Applied Energy 213 (2018) 343–352

Contents lists available at ScienceDirect

Applied Energy
journal homepage: www.elsevier.com/locate/apenergy

Performance study on a passive solar seawater desalination system using T

multi-effect heat recovery
⁎
Shuang-Fei Lia, Zhen-Hua Liua, , Zhi-Xiong Shaob, Hong-shen Xiaoc, Ning Xiab
a
School of Mechanical Engineering, Shanghai Jiaotong University, Shanghai 200240, China
b
Sunshore Solar Energy Co., Ltd, Nantong 226000, China
c
School of Mechanical Engineering, Nantong Vocational University, Nantong, China

H I G H L I G H T S

• ASystem
solar seawater desalination system with multi-effect heat recovery processes.
• Mechanismuses all-glass evacuated tube absorber as heat collector with a simplified CPC.
• The system operated
of its operation are multi-stage distillation and heat recovery.
• under barotropic or atmospheric pressure without power consumption.

A R T I C L E I N F O A B S T R A C T

Keywords: A novel small-sized solar seawater desalination system with multi-effect heat recovery processes using all-glass
Solar energy evacuated tube absorber as heat collector, in which there is no power pump and the steam and freshwater flow
Seawater desalination are driven only by pressure drop, was designed and tested. The whole system consists of 7 heat collecting/heat
Heat recovery recovery integration units, which were divided into 7 temperature/pressure states. Each unit has a heat collector
CPC
which consists of a simplified CPC panel, an all-glass evacuated tube absorber, a seawater tank and a bar heat
pipe that connects the absorber and seawater tank to transfer heat from the absorber to the seawater tank. Every
unit operated under barotropic or atmospheric pressure. Meanwhile, a stepwise heat recovery method was
adopted to recycle the sensible heat and latent heat of the steam generated. In order to investigate the effects of
operating parameters on system performance, including freshwater yield, solar collecting performance and heat
recovery performance, a series of experiments were conducted under different weather condition. It can be found
that the all-day freshwater yield of unit area can reach 4.23 kg/m2 on the sunny day and 3.03 kg/m2 on the
cloudy day. Meanwhile the collecting efficiency and comprehensive thermal coefficient can reach 0.41 and 1.39
respectively. The experiment results confirm that the designed system has a superior performance in seawater
desalination without power consumption.

1. Introduction sustainable development and long term stability. Therefore, it is very

Water is the most active and influential factor in the ecological
environment and it is also one of the most valuable resources in the
world. However, about 97% of the Earth’s water is salt water in the
ocean, and a tiny 3% is fresh water [1]. At present, about 1.5 billion
people in more than 80 countries around the world face the shortage of
fresh water. By 2025, it is projected that there will be 3 billion people
lack of water, involving more than 40 countries and regions. Water
resource is becoming a valuable resource in the twenty-first Century; as
a result, the problem of water resource is not only a resource problem,
but also a major strategic issue related to the national economy, social
urgent to solve the problem of water shortage. Desalinating salt water
from the ocean, river, lakes is the best way to supply fresh water to
growing population.
Seawater desalination techniques are mainly divided into two ca-
tegories on the basis of different elements of fresh water production,
namely, the membrane processes, for which reverse osmosis (RO) and
electrodialysis (ED) were utilized [2,3] and the thermal processes
whose thermal energy may be obtained from a conventional fossil-fuel
source, nuclear energy or from a non-conventional solar energy, etc.
[4–6]. Moreover, the thermal processes are broadly classified into two
major categories, direct and indirect systems. The indirect method of

⁎
Corresponding author.
E-mail address: liuzhenh@sjtu.edu.cn (Z.-H. Liu).

https://doi.org/10.1016/j.apenergy.2018.01.064
Received 10 November 2017; Received in revised form 2 January 2018; Accepted 22 January 2018
0306-2619/ © 2018 Elsevier Ltd. All rights reserved.
S.-F. Li et al.
Nomenclature
Acpc horizontal projection area of CPC panel (m2)
Gm freshwater yield (kg/h)
Gm,all all-day freshwater yield of the system (kg)
Gm′ freshwater yield of unit collecting area (kg/(h m2))
Δh enthalpy difference (J/kg)
n number of collecting units in the system
PR performance ratio of desalination
qr irradiance (W/m2)
Q power (W)
Q̇ useful power extracted from collector (W)
ηre heat recovery efficiency
ηt heat collecting efficiency
ε comprehensive thermal coefficient
Subscripts
1 Unit 1
solar desalination plant consists of two subsystems, solar collector and
desalination unit. The different types of solar collector such as evac-
uated tube, flat plate and heat pipe can be used along with the thermal
desalination processes such as multiple effect evaporation (MEV), vapor
compression (VC) and multi-stage distillation (MSD) [7]. In addition,
the solar distillation systems can be classified as passive solar still and
active solar still. In a passive solar still, the solar radiation is received
directly by the basin water and is the only source of energy for raising
the water temperature. However, in an active solar still, an extra
thermal energy is supplied to the basin through an external mode to
increase the evaporation or a vacuum pump is used to maintain the
vacuum working condition.
The water demand is increasing rapidly than the sustainable level
and desalination is the best method to provide the shortfall of water.
From the perspective of energy consumption, most of the desalination
systems [8–11] are energy intensive, which consume high grade energy
like gas, electricity, oil and fossil fuels accelerating the global warming
that is the burning topic and becomes threat to life sustainability. Re-
newable energy is the alternative solution to decreasing consumption of
fossil fuels [12]. So far, there are already several kinds of renewable
energy applied to seawater desalination such as wind energy [13,14]
and solar energy. Considering that the solar energy can be efficiently
harvested for solar to implement the heat application for green and
environmental sustainability, it is one of the most promising applica-
tions of renewable energies in thermal desalination processes. So far,
there have been many solar desalination researches carried out as fol-
lows.
Many researchers have carried out research on solar humidifica-
tion/dehumidification desalination system (HDDS) [15,16]. However,
the performance of HDDS seriously depends on the thermal properties
of working medium and the humidification/dehumidification ability is
weak at low temperature difference. Besides the HDDS, Khamid Mah-
kamov et al. [17] developed and tested an innovative small dynamic
water desalination plant that is a combination of a heat pipe evacuated
tube solar collector, conventional condenser and novel fluid piston
converter. Young-Deuk Kim et al. [18] proposed a hybrid desalination
system consisting of vacuum membrane distillation (VMD) and ab-
sorption desalination (AD) units, designated as VMD-AD cycle. Nema-
tollahi et al. [19] carried out an experimental and theoretical energy
and exergy analysis for a solar desalination system consisting of a solar
collector and a humidification tower. In addition, the multi-effect de-
salination system attracts many researchers due to its high-energy ef-
ficiency. Zhili Chen et al. [20] designed a multi-stage stacked-tray solar
seawater desalination still and test water production performance in
344
Applied Energy 213 (2018) 343–352
2 Unit 2
3 Unit 3
4 Unit 4
5 Unit 5
6 Unit 6
7 Unit 7
Sys system
Ac actual
Th theoretical
Acronyms and abbreviations
CPC compound parabolic concentrator
MED multi-effect distillation
MEH/D multi-effect humidification/dehumidification
both transient and steady states. Meanwhile, a mathematical model of
heat and mass transfer was developed. Patricia Palenzuela et al. [21]
conducted work to analyze whether the integration of Multi-Effect
Distillation (MED) process into Concentrating Solar Power (CSP) plants
can be more competitive, under certain conditions, than the in-
dependent freshwater and power production by connecting a Reverse
Osmosis (RO) system to a CSP plant. Guo Xie et al. [22] presented a
novel conceptual design of a low temperature multi-effect desalination
(LT-MED) system and is composed of several modules of tubular solar
still (TSS) cells. Reddy et al. [23] developed a novel multi-effect evac-
uated solar desalination system utilizing latent heat recovery and in-
vestigated the effect of various design and operating parameters on the
system performance to optimize the configuration. Estahbanati et al.
[24] conducted experiment to investigate the effect of the number of
stages on the productivity of a multi-effect active solar still. Dayem [25]
conducted work to demonstrate experimentally and numerically the
performance of a simple solar distillation unit that is based on the
multiple condensation-evaporation cycle. Chorak et al. [26] presented
and discussed an experimental characterization of the solar MED system
under off-design conditions.
So far, a lot of research relate to the MED systems have been pub-
lished, however, many of them need auxiliary energy consumption
during runtime. In addition, most of them have large floor space and
high cost. In order to design a MED system that is suitable for the island
and small fishing boats where the solar energy is rich, but electricity,
fresh water and fossil fuel are rare, our research team developed a
principle passive solar desalination system with multi-effect evapora-
tion/heat recovery processes and verified the feasibility of the device
[27], which is the mechanism research for the system presented in this
paper. Furthermore, Liu ZH carried out some modification measures to
optimize the principle system and considered more operating para-
meters for analyzing the system performance [28]. This system in-
tegrated solar collecting, seawater evaporation, heat recovery and
condensation processes into the common evacuated tubular solar col-
lector to produce freshwater directly. The system operated under bar-
otropic and atmospheric pressure and adopted a stepwise heat recovery
method to recycle the sensible heat and latent heat of the steam gen-
erated. However, for this principle system, it is difficult to develop a
commercial product due to its especial fabrication.
In this research, based on the principle system, a small-sized solar
seawater desalination system with multi-effect heat recovery processes
using all-glass evacuated tube absorber as heat collector, in which there
is no power pump and the steam and freshwater are driven only by
pressure drop, is designed and tested. The originality of this system
S.-F. Li et al. Applied Energy 213 (2018) 343–352

