You are on page 1of 9

Desalination 427 (2018) 1–9

Contents lists available at ScienceDirect

journal homepage:

Desalination and energy consumption. What can we expect in the near MARK
Domingo Zarzoa,⁎, Daniel Pratsb
Valoriza Agua, Paseo de la Castellana, 83-85, 28046 Madrid, Spain
University Institute of the Water and the Environmental Sciences, University of Alicante, San Vicente del Raspeig s/n, Alicante, Spain


Keywords: Desalination technologies have become necessary tools for hydrological planning, along with conventional water
Desalination resources.
Reverse osmosis One of the main barriers to extend desalination is higher water costs, which are seriously influenced by
Energy energy consumption (represents > 50–60% of total costs).
This paper takes into account relevant energy and desalination aspects and different available technologies.
This work focuses mainly on reverse osmosis, which is the most widely used technology. It also considers
aspects like major consumers and their contribution to overall. The configuration of HP pumps, membranes and
ERDs is explored, as is what we can expect in the future, while we bear in mind that we are currently very close
to reaching the thermodynamic limits for energy consumption.
Special dedication will be taken to renewable energies and how they can be combined with desalination to
produce more efficient systems, though not necessarily directly coupled to desalination plants.
How to choose the most adequate energy rates and the production plus storage combination is another im-
portant consideration to optimize energy costs. An adequate production management strategy can significantly
reduce the cost of water.
To conclude, emerging technologies are analyzed, by looking at all the possible improvements and potential
generated uses.

1. Introduction Regarding membrane technologies, except for a few emerging

technologies (e.g., membrane evaporation or pervaporation), they all
Desalination has become one of the world's most important un- operate by supplying electrical energy, but in different ways; by
conventional water resources in recent years, and it is particularly re- membrane pressurization (Reverse Osmosis (RO) and Nanofiltration
levant in places where water is scarce. (NF)) or by passing a direct current between electrodes to separate ions
Together with the huge desalination benefits (increasing water re- by ionic membranes (Electrodyalisis Reversal (EDR)).
sources and improving water quality), there are still ample research and Other processes that can be used for salt removal, such as ion ex-
improvement opportunities, especially in those aspects related to re- change, precipitation or freezing, are not used for desalination on a
ducing energy consumption (EC). large scale.
According to IDA (International Desalination Association), desali- As evaporation technologies use much energy, RO is currently the
nation cumulative contracted capacity is estimated at 99.8 millions of most prevalent technology, with approximately 65% of the installed
m3/day worldwide (considering the installations built since 1965), capacity [2], and with the majority of new facilities. For this reason, the
while the capacity of operative plants is 92.5 millions of m3/day [1]. present paper focuses mainly on this desalination technology.
Among the different existing desalination technologies, there are An analysis by Acuamed [3] (the Spanish public company in charge
two main process groups; processes based on evaporation and processes of large desalination projects) with over 12 large SWRO (seawater RO)
based on membranes. Evaporation processes can operate by heat plants in Spain and a global capacity of 409 hm3/year, determined by
supply, as in Multistage Flash (MSF) and Multi Effect Distillation energy costs, represents around 60% of water production costs, with
(MED), or through the supply of electric energy by mechanical com- variable plant production (60% for 100% water production and 30% for
pression (Vapor Compression – VC). 20% water production).

Corresponding author.
E-mail address: (D. Zarzo).
Received 31 August 2017; Received in revised form 25 October 2017; Accepted 25 October 2017
0011-9164/ © 2017 Elsevier B.V. All rights reserved.
D. Zarzo, D. Prats Desalination 427 (2018) 1–9

Table 1 Table 3
Specific energy consumption in different large desalination plants (design). The EC breakdown at the Southern Seawater Desalination Plant (SSDP), Binningup,
Source: Valoriza Agua. Western Australia (SWRO, 306,000 m3/day).
Source: Valoriza Agua.
Plant Capacity Contract year SEC (kW-h/m3)
(m3/day) Process SEC (kW-h/m3) Percentage (%)

Skikda (Algeria) 100,000 2004 3.65 Intake-pumping station 0.33 8.9

Honaine (Algeria) 200,000 2005 3.5 Pretreatment (Ultrafiltration, UF) 0.05 1.3
Aguilas (Spain) 200,000 2008 3.3 (5.4, including RO 1st pass 2.3 (2.2 real operating 62
product water value)
pumping) RO 2nd pass 0.35 9.4
Southern Seawater 305,000 2012 3.1 (3.7, including Ancillary equipment 0.01 0.3
desalination plant product water Post-treatment 0.01 0.3
(Australia) pumping) Product water pumping station 0.58 15.6
Effluent treatment plant 0.02 0.5
Ancillary equipment and others 0.06 1.6
Specific energy consumption (SEC) in large SWRO plants is cur- TOTAL 3.71
TOTAL with no product pumping 3.13
rently in values of 2.5–4 kW-h/m3, as shown in some examples pre-
sented in Table 1.

