6]. One of the most important applications of solar ponds is to utilize thermal energy stored in LCZ to generate electrical power. For
this purpose, solar pond is usually combined with an organic
Rankine cycle (ORC) heat engine [7,8]. The ORC has been proven
to be a suitable method of converting low-temperature energies
into power because of its simplicity, high reliability and ease of
maintenance. Due to the advantageous features of the organic fluids of ORC such as low boiling point and high evaporation pressure,
the ORC is able to produce power using low-grade heat sources like
solar thermal energy [9,10]. Suarez et al. [5] have investigated the
main factors that result in differences between small- and largescale solar pond performances by using a new approach that combines high-resolution DTS data with computational fluid dynamic
simulations. Their investigation showed that experimental results

284

B.M. Ziapour et al. / Energy Conversion and Management 136 (2017) 283293

Nomenclature
A
ALCZ
ANCZ
AUCZ
Ir
K b1
K b2
Q_ lb
Q_ lsr
Q_ lt
Q_ lu
Q_ lw
Q_ e
hLCZ
Ug
Ab
Ibl
U wl
Awl
_b
m
C pb
T bw
T br
T amb
Tc
T c1
T ev a
T h1
T h2
T in
T LCZ
TULZ

T NCZ;b
T NCZ;t
T out
_ net
W
_ pump
W
_ turbin
W
CON
EVA
G

pond outer surface area (solar pond size), m2

mean area of the LCZ surface, m2
mean area of the NCZ surface, m2
area of the UCZ top surface, m2
radiation flux in water at the outer surface of pond,
W m
2
conductivity of pond liquid relation with the brine LCZ,
W m
1 K
1
conductivity of pond liquid relation with the brine UCZ,
W m
1 K
1
heat loss rate from the pond bottom surface, W
absorbed heat rate by the LCZ due to solar radiation, W
conduction heat loss through the LCZ top layers, W
useful heat extraction rate from the LCZ, W
heat loss rate from the pond wall surfaces, W
heat transfer rate to Rankine cycle, W
convective heat transfer coefficient in LCZ, W m
2 K
1
overall heat transfer coefficient, W m
2 K
1
pond bottom surface area, m2
distance between aquifer and LCZ, m
overall heat transfer coefficient, W m
2 K
1
area of pond in LCZ, m2
mass flow rate in in LCZ, kg/s
specific heat of salty water, J/(kg K)
temperature of outlet brine in LCZ, C
temperature of inlet brine in LCZ, C
ambient temperature, K
condenser section working fluid temperature, K
heat exchanger inlet cold side temperature, K
mean temperature of the evaporator section of the ORC
cycle, K
heat exchanger inlet hot side temperature, K
heat exchanger outlet hot side temperature, K
inlet temperature of the evaporator section of the ORC
cycle, K
LCZ temperature, K
ULZ temperature, K
bottom surface of NCZ temperature, K
top surface of NCZ temperature, K
outlet temperature of the evaporator section of the ORC
cycle, K
overall net work, W
ORC system pump work, W
ORC system turbine work, W
condenser
evaporator
generator

from small- scale solar pond experiments can be used to investigate the expected performance of large-scale solar ponds, as well
as the main issues that can decrease the thermal performance of
solar ponds. Boudhiaf and Bacar [11] have numerically studied
the problem of double-diffusive natural convection in a two
dimensional salinity gradient solar pond to analyze the complex
flow structure velocity, temperature and concentration distributions in transient regime. They have concluded that the internal
Rayleigh number has a very important effect on the temperature
of UCZ, NCZ and LCZ. Moreover, the solar heating effect has an
important influence on the transient evolution of velocity, temperature and concentration fields.
Ziapour et al. [12] theoretically proposed and investigated an
enhanced design of a large scale salinity-gradient solar pond power
plant. In their proposed model, some ORC systems and two-phase

