Solar Energy for Water Desalination

Abstract

This paper presents the different solutions to the most commonly used desalination process (RO, MSF, MED), and solar energy production technology compatible with desalination. The goal is to assess the feasibility and profitability of the substitution of fuel energy used for desalination plants with renewable energy. A review of various technologies will define broadly features associated to each technology and range of cost that are expected. Finally, a review of various projects will detail the practical aspects of floor space and actual production costs of fresh water.

Full text

Procedia Engineering 46 ( 2012 ) 220 – 227
1877-7058 © 2012 The Authors. Published by Elsevier Ltd. Selection and/or peer-review under responsibility of the Scientifi c Committee of SYMPHOS 2011
doi: 10.1016/j.proeng.2012.09.468
1st International Symposium on Innovation and Technology in the Phosphate Industry
[SYMPHOS 2011]
Solar Energy for Water desalination
Pascale Compain
a
*
Bertin Technologies, 10, Avenue Ampère, 78180 Montigny-le-Bretonneux, France
Abstract
This paper presents the different solutions to the most commonly used desalination process (RO, MSF, MED), and solar energy
production technology compatible with desalination. The goal is to assess the feasibility and profitability of the substitution of fuel energy
used for desalination plants with renewable energy.
A review of various technologies will define broadly features associated to each technology and range of cost that are expected. Finally, a
review of various projects will detail the practical aspects of floor space and actual production costs of fresh water.
© 2012 The Authors. Published by Elsevier Ltd. Selection and/or peer-review under responsibility of the Selection and /or
peer-review under responsibility of the scientific committee of SYMPHOS 2011.
Keywords : Desalination, RO, MSF, MED, Solar energy
Nomenclature
RO Reverse Osmosis
MSF Multi Stage Flash Distillation
MED Multi Effect Distillation
PV PhotoVoltaic
LFR Linear Fresnel Reflector
CSP Concentrating Solar Power
1. Introduction
It’s true that freshwater scarcity is associated with large quantity of solar resource. It seems also logical and attractive to
associate those two parameters for countries where grid electricity is not spread widely and with easy access to seawater or
brackish water.
Solar desalination is not a new idea: it has been known for ages, antique sailors used to desalt water with simple and
small sized solar stills.
It’s also a fact that production of fresh water requires a large amount of energy: 1000 m3 of freshwater per day requires
10 000 tons of oil per year [1]. Though solar energy is often labelled as ‘free energy’, it’s not so simple to evaluate
feasibility and cost for solar desalination
Some technologies will not be taken in account in this paper: solar ponds, which are a direct desalination method, as well
as desalination with electrodialysis (whose application is restricted to low salinity water).
* Corresponding author. . Tel.: +33 (0)442 604 634; fax: +33 (0)442 600 013.
E-mail address: pcompain@bertin.fr.
Available online at www.sciencedirect.com
© 2012 The Authors. Published by Elsevier Ltd. Selection and/or peer-review under responsibility of the Scientifi c
Committee of SYMPHOS 2011

