AIChE Journal
DOI 10.1002/aic.15726
Perspective
A Perspective on
Reverse Osmosis Water Desalination: Quest for Sustainability
Yoram Cohen(a), Raphael Semiat(b), and Anditya Rahardianto(a)
(a) Department of Chemical and Biomolecular Engineering, and Institute of the
Environment and Sustainability, University of California, Los Angeles, CA, USA
(b) Wolfson Faculty of Chemical Engineering, Technion – Israel Institute of Technology,
Technion City, Haifa 32000, Israel
Submitted to AIChE Journal
This article has been accepted for publication and undergone full peer review but has not been
through the copyediting, typesetting, pagination and proofreading process which may lead to
differences between this version and the Version of Record. Please cite this article as
doi: 10.1002/aic.15726
© 2017 American Institute of Chemical Engineers (AIChE)
Received: Mar 27, 2017
This article is protected by copyright. All rights reserved.
Table of Content
1
Introduction – Water Scarcity and the Rationale for Water Desalination........................................................... 3
2
Water Desalination ................................................................................................................................... 6
2.1
Overview ....................................................................................................................................... 6
2.2
RO Desalination............................................................................................................................. 8
2.2.1
RO Emergence as the Standard Technology for Water Desalination ................................... 8
2.2.2
Mitigation of Fouling and Mineral scaling .......................................................................... 11
2.2.3
Concentrate Management .................................................................................................. 12
3
Configurations of RO Desalination Systems........................................................................................ 13
4
The Cost of RO Desalination ............................................................................................................... 18
5
The Energy Cost of RO Desalination ........................................................................................................... 21
5.1
Minimum energy consumption for a reversible desalination process ....................................... 21
5.2
RO Energy Consumption and the Thermodynamic Limit............................................................ 22
5.3
Multi-Stage RO Desalination ....................................................................................................... 26
5.4
Concentrate recirculation ........................................................................................................... 27
6
The Path To Improving RO Desalination ............................................................................................. 29
7
Summary ............................................................................................................................................. 31
8
References .......................................................................................................................................... 33
2
This article is protected by copyright. All rights reserved.
1
INTRODUCTION – WATER SCARCITY AND THE RATIONALE FOR WATER DESALINATION
Water is needed at different qualities for various types of municipal/public, agricultural, and
industrial consumption which account for about 23%, 67%, and 9%, respectively, of total fresh
water withdrawals in the US, excluding thermoelectric use [1]. Although water reclamation,
recycling, and reuse could be advanced so as to provide a significant percentage (20%–50%) of
the water use portfolio, at present it remains in the single digits in most regions of the United
States [2]. Water and energy are inextricably linked; nationwide, there is an urgent need to
reduce the energy cost of water delivery, which represents a major portion of the energy usage
in various sectors [3]. For example, the total energy consumption for direct water services in
the U.S. by the U.S. residential, commercial and industrial sectors is estimated to be about
20.2%, 12.7%, and 4.1%, respectively, in 2010 [3]. Population growth, aging water infrastructure
requiring significant upkeep and retrofitting investment, increased demand for crop production
(for both food and biofuels), shrinking freshwater resources, increased soil, groundwater, and
surface water salinity, contamination of groundwater resources, fossil fuel extraction, and
increased security concerns have also created multiple threats to the water sustainability in the
U.S. and globally [4-8]. Given the above realities, along with the impending impact of global
climate change, it is not surprising that the development of new water resources is high on the
priority list of various water agencies around the U.S. and the globe.
In addition to the various approaches to increasing the utilization of water resources (e.g.,
water use efficiency, water recycling and reuse), there has been increased interest in
augmenting potable water supplies through desalination of seawater and inland brackish water
3
This article is protected by copyright. All rights reserved.