H.E
TUR
Q_ uu
Q_ usr
Q_ ub
Q_ uw
Q_ uc
Q_ ur
Q_ ue
hxu
h
hc

r
ES
he

c
V
Q_ nu
Q_ nsr
Q_ nb
Q_ nt
Q_ nw
hx
U wn
T aquifer
ti
ki
lwn
Kg
Z
TH
TC
K teg

heat exchanger
turbine
the useful heat transfer rate from the UCZ, W
absorbed heat rate by UCZ due to solar radiation, W
heat loss rate from bottom layer to UCZ, W
heat loss rate from the pond wall surface, W
convection heat loss through the UCZ to ambient, W
radiation heat loss through the UCZ to ambient, W
heat loss due to evaporation, W
the value of solar radiation in UCZ, %
radiation incident angle, deg
convective heat transfer coefficient, W m
2 K
1
constant of Stefan-Boltzman, W/(m2 K4)
emissivity of the water surface
evaporation heat transfer coefficient, W m
2 K
1
relative humidity, %
wind velocity, m/s
the useful heat transfer rate from the NCZ, W
absorbed heat rate by NCZ due to solar radiation, W
heat loss rate from bottom layer to NCZ, W
conduction heat loss through the NCZ to UCZ, W
heat loss rate from the pond wall surface, W
the value of solar radiation in NCZ, %
overall heat transfer coefficient, W m
2 K
1
aquifer temperature near the NCZ, C
insulation thickness, m
conductive heat transfer coefficient of insulation,
W m
1 K
1
distance between aquifer and NCZ, m
conductive heat transfer coefficient of insulation of soil,
W m
1 K
1
figure of merit, K
1
the hot- side temperature of TEM, K
the cold-side temperature of TEM, K
thermal conductivity of TEG, W m
1 K
1

Greek letters
dLCZ
LCZ thickness, m
dNCZ
NCZ thickness, m
dUCZ
UCZ thickness, m
e
effectiveness of the heat exchangers used in ORC systems
gP
pond heat collection efficiency, %
g0
overall thermal efficiency of the solar pond power plant,
%
a
seebeck coefficient, V K
1

closed thermosyphons are utilized to generate electricity. They

have proved that the larger SGSP is more economically feasible.
Their results shows that: (1) The size of the wickless heat pipe
was decreased for water as a working fluid within it. (2) The selection of isobutane inside ORC system as a working fluid increases
the solar pond power plant performance. (3) The change of pond
size did not effect on the overall thermal efficiency of the solar
pond power plant.
Agha [13] developed a simulation model describing the thermal
behavior and economic feasibility of a solar pond coupled with
multi-stage desalination (MSF) system under the conditions prevailing on Tripoli-Libya. Kumar and Kishore [14] reported the technical and economic viability of a 6000 (m2) solar pond constructed
at Bhuj in India. The pond attained a maximum temperature of
99.8 C in May. The total cost of construction of the solar pond

B.M. Ziapour et al. / Energy Conversion and Management 136 (2017) 283293

was $90,000 (1997 prices), corresponding to a unit cost of 15 ($/

m2). Some others works have been conducted numerically and
experimentally in order to improve solar pond efficiency [1519].
Another possibility way to produce the electricity in a salinitygradient solar pond (SGSP) is to use thermo electric generator
(TEG) directly or with ORC cycle. Omer and Infield have studied theoretically and experimentally application of TEGs for solar energy
conversion [20]. Chavez et al. [21] have investigated the possibility
of using of thermoelectric generators in solar hybrid systems. In
their work, four different systems were examined. They studied
experimentally the TEGs based on traditional semiconductor material Bi2Te3 and designed for temperature interval of 50200 C.
Ding et al. [22] examined solar pond in generating electricity
with the aid of using TEGs for converting the heat available at
LCZ into electricity. Their system has covered potential of generating electricity for the solar pond operates in the different climate
classification. Also, they investigated the effect of heat extraction,
climatic variation, temperature polarization, and the conversion
efficiency of TEG on the thermal performance and electrical performance of the system. The thermal-electrical conversion efficiency
of this system is in the range of 11.5% from the heat extracted.
Singh et al. [23] presented the potential of thermoelectric generators as a power generation system using heat from the salinity gradient solar pond. They obtained that this system is capable of
producing electricity even on cloudy days or at night as the salinity
gradient solar pond acts as a thermal storage system.
In this paper, an enhanced design of a salinity-gradient solar
pond power plant is proposed and simulated, by helping the EES
software. In this work, an alternative method of power generation
is investigated. This method is to use of TEG instead of condenser
in ORC cycle with the aim increasing overall efficiency of ORC cycle
and optimum utilization of heat from salinity-gradient solar pond.
Two new models of SGSP have been presented as: (1) SGSP using
TEG in condenser of ORC without heat exchanger and (2) SGSP
using TEG in condenser of ORC with heat exchanger. These systems
are more cost effectiveness than SGSP with ORC cycle without TEG
due to prevention of wasting of heat source energy through condenser and production more power from it.