consists of the following aspects: (1) The most originality is that this is a
novel passive system without pumps. (2) The heat pipe was utilized as
the heat transfer element, which realizes the rapid heat transfer be-
tween the collector section and the desalination unit. (3) A stepwise
heat recovery method was adopted to recycle the sensible heat and
latent heat of the steam generated, which is realized by several pressure
regulating valves.
In order to investigate the effects of operating parameters on system
performance, including freshwater yield, solar collecting performance
and heat recovery performance, a series of experiments were conducted
under different weather and irradiance conditions. Through the ex-
periments, the heat collection performance, heat recovery performance
and desalination performance have been discussed and analyzed. The
experiment results confirm that the designed system has a superior
performance in seawater desalination without power consumption. The
experimental research has a significant reference value for the devel- Fig. 2. A photograph of solar seawater desalination system with 7 collecting units.
opment of a passive small-sized solar desalination system.

2. Experimental apparatus and procedure

2.1. Experimental apparatus

The designed seawater desalination system is made up of 7 heat

collecting/heat recovery integration units, which were divided into 7
temperature/pressure states. Fig. 1 gives out a schematic view of the
whole system for every units and connection relationship in which the
saturated temperature of steam generated at every unit is illustrated,
and Fig. 2 shows a photograph of the actual experimental apparatus.
As shown in Fig. 1, the designed system mainly consists of seven
linked units for heat collecting/recovery, air cooler and freshwater
collector. The apparatus is located in Tongzhou District, Nantong,
Jiangsu, China. The precise location is “N32°06′45.38″, E121°0′17.36″”.
In addition, the average local irradiance is above 500 W/m2 during
work time (from 9:00 to 15:00). In the first unit, the energy used to
evaporate seawater is only obtained from solar energy. However, in the
remaining 6 units, the steam produced at the upper unit releases the
latent heat to the seawater in the next unit through the heat recovery
coil tube. The steam/water mixture outlet of the last unit is connected
with the naturally convective air cooler that is a passive cooler and has
no energy consumption to condense the final steam-water mixture. And
then, the condensed freshwater was stored in the freshwater collector.
The Units 2–7, as depicted in Fig. 3, have the same structure con- Fig. 3. Structures of the unit (red point: measuring point of thermocouples). (For inter-
taining a seawater tank, a pressure reducing valve, a simplified CPC pretation of the references to colour in this figure legend, the reader is referred to the web
panel, a bar heat pipe, an all-glass evacuated tube, a seawater inlet, a version of this article.)

steam/water mixture inlet, a steam outlet tube, a steam/water mixture

outlet and a heat recovery coil tube. In every unit, evaporation section
of heat pipe is concentrically mounted into the all-glass evacuated tube,
meanwhile, the condensation section of heat pipe is concentrically
mounted into the heating tube fixed in the center of the seawater tank.
The first unit (Unit 1), as a simple heat collector without heat recovery
process, doesn’t have a heat recovery coil tube installed inside.
By such arrangement, the collected solar energy can be rapidly
transferred into the seawater tank by using a bar heat pipe to heat the
seawater under small temperature difference. The steam generated
from unit 1 flows into the heat recovery coil tube fixed in the seawater
tank of unit 2 through the steam/water mixture inlet. Similarly, from

Fig. 1. Designed system for every units and connection chart Blue line: seawater, red line: saturated water, green line: vapor. (For interpretation of the references to colour in this figure
legend, the reader is referred to the web version of this article.)