[9-12]. Water salinity is typically categorized based on the concentration of total dissolved
solids (TDS; i.e., mineral salts) as shown in Table 1. The recommended salinity level for drinking
water is ≤ 500 mg/L TDS [13], while seawater salinity is typically in the range of ~33,000 mg/L –
42,000 mg/L TDS. Brackish water is classified as having salinity above 1,000 mg/L but below that
of seawater, with further classification from mild brackish to heavily brackish water [14]. In
assessing the potential availability of saline water resources for desalination, it should be
recognized that about 97% of the world water is saline. About 1.74% of the word’s water is
trapped in ice caps and glaciers and the remainder as groundwater (1.69%), surface water
(~0.014%), and moisture (0.002%) [15]. Clearly, freshwater supplies (for potable and nonpotable use) are limited and constitute but a small fraction of the world water resources. While
the majority of the world’s saline water volume is in seas and oceans, inland water (e.g.,
groundwater aquifers, lakes, wastewater) resources can also be saline, but are typically at
lower salinity levels (i.e., brackish) relative to seawater. The potential contribution of desalted
brackish groundwater to the overall potable water supply can be significant considering that
subsurface saline aquifers are abundant throughout much of the continental United States and
many parts of the world [10, 12, 16]. Thus, it is not surprising that in order to augment
dwindling freshwater supplies, various inland communities across the U.S. and globally have
resorted to desalting brackish groundwater reserves (1,000-10,000 mg/L TDS) [9].
The process of desalination involves the treatment of a saline water source to separate
dissolved salts from the feed stream in order to produce a low salinity product water and a
residual stream of high salinity water. Given major advances in both thermal and non-thermal
based membrane desalination technologies [11, 12, 16, 17], seawater desalination has
4
This article is protected by copyright. All rights reserved.
developed rapidly in a number of countries (e.g., Singapore, Australia, Israel and Spain, Saudi
Arabia, Gulf-States), with others also adding desalination to supplement freshwater supplies
[9]. In addition to its use for desalting brackish surface and groundwater, membrane based
desalination technology is utilized in indirect potable water reuse applications (e.g., aquifer
recharge, industrial use, and irrigation) to reduce salinity and provide a barrier against multiple
contaminants (e.g., organics, viruses, bacteria, in addition to toxic inorganics such as boron [18]
and selenium), as well as for reclamation of industrial, mining, and agricultural wastewaters
[10, 12, 19, 20].
Table 1. Characterization of Water Salinity
Typical TDS levels of different water sources
Type
TDS (mg/L)
Drinking water
≤ 500
Fresh water
< 1,000
Mildly brackish water
1,000 - 5,000
Moderately brackish water
5,000 - 15,000
Heavily brackish water
15,000 - 35,000
Seawater
≥ 35,000
Source: Watson et al. [14]
Water desalination is practiced via various approaches of which reverse osmosis (RO)
desalination has become the dominant technology [16, 21-24]. Accordingly this perspective
presents an overview of water desalination and comparison of energy requirements by the
main desalination technologies practiced today, followed by a brief account of the basics of RO
desalination process configuration and the challenges imposed by RO concentrate
management, membrane fouling and mineral scaling, and feed pretreatment. The various
5
This article is protected by copyright. All rights reserved.
factors that contribute to the overall cost of RO desalination are then introduced and the
energy cost of RO desalination is assessed considering the thermodynamic limit and the
implications for feasible future reductions in energy utilization and other aspects of RO process
operation that could provide further reduction in the cost of RO desalination.
2
WATER DESALINATION
2.1
Overview
The total number of desalination (seawater and brackish water) plants in the world (in
2015) is reported to be 18,426 with a total water desalination capacity of about 86.5 million
m3/day [9]. The majority of the world desalination installation capacity (Fig. 1a) is for desalting
of seawater (59%), followed by brackish water (22%), with the remainder (19%) being for river
water, wastewater and brackish water [9]. In terms of worldwide desalination capacity by
technology (Fig. 1b), RO has emerged as the leading desalination technology practiced today
(65%), followed by multi-stage flash evaporation (21%). The remaining 14% of the capacity
include multiple effect distillation (MED), vapor compression (VC), and electrodialysis (ED)/ED
reversal (EDR). Other desalination technologies have also been proposed in recent years
including, at a smaller scale, membrane distillation [25] and forward osmosis [26-28].