285

2. Descriptions of the two proposed models

Figs. 1 and 2 present the schematics of the two proposed models of the solar pond power plants. As shown in Figs. 1 and 2, each
ORC unit consists of an evaporator section (denoted by EVA), a turbine (TUR) coupled with a generator (G), thermoelectric generator
(TEG), and a pump (Pump 2). In Fig. 2, an ORC cycle with a heat
exchanger (HE) is used (model 2). The cold water in the UCZ of
pond can be used for cooling the working fluid inside the TEG using
pump 3.
As shown in Fig. 1 (model 1), the pump of ORC (pump 2) supplies the organic fluid to the evaporator where the organic fluid
is heated. The working fluid is then vaporized in the evaporator
by the heat source from solar pond. The high-pressure vapor is
directed to the turbine where useful work is attained. The lowpressure vapor is cooled directly by transferring its heat through
TEG to UCZ and the organic fluid is liquefied. In addition, TEG generates electricity with temperature difference. A circulation pump
(pump 3) also pumps the cooling water from the UCZ. The liquid is
pumped back to the EVA and a new cycle starts.
As shown in Fig. 2 (model 2), the total of above steps is done
with this difference that outlet organic working fluid from turbine
gives its heat to cooling water in heat exchanger that this cooling
water is circulated by pump 4 and then this cooling water cooled
directly by transferring its heat through TEG to UCZ. The cooling
water is also pumped from the UCZ by a circulation pump (i.e.
pump 3).
Moreover, the following assumptions are applied in the analysis
of the solar pond power plants:
 A steady state simulation program for all parts of the solar pond
power plant is developed.
 Changes in the potential and kinetic energies of streams are
negligible.
 All components of the power generation systems operate
steadily.
 The turbine and all the pumps are adiabatic devices.
 Negligible pressure drops occur in the piping systems.

Fig. 1. Solar pond power plant (model 1) using TEG without H.E.

286

B.M. Ziapour et al. / Energy Conversion and Management 136 (2017) 283293

Fig. 2. Solar pond power plant (model 2) using TEG and H.E.

3. The proposed power plant formulations

3.1. The pond formulation
In order to study solar pond, energy balance is performed on the
energy components in three layers of UCZ, NCZ and LCZ, as shown
in Figs. 3, 4 and 5, respectively.
The useful heat extraction rate from the UCZ can be obtained as
follows [2429]:
Fig. 3. Energy balancing on the energy components in the UCZ.

Q_ uu Q_ usr Q_ ub
Q_ uw
Q_ uc
Q_ ur
Q_ ue

Fig. 4. Energy balancing on the energy components in the NCZ.

Fig. 5. Energy balancing on the energy components in the LCZ.

B.M. Ziapour et al. / Energy Conversion and Management 136 (2017) 283293

In Eq. (1), we have as:

3.2. The ORC cycle formulation

Q_ usr 1
hxu AUCZ Ir ; hux 0:36
0:08lndUCZ = cos h;

Q_ ub

287

The Low-temperature organic Rankine cycle (ORC) seems a

desirable technology for conversion of the low-temperature heat
into useful work. In an ORC cycle, the working fluid is an organic
fluid with a low boiling point. In this work, has been used water
pump for heat extraction from LCZ for evaporating organic fluid
in evaporator of ORC and then generating power.
Thermal efficiency of ORC cycle is equal with:



 



K b1 K b2
ANCZ AUCZ T NCZ
T UCZ
2

K b1
;
2
2
dUCZ dNCZ =2

Q_ uc hc AUCZ T UCZ
T amb ; hc 5:7 3:8V;

Q_ ur rES AUCZ T UCZ 4
T K 4 ;

_
W

T K 0:0552T amb 1:5 ;

Q_ ue he AUCZ 2:933T UCZ
T amb
39:115051
c and

gORC _ net;ORC
Q in;EVA

he 8:88
7:82V:

_ net;ORC W
_ TUR
W
_ Pump2
W

_ net;ORC ) can be defined as follows:

The net work of ORC cycle (W

Due to same temperature of UCZ and ambient, Q_ UW can be

ignored.
The useful heat extraction rate from the NCZ can be obtained as
follows [25,27,30]:

Q_ nsr hx
hxu ANCZ Ir ; hx 0:36
0:08 lndUCZ dNCZ = cos h;

_ TUR ) and the input

For obtaining the output turbine work (W
_
pump work (W Pump2 ), in Eq. (7), then the steady state form of the
mass and energy conservation equations neglecting the kinetic
and potential energies changes were applied to each of the ORC
system components. Here, both the turbine and the pump isentropic efficiencies were selected be as 0.8.
The overall thermal efficiency of the solar pond power plant (g0 )
can be defined as follows [12,33]:


K b1 K b2
T LCZ
T UCZ
T LCZ T UCZ
; T NCZ
ANCZ
;
Q_ nb
2
dNCZ
2

go _ net
Q usr

Q_ nu Q_ nsr Q_ nb
Q_ nt
Q_ nw

In Eq. (2), we have as:

_
W

The power conversion efficiency can be determined as [41,42]:

1
t i lwn

and
Q_ nw U wn Awn T NCZ
T aquifer ;
U wn K i K g
Q_ nt

_
W

gc _ net
Q Iu



 



K b1 K b2
ANCZ AUCZ
T NCZ
T UCZ
2

K b1
2
2
dUCZ dNCZ =2

_ net is the net work of the overall system which is deterwhere W

mined as:

The useful heat extraction rate from the LCZ can be obtained as
follows [12,25,27,30,31]:

_ net W
_ TUR W
_ TEG
W
_ Pump
W

Q_ lu Q_ lsr
Q_ lt
Q_ lb
Q_ lw
Q_ e

_ TUR is the net work of turbine, W

_ TEG is power generation
where W
_
with TEG and W Pump is required net work for circulating of total

In Eq. (3), we have as:

pumps in system.

Q_ lsr hx ALCZ Ir ;

3.3. The thermoelectric generator formulation


K b1 K b2
T LCZ
T UCZ
ANCZ
;
Q_ It
2
dNCZ
Q_ lb U g Ab T LCZ
T aquifer ;
1
1
ti
lbl

;
U g hLCZ K i K g
Q_ lw U wl Awl T LCZ
T aquifer ;
1
1
t i lwl

and
U wl hLCZ K i K g
_ b C Pb T bw
T br :
Q_ e m
Also, one can define the pond heat collection efficiency as follows [12]:

gP Q_ Iu

Q_ usr

For calculating of optimum size of NCZ, can be written as [32]:

dQ_ lu
0
ddNCZ

10

Thermoelectric is defined as the science and technology associated with thermoelectric electricity generation (Seebeck effect)
and refrigeration (Peltier effect) [34]. A thermoelectric generator
is a solid state heat engine in which electron gas serves as the
working fluid and converts a flow of heat to electricity. Thermoelectric generators have several major advantages comprising
being highly reliable, having no moving or complex parts, being
environmentally friendly, being maintenance free and silent in
operation, having long life cycle (more than 100,000 h steadystate operation),and having modular structure as well as adaptability for various sources and types of fuel [35,36]. Fig. 6 shows cross
section view of an ELEGANT (Efficient Liquid-based Electricity
Generation Apparatus inside Thermoelectric). As mentioned above
sections, thermoelectric generator was used instead of condenser.
Returned Hot organic fluid from turbine passes through one side
of TEG and cold water, as coolant, from UCZ passes through the
other side and heat transfer is done from hot side of TEG to its cold
side and consequently this temperature difference led to power
generation.
With energy balance for ELEGANT, we can write as:

_ ci mh
_ ho mh
_ co w
_
_ hi mh
mh
_ co
hci
That Q_ ELEGANT is equal with: Q_ ELEGANT mh

11

288

B.M. Ziapour et al. / Energy Conversion and Management 136 (2017) 283293

Fig. 6. Energy balancing on the energy components in the TEG.

Fig. 8. The comparison between the present and Ziapour et al. models for g0 vs. dNCZ
[12,39].

For thermoelectric generator we have as [21,37]:

p
1 ZT
1

gteg gcarnot pM

1 ZT M TTHC

Then,

gcarnot

DT
TH

in

Eq.

1

TC
,
TH

(12),
TM

or g
we

T C T H
,
2

_
W
_
Q ELEGANT

have
Pteg gteg

12
2

[21,37]: ZT M a KRT M ,

T H
T C
, a
DDVT .
R

as


heatteg

4. Results and discussion

4.1. Model validation
Salinity-gradient solar pond with ORC cycle and thermoelectric
generator equations that were indicated in previous section, were
planned in EES software [38]. To verify the present work, then the
present program code has been provided which was adjusted with
Ziapour et al. work [12,39] conditions, without TEG. Fig. 7 has been
drawn for g
and W_net vs. T LCZ at T amb 14  C, area of pond 2000

4.2. Optimum size of the solar pond area

The average of the monthly ambient conditions are given in
Table 1 that prepared from Urmia salty lake (with salt capacity
more than 7  1012 kg) in Iran [39]. Design parameters for proposed salinity-gradient solar pond have been given in Table 2.
According to above formulations, a steady-state simulation program for all parts of the solar pond power plant was developed.
Fig. 9 shows the effects of the solar pond area on its efficiency vs.
solar radiation. This figure has been drawn for condition of solar
pond as: T amb 25  C, TUCZ = 14 C, TLCZ = 90 C, the UCZ salinity
about 5% and the LCZ salinity about 25%. Based on this curve, it
is evident that with increasing of A, then gP is increased. As seen
in Fig. 9, between 100340 W/m2, when the A exceeds 50,000
(m2), then the increasing of gP is stopped. In the following results,
the area of solar pond were optimally selected 50,000 (m2).