221
Pascale Compain / Procedia Engineering 46 ( 2012 ) 220 – 227
2. Desalination technologies
Desalination is by definition a process removing minerals and salts from saline water to produce freshwater, that can be
used for human use or irrigation. It’s applied to seawater and brackish water with different performances criteria.
It’s normally considered that salinity below 500 ppm is suitable as drinking water. Basically, a complete desalination
process includes 3-4 steps with, first pumping water (from sea, estuaries or saline aquifers), pretreatment of pumped water
(filtration, chemical addition) desalination process ‘stricto sensu’ and last, a post treatment if necessary (in some case,
adding few minerals).
There are different ways to produce freshwater with desalination technologies. More common technologies are:
• Reverse Osmosis
• Multi-stage Flash Process
• Multi Effect distillation.
As shown in table 1, those technologies have been developed worldwide with MSF and RO being dominant.
Table 1. Distribution of solar technologies in existing installations [2]
An example of a column heading Total desalination plants (seawater +
brackish water as feed in)
Desalination plants with seawater as
feed in
Multi Stage Flash Process 43,5% 66,3%
Reverse osmosis 43,5% 22,4%
Multi effect distillation ~ 10%
Strong increase in worldwide installations
2.1. Reverse Osmosis
The RO technology is based on the properties of semi-permeable membranes which can separate water from a saline
solution, when excess of osmotic pressure is applied on the membrane systems. Pressure is applied with a high pressure
pump (approximately 70 – 90 bars). Part of the flow (35 to 50%) goes through the membrane, with a salt concentration less
than 500 ppm, rest of the flow called retentate (50 to 65%), containing high concentration of salts, doesn’t pass through the
membrane and is directly rejected at a high pressure.
RO can be applied to different types of water: seawater as well as brackish water, with the equivalent objectives
depending on the pressure applied to the membrane
Reverse osmosis has known a great development over the last twenty years due to its easy and rather low cost technology
and great improvement on membrane quality.
Key features of the RO process are the following:
• Low energy consumption
• Easy and ready to use : immediate stop and start
• Needs important pre-treatment : pre-filtration and chemical (anti-scalant) to avoid fouling on the membrane
• Outlet salt concentration around 500 ppm
Fig. 1. Schematic of membrane RO systems
RO capacity has increased in the last 30 years to reach approximately 45 – 50 % of the desalinated total capacity.
Pre-treated
feed water
2/3 Waste Stream
(high concentration in salts)
1/3 drinking water

222 Pascale Compain / Procedia Engineering 46 ( 2012 ) 220 – 227
Many developments contributed to the reduction of unit water cost (material, membrane module performances, reduction
in energy consumption), with now a unit water price around 0,5 US$/m3
2.2. Multi Stage flash
In the MSF process, seawater is heated in a vessel called the brine heater, up to a temperature of 120°C, and then flows
into another vessel, called a stage, where the ambient pressure is lower, causing water to boil. Steam is then condensed on a
range of tubes passing through the vessel.
Low pressure ensures seawater cooling down to 40°C.
This well known technology has been used on large installations (more than 50 000 m3/day), with coupling of heat
generation from a power plant, but MSF process can also be used with solar power.
Key features for MSF are the following:
• High reliability
• No need for complicated pre-treatment due to very limited scaling: simple filtration, and anti-scalant
• High investment cost
• High quality of produced freshwater (depending on the number of cells): salts concentration below 50 ppm
• Low running flexibility (low variation in flowrate)
•
Fig. 2. Schematic drawing of MSF process
MSF plants are now installed worldwide, even though performance is not high and represent 40 – 45% of the total
capacity in desalination.
Over the last 40 years, cost of desalinated water with MSF technology decreased by a factor 10 (from 10 $/m3 in 1960
down to less than 1 $/m3 in 2002) [3].
Hot fluid (oil, steam)
Seawater feed
Flash VaporisationVapor
Water pressure
Salt concentrated brin
e
Distillate
Concentrated waste
discharge
Vacuu
m
Concentrated waste
discharge
Flash Vaporisation
Salt concentrated brin
e
Vapor
Flash Vaporisation
Salt concentrated brin
e

223
Pascale Compain / Procedia Engineering 46 ( 2012 ) 220 – 227
2.3. Multi Effect Distillation : MED
Multi Effect Distillation (MED) process is based on using latent heat of condensation of the vapor from the first cell to
provide heat to a second cell. The evaporation takes place in cells where equilibrium temperature (Te) liquid / vapor is
between 40 ° C and 68 ° C.
In the first cell, the steam produced is injected into the second effect, in order to ensure the evaporation and condensation
at a lower temperature. This is then repeated in all following cells
In case of using solar source as heat source, hot water from the solar collector is introduced at the bottom tray, either
directly or through a heat exchanger.
MED units with horizontal sprayed tubes are generally made with materials like, aluminum brass (AlBr) for tubes and
stainless steel 316L for the casing. Sometimes, vertical heat exchanger can be used.
Key features for MSF are the following:
• High reliability, easy start of the system (less than 1 hour)
• High quality of produced freshwater (depending on the number of cells): salts concentration below 50 ppm
• No need for complicated pre-treatment due to very limited scaling: simple filtration, and anti-scalant
• Can be used on low temperature heat (from 60°C), which can be easily recovered as by-product in industrial plants.
MED is also suitable for small capacity installations and now represents more than 10% of the total capacity in
desalination, with still promising developments. Use of coupling MED with Thermal vapor compression helped in
decreasing running costs, as well as increase in unit capacity and heat transfer coefficients.
Water cost with recent MED is 0,7 US$/m3 which is close to RO [3].
Fig. 3. Schematic drawing of MED process
Heating fluid (steam, oil)
Seawater /
Brackish Water
Feedwater
vaporisation
Steam
Condensation
Steam transfer in next cell
(lower T)
Final
Condenser
High concentrated
salts wastewater
Air Blower
Steam transfer in next cell
(lower T)
Feedwater
vaporisation
Steam
Condensation
Temperature