The total water desalination capacity in the U.S. is about 4.5 million m3/day [29] with most
being inland brackish water desalination with the majority via RO technology. More than 300
municipal brackish water (BW) desalination plants operating in the U.S. are primarily located in
Florida (45 %), California (14 %), and Texas (9%) [30]. The largest online brackish RO water
desalination plants in the U.S. is the Orange County Groundwater Replenishment System
6
This article is protected by copyright. All rights reserved.
(GWRS) having a capacity of 378,000 m3/day with planned expansion to 492,000 m3/day. The
largest operating SWRO plants in the U.S. are: (a) the Tampa Bay plant in Florida (capacity of
94,500 m3/day), and (b) the new plant in Carlsbad, California (capacity of 189,000 m3/day)
began its operation in late 2015.
Figure 1. (a) Desalination installation capacity by water source , (b) Desalination Capacity by Technology
(Total: 86.5 million m3/day) [9].
Comparison of the energy cost of different energy desalination methods based on the
published literature is not trivial given the variability of feed salinity and quality encountered in
the various operating plants. Therefore, it is not surprising that the reported specific energy
consumption (SEC, i.e., energy per volume of desalted water product) can vary by up to an
order of magnitude in some cases (Table 2). In general, the SEC of desalination via RO is lower
for seawater than by thermal methods [11, 31, 32]. Brackish water desalination via RO and EDR
can be competitive depending on the salinity level [10, 11]. Both RO and EDR require only
electrical energy, while processes based on evaporation/distillation also require thermal
7
This article is protected by copyright. All rights reserved.
energy. It is noted that heat- and osmotically-driven (with thermolytic draw solutions, e.g. FO)
desalination have been reported to be cost-effective only if low-cost heat source (or “waste
heat”) is readily available locally to drive the process or to provide for regeneration of a process
stream (e.g., in FO) [27, 33-35]
Table 2. Summary of Reported Specific Energy Consumption (SEC) for Seawater and Brackish
Water (a)
Specific Energy
Total Specific
Total Water
(b)
Desalination Technology
Consumption
Energy
Production
(KWh/m3)
Consumption(c)
Cost(d)
(KWh/m3)
($/m3)
Electrical
Thermal
Reverse Osmosis
1.58 – 7.5
NA
1.58 – 7.5
0.52 – 1.72
(Seawater)
Reverse Osmosis
0.3 - 3
NA
0.3 -3
0.2 – 1.33
(Brackish Water)
Electrodialysis
0.5 – 1.8
NA
0.5 – 1.8
0.6 – 1.05
(Brackish Water)
Multi-effect distillation
4 – 20.2
1.5 – 2.5
5.5 – 22.7
0.52 – 1.5
Multi-stage flash
(Brackish Water)
Thermal Vapor
Compression
Mechanical Vapor
Compression
7.5 – 30.3
2.5 - 5
10 – 35.3
0.56 – 1.75
-
16.3
16.3
0.27 – 1.6
7 - 12
-
7 - 12
-
(a) Source: [36]. Note: reported energy consumption (SEC) is for 1 m3 of desalted product water. The
wide range of costs is due to differences in source water salinity, plant size, pump and other system
components efficiencies, product water recovery (i.e., product volume/feed volume), and heat quality
(for thermal desalination processes); (b) Thermal desalination processes require thermal energy and
electrical energy. In the absence of available thermal energy source electrical energy is used (converted)
to provide the needed thermal energy; (c) Reported range of total SEC for the desalination process(d)
Total water cost per 1 m3 of desalted product water includes energy, capital and O&M costs.