ORC

(m2), dLCZ = 1.2 (m), dUCZ = 0.2 (m) and Ir 428 (W/m2). As shown in
Fig. 7, the results of the present and Ziapour model are so close
together. Similar to Fig. 7, the comparison between the present
and Ziapour models was done for the overall thermal efficiency
of the solar pond power plant (g0 ) vs. dNCZ at different LCZ temperature (or heat extraction temperature) as 60, 80 and 100 C, as
shown in Fig. 8. In addition, the other conditions are same as
Fig. 7 that mentioned above. It can be seen that, the attained present model results were similar to the results of our previously
cited work.

4.3. Optimum size of pond layer thickness

The thermal efficiency of pond depends on the thickness of the
various zones in the solar pond. Fig. 10 indicates the effects of the
change of the pond layers thickness on the pond heat collection
efficiency of the solar pond power plant (gP ). This figure has been
sketched for conditions of the solar pond as: T amb 25  C
TLCZ = 90 C, the UCZ salinity about 5% and the LCZ salinity about
25% [45]. As seen, the increase of the UCZ thickness (dUCZ ) leads

_ net vs. T LCZ [12,39].

Fig. 7. The comparison between the present and Ziapour et al. models for gORC and W

289

B.M. Ziapour et al. / Energy Conversion and Management 136 (2017) 283293
Table 1
The average of the monthly ambient conditions at Urmia Lake in Iran [39].
Month

Wind velocity (m/s)

Relative humidity (%)

Ambient temperature (C)

Solar radiation (W/m2)

21 March20 April (labeled as Far)

21 April21 May (Ord)
22 May21 June (Khor)
22 June22 July (Tir)
23 July22 August (Mor)
23 August22 September (Shah)
23 September22 October (Mehr)
23 October21 November (Aban)
22 November21 December (Azar)
22 December20 January (Dey)
21 January19 February (Bah)
20 February20 March (Esf)
Annual average

3.40
2.85
2.76
2.74
2.46
2.38
2.45
2.53
1.83
2.09
3.25
2.85
2.63

52.36
50.52
31.59
26.63
32.97
36.22
42.92
75.39
73.77
82.56
63.45
56.17
52.05

13.1
17.8
24.7
29.0
27.8
22.5
14.7
3.1
0.6
-1.5
2.6
6.8
13.4

221.51
303.27
335.32
334.92
306.28
254.6
172.07
101.8
90.47
96.53
144.2
208.65
214.13

Table 2
Design parameters for proposed salinity-gradient solar
pond.
Parameter

Value

T UCZ
T LCZ
SUCZ
SLCZ
A
dUCZ
dNCZ
dLCZ

14 C
90 C
5%
25%
2000 m2
0.15 m
1.2 m
1.4 m

to a reduction in gP . Reversely, the increase of the LCZ thickness

(dLCZ ) leads to an increase in gP . As shown in Fig. 10, there is a peak
value of gP (about 9.9%) for NCZ thickness curve (dNCZ 1:2 m). By
considering this peak value, then the optimum size of the pond layers thickness are obtained as: dUCZ 0.15 (m), dNCZ 1.2 (m) and
dLCZ 1.4 (m). The values of the design parameters have been given
in Table 2.

4.4. Results for performance of the model 1

This model has been shown in Fig. 1. Since boiling point of the
organic fluid is lower than that of water, The ORC cycle could be
used with various kinds of low temperature heat sources such as
solar energy [40]. The ORC system has been proven to be an effective method for the low grade energy utilization. Fig. 11 exhibits
the investigation results of the present solar pond power plant
_ power vs. TLCZ. In this examination, four
for the variation of g and W
o

Fig. 10. The effects of the pond layers thickness variation on the pond heat
collection efficiency of the solar pond power plant (gP ).

organic fluids were studied [43,44]. As shown, the isobutane has

higher performance than the other fluids. Among these fluids
toluene has the most undesirable performance. So, isobutane has
been used as organic Rankine fluid in this work. The variation of
g and W_ TEG vs. Z has been shown in Fig. 12. It is obvious that
TEG

_ TEG is increased.
with increasing of Z, then the value of gTEG and W
In Fig. 13 the variations of both the total efficiency and the total
net work system vs. the pond outer surface area (A) for cases of
with TEG and without TEG have been shown. As seen in Fig. 13,
in model 1 with TEG and for different A, both the total efficiency

Fig. 9. The variation of gp vs. Ir for different pond area outer surface (A).