3. Sources of energy consumption in RO desalination plants

2. Energy consumption in desalination
There are different sources of energy consumption in RO plants,
The minimum energy required for desalination is the equivalent to where it is possible to act to optimize or reduce energy use. A real
the energy produced when salts are dissolved in pure water. example of an energy use breakdown in a SWRO plant is found in
Table 3. Of course, high pressure systems for RO stage form the ma-
Energy produced = R T lnaW jority of uses, and are some processes that are strongly influenced by
local conditions, such as product water pumping intake.
where The most relevant aspects of the different processes from an energy
use point of view are commented on below.
R = constant and
T = absolute temperature
3.1. Intake and product water pumping stations
aW = activity of Water
Considering seawater with 35,000 ppm of TDS (total dissolved so-
The location of desalination plants and the delivery point are often
lids), at 25 °C the minimum required separation energy is 13.6 cal/mol,
predetermined by the promoter. Then there is not much left to do to
which is the approximate equivalent to 0.9 kW-h/m3. Real consumption
reduce energy use, except for pump selection (efficiency), motor se-
is higher and also depends on other factors, such as operating condi-
lection (induction motors, permanent magnetic motors, etc.) and re-
tions, temperature, recovery, etc.
ducing pressure losses in the pipe layout and installation of VFDs
This means that we are currently close to the thermodynamic limit
(Variable Frequency Drives) to adapt the working point of pumps to
(considering that it is not an ideal system), and it is becoming in-
real needs at any given time.
creasingly difficult to reduce energy consumption in desalination
Intake or product water pumping entails different casuistries, based
technologies below current values. However, the evolution of energy
on differences in heights and distances, ranging from 0.1–0.2 kW-h/m3
consumption in recent years has been significant (see Table 2), which
to much higher values.
represents the historical evolution in desalination plants in Spain,
In some cases, product water pumping can use even more energy
published in 2009 by AEDyR (Spanish Desalination and Reuse Asso-
than the desalination process itself. By way of an example, the SWRO
ciation) [4]. The data table also shows the historical reduction of SWRO
plant called Mantoverde, installed on the coast of Chile for the mining
energy consumption due to the inclusion of the new incoming ERDs
industry, feeds water to the industry located at a height of 700 m and a
(Energy recovery devices, such as Francis turbine, Pelton turbine, hy-
distance of 35 km, which uses 2.5 kW-h/m3 in desalination and 2.5 kW-
perbaric recovery devices, etc.).
h/m3 in product water pumping (data from Valoriza Agua).
Height differences in brine discharge (discharge height-seawater
Table 2 level) can also be taken advantage of, and it is possible to use a turbine
Evolution of average SEC in Spanish seawater desalination plants. to produce energy, which is the case of the SWRO plant in Adelaide
(AEDyR, 2009) [4]. (Australia). In this plant (300,000 m3/day production) taking ad-
vantage of 40 m height difference, 685 kW are produced at 100% ca-
Year Technology kW-h/m3
pacity (using Francis turbine), representing 3% of the energy con-
1970 MSF 22 sumption [5]. Valdelentisco SWRO plant in Murcia (Spain) also
1980 MSF 18 produces 133 kW at 100% capacity (140,000 m3/day) [6]. The main
1985 VC 15
problem with this kind of turbines is the reduction of efficiency for
1988 VC 13
1990 RO 8.5
lower flows and foam generation. Other option for smaller plants can be
1994 RO 6.2 the installation of microturbines.
1996 RO 5.3
1998 RO 4,8
1999 RO 4,5 3.2. Pretreatment
2000 RO 4,0
2001 RO 3,7 Although pretreatment is not the major energy user in RO plants,
2002 RO 3,5
some design aspects can determine the most efficient process; it is
2005 RO 3
2009 RO <3 possible to choose between conventional or membrane treatments,
work by gravity or pressurized in different stages, and intermediate

D. Zarzo, D. Prats Desalination 427 (2018) 1–9

Table 4
Comparison of different pretreatment alternatives.
(Adapted from Buendia et al. [7]).

Case Variable Parameter SEC (kW-h/m3)

Gravity filtration + pressurized filtration. Filter backwash made with filtered water Pretreatment recovery (92%) 0.620
Gravity filtration + pressurized filtration. Filter backwash made with brine Pretreatment recovery (96%) 0.613
Double stage filtration (pressurized). Filter backwash made with brine Pretreatment recovery (98%) 0.680
1 stage pressurized filtration. Filter backwash with brine Pretreatment recovery (98%) 0.585
1 stage pressurized filtration. With no low pressure pumps. Filter backwash with brine Pretreatment recovery (98%) 0.576
Double stage filtration (gravity). Filter backwash made with brine Pretreatment recovery (96%) 0.543

pumping also impacts energy use.