345
S.-F. Li et al. Applied Energy 213 (2018) 343–352

unit 2 to unit 7, in each seawater tank, there is a steam/water mixture experiment, the steam temperature and pressure were changing all the
inlet and a steam/water mixture outlet to conduct both the steam time with this control method. As a compromise, this control method is
generated and the water condensed into the heat recovery coil tube feasible and has the advantage of low cost and simple operation.
fixed in the seawater tank of next unit. Therefore all latent heat of the In the present stage, a naturally convective air cooler was used for
steam can be recovered in the last six units for the evaporation of final steam-water mixture condensation and there is not energy con-
seawater. In this designed system, a steam condenser that consists of sumption in this passive heat exchanger.
coil tubes and use natural convective heat transfer of air to cool fluid in In this experimental system, the steam temperatures, the wall
tube, is installed at the final stage in the system to fully condense the temperatures of heat pipe, the temperatures of steam/water mixture
rest steam from the last unit and the freshwater is finally stored in the inlet and outlet were measured for investigating both the solar col-
freshwater collector. In addition, the cumulative freshwater influx was lecting and heat recovery performances. The measuring positions of
measured every 15 min to get the mean freshwater yield. thermocouples are displayed in Fig. 3. For each heat pipe, the tem-
The irradiance, steam temperatures and wall temperatures of the peratures at evaporation section, adiabatic section and condensation
tubes in each unit were measured, collected and processed by the data section were monitored to ensure the heat pipe is in normal working
acquisition system. Data were logged using Agilent 34970A data-logger condition. For each unit, the steam temperatures were measured to
connected to the computer. monitor its operating temperatures. From unit 2 to unit 7, the tem-
peratures of steam/water mixture inlet and outlet were measured to
2.2. Geometry design of simplified CPC study the heat recovery performance. In addition, a TES-133R solar
radiometer was utilized to measure the solar irradiance. As long as the
For obtaining a relatively high collecting temperature and steam measuring surface of the solar radiometer was placed in parallel with
pressure, a CPC-type concentrator was utilized in this study which is the the heat collecting unit, the real-time solar irradiance perpendicular to
same as that in the authors’ principle study [28]. The design method of the CPC plate can be obtained.
standard CPC for tubular all-glass evacuated tube comes from Winston
[29] and Rabl [30–32]. In addition, Alaydi [33] and Kalogirou [34] 2.4. Uncertainty analysis
applied parabolic solar-collector for seawater desalination, and Lillo
et al. [35] evaluates the potential of solar concentration technologies as Five main parameters are considered during the test, including in-
an alternative to conventional sources of energy for industrial processes stantaneous freshwater yield, performance ratio, heat collecting effi-
in Latin America. For reducing CPC material and increasing collecting ciency and heat recovery efficiency, which are calculated as follows:
time, the standard CPC was cut to a truncated CPC. Then, for further
reducing machining cost, bottom involute shape was cut to flat shape. m
Gm =
Fig. 4 presents a finally simplified CPC profile after truncation and in- t (1)
volute shape cutting. According to the calculation and test results, the
Gm Δh
concentration efficiency of the simplified CPC with a flat curve bottom PR =
Q̇ (2)
decreases by about 15% compared with the truncated CPC with an
involute bottom [36].
Gm′ h
The CPC is made of a stainless steel mirror plate welded on the steel ηt ,exp =
Acpc qr (3)
shelf directly. For the present tubular absorber, the concentration ratio
of the CPC is defined as the ratio of the aperture area of the CPC to the
Gm Δh
peripheral area of the tubular absorber. According to this definition, the ε=
concentration ratio of the present CPC is 2.16 which does not depend on ACPC nqr (4)
the acceptance half angle of standard CPC.
Gm,ac Gm,all−7Gm,1
The characteristics of the presented collector part including CPC, ηre = =
all-glass evacuated tube and bar heat pipe are shown in Table 1. Gm,th 21Gm,1 (5)

where m is the freshwater yield within 15 min, t is the time interval,

2.3. Experimental procedure
t = 15 min = 0.25 h.
According to the relative standard deviation of the transfer function
In the experimental process, NaCl solution with a mass fraction of
theory, the maximum relative deviation of instantaneous freshwater
3.5% and temperature of 20 °C–25 °C is used to substitute sea water. In
yield, performance ratio, heat collecting efficiency and heat recovery
addition, the ion level of Na in the freshwater and NaCl solution have
efficiency can be derived from Eqs. (1)–(4) as follows:
been measured by Inductively Coupled Plasma Optical respectively and
the value is 0.0138% for the freshwater and 3.414% for the NaCl so- σGm 2 2
= ⎛ ∂lnGm ⎞ σm2 + ⎛ ∂lnGm ⎞ σt2
lution. Then, the temperature and pressure of the saturated steam in Gm ⎝ ∂m ⎠ ⎝ ∂t ⎠ (6)
each unit were controlled by the pressure reducing valves. All these
valves were regulated only once at noon when the direct irradiance was
approximately strongest. And then, the highest temperatures of the
steam in each unit were dependent on their own without the further
adjustment of the valves. Because the designed solar desalination
system was demanded to run all year round, the highest collecting
temperature should not be too high and could be achieved even in the
winter. Meanwhile, it should achieve six-effect heat recovery, and the
temperature gradient of two adjacent units should not be too large.
Since the steam temperature in unit 7 is 100 °C corresponding to the
atmospheric pressure, the ideal distribution of the steam temperature is
that the steam temperature in unit 1 is controlled at about 150 °C and
decreased by 8 °C at each next unit. According to the preparatory ad-
justment, the maximum difference between the two adjacent units was
Fig. 4. Simplified CPC profile for an all-glass evacuated tube (unit: mm).
set at 8 °C. As a matter of fact, with the irradiance changing during the