224 Pascale Compain / Procedia Engineering 46 ( 2012 ) 220 – 227
Central Tower
(CNRS Promes)
Linear Fresnel
(CNIM)
Parabolic troughs
(NEP Solar)
2.4. Cost comparison of the different desalination technologies
As a brief outline, we can indicate energy consumption for the different desalting technologies above.
Table 2. Comparison of running cost in the different desalination technologies (per m3 of produced freshwater)
Consumption of desalination
technologies
MSF MED RO
Heat power consumption
80 -120 kWth/m3 50 - 90 kWth /m3 N/A
Electricity power
consumption
2.5 kWeh/m3 1 - 3 kWeh/m3 3 - 6 kWeh/m3
3. Solar Technologies
All desalination technologies described here mainly use thermal energy. Knowing that desalination needs are mainly in
dry countries receiving huge intensity solar radiation, it comes as an evidence to use solar power for the running of the
plants. Solar technologies can produce heat and thus electricity through a turbine (CSP), or directly electricity (PC and
CPV). Growing interests and developments on CSP as well as PV tend to make those technologies more and more attractive
Solar technologies can be divided in two categories:
concentrating solar power (CSP) technology and photovoltaic
(PV) technology.
Concentrating Solar Power Technologies mainly include
• Parabolic trough,
• Linear Fresnel reflector systems.
• Central tower receiver
These technologies concentrate solar radiation onto an
absorptive pipe (receiver) which contains heat transfer fluid
(water, oil or salt). When water is used, heat from the sun’s
radiation converts the water to steam by direct steam
generation. In case of oil or salt, fluid acts as an intermediate
thermal energy carrier, which passes through different heat
exchanger converting water to steam.
These CSP systems generate power using a steam-driven
turbine (with a Rankine cycle, for example)
Photovoltaic technology has been developed as flat-plate PV module, which is the most commonly used technology
(PV), and concentrating photovoltaic (CPV) technology, which is a developing technology. The main difference between
the two technologies is that CPV technologies use DNI as a solar source instead of Global Irradiation.
Recent developments in solar technology (material, geometry design of solar collector, sun tracking systems) helped
increasing efficiency in performances of the global systems.
We’ll go in more details on the different technologies, especially the most common ones which are parabolic trough,
linear fresnel reflector system, as well as photovoltaic modules. On a practical point of view, each one of the different
technologies will be selected and installed depending a list of criteria: technical, financial and site related constraints

225
Pascale Compain / Procedia Engineering 46 ( 2012 ) 220 – 227
3.1. Parabolic troughs
CSP parabolic troughs produce steam and electricity through the use of a parabolic reflecting surface that concentrates
direct normal solar radiation (DNI) onto a receiving tube surrounded by a glass element.
The solar collectors track the sun from east to west during daytime, allowing the continuous focus of sun on the solar
collectors. Thermal fluid is transferred through the receiving tube where it is heated to approximately 350 - 400°C. It is
then used as heating fluid to the power plant for generation of high-pressure steam. Thermal storage (on sensible or latent
heat) can be added to the system to ensure continuity during low solar radiation period (nighttimes or cloudy days).
Parabolic trough is the most developed of the CSP technologies and is now commercially available for industrial heat
production purpose.
3.2. Linear Fresnel Reflectors Systems
A linear Fresnel reflector (LFR) technology uses thin segments of mirrors arranged in long parallel lines to reflect
sunlight onto a fixed receiver, allowing thus to transfer energy through the absorber into some thermal fluid (water or oil).
Concentration capacity of the mirrors is approximately 30 times sun’s energy normal intensity. As for parabolic troughs,
thermal storage can be added to the system.
The fluid then goes through a heat exchanger to power a steam generator, thus producing electricity and heat as a by-
product.
LFR technology has great advantages on parabolic troughs and is the most promising technology in CSP. First,
manufacturing process doesn’t need high precision in bending mirrors, keeping Fresnel mirrors at a low cost, with
manufacturing sites close to the installation location; then, structure and equipment are much lighter than parabolic troughs,
which is another cost saving argument.
Fig. 4. Schematic drawing of LFR with different uses (CNIM)
Fig. 5. Details of LFR installation 0,7 MW (CNIM – F - La Seyne)

