Desalination 396 (2016) 7078

Contents lists available at ScienceDirect

Desalination
journal homepage: www.elsevier.com/locate/desal

Concentrating solar power for seawater thermal desalination

Osman Ahmed Hamed a,, Hiroshi Kosaka b, Khalid H. Bamardouf a, Khalid Al-Shail a, Ahmed S. Al-Ghamdi a
a
b

Desalination Technologies Research Institute (DTRI), Saline Water Conversion Corporation (SWCC), P.O. Box 8328, Al-Jubail 30951, Saudi Arabia
Hitachi Zosen Corp., Osaka, Japan

H I G H L I G H T S
Operational performance of Fresnel concentrating solar power (CSP) system.
Cost effectiveness of a commercial solar assisted thermal desalination plant
Impact of DNI, thermal energy storage and fuel cost on the feasibility of CSP assisted thermal desalination plant.

a r t i c l e

i n f o

Article history:
Received 29 December 2015
Received in revised form 18 May 2016
Accepted 11 June 2016
Available online 17 June 2016
Keywords:
CSP
Fresnel
Desalination
Cost effectiveness

a b s t r a c t
Extensive pilot plant experimental studies for a period of one year were carried out to study the impact of climatic
conditions on the operational performance of an innovative Fresnel solar collecting system. The solar measurements revealed that the total yearly Direct Normal Irradiance (DNI) on the tested site amounts to 1132 kWh/
m2. The thermal collector efciency, which depends on climatic conditions such as solar insolation, ambient temperature, receiver temperature as well as heat losses, ranges from 60% to 80%. The cost effectiveness when the
tested Fresnel solar collection system with solar multiple of 1.0 (limited to day time operation) is combined
with a commercial thermal desalination plant is compared with one completely run by fossil fuel. The breakeven
fuel cost whereby the levelized cost of water of the two cases will be equal is yielded at a fuel cost of $92/bbl.
When the tested Fresnel solar collection system is run at a location with a relatively high annual DNI level
(1937 kWh/m2), the fuel breakeven cost falls to $52/bbl. This study also revealed that combining a Fresnel
solar collection system with an MED thermal desalination plant under specic climatic conditions is considered
more cost effective when operated without thermal energy storage.
2016 Elsevier B.V. All rights reserved.

1. Introduction
The Saline Water Conversion Corporation (SWCC) of Saudi Arabia is
currently operating small scale single purpose thermal desalination
plants with water production capacities ranging from 250 to 9000 m3/
day. One of the major problems that impede the cost effectiveness of a
single purpose thermal desalination plant is its high fuel energy consumption. Techno-economic feasibility of small scale MSF or MED thermal desalination plants driven directly by boilers can be greatly
enhanced when solar energy is employed to provide all or part of the
thermal energy consumption.

Part of the paper is presented at the International Desalination Association (IDA)

Conference held at San Diego, USA August 304 September, 2015.
Corresponding author.
E-mail addresses: ohamed@swcc.gov.sa, oahamed@yahoo.com (O.A. Hamed),
kosaka@hitachzosen.co.jp (H. Kosaka).

http://dx.doi.org/10.1016/j.desal.2016.06.008
0011-9164/ 2016 Elsevier B.V. All rights reserved.

A solar distillation plant may consist of one integrated system (direct

solar desalination) or two separate devices, the solar collector/accumulator and distiller (indirect solar desalination). Direct use of solar energy
is through the direct heating of salty water by the sun through conventional solar stills for low water production [1]. Another example of direct use of solar energy is achieved by combination of the principle of
humidication-dehumidication with solar desalination using air as a
heat carrier [2,3]. For relatively large water production capacities, the
solar energy is indirectly used to drive thermal desalination plants by
capturing solar radiation through one of the modern technologies
which transform the solar energy into heat using means such as parabolic trough and linear Fresnel collectors, evacuated tube collectors
and salinity gradient solar ponds [420]. Concentrated solar power
(CSP) technology stores the energy from solar radiation in a working
uid in the form of heat. This heat can then be used directly to run a conventional thermal desalination multistage ash ((MSF), multi-effect
distillation (MED), multi-effect distillation with thermal vapor compression (MED/TVC) plant. The thermal energy can alternatively be converted into electrical energy through a conventional power generation

O.A. Hamed et al. / Desalination 396 (2016) 7078

Nomenclature
A
total collector aperture area (m2)
ATAN
inverse tangent value
CAPEX
total capital expenditure of solar eld ($)
COSd cosine absolute value of sun's direction angle
Cp
specic heat (kJ/(kg C))
CRF
capital recovery factor
DNI
direct normal irradiation (kW m2)
DNIeff
actual or effective received irradiation (kW/m2)
T
oil temperature rise (C)
K_(_) longitudinal correction factor
K_(_) transversal correction factor
LCOW
levelized cost of water ($/m3)
n
amortization period (years)
OPEX
total annual operational and maintenance expenditure
($)
Ploss
heat loss from receiver per unit length (W/m)
Pth
net (useful) thermal energy absorbed by the receiver
(kW)
q
oil volumetric ow rate (m3/s)
SINd sine absolute value of sun's direction angle
Tabs
average of inlet and outlet temperature of heat transfer
uid (C)
TAN
tangent value
Wc
MED annual water production (m3/year)
Z
discount rate (%)
0
optical efciency at zero
sy
overall solar system efciency (%)
th
thermal efciency (%)
a
sun's measured height angle ()
d
sun's direction angle ()
II
sun irradiation longitudinal angle ()
L
sun irradiation transversal angle ()