346
S.-F. Li et al. Applied Energy 213 (2018) 343–352

Table 1 long term test data of all-day freshwater yield of unit collecting area
Characteristics of the collector system. where the temperature represents the average environment tempera-
ture during the test process. For the laboratory prototype stage, even if
Parameter Value
the device runs continuously, it is not necessary to carry out perfor-
CPC mance tests every day. The performance tests were conducted randomly
Width of CPC: 330 mm according to the weather condition. In addition, the latest 5 sets of test
Effective length of CPC 1600 mm
data are adopted to further analyze the performance of freshwater
Height of CPC: 146 mm
Aperture area of a CPC plate 0.512 m2 production including freshwater yield and performance ratio, the steam
Radius of absorber 24 mm temperatures and the system efficiency including heat collecting effi-
Distance of CPC cusp and absorber 6 mm ciency, comprehensive coefficient and heat recovery efficiency have
Acceptance half angle of standard CPC 18o been discussed.
Edge-ray angle of truncated CPC 65o
Reflectivity of CPC 0.85
Concentrating ratio 2.16 3.1. Freshwater production performance
All-glass evacuated tube
Cover glass tube diameter 58 mm
The amount of water production is obtained by weighing the
Inner glass tube diameter 47 mm
Effective length glass tube 1600 mm freshwater collected in a bucket. In order to improve measurement
Bar heat pipe accuracy, the actual measurement time interval is 15 min. Therefore,
Outer dia. of bar heat pipe 45 mm the instantaneous value of the freshwater yield is the average value
Inner dia. of bar heat pipe 41 mm
within 15 min and calculated by Eq. (1). The instantaneous value of the
Effective length 2800 mm
Wall thickness 2 mm freshwater yield of unit collecting area is obtained by Eq. (11).
Working fluid Water
Gm
Stainless steel screen Gm∗ =
Material 304 stainless steel Acpc × n (11)
Wire diameter 0.12 mm
Mesh number 150 meshes The calculation of all-day freshwater yield and all-day freshwater
yield of unit collecting area is similar to Eqs. (1) and (11), the difference
is that m is the value of all-day freshwater production; t is the all-day
2 2
working hours.
σPR
= ⎛ ∂lnPR ⎞ σ 2 + ⎛ ∂lnPR ⎞ σ 2
⎜ ⎟ ⎜ ⎟
Gm Q̇
PR ⎝ ∂Gm ⎠ ⎝ ∂Q̇ ⎠ (7)
3.1.1. Analysis of freshwater yield
2 2 Fig. 6 shows the data summary of all-day freshwater yield of unit
σηt,exp
= ⎛ ∂lnηt ,exp ⎞ σG2m′ + ⎛ ∂lnηt ,exp ⎞ σq2r
⎜ ⎟ ⎜ ⎟ area, all-day freshwater yield and the average irradiance for unit 1
ηt ,exp ⎝ ∂Gm′ ⎠ ⎝ ∂qr ⎠ (8) under different weather conditions to shows the effect of alone heat
2 2 colleting capacity on freshwater yield. Meanwhile, Fig. 7 shows the
σε
= ⎛ ∂lnε ⎞ σ 2 + ⎜⎛ ∂lnε ⎟⎞ σ 2
⎜ ⎟
data summary of all-day freshwater yield of unit area, all-day fresh-
Gm qr
ε ⎝ ∂Gm ⎠ ⎝ ∂qr ⎠ (9) water yield and average irradiance for the whole system to show the
multiple effect of both heat collecting and heat recovery on freshwater
2 2
σηre yield. As manifested in Figs. 6 and 7, the solar radiation intensity was
= ⎛⎜ ∂lnηre ⎞⎟ σ 2 ⎛ ∂lnηre ⎞⎟ σ 2
Gm,all + ⎜ Gm,1 closely related to weather conditions. The maximum average irradiance
ηre ∂G
⎝ m,all ⎠ ⎝ ∂Gm,1 ⎠ (10)
during sunny days was about 769.69 W/m2, 26.7% higher than that
In the experiment process of the designed system, the parameters during the cloudy day (607.52 W/m2). Furthermore, it is obvious that
that need to be measured include temperature, fresh water mass and the all-day freshwater yield depends on the average irradiance and is
solar irradiance. In addition, the time also need to be recorded. For the basically proportional to the average irradiance.
measurement of temperature, GG-K-30 thermocouples were utilized For unit 1, there is no heat recovery process, that is to say, the
with ± 0.1 K measurement errors. An electronic scale with ± 0.005 kg freshwater was only produced by the effective heat collection of CPC
measurement errors was adopted to measure the mass of freshwater panel. According to Fig. 6, it can be found that the maximum all-day
obtained in a measuring interval of 15 min. A TES-133R solar radio- freshwater yield can reach 1.2 kg on the sunny day and 0.95 kg on the
meter was utilized to measure the solar irradiance. As long as the cloudy day for one collecting glass tube. Correspondingly, the max-
measuring surface of the solar radiometer was placed in parallel with imum all-day freshwater yield of unit area can reach 2.3 kg/m2 on the
the heat collecting unit, the real-time solar irradiance perpendicular to sunny day and 1.98 kg/m2 on the cloudy day. It can be estimated by
the CPC plate can be obtained and the measurement error was ± 4%. this data that the maximum heat collecting efficiency can reach 41% on
For the record of time, it was achieved by the system time of computer the sunny day and 34% on the cloudy day for even a high collecting
which was consistent with “Beijing time”. In addition, because the re- temperature of 140 °C as shown in Fig. 13. It is obvious that the de-
cord of fresh water mass was manual operation consuming time, there signed system has a good heat collecting performance. For the whole
existed a time measurement error with about ± 3 s. The measurement system, there is a heat recovery process in each unit from unit 2 to unit
errors of the parameters in the experiment are listed in Table 2. 7 to recover the latent heat of steam generated in the previous unit. As
According to Eqs. (5)–(8), the relative errors of instantaneous manifested in Fig. 7, the maximum all-day freshwater yield can reach
freshwater yield, performance ratio, heat collecting efficiency and heat
recovery efficiency are less than 4.3%, 6.1%, 7.4%, 5.9% and 9.7%, Table 2
respectively. Measurement errors for experimental parameters.

Parameters Measurement errors

3. Results and discussion
Temperature ± 0.1 K
Fresh water ± 0.02 kg/h
In order to verify the superior performance of the designed solar
Solar irradiance ± 4%
seawater desalination system, a series of experiments have been con- Time ±3s
ducted under different weather conditions. Fig. 5 shows some typical

347
S.-F. Li et al. Applied Energy 213 (2018) 343–352

8 800 Fig. 5. The long term test data of all-day freshwater

yield of unit collecting area.
2
All-day freshwater yield of unit collecting area /kg/m

7 700

Average solar radiation intensity /W/m

6 600

Average solar radiation intensity

5 500
All-day freshwater yield of unit collecting area
4 400

3 300

2 200

2
1 100

0 0
17-10-15 7-12-15 8-1-16 14-3-16 21-4-16 17-8-16 21-7-16 20-7-16 23-7-16 22-7-16
Autumn Winter Winter Spring Spring Summer Summer Summer Summer Summer
Sunny Sunny Sunny Cloudy Cloudy Sunny Sunny Sunny Cloudy Cloudy
20.7 °C 8.2 °C 3.5 °C 7.6 °C 19.8 °C 30.3 °C 31.1 °C 31.8 °C 32.3 °C 31.5 °C

900 20 900
2.6 Average solar irradiance All-day freshwater yield (kg)
2
2.4 All-day freshwater yield of unit area (kg/m ) 800 18 2
All-day freshwater yield of unit area (kg/m ) 800
All-day freshwater yield of Unit 1

Average solar irradiance /(W/m )

Average solar irradiance /(W/m 2 )

2.2 All-day freshwater yield (kg)

16 Average solar irradiance 700
All-day freshwater yield

700
2.0
14
1.8 600 600
1.6 12
500 500
1.4 10
1.2 400 400
8
1.0 300
300 6
0.8
0.6 200 4 200
0.4