Table 4 provides a comparison for different pretreatment systems
(adapted from Buendia et al. [7]). Data were obtained by theoretical
calculations based on the following assumptions:
Production: 210,000 m3/day
Expected water product quality: Boron < 0.5 mg/l (which implies a
2nd pass)
Height level at site: + 20 m.w.c. Fig. 1. The hybrid-split concept.

Ancillary equipment considers a fixed consumption

Efficiency of pumps: 85% The hybrid concept is based on using different membranes inside
Efficiency of motors: 93% pressure vessels, while the split is the product water extraction from
First pass recovery: 45% both sides of pressure vessels, as shown in Fig. 1.
Second pass recovery: 90%. The main advantages of a hybrid configuration are the flux balance
At the time of this study, not many large facilities had been installed and water quality along the pressure vessels, and fouling is less likely.
with Ultrafiltration, which is why the authors did not include this al- This configuration combined to a split design maximizes the production
ternative at that time. In recent times, and with the development of new of a high-quality permeate, which can be sent directly to the product
UF modules, the SEC of membrane pretreatments have been reduced, water tank to thus reduce the size of the 2nd pass and to operate with
being similar to conventional ones, so this is not usually the key factor lower pressure, shown in Table 6.
for making decisions about the pretreatment kind (Capex, Opex,
membrane fouling, water quality, etc.).
Table 5 offers a comparison for different pretreatments at the SSDP 3.4. High pressure pumps
Plant in Australia.
Considering membrane pretreatment systems, the installation High pressure (HP) pumps are the major energy consumers on all
without intermediate tanks in another option which saves energy sig- RO desalination plants. The search for this equipment's (pumps and
nificantly, although it requires a number of standby systems (higher motors) maximum efficiency lies with manufacturers, and then there is
Capex) and a more complex automation and control system to couple not much left that designers can do, except to select the best available
UF and RO in continuous way. commercial equipment at the optimum working point.
As required pressure depends on temperature, recovery, membrane
fouling and aging, etc., it is necessary to design HP pumps that are able
3.3. Membranes
to work in different scenarios. Leaving to one side the effect of control
valves, which implies wasting energy, installing VFDs, has become an
Membranes are the core of RO plants, and R & D in this field can
essential tool to manage the working point and curves of pumps.
produce the most important results in energy reduction. Developments
Regarding HP pumps, there are two ways to optimize SEC; install
can be seen from both the manufacturer and industry points of view.
VFDs in HP pumps and the booster pump in the energy recovery loop
From the membrane manufacturers perspective, the main research
(Fig. 2); install a booster pump with a VFD before the HP inlet (Fig. 3).
involves searching for new chemicals for active layers and their su-
The advantages of this last option (booster pump with a VFD) is a
perficial modification (nanoparticles, tubes, fibers, etc.), as well as
reduced CAPEX (VFDs for HP pumps are very expensive because of high
membrane configuration (modules, engineering, spacers).
electrical power) and it guarantees that the high pressure pump always
Research into new membranes aims to not only increase flow and
works at its most efficient working point (fixed), while a booster pump
rejection, but to also reduce pressure with better product water quality.
is that which manages the different pressure requirements.
Other innovations could be related to construction methods (e.g., 3-
Another decision about installing HP pumps can be made: installing
D printing) or to the development of “smart membranes”.
a dedicated pump (each train has its own HP pump) (Fig. 4) or a
From the industry point of view, some decisions can be made by
pressure center (some HP pumps supply different RO trains in the
playing with commercial products, such as membrane and module se-
lection, configuration (stages, passes), and hybrid and split-hybrid
Table 6
systems. Design values for the hybrid-split system at the SSDP (Australia). Values for 20 °C and a 5-
year projection. Hybrid: 2 high rejection membranes + 5 high rejection membranes.
Table 5 Source: Valoriza Agua.
Comparison of EC for different pretreatments at the SSDP (Australia).
(Source; Valoriza Agua). Design Pressure (bar) Blending

Case Configuration SEC (kW-h/m3) 1st pass 2nd pass

1 Double stage filtration (pressurized) 0.55 Conventional 57–62 12.5 12%

2 Gravity filtration + pressurized filtration 0.54 Hybrid-split 52 11.5 25%
3 Ultrafiltration 0.46 Reduction 9% 8%

D. Zarzo, D. Prats Desalination 427 (2018) 1–9

“common rail” concept) (Fig. 5).

The pressure center concept can be applied to different process
stages (pretreatment, HP pumps, Energy recovery devices (ERDs), etc.).
The main advantage of pressure centers in HP pumps is that there is a
lower Capex (fewer HP units) and higher efficiency in pumps due to
their larger size. Although this theoretically implies less energy con-
sumption, lack of flexibility when working with capacities below 100%,
plus a very complicated hydraulic distribution and control, mean that
many designers do not prefer this option.