numerical value of PI (3.142)

density (kg/m3)

plant and can then be used to run a reverse osmosis (RO), electrodialysis
(ED) or mechanical vapor compression (MVC) desalination plant.
CSP collectors developed and tested so far can be broadly divided
into two categories: Line focus collectors and point focus collectors
[21]. Line focus collectors include both the parabolic trough (PT) and
Linear Fresnel (LF) types which use a single axis tracking system and
can yield moderate temperatures up to 400 C, while point focus collectors include parabolic dish and central receiver collectors with dual-axis
tracking systems generating temperatures as high as 1000 C or more.
CSP technology offers two main advantages [21]. First, all CSP technologies can be combined with thermal energy storage systems. Second, CSP
plants can be operated with fossil fuel backup (hybrid operation). The
use of thermal energy storage systems and/or hybrid operation provides
the possibility of continuous 24 h operation of solar assisted desalination plants.
A number of studies have been reported comparing between parabolic trough and linear Fresnel applications [2232]. Lined focused parabolic and linear Fresnel solar concentrators [22] both consist of a long
reector, which act as the only concentrator aligned on a northsouth
axis. One advantage of these systems is the tracking which is primarily
only in one dimension. The reector is rotated to track the sun's movement and it's reected solar energy is concentrated along a focal line
and captured by receiver tube containing a heat absorbing uid that absorbs the concentrated heat. One-axis solar concentration provides a
simple operation and highly reliable system to reach maximum operation temperatures about 400 C. Normally, medium concentration ratios

71

between 15 and 40 are attainable; therefore, one-axis sun tracking is required [23]. Synthetic oils are used as heat transfer uid in conventional
solar PT collectors, which limits the top temperature. Nevertheless, the
synthetic oil may be replaced by water in order to generate steam directly into the absorber pipe and temperatures up to 400 C may be
allowed. Direct steam generation (DSG) offers the potential for higher
performance of the plants and for cost reduction [24]. Parabolic trough
collector (PTC) using DSG has identical collector structures as for thermal oil. On the other hand, linear Fresnel collectors (LFC) with DSG
use a potentially cheaper design mainly due to the use of at mirrors
and structural advantages, however with a lower optical efciency.
DSG avoids the costs of heat transfer uid and the central oil heated
steam generator. The DSG system is not without its technical challenges,
with the risk of overheating tubes and potential ow instabilities [25].
Sophisticated controls are required to accommodate the use of the
two-phase ow of water and steam.
Compared with parabolic troughs, linear Fresnel collectors suffer
from lower optical efciency [26]. However, the low-prole setting of
the linear Fresnel collector poses no mechanical difculty to maximize
the collector geometrical concentration ratio (the ratio of mirror aperture to receiver aperture), which enables high temperature output.
The high temperature output would give rise to high power cycle efciency and accordingly a great reduction of storage system cost. The
low-prole setting of the mirrors also leads to a lower wind load requirement and thus lower-cost mirror assembly design. Further, it will
also help in lowering the O&M cost for a power plant. The xed receiver
assembly greatly reduces the risk of heat transfer uid (HTF) leakage
and the resulting maintenance labor. A comparison has been made between the optical performance of parabolic trough collectors and linear
Fresnel reectors using multi-tube receivers and secondary [27]. The results reveal that PTC efciency is higher than the efciency of LFCs, either with multi-tube or secondary reector receiver. This was due to
the fact that PTCs conform a perfect parabola with its aperture perpendicular to the impinging beam, in the transversal plane, at all moments.
However, LFC are characterized by a simpler conguration: narrower
mirrors, and thus lighter structure, xed receiver, and leakages avoidance. Comparison of the annual performance and economic feasibility
of Integrated Solar Combined Cycles (ISCC) using two solar concentration technologies: parabolic trough collectors and linear Fresnel reectors, is reported [28]. Results show that the thermal contribution is
higher with PTC, but LFR may improve the economic feasibility of the
plant.
Existing commercial CSP plants are mainly used for electricity generation rather than water production. The National Renewable Energy
Laboratory (NREL) [33], has compiled data on concentrating solar
power (CSP) projects around the world that are either under operational, construction, or development stage. CSP technologies include parabolic trough, linear Fresnel reector, power tower, and dish/engine.
The majority of solar assisted power generation plants are using parabolic trough collectors with planned electricity generation per plant in
the range 1800 to 175,000 MWh/year and are equipped with thermal
storage system of molten salt (60% sodium nitrate, 40% potassium nitrate). There are seven operational solar power projects which are
employing Fresnel solar collectors with plant generated electricity in
the range of 2000 to 280,000 MWh/year and the majority of which are
without storage. Around 24 concentrating solar power (CSP) projects
that are either operating or under construction use power tower systems and molten salt for storage. Only one CSP project is under construction that uses dish/engine systems with at turbine capacity of
1 MW and without storage system.
A number of parabolic collector desalination demonstration plants
have been implemented and tested [6]. At the plataforma Solar de
Almeria, Spain, a parabolic trough collector eld was connected to an
MED plant with a water thermal storage system. At the second phase
of the project, a double-effect absorption heat pump was coupled with
the solar desalination plant. Subsequently, the thermal energy