290

B.M. Ziapour et al. / Energy Conversion and Management 136 (2017) 283293

_ power vs. TLCZ for the four organic fluids.

Fig. 11. The variation of go and W

_ TEG vs. figure of merit (Z) in model 1.

Fig. 12. The variation of gTEG and W
_ power vs. m
_ for model 1.
Fig. 14. The variation of gpower and W

and the total net work system are higher than system without TEG.
This is due to this fact that the TEG prevents the heat loss and helps
to generate power from potential energy of it. Also, for model 1, the
_ of isobutane on both the total effieffects of the mass flow rate (m)
ciency and the total net work system has been sketched in Fig. 14.
_ then both the total effiAs seen from this figure, with raising of m,
ciency and the total net work system are increased.

4.5. Results for performance of the model 2

This model has been shown in Fig. 2. In Fig. 15 the variations of

gORC (Rankine cycle efficiency) and the overall net work system
_ net ) vs. T LCZ have been shown both for the model 1 and the model
(W
2. As shown, for each two models 1 and 2, with increasing T LCZ ,

_ power vs. different pond area outer surface (A) for cases as: with TEG (model 1) and without TEG.
Fig. 13. The variation of gpower and W

B.M. Ziapour et al. / Energy Conversion and Management 136 (2017) 283293

291

_ net vs. T LCZ for model 1 and model 2.

Fig. 15. The variation of gORC and W

Fig. 16. Variation of the heat exchanger effectiveness (e) vs. T LCZ .

Fig. 18. The comparison between three cases as: (1) the model 1, (2) the model 2,
and (3) the solar pond with ORC and without TEG, for gpower vs. T LCZ .

ciency for model 2 at higher temperature. In Fig. 16, we have

sketched variation of the heat exchanger effectiveness (e) vs. T LCZ .
As shown, the increase of T LCZ is resulted to the increase of the heat
exchanger effectiveness. But with increasing T LCZ , the net work system is decreased for each two models 1 and 2. This due to this fact
that in constant solar irradiance, the conduction heat loss through
the top layers of LCZ decreases. Here, the ORC with heat exchanger
has higher efficiency than the ORC without heat exchanger.
Finally, we try to compare the overall thermal efficiency (go )
_ net ) of the solar pond power plant,
and the overall net work (W
between three cases as: (1) the model 1, (2) the model 2 and (3)
the ORC cycle without TEG. Our comparison results have been
shown in Figs. 17 and 18. As shown from these figures curves, it
_ net
is obvious that the model 1 has averagely higher both W

Fig. 17. The comparison between three cases as: (1) the model 1, (2) the model 2,
_ power vs. T LCZ .
and (3) the solar pond with ORC and without TEG, for W

then the values of the efficiencies increase since the turbine inlet
temperature and enthalpy are increased. Here, for model 1 has
been obtained better efficiency at lower temperature, and for
model 2 has been obtained better efficiency at higher temperature.
With referring to Fig. 16, one can find the reason of this better effi-

(Fig. 17) and the total efficiency (Fig. 18) than other two cases.
_ net for model 1, 1700 W and for
From Fig. 16, it is obtained that W
model 2, 1620 W are averagely more than ORC without TEG in
T LCZ = 90 C. Also, from Fig. 17, one can obtain that go for model
1, 0.21% and for model 2, 0.2% are averagely more than ORC without TEG in T LCZ = 90 C.
4.6. More studying on the SGSP system
In Fig. 19, the variation of the heat transfer vs. different area of
the SGSP system has been shown, through the LCZ and for

292

B.M. Ziapour et al. / Energy Conversion and Management 136 (2017) 283293

_ TEG vs. figure of

 With investigation of variation of gTEG and W
merit (Z), has been depicted that with increasing of ZT, the
_ TEG is increased.
and W
value of g
TEG

_ of
 With investigation of the effects of the mass flow rate (m)
isobutane on both the total efficiency and the total net work
_ then both the total effisystem, it is seen that with raising of m,
ciency and the total net work system are increased.
Simulation results show that the model 1 has averagely higher
_ net ) and the overall thermal efficiency
both the overall net work (W
(g0 ) than others cases.

References

Fig. 19. variation of heat transfer vs. different pond area outer surface (A) for LCZ in
Ir = 214 W/m2.

Fig. 20. The variation of gP vs. Q_ irradiation .