3.5. Energy recovery devices

3.5.1. Seawater
Fig. 2. A HP pump with a VFD (represented in orange). (For interpretation of the re-
The use of ERDs for seawater RO was established entirely many
ferences to color in this figure legend, the reader is referred to the web version of this years ago. Due to the high pressure supplied to membrane remains in
article.) brine, with the reduction caused by pressure losses in membranes,
pipes, valves, etc., this pressure can be used to reduce the plant's energy
use. For this purpose, a number of different devices has been histori-
cally used, such as turbines (Pelton, Francis) or current pressure ex-
changers (aka isobaric, hyperbaric), with different designs (displace-
ment, rotating), which are widespread in all modern seawater plants.
Logically, these systems are very efficient with seawater given
higher pressures and higher brine flows, while their use and effective-
ness are lower and less frequent with brackish water plants (they work
with lower pressures and brine flows).
In the study previously mentioned in Section 3.2. (Buendia et al.
[7]), a comparison was also made of the different options for the con-
figuration of high pressure pumps and ERDs, as shown in Table 7.
Further to the premises described in Section 3.2., the following data
were assumed:
First pass RO inlet pressure: constant 67 bar
Second pass RO inlet pressure: constant 12.5 bar
Fig. 3. A HP pump with a previous booster pump with a VFD (VFD in orange). (For
Efficiency of ERDs (hyperbaric type): 94%.
interpretation of the references to color in this figure legend, the reader is referred to the
web version of this article.)
For the comparative calculations, the same temperatures, mem-
branes and membrane agings were used.
Despite the simulation dating back to 2011, not many advances
have been developed in recent years for it to be rendered obsolete.

3.5.2. Brackish water

As previously mentioned, ERDs are less frequently used with
brackish water treatment plants due to lower flows and pressures.
However in some cases, where the plant's size and working pressure is
large/high enough, it is possible to use some of these ERDs.
By way of example, the BWRO Cuevas de Almanzora, in Almeria,
Spain, has installed interstage turbochargers (Fig. 6) to recover > 30%
of energy with a SEC of 0.9 kW-h/m3 (product water pumping not in-
cluded) at 15,000 μS/cm conductivity in raw water, as shown in Table 8
Fig. 4. Dedicated pump concept. The ERD could have been installed by feeding the HP Pump, but the
interstage configuration has the advantage of producing an additional

Table 7
Comparison of different HP pumps, ERDs and RO trains.
(Adapted from Buendia et al. [7]).

Case Configuration SEC (kW-


1 High pressure pump 1st pass with Pelton wheel as an ERD, 3.31
high pressure pump 2nd pass with a VFD
2 High pressure pump 1st pass with a booster pump with a 2.960
VFD. Hyperbaric ERD system with a booster pump with a
VFD, high pressure pump 2nd pass with a VFD
3 Case 2 + interstage turbocharger in 2nd pass 2.923
Fig. 5. Pressure center concept. 4 Pumps in pressure center 2.835
5 Case 2 + Split-hybrid concept in membranes 2.713
6 Case 6 + SWRO membranes in 2nd pass 2.667
7 Case 2 without 2nd pass 2.248

D. Zarzo, D. Prats Desalination 427 (2018) 1–9

Fig. 6. Configuration of an interstage turbocharger at the

BWRO plant Cuevas de Almanzora.