72

O.A. Hamed et al. / Desalination 396 (2016) 7078

consumption of the desalination plant decreased from 63 kWh/m3 to

36 kWh/m3. A comprehensive review of the salient design and operating features of solar assisted thermal desalination plants were reported
[6,20].
Linear Fresnel systems have recently been developed to achieve a
more simple design and lower cost than the parabolic trough. Compared
to the existing parabolic trough, the linear Fresnel collector system designed by Novatec-Biosol [34] shows a weight reduction per square
meter of 80% resulting in both lower capital cost and reduces life cycle
emissions. Also, land use efciency with a linear Fresnel is about 2
times higher than that of a parabolic trough. The use of Fresnel collector
in desalination is limited. One of the problems experienced with conventional slightly curved Fresnel mirrors is that the focal point of
reected sunlight reaching the receiver can vary depending on mirror
location and sun angle.
This study was carried out jointly by the Desalination Technologies
Research Institute (DTRI) of SWCC and Hitachi Zosen Corporation
(Hitz) of Japan to explore the prospects of the application of an improved Fresnel CSP collector system when integrated with a thermal desalination plant under the climatic conditions of Saudi Arabia. The
improved Fresnel CSP collector is equipped with a new mechanism
that controls the curve of each mirror, together with sun tracking to
focus all reected sunlight on the receiver tubes and consequently improve the solar collector thermal and optical efciencies. A fully controlled innovative Fresnel solar collector system was constructed and
installed at DTRI premises in Al-Jubail city, Saudi Arabia. Extensive experimental tests covering a one year period were carried out to assess
the operational performance of the solar collecting system under the
Saudi Arabia site specic solar radiation conditions and ambient temperatures. The experimental test results are then utilized to explore
both the technical and economic potential of a commercial solar
assisted multi-effect desalination plant.
2. Plant description
As shown in Fig. 1, the CSP demonstration plant consists of a solar
concentrating system, a steam generator and a condenser. The solar
concentrating and collecting plant consists of innovative solar Fresnel
reectors, receiver tube, and low and high temperature heat transferred
uid oil accumulator tanks. The solar reector consists of 6 parallel
rows. Each row consists of 92 controlled curved mirrors positioned in
south-north horizontal axis and with a total aperture area of 662.4 m2.
The mirrors focus the sun's direct beam radiation onto a linear absorber
pipe at the focus of the Fresnel mirrors. The absorber pipe is formed
from 24 SCHOTT PTR70 receivers which are designed to operate
with oil-based heat transfer uids at temperatures up to 400 C and having more than 95% of heat absorption coefcient and less than 10% of
emissivity characteristics. Sun.'s rays on Fresnel reectors are concentrated more than seventy times on the receiver tube positioned at a

height of 3.8 m and then transferred to the oil-based heat transfer

uid (HTF) and owing inside the receiver tube. Heated HTF is accumulated in the high temperature tank and sent to steam generator as a heat
source. Low temperature HTF discharged from the steam generator is
returned to the low temperature tank and sent to the receiver tube
again by thermal HTF circulation pumps. The steam generated can
then either be passed to a condenser that simulates an MSF brine heater
or passed directly to an existing MSF pilot plant to supply part of its
steam requirements. A pyrheliometer with a sensitivity of
(8.24 106 V) / (W/m2) is used for measurement of direct normal irradiance (DNI) and it is linked with a tracking system to follow the apparent movement of the sun and monitor sun direction and height
angles with a 0.01 C accuracy. Temperatures are measured by platinum
thin lm RTD class B sensors with tolerance of 0.3 C. The oil ow rate
is measured with a ow meter with a measured error of 0.05%. The
pressure transmitters are having an accuracy of 0.25%. Steam ow
rate and pressure are measured by a ow meter equipped with dual
sensors measuring pressure differential and pressure simultaneously
and separately with a high accuracy specication.
3. Methodology
The main objective of the experimental study is to assess the thermal
and optical performance characteristics of the modied Fresnel and to
what extent it is inuenced by the local site climatic conditions. Site
measurements which are utilized to assess the operational performance
of the Fresnel solar collecting system included daily variation in ambient
conditions such as temperature, direct normal irradiation, sun direction
and height angles. The inlet and outlet temperature of the oil passing
through the receiver as well its volumetric ow rates were also monitored together with the quantity, pressure and temperature of the generated steam. A data acquisition system integrated with solar system
collects and stores all the monitored and measured data every 10 s.
The 10 s readings are then used as input to an excel program to calculate
the quantity of thermal energy absorbed and lost by the receiver and the
actual amount of solar incidence received by the collector. For each calculated parameter, an average daily value of the 10 s readings is obtained by dividing the summation of all the daily 10 s readings by the
number of readings. The monthly readings are then obtained by the
summation of the corresponding daily readings.
3.1. Overall solar system efciency
The overall solar system efciency (sy) represents the ratio of total
thermal energy absorbed by the receiver (Pth/A) to the incident direct
normal solar irradiation (DNI). It excludes sun position and height by
assuming that all mirrors are perpendicular to the impinging solar
beam and disregard all other imperfections.
Overall solar system efciency = (useful absorbed energy / received
sunlight)
sy Pth =A=DNI

A = Total collector aperture area.