0.2
0.0
17-8-16
Summer Summer Summer Summer Summer
Sunny Sunny Sunny Cloudy Cloudy
Fig. 6. Data summary of all-day freshwater yield of unit area (black column), all-day
freshwater yield (red column) and the average irradiance for unit 1. (For interpretation of
the references to colour in this figure legend, the reader is referred to the web version of
this article.)
17 kg on the sunny day and 12 kg on the cloudy day. Correspondingly,
the maximum all-day freshwater yield of unit area can reach 4.23 kg/
m2 on the sunny day and 3.03 kg/m2 on the cloudy day.
Based on the comparison of the freshwater yield of unit 1 and whole
system, it can be found that the maximum all-day freshwater yield of
unit area can be increased by about 2 times, the mean all-day fresh-
water yield of unit area of whole system can be increased by 1 times for
all days, which is attributed to the heat recovery process. It can be
confirmed that the heat recovery effect was relatively ideal whether the
weather was sunny or cloudy.
3.1.2. Analysis of instantaneous freshwater yield
Fig. 8 depicts the instantaneous freshwater yield of unit collecting
area and irradiance along with time under different weather conditions
to show the effect of instantaneous irradiance on the freshwater yield. It
is clear from Fig. 8 that the irradiance fluctuated violently in cloudy
weather, which leads to the lower average irradiance. On the contrary,
the irradiance fluctuated gently in sunny weather, which leads to
2
2 100
100
0 0
0 17-8-16 21-7-16 20-7-16 23-7-16 22-7-16
21-7-16 20-7-16 23-7-16 22-7-16
Summer Summer Summer Summer Summer
Sunny Sunny Sunny Cloudy Cloudy
Fig. 7. Data summary of all-day freshwater yield (black column), all-day freshwater yield
of unit area (red column) and the average irradiance for the whole system. summary of
all-day freshwater yield of unit area (black col
relatively higher average irradiance. It is also found that the change of
instantaneous freshwater yield was basically in accordance with that of
irradiance, which illustrates that the instantaneous freshwater yield is
very sensitive to the irradiance. The change of irradiance not only
causes change of the heat collecting efficiency, but also causes change
of the heat recovery efficiency which is more important for freshwater
yield. In addition, there was no steam generation before 11:15 whether
the weather was sunny or cloudy, because a large amount of seawater
accumulated in the seawater tank was being heated form a relatively
low temperature to the saturated temperature before noon. When the
weather was sunny, the maximum instantaneous freshwater yield ap-
peared at noon and the value was about 1.51 kg/(h m2). In the after-
noon, the instantaneous freshwater yield decreased gradually as the
time goes on until there was no freshwater generation. When the
weather was cloudy, the instantaneous freshwater yield fluctuated with
the change of irradiance and the maximum value was 1.5 kg/(h m2)
that was close to the value (1.51 kg/(h m2)) of sunny day, even so, it is
obvious that the average value of instantaneous freshwater yield in
cloudy day was much lower than that in sunny day.

348
S.-F. Li et al. Applied Energy 213 (2018) 343–352

2.0 1400 3.0 1400

1300
1.5 1200 2.5 1200
1.0 2.0 1100
Freshwater yield /(kg/h · m )
2

1000 1000

Solar irradiance (W/m 2 )

0.5 1.5

Performance ratio

Solar irradiance (W/m )

900
0.0 800 1.0
800
700
-0.5 600 0.5 600
500
-1.0 400 0.0 400
-1.5 17-8-2016 sunny 17-8-2016 -0.5 300
17-8-2016 sunny 17-8-2016
21-7-2016 sunny 21-7-2016
200 21-7-2016 sunny 21-7-2016 200
-2.0

2
-1.0 100
20-7-2016 sunny 20-7-2016 20-7-2016 sunny 20-7-2016
-2.5 23-7-2016 cloudy 23-7-2016 0 23-7-2016 cloudy 23-7-2016 0
-1.5
22-7-2016 cloudy 22-7-2016 22-7-2016 cloudy 22-7-2016 -100
-3.0 -200 -2.0 -200
11:15 11:45 12:15 12:45 13:15 13:45 14:15 14:45 11:15 11:45 12:15 12:45 13:15 13:45 14:15 14:45
Time /hr Time /hr
Fig. 8. Instantaneous freshwater yield of unit collecting area and irradiance changes with Fig. 9. Performance ratio and irradiance changes with time.
time.

the solar radiation was stronger, the temperature difference between

3.1.3. Analysis of performance ratio
The performance ratio (PR) is an important performance factor to
evaluate the performance of desalination system and it is defined as the
ratio of the total energy used to evaporate seawater into freshwater
including the energy recovered to the net energy absorbed from the
collector. PR represents the heat recovery performance of the system
and it would be 1 if there is no heat recovery effect. The theoretical
value of PR is determined by the system structure and calculated by the
Eq. (2).
Assuming that the collecting power of each panel, Q̇ , is the same and
the latent heat of vaporization of water can be fully recovered, then the
power used to evaporate seawater in each unit is as follows:
Q1 = Q̇
Q2 = Q̇ + Q1 = 2Q̇
adjacent units became larger, meanwhile, the steam flow rate was also
larger, the heat recovery performance was better, and therefore, the
value of PR was relatively large. On the contrary, when the solar ra-
diation was weaker, the temperature difference between adjacent units
became smaller, meanwhile, the steam flow rate was smaller, the heat
recovery performance deteriorated, which led to the rapid decrease of
PR.
According to the data from Fig. 9, it is found that the actual PR is
much less than the theoretical value and there is great improvement
space for the heat recovery capacity by increasing the heat transfer area
of coil heat exchanger installed in the seawater tank which will only
cause small cost increase. If the actual PR value can increase 50%, that
(12a) means the maximum instantaneous value reaching 4 and mean value of
all-day reaching 3, then, the freshwater yield will increase 50%.
(12b)

Q3 = Q̇ + Q2 = 3Q̇ (12c) 3.2. Analysis of steam temperature difference between each unit

Q4 = Q̇ + Q3 = 4Q̇
Q5 = Q̇ + Q4 = 5Q̇
Q6 = Q̇ + Q5 = 6Q̇
Q7 = Q̇ + Q6 = 7Q̇
where, Q1 ∼ Q7 are the total power used to evaporate seawater in unit
1 ∼ unit 7, respectively. Q̇ is the net power from collector in each unit.
Hence, the maximum theoretical value of the PR for the system is cal-
culated as follows:
Q1 + Q2 + Q3 + Q4 + Q5 + Q6 + Q7 28Q̇
PR = = =4
7Q̇ 7Q̇
Due to the existence of heat recovery process, the PR will be bigger
than 1, and the bigger the value is, the better the heat recovery effect
will be. In this designed system, the maximum theoretical value of PR is
4 according to Eq. (12a-h).
Fig. 9 shows the actual performance ratio curves and irradiance
curves along with time under different weather conditions. It can be
noticed that the fluctuation of instantaneous value of PR along with
time was similar to that of instantaneous freshwater yield that is closely
related to the irradiance. According to Fig. 9, on the sunny day, the
actual PR value increased gradually with the increase of irradiance
before 12:15 and almost remained stable during the 12:00 – 13:15
period, and then decreased gradually with the decrease of irradiance
after 13:15. The maximum instantaneous value of PR can reach 2.5 at
the noon, and mean value of PR for all-day is close to 2. The change of
PR mainly depends on the change of heat recovery performance. When
(12d) The designed system operates in passive mode and both freshwater
and steam flow depends on the steam temperature difference (the sa-
(12e) turated steam pressure difference) between each unit. Therefore, it is of
vital importance to understand the steam temperature difference be-
(12f) tween each unit.
Figs. 10 and 11 show the steam temperature curves of each unit and
(12g)
irradiance curves along with time on sunny and cloudy day in summer,
respectively. As depicted in Figs. 10 and 11, in the morning, the water
accumulated in the system was being heated from a relatively low
temperature to the saturated temperature, hence there was no steam
generation during this period and the steam temperature was below
100 °C. The amount of heat required to heat the total seawater with a
maximum value of 34 kg in seven seawater tanks from initial 20 °C to a
(12h)
mean saturated temperature of 120 °C of the whole system takes a
preheating time of about 2.4 h. This preheating time is too long for an
actual application. Since this study focused on the collecting and heat
recovery performance, thence had not used auxiliary preheating device.
When the steam temperatures of each unit reached above 100 °C at
about 11:15, the steam started to emerge. At noon, the steam tem-
peratures and steam pressures in each unit reached the maximum set
value and the temperature difference between two adjacent units can
reach the designed value of 8 K on sunny day. In the afternoon, the
temperature difference gradually narrowed with the decrease of solar
radiation.
In the next study, the system will install a passive type of seawater
preheater to preheat the cold seawater to hot seawater with about
90 °C–95 °C for experiment in the following day. If the temperature of
seawater in the preheater can hold at about 70 °C in the morning of the
following day, then about 1.2 h preheating time will be saved and the