Table 8 4. Energy recovery from brine

Pressure recovered by turbochargers in the four trains at the BWRO Cuevas de Almanzora
(Spain). By thermodynamic principles, the minimum energy needed for any
(Adapted from [8]).
desalination process equals the produced energy by dissolving salts in
Pressure Raw water Product water Recovered water. This means that if we are able to reverse the process by means of
(bar) conductivity (mS/ conductivity (μS/ pressure (bar) technologies such as Forward Osmosis (FO) (and the different variants
cm) cm) such as Pressure Retarded Osmosis (PRO)) we should have a positive
TRAIN 1 19.0 15.00 150 9.5
balance of energy. The practical idea for this process is to blend brines
TRAIN 2 18.5 15.00 145 8.5 with seawater, brackish water or wastewater to produce energy thanks
TRAIN 3 19.0 15.00 155 8.5 to the differences in osmotic pressures and chemical potential.
TRAIN 4 18.2 15.00 180 8.0 Although it is difficult to generalize and to quantify the produced
energy, for example Chung et al. [10] reported a recovery of 0.42 kW-
h/m3 in a PRO system combined with desalination and Ordonez et al.
hydraulic balance between stages, which are unbalanced for high raw
[11] reported 0.78 kW-h/m3 in experiences blending seawater with
water salinities. This imbalance can be corrected by a permeate valve
(backpressure), which means wasting energy, with an interstage
This idea is apparently new, but references that have explored this
booster pump (which implies further use of electricity) or by this kind
possibility can be found; e.g., Pattle in 1954 [12], and in further works
of ERDs.
by Loeb in 1974 [13], by means of what he called pressure retarded
3.5.3. Brine concentrator concept The current difficulties of this application lie in developing new
A different configuration to have been installed in some plants is the membranes (asymmetric RO membranes are not usable due to the op-
brine concentrator concept. In this configuration, seawater plants work posite flux direction), low flows and productivity, membrane fouling,
at higher recoveries (typically above 50%), which implies installing a and for high osmotic pressure solutions (called draw solutions), in
second stage fed by a booster pump, and one that requires special finding the most adequate (and their compatibility with the use of
membranes to work at very high pressure. This configuration has been product water) and further separation from water.
typically installed as a way to extend old plants to produce more water. Other technologies such as Electrodyalisis Reversal can also be used
However, it has not been demonstrated as a way to save energy in new to produce energy based on salinity gradients [14]. EDR can be used
plants. directly or in combination with other processes, as described in a work
by the University of the State of Pennsylvania [15], where a new
technology for producing energy was developed (MRC, microbial re-
3.6. Post-treatment
verse-electrodialysis cell). Strictly speaking, although the application
was developed by capturing the salinity gradient energy from ammo-
The different post-treatment alternatives have no significant impact
nium bicarbonate salt solutions, and not from desalination brines, this
on energy consumption in RO plants. All that mentioned before about
is a potential use to be explored.
pump efficiencies, height differences, etc., also applies here.
Brines from seawater desalination plants can also be used as the
draw solution to produce energy by PRO by using different sources of
3.7. Ancillary equipment (compressors, CIP systems, instruments, lighting, low salinity water (river water, wastewater, etc.) as feed solutions [16].
etc.) PRO has also been tested to produce energy with brines from eva-
poration technologies [17] with the differences caused by the distinct
The SEC of ancillary equipment represents a very small percentage technologies applied (salinity, high temperature, etc.).
of the total use. Precisely because of this, there is room for, and it is Finally, another reported technology for energy production by a
typical of, using renewable energies to feed them. salinity gradient, which could be used with brines, is hydrocratic gen-
To illustrate this section, and also as an interesting real example, eration. It consists of using a hydrocratic generator without mem-
Table 9 offers the historical development of the EC at the Las Palmas III branes.
SWRO plant (Spain). This table was adapted from Lemes et al. [9]. This
example also shows how an old facility can be improved with new in- 5. Operational strategies. Electricity tariffs
coming developments to produce a higher permeate flow, better water
quality and lower energy use. Operational strategies can be a useful tool to reduce energy costs in

D. Zarzo, D. Prats Desalination 427 (2018) 1–9

Table 9
Historical development of the energy consumption at SWRO Las Palmas III (Spain).
Adapted from Lemes et al. [9].

Design Year SEC (kW-h/ Reduction in SEC from Plant capacity (m3/ Permeate conductivity (μS/ Plant recovery (%)
m3) origin (%) day) cm)

Original design: Francis turbines, 315 ft2 1996 6.67 – 36,000 2600 45.00
New Pelton wheels (six trains) and 315 ft2 1998 5.85 12.3 36,000 1000 47.90
Intermediate booster pumps in HP pumps. Brine 2001 5.11 23.4 50,000 1234 48.60
concentrator concept
7th train (new) and increase in PV and membranes 2001 5.11 23.4 57,800 1234 48.60
in all racks
8th train (new) 2003 4.76 28.6 66,000 1498 51.16
10th and 11th trains (new) 2006 4.63 30.6 80,000 1100 52.80
Installing new ERDs (ERI-PX) in trains 5 & 6 2007 4.5 32.5 80,000 1212 53.30
Installing new ERDs (ERI-PX) in trains 4 & 7 2009 4.33 35.1 85,000 518 50.75
New ERDs in all racks 2011 4.1 38.5 86,500 404 50.08

Fig. 7. Desalination technologies coupled to re-

newable energies.

Table 10
Example of calculations for the supply of a large SWRO plant by solar energy [18].

Case Description Installed power (MWp) Generation Capex (millions of €) Amortization (years) Footprint (Mm2)