The net (useful) thermal energy absorbed by the receiver and transferred to the oil passing inside the receiver tube (Pth) is determined
from the following equation:
Pth W q   Cp  T
Where

Fig. 1. Experimental setup of the solar collecting system.

Cp
T

oil volumetric ow rate (m3/s)

density (kg/m3)
specic heat (kJ/(kg C))
oil temperature rise (C)

O.A. Hamed et al. / Desalination 396 (2016) 7078

Fig. 2. Impact of variation of incidence angles in transversal and longitudinal dimension on

the correction factors for the energy yield at the tested Solar Fresnel collector.

73

Fig. 4. The daily variation of the overall solar system efciency (sy).

determined from the following relationship [35]:

Daily average overall thermal efficiency Pth =A=DNI

3.2. Collector thermal efciency

Out of the total amount of solar radiation incident on the solar collectors, only a fraction is converted into thermal energy while the rest
is lost to the environment by the reection of the incoming radiation
and to the ambient by various heat transfer mechanisms. The collector
thermal efciency (th) is dened as the ratio of the useful thermal
energy delivered to the actual energy incident on the collector aperture.
th collected thermal energy by HTF oil=actual received sunlight
4
th Pth =A  DNIeff

Where
DNIeff = The actual or effective received irradiation.
The effective received irradiation of the LFR system is limited by
the angle at which the sunlight strikes the reectors and because
the mirrors must be oriented to reect the irradiation to the receiver,
they most often do not directly face the sun. This non-normal orientation towards the incoming sunlight is the primary optical loss for
LFR, and the loss is incurred both with respect to the transversal
plane (perpendicular to the axis of the collector) and longitudinal
plane (parallel with the axis of the collector). DNIeff is conventionally

DNIeff DNI  K  K

Where K_(_) and K_(_) are transversal and longitudinal correction factors respectively and depend strongly on the position of the
sun with respect to the collector [35]. They are dependent on the sun's
measured height angle (a) and sun's direction angle (d).
The transversal K_(_) and longitudinal K_(_) correction factors are usually described respectively as a function of the transversal
(L) and longitudinal (II) solar incidence angles on the collector [35].
L and II are computed from the measured(a) and (d) solar angles
using the following equation:
L 90ATANTANa=180  =SINd=180  =  180

II 90ATANTANa=180  =COSd=180  =  180 8

The Transversal correction factors K_(_) which accounts to the
mirror shading and blocking and for the collector under study is computed from sun irradiation transversal angle (L) using the following
equation:










K 1  106  L3 4  107  L2
0:0001  L 0:9978

The longitudinal correction factors K_(_) which accounts for the

energy that does not impinge in the receiver due to the longitudinal
component of the irradiance [27] is determined from irradiation

Fig. 3. The monthly variation of direct and effective irradiance, thermal energy absorbed by receiver and receiver thermal losses.

74

O.A. Hamed et al. / Desalination 396 (2016) 7078

Fig. 5. The daily variation of the thermal collector efciency(th).

longitudinal angle (II) using the following equation:








K 7  1010  II5 9  108  II4












2  106  II3 3  105  II2 0:005  II 1:0001

Tabs is calculated from the average of inlet and outlet temperature of

heat transfer uid in C.

4. Results & discussion

10
4.1. Thermal and optical performance assessment
Fig. 2 shows the impact of variation of sun irradiation longitudinal or
transversal angle on the corresponding longitudinal or transversal correction factors. It also shows that the predicted values of the longitudinal or transversal correction factors for the Fresnel solar collector
tested in this study are to a great extent comparable with Novtec published correction factors [34].
3.3. Optical efciency
The optical efciency depends on the optical properties of the materials involved, the geometry of the collector and the various imperfections arising from the construction of the collector with regard to
receiver alignment to the focal line of the collector, and the system's
tracking precision [36]. It is dened as the ratio of energy absorbed by
the receiver to the energy incident on the collector's aperture as
shown in the following relationship:
0 Pth =A Ploss =DNIeff

11

Where Ploss is the heat loss from receiver per unit length and is determined from the following empirical equation [37]:
Ploss W=m 0:141Tabs 6:48E9Tabs 4

12

The experimental results revealed that at the examined site, the

maximum direct normal irradiation (DNI) from sun on the ground is
relatively low and in most cases range from 350 to 500 W/m2. As
shown in Fig. 3, the monthly measured DNI varies from as low as 65
kWh/m2 in January to about 125 kWh/m2 in June and the total yearly irradiation amounts to 1132 kWh/m2. The actual or effective energy incident on the collector's aperture is less than this maximum irradiation
because all the mirrors will not be perpendicular to the impinging
solar beam. As shown in Fig. 3, the monthly actual impinging solar irradiation ux varies from 32 to 107 kWh/m2 and around 13% to 57% of the
direct solar irradiation is dissipated. The actual total yearly solar energy
ux impinging on the solar collectors aperture is 835 kWh/m2 which
represents 73.76% of the maximum impinging solar power. Fig. 3 also
shows that the monthly thermal energy ux absorbed by oil varies
from 21 kWh/m2 in January to 58.3 kWh/m2 in June. The total yearly
thermal energy received by the oil is 414 kWh/m2 which represents
36.65 and 49.6% of the maximum solar power that would impinge
onto the reecting surface and the actual solar radiation striking the collector surface, respectively.
Fig. 4 Shows the variation of the overall solar system efciency (sy)
with the year days. The temperature of the HTF leaving the receiver is
initially set for the rst two months at 340 C. If the HTF outlet

Fig. 6. The daily variation of the optical efciency(0).