2.2 RO Desalination
2.2.1 RO Emergence as the Standard Technology for Water Desalination
The popularity of RO desalination technology and its growing market share [9] has been, in
part, due to its simplicity and reliable operations and maintenance support by diverse and well8
This article is protected by copyright. All rights reserved.
established supply chain of off-the-shelf components and consumables (e.g., membrane
elements, prefilters, compatible water treatment additives and membrane cleaning chemicals,
etc.) [10, 16, 17]. Moreover, the process is easily scalable whereby both small- and large scale
RO plants can use the similar types of membrane elements and pressure vessels (Fig. 2). RO
desalination relies on a semi-permeable membrane (housed in a membrane element inside a
pressure vessel) that allows water permeation but rejects dissolved solids. The feed side of the
membrane is pressurized to a level above the osmotic pressure so as to provide the desired
permeate water flux. The RO process is driven by a pump for fluid conveyance and for
generating the required feed pressure. The overall desalination process consists of a process
train that includes feed intake system, feed water pretreatment, the desalination separation
system, product water post-treatment and concentrate management (Fig. 3). An intake system
is required to convey the feed water from the source to the plant, and concentrate
management is required to handle the high salinity concentrate (or brine). Pretreatment of the
raw RO feed water is a crucial component of RO desalination technology (Fig. 3). The levels of
suspended solids, organics and microorganisms in the RO feed must be reduced to a sufficient
level in order to avoid fouling of the RO membranes [12, 16]. The extent of fouling has been
described to depend on a variety of factors including hydrodynamic forces, membrane
properties (e.g. surface charge, permeability, surface roughness), and solute on ionic matrix
[24, 37, 38]. RO operation can also be negatively impacted by mineral scaling which can occur
when the concentration of sparingly soluble mineral salts (e.g., gypsum (CaSO 4∙2H2O), BaSO4,
SrSO4, CaCO3, SiO2, etc.) in the RO feed rises, as product water is extracted, above their
solubility limit [39-45]. The consequence is mineral salt precipitation and scale formation on the
9
This article is protected by copyright. All rights reserved.
RO membrane. Membrane fouling and mineral scaling lead to membrane surface blockage and,
as a result, degradation of membrane performance (i.e., permeate flux decline and
deteriorating salt rejection) which scan shorten membrane longevity and increase operational
costs [43, 46].
Figure 2. (Left) Photo of the desalination plant at Orange County (Capacity of 378,000 m3/day)[47]
and (Right) a small mobile RO desalination plant (capacity of 45.4 m3/day) [48, 49] . The same size
membrane elements (8”) are used in both plant demonstrating the scalability of RO desalination.
10
This article is protected by copyright. All rights reserved.
Feed Intake
- Open Water
- Subsurface
- Power Station Collocation
Pretreatment
Removal of:
- Particulates/Colloids
- Organic Macromolecules
- Viruses/Bacteria
Desalination
Removal of:
- Salts/Inorganics
- Heavy Metals
- Small Organics
- Viruses/Bacteria
Product
Post-Treatment
- Micropollutants Removal
- Remineralization
- Disinfection
- Stabilization/Neutralization
Concentrate
- Volume Minimization
- Treatment
- Salt Harvesting
- Energy Conversion
- Disposal/Outfall
Figure 3. Process elements in water desalination.
2.2.2 Mitigation of Fouling and Mineral scaling
Current practice of membrane fouling prevention emphasizes the removal of potential
foulants, prior to any RO desalting unit operations, using conventional flocculation/coagulation,
disinfection (e.g., via chlorine dosing or UV pretreatment), media filters or membrane filtration
(microfiltration or ultrafiltration) processes for RO feed water pretreatment [16, 32, 43, 48, 50,
51]. In some cases, sodium metabisulfite, sodium bisulfite, or activated carbon may also be
used for dechlorination of the RO feed to avoid RO membranes degradation due to chlorine
[52, 53]; it is noted, however, that the use of chlorine in RO feed is now typically avoided in
modern plants [11]. These feed pretreatment processes, while effective for minimizing
membrane fouling due to particulate deposition, organic adsorption, and biological growth, do
not remove mineral scale precursors. Mineral scaling is typically mitigated via: (a) the addition
11
This article is protected by copyright. All rights reserved.