1 The whole plant fed by solar energy (SE) 350 100% 700 26 9
2 SE to compensate the peak consumption 255 100% 217 12 5
3 Legal auto-consumption 51.2 20% 46 11 1

desalination plants. In most countries, electrical tariffs depend on dif- and experiences of how to use renewable energies in desalination. Fig. 7
ferent periods (peak, off-peak, shoulder, etc.), which can be used to is a scheme of how to feed different technologies by distinct renewable
adapt the plant's production to the best tariffs. energy sources.
These strategies need some storage capacity to adapt production to The nature of the required energy depends on the desalination
requirements, and are more easily implemented in BWRO plants be- process, which can require different forms of energy (thermal, elec-
cause the frequent shutdowns of SWRO plants are not recommendable trical, etc.). The options for applying renewable energies in desalination
for membranes. are: 1) as a source of energy; 2) directly in the internal operation of
It is also possible to take advantage of this strategy if a plant has desalination systems.
seasonal production, but this can sometimes go against our interests; Although industry clearly intends to improve and implement re-
e.g., where energy costs are higher during peak water demands. newable energies in desalination, there are some difficulties for it to
As a relevant example, the BWRO plant Cuevas de Almanzora in spread:
Almeria (Spain) produces water for agriculture irrigation (30,000 m3/
day) with production automatically coupled to the electrical tariff - Production capacity: the energy demand from desalination plants is
timetable, and with automatic shutdown during peak periods and re- very high and the production capacity by renewables is generally
starts during lower price periods. Thanks to this measure, the average low
production cost is 0.34 €/m3 with an energy consumption of 2 kW-h/ - Location: the site selected for a desalination plant in not necessarily
m3 (0.9 in RO + 1.1 in product water pumping) for an average raw the best place to install a renewable energy system. This can be
water salinity of 15,000 μS/cm [18]. solved by locating the renewable energy far away from the desali-
nation plant and by supplying the electrical network which was the
6. Using renewable energies case of the SSDP plant in Australia, fed by the Mumbida wind farm
near Geraldton (400 km north of Perth, 22 turbines, 55 kW) and the
A complete book would be required to describe all the alternatives Greenough River 10 MW Solar farm (80 ha, 150,000 PV panels) [19]

D. Zarzo, D. Prats

Table 11
Comparison of the main emerging technologies.
(Adapted from [22]).

Technology FO PV MD CDI

Fundamentals A draw solution with high osmotic pressure is used Membrane separation process with nonporous Pre-heated saline water and permeate are on two An electrochemically induced alternative approach to
to extract the product water from seawater or high membranes applied to miscible liquids. Separation sides of a hydrophobic membrane, which maintains remove ions from concentrated aqueous solutions by
salinity water passing through a membrane. is produced by applying a vacuum on the the liquid streams out of the membrane. Due to ΔT forcing charged ions onto the electrical double layer at
Furthermore, water is separated from the draw membrane side where the permeate is collected as and the vapor pressure driving force, water is an electrode-solution interface when the electrode is
solution by means of thermal or membrane vapor, which is finally condensed as a product. vaporized from the feed side, diffused through the connected to an external power supply.
processes. membrane and finally condensed into the cold
permeate side, to leave salts on the feed side.
Strengths High rejections, low membrane fouling and Latent heat lower than evaporation techniques, Large membrane contact area, high salt rejection, Low operating cost. It is readily amenable to solar
potentially less operation energy useful for temperature- sensitive compounds small footprint and mild operation conditions, power for the photovoltaic production of the necessary

capable of integrating renewable energies from electrical power. It requires less pre-treatment of the
different sources feed water, up to 80% energy recovery, charge reversal
drives off any foulants adsorbed on the electrode
Challenges Reduced fluxes – increased membrane area. Lack of Low permeate flow rate, water flux and membrane Low permeate flow rate and water flux, membrane The most critical component is carbon electrode
effective draw solutions. Separation techniques. stability fouling and membrane pore wetting, long-term materials; electrosorptive capacity strongly depends on
performance, and uncertain economics and energy physical properties, such as surface area and
costs conductivity of the electrode.
Processes -Pressure- enhanced osmosis (PEO) -Vacuum PV -Direct contact MD –
-PRO -PV with carrier gas -Air gap MD
-Integrated FO-RO systems -Vacuum MD
Possible uses Desalination, osmotic power generation and others Desalination, dehydration of organic solvents, Desalination, in general, membrane contactors for Desalination
(osmotic MBR, landfill leachate treatment) separation of azeotropic solutions (ethanol/water) providing gas/liquid or liquid/liquid mass transfer
areas (separation of the aqueous solutions of
organics), water degasification
Current situation Pilot and small plants, some manufacturers produce Commercially used for the dehydration of ethanol Laboratory, pilot and small plants Pilot and small plants
membranes. and isopropanol. For desalination, pilot and small
Desalination 427 (2018) 1–9
D. Zarzo, D. Prats Desalination 427 (2018) 1–9

Table 12
(Relatively) new trends in desalination plants (own making).