349
S.-F. Li et al. Applied Energy 213 (2018) 343–352

190 efficiency of the unit 1 which is a simple heat collector without heat
17-8-2016 900
180 Sunny recovery part was used for representing the system collecting efficiency
170 and its efficiency may be slightly lower than the actual system effi-
600 ciency due to its higher collecting temperature compared with other
Steam temperature (

Solar irradiance (W/m )

160 Unit 3 Unit 4
Unit 2
Unit 5 units. The heat collecting efficiency is defined as the ratio of the net
150 Unit 6
300 energy absorbed by the seawater in seawater tank to the total energy
140 Unit 1
Unit 7 incident on collector, and its instantaneous value is calculated by Eq.
130 (3).
0
120 Fig. 12 plots the instantaneous heat collecting efficiency curves of
110 unit 1, which is a mean measured value in 15 min and may represent

2
-300 mean heat collecting efficiency of system, and instantaneous irradiance
100
curves along with time under different weather conditions. According
90
11:00 11:30 12:00 12:30 13:00 13:30 14:00 14:30 15:00 to Fig. 12, on the sunny day, the heat collecting efficiency increased
Time /hr gradually in the morning, on the contrary, it decreased gradually in the
afternoon. On the cloudy day, the heating collecting efficiency fluc-
Fig. 10. The curves of steam temperatures in each unit changing with time on the sunny tuated acutely with the fluctuation of irradiance. The maximum value
day.
of the instantaneous heat collecting efficiency can reach 0.42 on the
sunny day and 0.34 on the cloudy day. It is obvious that the designed
180 900 system possesses superior instantaneous heat collecting performance
20-7-2016
170 Cloudy during a short noon time. However, the heat collecting efficiency de-
160 creased rapidly in the afternoon. One of the reasons is that the solar
600 radiation decreased in the afternoon. The second reason is the structure
Steam temperature (

Unit 2 Unit 3
Solar irradiance (W/m )

150
Unit 4 of CPC can’t make all the solar energy be projected to the CPC panel,
140 which reduces the collecting efficiency and it is also a limitation of non-
Unit 1 Unit 5 Unit 6
130 Unit 7 300 tracking type collectors. But in terms of practical and civilian applica-
120 tions, this limitation is acceptable. In the present system, the acceptance
half angle of standard CPC and the edge-ray angle of truncated CPC are
110
0 18° and 65°, respectively. If the acceptance half angle of standard CPC
100
2

90
80
11:00 11:30 12:00 12:30 13:00 13:30 14:00 14:30 15:00
Time /hr
Fig. 11. The curves of steam temperatures in each unit changing with time on the cloudy
day.
effective steam generating time will prolong 1.2 h, which means the
freshwater yield will increase about 30% according to the present
system performance.
By comparing Fig. 8 with Fig. 10 and Fig. 11, it is noticed that the
changing regularity of steam temperature difference was similar to that
of instantaneous freshwater yield. When the steam temperature differ-
ence kept stable, the instantaneous freshwater yield was also basically
stable. And when the steam temperature difference fluctuated, the
fluctuation also occurred in the instantaneous freshwater yield. This is
because the instantaneous freshwater yield is closely related to the heat
recovery capacity, which strongly depends on the steam temperature
difference. Furthermore, the collecting power reached the maximum at
noon when the irradiance reached peak and at this time, the steam yield
was the largest and the steam flow rate was large enough to ensure the
set pressure difference. However, in the afternoon, with the irradiance
gradually decreasing, the steam temperatures in the upper collecting
increase by 1 time with the same concentration ratio, the highest col-
lecting temperature at noon will have a decline, but, both the all-day
-300 collecting time and the mean heat collecting efficiency of all-day will
significantly increase. According to the experimental results, the
highest collecting temperature was not the key issue for the designed
system; therefore, it’s meaningful to further improve CPC design.
3.3.2. Comprehensive thermal coefficient
Besides the heat collecting efficiency, the comprehensive thermal
coefficient is also an important parameter for the solar desalination
system with heat recovery process which reveals the total thermal
performance of the system and is defined as the ratio of the total energy
used to evaporate seawater into freshwater to the total energy incident
on the solar collector. The parameter can be seen as an integrated in-
dicator to judge the solar energy collection and heat recovery perfor-
mance of the system, in addition, the value of comprehensive thermal
coefficient may be greater than 1 due to the existence of heat recovery
process and is calculated by Eq. (4).
Fig. 13 shows the comprehensive thermal coefficient curves and
irradiance curves along with time under different weather conditions. It
1400
0.40
1300
0.35 1200
1100
Collecting efficiency

Solar irradiance (W/m )

units gradually decreased, which led to the rapid decrease in the steam 0.30
1000
temperature differences between the two adjacent units. Therefore, it is
0.25 900
very difficult that actual mean PR of all-day reaches the theoretical
value. 0.20 800
700
0.15 600
3.3. Analysis of system efficiency 17-8-2016 17-8-2016 exp.
0.10 500
2

21-7-2016 21-7-2016 exp.

20-7-2016 20-7-2016 exp. 400
3.3.1. Heat collecting efficiency 0.05 23-7-2016 23-7-2016 exp. 300
The system heat collecting efficiency is an important indicator to 22-7-2016 22-7-2016 exp.
judge the solar energy collecting performance of a solar collecting 0.00 200
11:15 11:45 12:15 12:45 13:15 13:45 14:15 14:45
system. Because the present system has heat recovery process in the Time /hr
whole heat transfer process, the system collecting efficiency cannot be
Fig. 12. The collecting efficiency changing with time.
simply measured by using the final steam yield. The collecting

350
S.-F. Li et al. Applied Energy 213 (2018) 343–352

1000 transfer area and improvement of CPC design will be our priority in the
2.4
900
Comprehensive thermal coefficient

2.2 800 following researches.