Ir 214 W=m2 . As seen from this figure, the heat loss rate through
the LCZ to top layer (Q_ lt ) is the highest value and the heat loss rate

from the bottom (Q_ lb ) is the lowest value.

The effect of the absorbed solar irradiance on performance of
the SGSP system has been conducted, as shown in Fig. 20. It is seen
that with increasing of Q_ irradiation from 0 to 1400 W/m2, then the
value of gP increases.
5. Conclusions

The simulation of an economically SSGP system [46] combined

with TEG was investigated in this paper. Thermoelectric generator
was used instead of condenser of ORC with the purpose of improving the performance of system. Two new models of SGSP have been
presented as: (1) SGSP using TEG in condenser of ORC without heat
exchanger and (2) SGSP using TEG in condenser of ORC with heat
exchanger. Based on the simulation results, the following conclusions are drawn:
 The optimum size of area of solar pond based on its effect on
efficiency is 50,000(m2) between 100 and 340 W/m2.
 The optimum sizes of the pond layers thickness were obtained
as: dUCZ 0.15 (m), dNCZ 1.2 (m) and dLCZ 1.4 (m).
 The selection of isobutane inside ORC system as a working fluid
enhances the SGSP performance.

[1] Aleklett K, Hk M, Jakobsson K, Lardelli M, Snowden S, Sderbergh B. The

peak of the oil age-analyzing the world oil production Reference Scenario in
World Energy Outlook 2008. Energy Policy 2010;38:1398414.
[2] Lonngren KE, Bai EW. On the global warming problem due to carbon dioxide.
Energy Policy 2008;36:15678.
[3] Atiz A, Bozkurt I, Karakilcik M, Dincer I. Investigation of turbidity effect on
exergetic performance of solar ponds. Energy Convers Manag 2014;87:3518.
[4] Karakilcik M, Dincer I, Bozkurt I, Atiz A. Performance assessment of a solar
pond with and without shading effect. Energy Convers Manag
2013;65:98107.
[5] Suarez F, Ruskowitz JA, Childress AE, Tyler SW. Understanding the expected
performance of large-scale solar ponds from laboratory-scale observations and
numerical modeling. Appl Energy 2014;17:110.
[6] Akbarzadeh A, Andrews J. Integration of solar ponds in salinity mitigation
schemes to produce low grade heat for industrial process heating, desalination
and power. In: Solar world congress, Orlando, Florida, USA.
[7] Singh R, Tundee S, Akbarzadeh A. Electric power generation from solar pond
using combined thermosyphon and thermoelectric modules. Sol Energy
2011;85:3718.
[8] Date A, Alam F, Khaghani A, Akbarzadeh A. Investigate the potential of using
trilateral flash cycle for combined desalination and power generation
integrated with salinity gradient solar ponds. Procedia Eng 2012;49:429.
[9] Song J, Gu C, Ren X. Parametric design and off-design analysis of organic
Rankine cycle (ORC) system. Energy Convers Manag 2016;112:15765.
[10] Kang SH. Design and preliminary tests of ORC (organic Rankine cycle) with
two-stage radial turbine. Energy 2016;96:14254.
[11] Boudhiaf R, Baccar M. Transient hydrodynamic, heat and mass transfer in a
salinity gradientsolar pond: a numerical study. Energy Convers Manag
2014;79:56880.
[12] Ziapour BM, Shokrnia M, Naseri M. Comparatively study between single-phase
and two-phase modes of energy extraction in a salinity-gradient solar pond
power plant. Energy 2016;111:12636.
[13] Agha KR. The thermal characteristics and economic analysis of a solar pond
coupled low temperature multi stage desalination plant. Sol Energy
2009;83:50110.
[14] Kumar A, Kishore VVN. Construction and operational experience of a 6000 m2
solar pond at Kutch, India. Sol Energy 1999;65:23749.
[15] Bozkurt I, Karakilcik M. The effect of sunny area ratios on the thermal
performance of solar ponds. Energy Convers Manag 2015;91:32332.
[16] Khalilian M. Erroneous equations used for energy and exergy analysis of a
solar pond. Energy Convers Manag 2016;120:4668.
[17] Karakilcik M. Reply to Erroneous equations used for energy and exergy
analysis of a solar pond by M. Khalilian. Energy Convers Manag
2016;120:46970.
[18] Khalilian M. Erroneous equations used to assess the performance of a solar
pond. Energy Convers Manag 2016;114:3948.
[19] Bozkurt I. Reply to Erroneous equations used to assess the performance of a
solar pond by Morteza Khalilian. Energy Convers Manag 2016;114:399.
[20] Omer SA, Infield DG. Design optimization of thermoelectric devices for solar
power generation. Solar Energy Mater Solar Cells 1998;53:6782.
[21] Urbiola EAC, Vorobiev YV, Bulat LP. Solar hybrid systems with thermoelectric
generators. Sol Energy 2012;86:36978.
[22] Ding LC, Akbarzadeh A, Date A. Transient model to predict the performance of
thermoelectric generators coupled with solar pond. Energy 2016;103:27189.
[23] Singh B, Gomes J, Tan L, Date A, Akbarzadeh A. Small scale power generation
using low grade heat from solar pond. Procedia Eng 2012;49:506.
[24] Ali HM. Potential of solar ponds in hot climates. Solar & Wind Technol
1989;6:13741.
[25] Bryant HC, Colbeck I. A solar pond for London? Sol Energy 1977;19:3212.
[26] Bryson K. Experimental study of the stability of volatiles on Mars PhD
Thesis. Arkansas, USA: University of Arkansas; 2008.
[27] Holman JP. Heat Transfer. New York: McGraw-Hill Inc.; 1997.
[28] Sodha MS, Kaushik ND, Rao SK. Thermal analysis of three zone solar pond. Int J
Energy Res 1981;5:32140.
[29] Swinbank WC. Long wave radiation from clear skies. Quarter J Roy Meteorol
Soc 1963;89:3358.