Trends Remarks

Use of pressure exchange energy recovery systems Absolutely extended in SWRO plants. Less extended in brackish water systems.
Hybrid/split membrane systems Implemented in many large facilities. Reduced energy consumption.
UF as a pretreatment Increasingly extended in large plants. Unfortunately it is not valid for any plant (depends on water quality, fouling,
etc.), and some problems have arisen in large plants. Relatively little impact on energy consumption.
New UF membranes Continuously being developed. Comes close to the commercialization of new UF membranes close to NF. Relatively
little impact on energy consumption.
Use of larger membranes (16–18″) Installed in some large facilities to save the Capex, but not very extensive given membrane weight and management
difficulties. Relatively little impact on energy consumption.
New membranes with a modified surface (nanotubes, Being developed. Slowly commercialized until the appearance of new membrane companies with some large contracts.
nanoparticles) Apparently new developments are reducing energy consumption although in less extent that it was expected.
Large installations versus small installations Economy of scale versus localized energy consumption and brine discharge, proximity to end users and production
management. From the energy consumption point of view large installations are more efficient.
Reduced chemical dosing – zero chemicals A rising trend; reduction in production costs and fouling. Little impact on energy consumption.
Modularity, train sizes, pressure centers Energy optimization versus flexibility in operation.

- Intermittent production: it requires storage options - Membrane distillation (MD)

- Space for implementation - Capacitive deionization (CDI)
- Favorable environmental conditions - Use of nanoporous graphene
- Legal framework in different countries - Biomimetic membranes
- Capex: subsidies or penalties in different countries. - Aquaporins
- Microbial fuel cells
Currently, Gulf countries lead the use of renewables in desalination - Bioelectrogenesys
(e.g., the MASDAR project), and surprisingly and incomprehensibly, a - Electrodialysis metathesis and electrodialysis with bipolar mem-
Mediterranean country like Spain is currently penalizing its use by branes
implementing different taxes. - Others (ultrasound, cavitation, etc.).
Here is an example of economical calculations: a Spanish company
specialized in solar energy did a theoretical exercise by calculating the Table 11 shows the characteristics, advantages and disadvantages of
energy (investment, cost, footprint) needed to feed the SWRO plant in the most important emerging technologies.
Torrevieja (Spain), with the following data: 240,000 m3/day, 51.2 MW, All these technologies have arisen in recent years to develop desa-
SEC 4.78 kW-h/m3 [20]. The results of this work are shown in Table 10, lination processes with lower energy consumption. It is not expected
which indicates the difficulty of this application. Evidently, this is only that this energy consumption is not expected to significantly reduce
an academic exercise and the results will depend on the country, solar with new technical advances (membranes, more efficient ERDs), and a
radiation, price of energy, etc. point that comes close to the thermodynamic limit has been reached.
Another option to reduce energy use is to use dual and hybrid Therefore, it is not foreseeable that any of these technologies will re-
plants; the combined production of water and energy. The options in- place RO as the main mid-term desalination technology.
clude: However thanks to research into these technologies, interesting
potential applications have been found, such as the following:
- Direct use of steam to evaporate water (using power plants–dual
plants) - Energy production using a salinity gradient
- Use of steam to produce energy by means of a turbine, and further - Improving the efficiency of current technologies
water production by RO - Hybrid systems
- Mixed or hybrid plants; combined evaporation/RO processes. - Brine treatments.

Another problem of renewable energies is that they can be affected Evidently, the main potential use of these technologies is to produce
by climate change, having impact over wind, solar radiation, rain, energy using the osmotic potential, which has been previously de-
surface water flow, etc. scribed.
In a recent study [21], three hydroelectrical power stations were In a study which analyzed most of these technologies, along with
analyzed in Spain by modeling the energy production during the their advantages and disadvantages, Morillo et al. [23] concluded that
1930–2098 period by considering the predictable effect of climate membrane distillation, FO, electro-separation processes and metal re-
change. covery were the most promising technologies to treat brines from de-
According to this study, energy production is expected to lower, salination plants.
with values ranging from 33% to 73% depending on the power plant, Finally, Table 12 summarizes the state of current trends in RO
water flow, etc. This was studied in different scenarios by presenting plants.
lack of economical feasibility for the three power plants in the future.

7. Emerging technologies 8. Conclusions

By emerging technologies we mean those that are still in the re- Desalination has grown extraordinarily in recent years due to im-
search phase, or have not passed pilot plant stages or have not devel- proved technologies and cost reductions, but limits in energy use op-
oped prototypes. In the desalination field, the best-known are: timization still seem to exist.
Different processes and stages can be optimized when designing
- Forward osmosis (in its various forms: FO, PRO, etc.) desalination plants to accomplish a more efficient plant, besides new
- Pervaporation (PV) developments in membrane manufacturing and new ERDs.
A consensus has been reached in the industry about the need to use