O.A. Hamed et al. / Desalination 396 (2016) 7078

75

Table 1
Performance characteristics of reported Fresnel solar collectors.

Collector reference condition

External temperature
Inow temperature
Outow temperature
No wind
Direct normal irradiation
Transversal angle _
Longitudinal angle _//
Optical efciency at zero
Thermal power output per unit area W/m2

The anticipated performance of the Fresnel collector evaluated in this study

Optical efciency at zero of Fresnel collector evaluated in this study
Thermal power output per unit area W/m2
Percentage increase in thermal power output

Case 1
Novatec Solar [34]

Case 2
Novatec Solar [34]

Case 3
Industrial Solar GmbH reported [38]

40 C
100 C
300 C
No wind
900 W/m2
30
10
0.67
Novatec solar
537 W/ m2

40 C
300 C
500 C
No wind
900 W/m2
30
10
0.65
Novatec solar
525 W/m2

30 C
160 C
180 C
No wind
900 W/m2
30
0
0.663
Industrial solar
562 W/m2

0.73
Present study
620 W/m2
15.5%

0.73
Present study
567 W/m2
8.0%

0.73
Present study
632 W/m2
12.5%

temperature is below the set value, heated HTF oil is recycled and
returned to the receiver tube. Otherwise it is passed to the steam generator. As a result of the prevailing site solar irradiance, it has been found
that it is more appropriate to set the outlet temperature of the oil leaving the receiver at a relatively low value. It was rst reduced from 340 C
to 250 C and then nally to 225 C. The overall system efciency during
the period when the outlet temperature is set to 225 C, ranges from
30% to 63%.
As shown in Fig. 5, the collector thermal efciency which depends on
climatic conditions such as solar insolation, ambient temperature, receiver temperature as well as heat losses ranges from 30 to 80%.
When the receiver outlet temperature is set to 225 C, the collector thermal efciency improves and, in most cases, ranges from 60% to 80%.
The optical efciency at zero (0 (eita zero) is an important design
characteristic factor and is normally used to determine the energy
yield of the solar collector eld. The energy yield of the solar eld
(Pth) is the product of the primary reector surface, the Direct Normal
Irradiance (DNI) and the optical efciency factor (0), minus the heat
losses from the receiver radiation (P).
Pth A  DNIeff  0 Ploss

13

As shown in Fig. 6, the optical efciency during the period when the
receiver outlet temperature is set at 225 C, ranges from 60% to 80% and
with about 73% average optical efciency. The optical performance of
the linear Fresnel collector (LFC) tested in this study can be compared
with commercial LFC developed by Novatec Solar [34] and Industrial
Solar GmbH [38]. Novatec Solar has developed a patented solar reector
design based on Fresnel collector technology. The solar boiler uses parallel rows of at glass mirrors to focus direct solar irradiation onto a linear receiver. Feed water is conveyed through an absorber tube to
generate steam. The operating performance of the Novatec solar

collector for two different cases is shown in Table 1. In case 1, the LFC
is operated at low evaporating temperature of 300 C generating a thermal output of 537 W/m2, while in the second case it is operated at a high
temperature of 500 C, producing superheated steam using a vacuum
receiver and with a thermal output of 525 W/m2.
Under the same solar operating conditions the thermal power output per unit area expected to be produced by the modied Fresnel collector examined in this study, can be determined using the following
relationship:
Pth A  DNI  0 K  K Ploss

14

Using the same operating conditions reported by Novatec and

shown in Table 1, the performance characteristics of the collector system tested in this study K_(_),K_(_ and P loss are determined
using Eqs. (9), (10) and (12) respectively. For each case as shown in
Table 1, the thermal power output expected to be produced using the
tested collector with optical efciency (_0) value of 0.73, is then obtained using Eq. (14). For case 1, the calculation results revealed that
the thermal power expected to be generated by the tested collector is
620 W/m2 which is about 15.5% higher than that yielded by the Novatec
solar collector. On the other hand, when the Novatec solar collector is
operated at the conditions as shown in Case 2, the expected thermal
yield is around 8.0% higher than that generated by the Novatec solar collector. Industrial Solar GmbH also reported [38] the operating conditions of linear solar Fresnel collector LF-11 as shown in case 3 of Table
1 for generating 562 W/m2 of process heat. Under the same operating
conditions, it has been determined that the modied Fresnel collector
generates thermal output of 632 W/m2, which is about 12.5% higher
than that generated by the Industrial Solar Fresnel collector. It has
thus been conrmed that the modied Fresnel collector tested in this
study yields relatively high thermal power output per unit area as compared to the reported performance of industrial Fresnel solar collectors.
This can be attributed to its high optical efciency and low thermal loss.
Industrial Fresnel solar collectors which are slightly curved can focus
sunlight on receiver, but the focal point varies depending on mirror location and sun angle. However the modied Fresnel collector system

Table 2
Cost data.