226 Pascale Compain / Procedia Engineering 46 ( 2012 ) 220 – 227
3.3. Flat Plate Photovoltaïc
Photovoltaic panels generate electrical power through the use of semiconductors, converting both DNI and diffuse
irradiation into electricity. The most efficient photovoltaic technology is a flat-plate module with a core composed of mono-
or polycrystalline silicon cells, which. A less efficient panel uses a thin film of cadmium telluride deposited onto a substrate.
The cells or film absorb both direct and indirect solar radiation, which excite the electrons and induce an electric current.
The panels are interconnected into a circuit to convert the direct current into an alternating current for the grid.
With diffuse collection, flat plate PV systems can operate effectively under conditions of light cloud cover
3.4. Concentrating Photovoltaic (CPV)
Almost similar to the Flat Plate PV, CPV technology is the most promising one with the following differences:
• CPV technologies use DNI as a solar source instead of Global Irradiation.
• CPV systems incorporate optical components such as mirrors combined with optical devices, to concentrate the
DNI onto the photovoltaic cell, thus improving significantly the solar energy reaching the surface.
• CPV technology uses multi-junction cells designed to convert a whole spectral region of solar radiation to reach
great efficiencies, almost double the efficiency of the conventional PV solar cell.
• CPV technology requires a highly accurate two-axis tracking system with tracker control units to continuously
track the sun for a maximum DNI during daytime.
CPV technology may need cooling systems to disperse heat due to high concentration of solar radiation on cells.
4. Coupling solar energy and desalination
When coupling desalination methods and solar power, it can result in many combinations. In that part, we will describe
and evaluate here projects already realised (mainly PV/RO, parabolic troughs/MSF, parabolic troughs/MED).
4.1. Photovoltaic / Reverse Osmosis
Many demonstration plants were carried out coupling RO systems with solar PV electricity. Most common way is to
convert DC from PV panels through an inverter, to produce AC immediately used in pumps.
Some developments anyway were done on connecting panels to a brushless DC motor, powering a low pressure pump,
thus allowing direct use from PV panel to pump [4].
Furthermore, battery banks are added to the system to store energy during night time to allow continuous running of RO
operation.
PV-RO systems od capacities ranging form 0,5 to 50 m3/d have been installed as demonstrators.
Regarding costs, investment cost is higher than conventional RO, but also vary on location, quality of saline water and
plant capacity.
In Ref [5] a complete table lists the different characteristics of PV coupled RO installations. Cost is extremely dependant
from one location to the other, ranging from 30 US$/m3 down to 3 US$/m3
Cost reductions have been investigated, smartest approach is to eliminate storage battery (giving 15-20% of cost
reduction) with varying flow of seawater through membrane with regards to available energy [6]
4.2. Parabolic troughs / MSF
The MSF process as a thermal process can also use solar power with parabolic troughs. Steam produced by parabolic
troughs is used as heat source through a heat exchanger for MSF inlet. A thermal storage system can be added to the system
to smooth variation of thermal energy supply and allow the continuous production of fresh water (during nighttime or low
radiation period)
Demonstrator realised in Kuweit [7] showed a capacity of 10 m3/d, for a surface parabolic trough collector of 220 m².
This corresponds to the average 10 – 60 l/m².day for solar powered MSF, as referenced in [8]
There are commercially available small-scale units combining MSF process with steam generating parabolic troughs,
whose approximate costs is 7-9 US $/m3 of produced freshwater