B.M. Ziapour et al. / Energy Conversion and Management 136 (2017) 283293
[30] Vergara J, Garrido F. Design of solar pond for water preheating used in the
copper cathodes washing at a mining operation at Sierra Gorda, Chile. J Renew
Sustain Energy 2013;5:116.
[31] Alenezi I. Salinity gradient solar ponds: theoretical modelling and integration
with desalination. Guildfold (Surrey, England): Chemical Engineering Faculty
of Engineering and Physical Sciences University of Surrey; 2012.
[32] Husain M, Patil PS, Patil SR, Samdarshi SK. Optimum size of non-convective
zone for improved thermal performance of salt gradient solar pond. Sol Energy
2003;74:42936.
[33] Trieb F, Langniss O, Klaiss H. Solar electricity generation- a comparative
view of technologies, costs and enviromental impact. Solar Energy 1997;59:
8999.
[34] Xi H, Luo L, Fraisse G. Development and applications of solar-based
thermoelectric technologies. Renew Sustain Energy Rev 2007;11:92336.
[35] Niu X, Yu J, Wang S. Experimental study on low-temperature waste heat
thermoelectric generator. J Power Sources 2009;188:6216.
[36] Faraji AY, Date A, Singh R, Akbarzadeh A. Base-load thermoelectric power
generation using evacuated tube solar collector and water storage tank. Energy
Procedia 2014;57:211220.
[37] Vorobiev Y, Hernandez JG, Vorobiev P, Bulat L. Thermal-photovoltaic solar
hybrid system for efficient solar energy conversion. Sol Energy 2006;80:
1706.
[38] Engineering Equation Solver (EES), Academic Version, <www.fchart.com>.

293

[39] Saadat M. Power generation enhancement in a salinity gradient solar pond

using thermoelectric generator Master thesis. Ardabil (Iran): University of
Mohaghegh Ardabili; 2016.
[40] Mohammadi A, Kasaeian A, Pourfayaz F, Ahmadi MH. Thermodynamic analysis
of a combined gas turbine, ORC cycle and absorption refrigeration for a CCHP
system. Appl Therm Eng 2017;111:397406.
[41] Hawas MM, Elasfouri AS. Performance of a salt-gradient solar pond for power
production. Energy Convers Manag 1985;3:32330.
[42] Akbarzadeh A, Johnson P, Singh R. Examining potential benefits of combining a
chimney with a salinity gradient solar pond for production of power in salt
affected areas. Sol Energy 2009;83:134559.
[43] Ziapour BM, Shaker H. Heat transfer characteristics of a two-phase closed
thermosyphon using different working fluids. Heat Mass Transf
2010;46:30714.
[44] Sureshkumar R, Mohideen ST, Nethaji N. Heat transfer characteristics of
nanofluids in heat pipes: a review. Renew Sustain Energy Rev 2013;20:
397410.
[45] Leblanc J, Akbarzadeh A, Andrews J, Lu H, Golding P. Heat extraction methods
from salinity-gradient solar ponds and introduction of a novel system of heat
extraction for improved efficiency. Sol Energy 2011;85:310342.
[46] Ziapour BM, Shoknia M. Exergoeconomic analysis of the salinity-gradient solar
pond power plants using two-phase closed thermosyphon: a comparative
study. Appl Therm Eng 2017;115:12333.