Fig. 7. Conguration of solar assisted thermal desalination plant.

Specic CAPEX solar eld US$/m2

279 US$/m2

CAPEX MED desalination plant

CAPEX Back-up boiler
OPEX MED without energy
OPEX Solar eld
OPEX boiler

$ 10,000,000
$600,000
$0.28/(m3 distillate)
$5.4/(m2 year)
$0.1/(m3 distillate)

76

O.A. Hamed et al. / Desalination 396 (2016) 7078

Fig. 8. Impact of fossil fuel cost on water production cost without storage.

examined in this study has the unique capability of controlling the curvature of each mirror together with sun tracking to focus all reected
rays on the receiver.
4.2. Potential of coupling the solar collecting system with a commercial desalination plant
The solar eld performance test results are utilized to assess the economic feasibility of a commercial solar assisted thermal desalination
plant operating under the same specic climatic conditions. A small capacity MED-TVC standalone desalination plant as normally operated at
relatively low top brine temperature (TBT) of 65 C and water production capacity of 1 MIGD (4546 m3/d) is selected. The MED-TVC distiller
will be integrated with the modied Fresnel CSP solar collector and
combined with a fossil fuel run boiler to supply part of its thermal energy requirements. When the supply of the solar energy is interrupted,
thermal energy requirements of the MED TVC desalination plant is supplied by the fossil fuel backup boiler to maintain constant heat supply
and 97% plant availability. The MED-TVC pumping and electrical energy
requirement which is around 2.0 kWh/m3 is supplied from the national
grid as shown in Fig. 7.
Based on an MED-TVC distiller with a performance ratio of 9 kg/
2326 kJ, the specic energy requirement will be 258.44 kJ/kg. The
total thermal energy requirements of 4546 m3/d (1 MIGD) distiller
will be 13.6 MWth. The monthly average thermal energy absorbed by
oil from the tested Fresnel CSP system under the specic site conditions
with annual DNI of 1132 kWh/m2 and transferred to the steam generator as obtained from Fig. 3, is 40 kWh/m2 month. The average sunshine
period at the tested site to maintain this value is around 5 h. Subsequently, the solar full load hours are 1770 h/year with a 20.8% solar
share without thermal energy storage. The remaining 79.2% of the
MED thermal requirements is covered by fossil fuel run boiler. The thermal energy ux that will be supplied by the solar eld and gained by
generated steam is 254 W/m2 assuming 90% thermal efciency. Accordingly, the total Fresnel minor area required with a solar multiple of 1.0
and without thermal energy storage to supply the MED thermal energy
requirements of 13.6 MWth is 55,737 m2.
The levelized cost of water (LCOW) is determined from the following relationship:

CAPEX is total capital expenditure of solar eld ,MED plant and backup fossil fuel boiler,
OPEX is total annual operational and maintenance expenditure of
solar eld, MED plant and fossil fuel boiler as well as the annual energy
cost of fossil fuel.
WC = MED annual water production
CRF is capital recovery factor (CRF) and determined from:
CRF

z1 zn
1 zn 1

Where is Z discount rate and (n) amortization period in years.

The levelized cost of water (LCOW) is calculated based on discount
rate of 5%/y, 30 years amortization period, 0.0751 capital recovery factor, 97% plant availability and $0.04/kWh electricity cost. A sensitivity
analysis is carried when fuel cost is varied between $10 to $120 oil per
barrel equivalent. The capital expenditure (CAPEX), operational and
maintenance expenditure (OPEX) of solar eld, MED plant and backup fossil fuel boiler are estimated and shown in Table 2.
Fig. 8 shows the impact of variation of fuel cost on the levelized cost
of water (LCOW) for two different scenarios. In the rst scenario as
shown by the dotted line, the MED-TVC thermal requirements are solely
supplied by back-up fossil fuel run boiler. The LCOW ranges from $1.43/
m3 to $7.1/m3 as the fossil fuel price is increased from $10/barrel oil
equivalent (bbl) to $120/bbl. In the second scenario, part of the MEDTVC system thermal energy requirement is supplied by the Fresnel
solar collector, (which is not oversized) with a solar multiple of 1.0
and the remaining part is provided by a fossil fuel run boiler. The solar

0
B
B
LCOW B
BUS
@
m3

Where

X

CAPEX  CRF OPEX

16

15

WC

Fig. 9. Impact of fossil fuel cost on water production cost without storage.

O.A. Hamed et al. / Desalination 396 (2016) 7078

77

levelized cost of water of the two cases will be equal is yielded at a

fuel cost of $92/bbl. The high value of the breakeven fuel cost is attributed to a low total yearly irradiance of 1132 kWh/m2 experienced at the
specic test site. When the tested Fresnel solar collection system is operated at a location of a relatively high annual DNI level of 1937 kWh/
m2, the fuel breakeven cost at this location falls to $52/bbl. The economic study also revealed that combining a Fresnel solar collection system
with an MED thermal desalination plant under the specic climatic conditions is more cost effective when operated without thermal energy
storage.