D. Zarzo, D. Prats Desalination 427 (2018) 1–9

renewable energies in desalination. However some difficulties, such as [9] R. Lemes, J.L. Perez, L. Lorenzo, D. Zarzo, R. Falcon, R. Arocha, et al., Evolution of
production and energy savings in SWRO plant of Las Palmas III, IDA World
capacity, intermittent production or legal framework, limit this appli- Congress on Desalination and Water Reuse 2011. Perth (Australia), 2011.
cation. [10] T. Chung, L. Luo, F.W. Wan, Y. Cui, G. Amy, What is next for forward osmosis (FO)
Although this is not strictly reduction in energy consumption, the and pressure retarded osmosis (PRO), Sep. Purif. Technol. 156 (2) (2015) 856–860.
[11] A. Ordonez, B. Gutierrez, El aprovechamiento energético de la salmuera mediante
flexibility and management of production and shutdown periods can, osmosis directa (in Spanish), AEDyR (Spanish Desalination and Reuse Association),
besides offering adequate electrical tariffs, reduce energy-related costs. Madrid (Spain), 2012.
Unfortunately, the development of emerging technologies re- [12] R.E. Pattle, Production of electric power by mixing fresh and salt water in the hy-
droelectric pile, Nature 174 (1954) 660.
presents small slow steps for industry, but has promising applications, [13] S. Loeb, Osmotic power plants, Science 189 (1974).
such as energy production using salinity gradients. [14] R.A. Tufa, E. Curcio, W.V. Baak, J. Veerman, S. Grasman, E. Fontananova,
A radical change in technology is not expected in the next few years G.D. Profio, Potential of brackish water and brine for energy generation by salinity
gradient power-reverse electrodialysis (SGP-RE), RSC Adv. 4 (2014) 42617–42623.
and, in any case, coming close to a thermodynamic limit makes a sig-
[15] R.D. Cusick, Y. Kim, B.E. Logan, Energy capture from thermolytic solutions in mi-
nificant reduction in energy use a difficult task. crobial reverse-electrodialysis cells, Science 335 (2012) 1474–1477.
[16] F. Helfer, O. Sahin, C.J. Lemckert, Y.G. Anissimov, Salinity gradient energy: a new
References source of renewable energy in Australia, Water Util. J. 5 (2013) 3–13.
[17] W. Akram, M.H. Sharqawy, J.H. Leinhard, Energy utilization of brine from an MSF
desalination plant by pressure retarded osmosis, Proceedings of the International
[1] IDA, GWI, IDA Desalination Yearbook 2017–2018, Ed. Media Analytics Ltd, UK, Desalination Association World Congress on Desalination and Water Reuse. Tianjin
2017. (China), 2013.
[2] IDA, GWI, IDA Desalination Yearbook 2014–2015, Ed. Media Analytics Ltd, UK, [18] F. Molina, E. Campos, D. Zarzo, Energy recovery and optimization in a brackish
2015. water desalination plant with variable salinity, IDA World Congress on Desalination
[3] F. Lopez, Acuamed, the Spanish approach – the public corporation model, Regional and Water Reuse 2015. San Diego (USA), 2015.
Mediterraean Workshop on Desalination, Non Revenue Water Reduction and [19] M.A. Sanz, D. Zarzo, Australian Desalination plants. PPP models and solutions for
Public-Private-Partnership under Water Scarcity 2016, World Bank, Marseille, the environment, 2 case studies, World Bank Workshop 2016. Marseille, France,
France, 2016. 2016.
[4] AEDyR, Spanish Desalination and Reuse Association, (2009). [20] D. Lopez, Is it possible to obtain desalinated water from the sun? Water and Energy
[5] E. Crespo, Recuperacion de energia mediante turbinado de salmuera en plantas Workshop, University of Alicante, Spain, 2014.
desaladoras, Jornada Tecnica Servicios energéticos y las ventajas de la desalacion y [21] K. Solaun, E. Cerda, The impact of climate change in hydroelectrical generation.
reutilizacion del agua. Las Palmas de Gran Canaria (Spain) October 2017, 2017. Pilot study of 3 power plants, Aquaenergy Conference 2017, Madrid, Spain, 2017.
[6] J. Arrieta, Global partnerships for SWRO & energy recovery devices, Texas Desal [22] D. Zarzo, Problematica Y Soluciones para la Gestion Y Tratamiento de Salmueras
2017. Austin (Texas, USA) September 2017, 2017. Procedentes De Desaladoras (Problem and Solutions for the Management and
[7] R. Buendia, A. Diaz, C. Garcia, D. Zarzo, SWRO design; evolution of energy and Treatment of Brines from Desalination Plants) (PhD Thesis, 2017), University of
chemical consumption and future trends, IDA World Congress on Desalination and Alicante, Spain, 2017.
Water Reuse 2011. Perth (Australia), 2011. [23] J. Morillo, J. Usero, D. Rosado, H. El Bakouri, A. Riaza, F.J. Bernaola, Comparative
[8] C. Garcia, F. Molina, D. Zarzo, 7 year operation of a BWRO plant with raw water study of brine management technologies for desalination plants, Desalination 336
from a coastal aquifer for agricultural irrigation, Desalin. Water Treat. 31 (2011) (2014) 32–49.