2.0 700
1.8 600 3.4. The comparison of performance between the tested system and

Solar irradiance (W/m )

1.6 500
1.4 conventional system
400
1.2 300
1.0 200 Table 3 shows the comparison of performance between the tested
0.8 100 system and conventional system and Table 4 lists the main efficiencies
0.6 0
17-8-2016 17-8-2016 -100 of the tested system. It can be noted from Table 3 that the conventional
0.4
21-7-2016 21-7-2016 -200 system have the disadvantages of complex structure, high energy-con-
0.2
20-7-2016 20-7-2016 -300

0.0
-0.2 23-7-2016
-0.4 22-7-2016
11:30 12:00 12:30 13:00
Time /hr
Fig. 13. The comprehensive thermal coefficient for the system.
is affected by heat collection and heat recovery performance of the
system, and the irradiance. For the designed system in this study, the
temperatures of each unit were getting lower and lower with the de-
crease of solar irradiance as the time went on (as shown in Figs. 10 and
11). As a result, the total heat used to evaporate the seawater and the
amount of steam decreased, which lead to the decrease of latent heat of
steam and the deterioration of heat recovery performance. However, it
is worth mentioning that the comprehensive thermal coefficient is not
proportional to the irradiance or the collecting efficiency as shown in
Fig. 13, and it holds a long stable period for 90 min in which the
maximum value is 1.32 on the sunny day and 1.39 on the cloudy day.
Though the maximum comprehensive thermal coefficient on the sunny
day is smaller than that on the cloudy day, it is obvious the average
comprehensive thermal coefficient on the sunny day is much larger
than that on the cloudy day. In addition, compared with the maximum
heat collecting efficiency of 0.42 on the sunny day shown in Fig. 12, the
comprehensive thermal efficiency increases 2 times.
3.3.3. Heat recovery efficiency
The heat recovery efficiency is an important indicator to judge the
heat recovery performance of a solar seawater desalination system. For
the designed system in this research, the heat used to evaporate sea-
water comes from two parts including heat collection and heat recovery
in each unit. The heat recovery efficiency is defined as the ratio of
actual freshwater yield (Gm,ac ) generated by heat recovery to the theo-
retical freshwater yield (Gm,th ) generated by heat recovery and calcu-
lated by Eq. (5).
For instantaneous heat recovery efficiency, Gm,all and Gm,1 denote the
freshwater yields of the whole system and the unit 1, respectively, in
unit time. For all-day mean value, Gm,all and Gm,1 denote the freshwater
yields of the whole system and the unit 1, respectively, in all-day. In Eq.
(7), the freshwater yield of each unit is similarly substituted by that of
2
sumption, large size and cost, etc. Though most of them possess high
23-7-2016 -400
-500
freshwater capacity, they are not suitable for the area where energy is
22-7-2016
-600 only solar energy. For the MED system presented in this paper, it is
13:30 14:00 14:30 15:00 designed to serve island and small fishing boats where solar energy is
rich, but electricity, fresh water and fossil fuel are rare. Though the
freshwater capacity of the presented system is lower than that of con-
ventional system, it possesses more practical value for island and small
fishing coats due to its simple structure, low cost, small floor space and
no auxiliary energy consumption. In addition, the performance im-
provement and structure optimization will be conducted in the further
study to promote its commercialization.
4. Conclusions
A passive type of small-sized solar seawater desalination system
with multi-effect heat recovery processes using all-glass evacuated
collector was designed and tested in this study. According to the ex-
perimental results, the designed seawater desalination system reached
primary demand and showed good freshwater production performance.
(1) The all-day freshwater yield of unit area can reach 4.23 kg/m2 on
the sunny day and 3.03 kg/m2 on the cloudy day without power
consumption. Furthermore, the instantaneous system collecting
efficiency can reach 0.41 at the collecting temperature of 140 °C on
the sunny day and 0.34 on the cloudy day. In addition, the per-
formance ratio can reach 2.5 on the sunny day and 2.3 on the
cloudy day. The comprehensive thermal coefficient can reach 1.32
on the sunny day and 1.39 on the cloudy day. In addition, the all-
day mean recovery efficiency can reach 0.38 on the sunny day and
0.29 on the cloudy day.
(2) Experimental study provides useful guidelines for the development
of passive type of high effective solar desalination system. The
present system can still be improved a lot due to the various design
defects found in the experiment. The freshwater yield may be
greatly increased through the increase of the heat transfer area in
the heat recovery exchanger. In addition, design improvement of
the CPC to increase collecting time. As mentioned above, preheat of
the seawater in the seawater tank using passive steam condenser in
All-day average efficiency of heat recovery

0.60 800
unit 1.
0.55
Average solar irradiance (W/m )

The all-day freshwater yields of unit area for the unit 1 which has 700
0.50
not heat recovery equipment and the whole system (7 units) have been
0.45 Average solar radiation intensity 600
shown in Figs. 7 and 8; therefore, the all-day mean value of heat re-
0.40 All-day average efficiency of heat recovery
covery efficiency may be obtained from the data management. 500
0.35
Fig. 14 illustrates the all-day mean value of heat recovery efficiency
0.30 400
and the mean solar irradiance for five typical weather conditions. Ac-
0.25
cording to Fig. 14, the maximum heat recovery efficiency can reach 300
0.20
0.38 on the sunny day and 0.29 on the cloudy day, which are much less
0.15 200
than the theoretical value of 1. It is obvious that there is a great gap
0.10
between the actual value and the ideal value due to the designed heat 100
0.05
2

transfer area of coil heat exchanger for heat recovery is much less than
0.00 0
the needed heat transfer area. By doubling the heat transfer area, the 17-08-16 21-07-16 20-07-16 23-07-16 22-07-16
Summer Summer Summer Summer Summer
all-day mean values of heat recovery efficiency will be approximately Sunny Sunny Sunny Cloudy Cloudy
doubled without great cost increase. How to improve the all-day mean
Fig. 14. All-day average efficiency of heat recovery and average irradiance.
values of heat recovery efficiency through the increase of the heat

351
S.-F. Li et al. Applied Energy 213 (2018) 343–352

Table 3
The comparison of performance between the tested system and conventional system.

Equipment type Advantages Disadvantages Fresh-water Auxiliary energy consumption

production rate

MEH/D [25] Keep running all day Complex structure; need auxiliary energy; H/D 7.74 L/(d m2) 2000–14,000 kJ/h
ability is weak at low temperature difference
MED-14 effects [26] High freshwater yield High energy-consumption; equipment with large 127.29 L/(d m2) 190 kW
size and cost
2
Hybrid solar/gas MED 24-h operation; operate in different High energy-consumption; equipment with large 5.6 L/(h m ) 150 kW
[37] modes; high freshwater yield size and cost
The present study Simple structure; low cost; no auxiliary Short effective working time; sensitive to weather 4.23 L/(d m2) 0
energy consumption conditions

Table 4 Hydrogen Energy 2016;41(45):20818–22.