References
Fig. 10. Impact of fossil fuel cost on water production cost with 8 h thermal energy storage.

full load hours are 1770 h/year and fossil fuel load hours are 6727.2 h/
year.
The LCOW varies from $2.24/m3 to $6.8/m3 for the same range of
fuel cost. The breakeven fuel cost whereby the LCOW of the two scenarios will be equal is yielded at a fuel cost of $92/bbl. The obtained high
breakeven fuel cost is attributed to the low total yearly irradiance of
1132/m2, the level which is exhibited at the specic test site. When
the tested Fresnel solar collection system will be operated at a location
with a relatively high annual DNI of 1938 kWh/m2, the fuel breakeven
cost at this location reduces to $52/bbl as shown in Fig. 9.
The cost effectiveness of integrating thermal energy storage (TES)
with an oversized Fresnel solar eld is also examined. An eight hour
storage system is assumed to be integrated with the solar collection system. The solar collection area has to be selected to generate the required
thermal energy to operate the MED TVC desalination plant for a 13 h
period at the rated capacity. A fossil fuel run boiler will supply the additional heat to operate the desalination plant continuously. The required
oversized solar collection area, which corresponds to a solar multiple
of 2.6, is 144,917 m2 when the annual DNI is 1132 kWh/m2. Fig. 10
shows a comparison between a solar assisted desalination plant
equipped with an eight hour thermal energy storage (TES) system
with a100% fossil fuel operated desalination plant assuming that
the specic capital cost of the thermal energy storage system is
$35/(kWhth ). The solar full load hours are 4603 h/year and fossil
fuel load hours are 3895 h/year In this case, 54% of the thermal energy requirements of the 1 MIGD MED desalination plant will be supplied by the solar system and the remaining 46% by fossil fuel. The
breakeven fuel cost is found to be $104/bbl. Meanwhile the breakeven fuel cost without TES system as shown above is $92/bbl at annual DNI of 1132 kWh/m2. It can be concluded that combining the
Fresnel solar collection system with MED thermal desalination
plant under the specic climatic conditions considered in this study
is more cost effective when operated without thermal energy
storage.
5. Conclusions
Extensive pilot plant tests covering one year period were carried out
to assess the optical and thermal performances of an innovative conguration of a Fresnel CSP collector under Saudi Arabia climatic conditions.
Performance comparison of the modied Fresnel collector tested in this
study with those of industrial Fresnel solar collectors reported in literature reveals that it yields a relatively high thermal power output per
unit area. This can be attributed to its relatively high optical efciency
and low thermal loss.
The cost effectiveness when the modied Fresnel solar collection
system with a solar multiple of 1.0 is combined with a commercial 1
MIGD MED-TVC thermal desalination plant is compared with one
completely run by fossil fuel. The breakeven fuel cost whereby the