227
Pascale Compain / Procedia Engineering 46 ( 2012 ) 220 – 227
4.3. Parabolic troughs / MED
We can find more installations of MED worldwide, coupled with solar energy from different technologies (tower,
parabolic troughs, others…).
Water production cost for seawater desalination with MED coupled with a solar field has been proven to be quite
dependant from plant capacity: for large planst (5000 m3/d), cost is around 2 US $/m3 and is increasing up to 3,2 US$/m3
for smaller plants (500 m3/d) [9].
Also as an example, Abu Dabhi installation ref [10], can run 85 m3/day of freshwater; in that case, heat source is water
passing through evacuated-tube solar collectors (1862 m2). After several years of operation, critical point of the installation
was maintenance of the pumps due to silt removal.
Some demonstrators were also realized to prove a total autonomy on energy with both thermal and electric energy
(Parabolic troughs and PV panels), ref [11].
Aquasol is a hybrid project running with both gas and solar energy. Now in operation for several years, freshwater
production is 3 m3/h of freshwater, combined to parabolic troughs with a surface of 500 m2, ref [12].
We can also point as an example, a project with combination of a tower with small scale MED producing 25 l/m2day for
approximately 5 Wxh/m2d of solar radiation.
5. Conclusion
Solar MED and MSF, though appearing to be natural and tempting solutions, cannot be taken as proven technologies.
More and more developments in both solar power and desalination technologies are expected to keep these solutions
competitive compared to RO systems coupled with conventional power plant.
Somewhere, solution lies in combination: combination of heat sources and combination of processes. Recent
developments present desalination systems including MSF combined with reverse osmosis, heat source being a thermal
power plant, but coupling with solar resource can be imagined
Main advantage of that installation is that RO can run during nighttimes when electricity costs are low and MSF work
during daytime with low running costs due to low pressure steam. Water storage offers flexible solution with energy
consumption optimised. Addition of solar energy from start of the project (with parabolic troughs) would have low cost
impact but no GHG emissions.
We can note that the range of possibilities is widely open in desalination. Need for fresh water will always be present,
therefore desalination technologies must be enhanced to become cleaner, more efficient and more virtuous.
References
[1] Seawater desalination using renewable energy sources - Kalogirou - Progr Energy Combustion Science, 31 (2005) 242 – 281
[2] Precise way to select desalination technologies - Khalid Al Subaie – Desalination 206 (2007) 29-35 - Elsevier
[3] Overview of the cost of desalinated water and costing methodologies - K.V.Reddy, N.Ghaffour - Desalination 205 (2007) 340 - 353 - Elsevier
[4] A photovoltaic-powered seawater reverse-osmosis system without batteries – M. Thomson, D. Infield - Desalination 153 (2002) 1–8
[5] Technical and economic comparison between PV-RO system and RO-Solar Rankine system - D. Manolakos et al - Desalination 221 (2008) 37–46
[6] Solar-driven desalination with reverse osmosis: the state of the art – A. Ghermandi, R. Messalem - Desalination and Water Treatment - 7 (2009) 285–
296
[7] Performance of a self-regulating solar multistage flash desalination system- S.M.A. Mustafa, D.I. Jarrar and H.I. Mansy - Solar Energy, 35 (1985) 333.
[8] Solar Desalination of Water - D.L. Block -FSECRR-14-89, Florida Solar Energy Center, Cape Canaveral, February 1989.
[9]) Techno-economic evaluation of a solar powered water desalination plant - G. Fiorenza, - V.K. Sharma - Energy Conversion and Management, 44,
(2003) Pages 2217–2240
[10] The solar desalination plant in Abu Dhabi: 13 years performance and operation history - A. El-Nashar and M. Samad -Renewable Energy, 14 (1998)
263–274.
[11] Proving test for a solar-powered desalination system in Gaza-Palestine -S. Abu-Jabal Moh’d, I. Kamiya and Y. Narasaki - Desalination, 137 (2001)
1–6.
[12] First experimental results of a new hybrid solar/gas multi-effect distillation system: the AQUASOL project – D.C Alarcón-Padillaa, J. Blanco-
Gálveza, et al - Desalination 220 (2008) 619–625.