The main efficiencies of the tested system.
Heat collecting Comprehensive thermal Heat recovery
efficiency coefficient efficiency
42% 139% 38%
the final stage of the system can also significantly increase fresh-
water yield. These design improvements will be our priority in the
following researches.
References
[1] Glenn EP, Jed Brown J, O'Leary JW. Irrigating crops with seawater. Sci Am-Am Ed
1998;279:76–81.
[2] Garcia-Rodriguez L, Romero-Ternero V, Gomez-Camacho C. Economic analysis of
wind-powered desalination. Desalination 2001;137:259–65.
[3] Miranda MS, Infield DA. wind-powered seawater reverse-osmosis system without
batteries. Desalination 2003;153:9–16.
[4] El-Ghonemy AMK. Water desalination systems Powered by Renewable energy
sources: Review. Renew Sustain Energy Rev 2012;16(3):1537–56.
[5] Li CN, Goswami Y, Stefanakos E. Solar assisted sea water desalination: A review.
Renew Sustain Energy Rev 2013;19:136–63.
[6] Sagie D, Feinerman E, Aharoni E. Potential of solar desalination in Israel and in its
close vicinity. Desalination 2001;139(1–3):21–33.
[7] Shatat M, Worall M, Riffat S. Opportunities for solar water desalination worldwide:
review. Sustain Cities Soc 2013;9:67–80.
[8] Li Weiyi, Krantz William B, Cornelissen Emile R, et al. A novel hybrid process of
reverse electrodialysis and reverse osmosis for low energy seawater desalination
and brine management. Appl Energy 2013;104(2):592–602.
[9] Li C, Besarati S, Goswami Y, et al. Reverse osmosis desalination driven by low
temperature supercritical organic rankine cycle. Appl Energy
2013;102(2):1071–80.
[10] Sayyaadi H, Saffari A. Thermoeconomic optimization of multi effect distillation
desalination systems. Appl Energy 2010;87(4):1122–33.
[11] Al-Sulaiman FA, Narayan GP, John HLV. Exergy analysis of a high-temperature-
steam-driven, varied-pressure, humidification–dehumidification system coupled
with reverse osmosis. Appl Energy 2013;103(1):552–61.
[12] Delyannis E. Historic background of desalination and renewable energies. Sol
Energy 2003;75(5):357–66.
[13] Miranda MS, Infield D. A wind-powered seawater reverse-osmosis system without
batteries. Desalination 2003;153:9–16.
[14] Garcia-Rodriguez L, Romero-Ternero V, Comez-Camacho C. Economic analysis of
wind-powered desalination. Desalination 2001;137:259–65.
[15] Rajaseenivasan T, Shanmugam RK, Hareesh VM, et al. Combined probation of
bubble column humidification dehumidification desalination system using solar
collectors. Energy 2016;116:459–69.
[16] Moumouh J, Tahiri M, Salouhi M, et al. Theoretical and experimental study of a
solar desalination unit based on humidification–dehumidification of air. Int J
[17] Mahkamov K, Orda E, Belgasim B, et al. A novel small dynamic solar thermal de-
salination plant with a fluid piston converter. Appl Energy 2015;156:715–26.
[18] Kim YD, Thu K, Ng KC, et al. A novel integrated thermal-/membrane-based solar
energy-driven hybrid desalination system: Concept description and simulation re-
sults. Water Res 2016;100:7–19.
[19] Nematollahi F, Rahimi A, Gheinani TT. Experimental and theoretical energy and
exergy analysis for a solar desalination system. Desalination 2013;31(7):23–31.
[20] Chen Z, Peng J, Chen G, et al. Analysis of heat and mass transferring mechanism of
multi-stage stacked-tray solar seawater desalination still and experimental research
on its performance. Sol Energy 2017;142:278–87.
[21] Palenzuela P, Zaragoza G, Alarcón-Padilla DC. Characterisation of the coupling of
multi-effect distillation plants to concentrating solar power plants. Energy
2015;82:986–95.
[22] Xie G, Sun L, Mo Z, et al. Conceptual design and experimental investigation in-
volving a modular desalination system composed of arrayed tubular solar stills.
Appl Energy 2016;179:972–84.
[23] Reddy KS, Kumar KR, O’Donovan TS, Mallick TK. Performance analysis of an
evacuated multi-stage solar water desalination system. Desalination
2012;288:80–92.
[24] Estahbanati MRK, Feilizadeh M, Jafarpur K, et al. Experimental investigation of a
multi-effect active solar still: The effect of the number of stages. Appl Energy
2015;137:46–55.
[25] Dayem AMA. Experimental and numerical performance of a multi-effect con-
densation–evaporation solar water distillation system. Energy
2006;31(14):2710–27.
[26] Chorak A, Palenzuela P, Alarcón-Padilla DC, et al. Experimental characterization of
a multi-effect distillation system coupled to a flat plate solar collector field: em-
pirical correlations. Appl Therm Eng 2017;120:298–313.
[27] Liu ZH, Hu RL, Chen XJ. A novel integrated solar desalination system with multi-
stage evaporation/heat recovery processes. Renew Energy 2014;64:26–33.
[28] Liu ZH, Guan HY, Wang GS. Performance optimization study on an integrated solar
desalination system with multi-stage evaporation/heat recovery processes. Energy
2014;76:1001–10.
[29] Winston R, Hinterberger H. Principles of cylindrical concentrations for solar energy.
Sol Energy 1975;17:255–60.
[30] Rab IA. Solar concentrators with maximal concentration for cylindrical absorbers.
Appl Opt 1976;15(7):1871–3.
[31] Rabl A. Comparison of solar concentrators. Sol Energy 1976;18:93–111.
[32] Rabl A, O'GaUagher J, Winston R. Design and test of non-evacuated solar collectors
with compound parabolic concentrators. Sol Energy 1980;25:335–51.
[33] Alaydi JY. Modeling of a parabolic solar-collector system for water desalination.
Global J Res Eng 2013.
[34] Kalogirou S. Use of parabolic trough solar energy collectors for sea-water desali-
nation. Appl Energy 1998;60(2):65–88.
[35] Lillo I, Pérez E, Moreno S, et al. Process heat generation potential from solar con-
centration technologies in Latin America: the case of Argentina. Energies
2017;10(3):383.
[36] Khonkar HEI, Sayigh AAM. Optimization of the tubular absorber using a compound
parabolic concentrator. Renew Energy 1995;6(1):17–21.
[37] Alarcón-Padilla DC, Blanco-Gálvez J, García-Rodríguezz L, et al. First experimental
results of a new hybrid solar/gas multi-effect distillation system: the AQUASOL
project. Desalination 2008;220(1–3):619–25.

352