[1] O.A. Hamed, E.I. Eis a, W.E. Abdallah, Overview of solar desalination, Desalination 93
(1993) 563579.
[2] S. Parekh, M.M. Farid, J.R. Selman, S. Al-Hallaj, Solar desalination with a humidication-dehumidication technique a comprehensive technical review, Desalination
160 (2004) 167186.
[3] K. Bourouni, M.T. Chaibi, L. Tadrist, Water desalination by humidication and dehumidication of air: state of the art, Desalination 137 (2001) 167176.
[4] L. Garcia-Rodriguez, A.I. Palmero-Marrero, C. Gomez-Camacho, Comparison of solar
thermal technologies for applications in seawater desalination, Desalination 142
(2002) 135142.
[5] H.M. Qiblaway, Fawzi, Solar thermal desalination technologies, Desalination 220
(2008) 633644.
[6] L. Garcia-Rodriguez, Seawater desalination driven by renewable energies: a review,
Desalination 143 (2002) 103113.
[7] H. Lu, J.C. Walton, A.H.P. Swift, Desalination coupled with salinity-gradient solar
ponds, Desalination 136 (2001) 1323.
[8] D.C. Alarcon-Padilla, J. Blanco-Galvez, L. Garcia-Rodriguez, W. Gernjak, L. RocaSobrino, Experimental results of a new hybrid solar-gas multi- effect distillation
system, Proc. IDA World Congress, Maspalomas, Grain Canaria-Spain, Oct. 21-26,
2007.
[9] O.A. Hamed, J.A.S. Al-Jabri, Simulation and performance of an MES solar distillation
system, Renewal Energy Technology and the Environment, 2nd World Renewable
Energy Congress, Solar Thermal Technology, vol. 2, 1992.
[10] A.M. El-Nashar, The economic feasibility of small solar MED seawater desalination
plants for remote arid areas, Desalination 134 (2001) 173186.
[11] A.M. EI-Nashar, Abu Dhabi solar distillation plant, J. Desalin. 52 (1985) 217234.
[12] A.M. EI-Nashar, Performance of the solar desalination plant at Abu Dhabi, J. Desalin.
72 (1989) 405424.
[13] A.M. EI-Nashar, Computer simulation of the performance of a solar desalination
plant, J. Sol. Energy 44 (4) (1990) 193205.
[14] A.M. EI-Nashar, A.M. El-Baghdadi, Seawater distillation by solar energy, J. Desalin. 61
(1987) 4966.
[15] A.M. El-Nashar, A.A. Qamhieyeh, performance simulation of the heat accumulator for the Abu Dhabi solar desalination plant, J. Solar Energy 44 (4) (1990)
183191.
[16] O.A. Hawaj, M.S. Darwish, Performance characteristics of a multi-effect solar
pond desalting system in an arid environment, Proc. of the IDA and WRPC
World Conference on Desalination and Water Treatment, Nov. 3-6, Yokohama,
Japan, vol. 1, 1993.
[17] P. Willson, D. Oliver, Changing perspectives on desalination by renewable energy,
Proc. in International Desalination Association World Congress: SP 05118, Singapore, 2005.
[18] A. Cipollina, C. Sommariya, G. Micale, Efciency increase in thermal desalination
plants by matching thermal and solar distillation: theoretical analysis, Desalination
183 (2005) 127136.
[19] G. Fiorenza, V.K. Sharma, G. Braccio, Techno-economic evaluation of a solar powered
water desalination plant, Energy Convers. Manag. 22 (2003) 22172240.
[20] M. Tauha Ali, H.E.S. Fath, P.R. Armstrong, A comprehensive techno-economic review
of indirect solar desalination, Renew. Sust. Energ. Rev. 15 (2011) 41874199.
[21] F. Verdier, H. Ludwig, J. Kretschmann, Solar powered seawater desalination A case
study, Proceedings at World Congress Perth Convention and Exhibition Center
(PCEC), Australia, September 49, 2011 (Ref. IDAWC/PER11165).
[22] J.H. Reif, W. Alhalabi, Solar-thermal powered desalination: its signicant challenges
and potential renewable and sustainable, Energy Rev. 48 (2015) 152165.
[23] L. Garcia-Rodriguez, A.I. Palmero-Marreroa, C. Gbmez-Camachob, Comparison of
solar thermal technologies for applications in seawater desalination, Desalination
142 (2002) 135142.
[24] G. Morin, J. Dersch, W. Platzer, M. Eck, A. Haberle, Comparison of linear Fresnel and
parabolic trough collector power plants, Sol. Energy 86 (2012) 112.
[25] J.D. Nixon, P.K. Dey, P.A. Davies, * Which is the best solar thermal collection technology for electricity generation in orth-West India? Evaluation of options using the
analytical hierarchy process, Energy 35 (2010) 52305240.
[26] G. Zhu, T. Wendelin, M.J. Wagner, C. Kutscher, History, current state, and future
of linear Fresnel concentrating solar collectors, Sol. Energy 103 (2014)
639652.
[27] R. Abbas, M.J. Montes, A. Rovira, J.M. Martnez-Val, Parabolic trough collector or linear Fresnel collector? A comparison of optical features including thermal quality
based on commercial solutions, Sol. Energy 124 (2016) 198215.

78

O.A. Hamed et al. / Desalination 396 (2016) 7078

[28] A. Rovira, R. Barbero, M.J. Montes, R. Abbas, F. Varela, Analysis and comparison of integrated solar combined cycles using parabolic troughs and linear Fresnel reectors
as concentrating systems, Appl. Energy 162 (2016) 9901000.
[29] A. Lewandowski, D. Simms, An assessment of linear Fresnel lens concentrators for
thermal applications, Energy 12 (5) (1987) 333338.
[30] N. El Gharbi, H. Derbal, S. Bouaichaoui, N. Said, A comparative study between parabolic trough collector and linear Fresnel reector technologies, Energy Procedia 6
(2011) 565572.
[31] H.H. Sait, J.M. Martinez-Val, R. Abbas, J. Munoz-Anton, Fresnel-based modular solar
elds for performance/cost optimization in solar thermal power plants: a comparison with parabolic trough collectors, Appl. Energy 141 (2015) 175189.
[32] V. Sharma, J.K. Nayak, S.B. Kedare, Comparison of line focusing solar concentrator
elds considering shading and blocking, Sol. Energy 122 (2015) 924939.
[33] National Renewable Energy Laboratory, [NREL], concentrating solar power
projectsavailable from: http://www.nrel.gov/csp/solarpaces/.

[34] Novatec Solar, Germany, technical data, www.novatecsolar.com Novatec Biosol Information Brochure, 2011. World's First Fresnel Solar Power Plant in Commercial
Operation.
bhttp://www.novatecsolar.com/les/mne0911_pe1_broschure_
english.pdfN
[35] M. Binotti, G. Manzolini, G. Zhu, An alternative methodology to treat solar radiation
data for the optical efciency estimate of different types of collectors, Sol. Energy
110 (2014) 807817.
[36] S.A. Kalogirou, Solar thermal collectors and applications, Prog. Energy Combust. Sci.
30 (2014) 231295.
[37] Technical Report, National Renewable Energy Laboratory, May 2009 (NREL/TP-55045633).
[38] Industrial solar, Technical data, www.industrial-solar.dee.