of polymeric antiscalants additives to the feed to suppress the nucleation and/or growth of
mineral crystals and promote dispersion of crystals [37, 41, 43, 46, 54], (b) feed pH adjustment
if calcium carbonate is an issue of concern [16, 54], (c) regulation of water recovery in order to
keep the level of retentate concentration within acceptable limits (as governed by mineral
scaling kinetics) [40, 43] , (d) periodic membrane cleaning (e.g., via fresh water flush, osmotic
backwash, feed flow reversal, or chemical cleaning [12, 16, 55, 56], and (e) removal of scale
precursors via crystallization, nanofiltration or ion exchange [12, 20, 57-59]. In certain cases
where boron and carbon dioxide removal are necessary and where silica scaling is to be
avoided (at high recovery), RO operation at high pH may be necessary [42, 44, 60-63]. Chemical
post-treatment of the product water may also be necessary and, depending on the end use, can
include disinfection, remineralization and pH adjustment [64].
2.2.3 Concentrate Management
RO desalination is typically carried out at a product water recovery of 40%-50% for
seawater desalting and in the range of 50% - 90% for brackish water desalting [10, 12, 16, 32] .
As a result a concentrate retentate (or brine) stream is generated which must be managed in an
environmentally compatible manner [20, 65, 66, 67 , 68, 69]. In seawater RO desalination, the
brine stream is discharged to the sea through an elaborate outfall system designed to disperse
the concentrate in a manner that reduce local elevated salinity impacts and precipitation along
the discharge pipes [16, 32, 70]. Environmental considerations with respect to the
implementation of ocean discharge of RO concentrate include, for example, the salinity
tolerance of marine organisms, discharge toxicity (due to possible chemical water
treatment/cleaning additives), long-term local salinity buildup and the need for meeting
12
This article is protected by copyright. All rights reserved.
effluent water quality standards [10, 11, 30]. At inland locations concentrate management
options are more limited [10, 12, 20, 71]. In some cases sewer discharge may be feasible (for
sufficiently low salinity concentrate), but may be restricted by regulatory restrictions. Disposal
of RO concentrate in evaporation ponds may be possible in some locations, but may be limited
due to large area requirement and constrained holdup and possible concerns with concentrate
toxicity. Deep well injection is another option which can be costly (in some cases, due to the
high injection pressure requirements) and in some areas deep geological injection may be
restricted. In some cases, ocean disposal via a long waste discharge pipeline may be feasible
but can be burdened by mineral scaling [70]. In principle, thermally-driven evaporative and
crystallization systems, the so-called zero-liquid-discharge (ZLD) technologies, are capable of
eliminating liquid RO concentrate discharge, but are often prohibitively energy intensive for
processing large volumes [20, 57-59, 63, 72]. Consequently, high recovery inland desalination (if
not limited by mineral scaling) is desirable in order to reduce the volume of generated brine
[57-59].
3
CONFIGURATIONS OF RO DESALINATION SYSTEMS
Reverse osmosis (RO) membrane desalination relies on a semi-permeable membrane that
allows water permeation but rejects dissolved solids (Fig. 4). The product water flux (Jw) is
dictated by the difference between the applied pressure ( p )and the osmotic pressure (  )
differences across the membrane as given by the classical expression, Jw  Lp (p   ) , in
which Lp is the membrane permeability and  is the so-called reflection coefficient [73], which
13
This article is protected by copyright. All rights reserved.
is in the range of [0,1] and nearly unity for high salt rejecting membranes. The solute flux
through the membrane, Jc, is typically described as Jc  JwC p  BC  (1   )JwC , in which Cp is
the permeate solute concentration, B is the solute (membrane) transport coefficient and C is
the average solute concentration across the membrane. As water permeates across the
membrane, the rejected solute accumulates near the membrane surface resulting in a
concentration polarization layer; thus, the osmotic pressure at the membrane surface increases
and thus the requirement for higher applied pressure for a given target water permeate flux.
Most RO systems employ membrane channels operated in a cross flow operation which
serves to reduce salt concentration buildup (i.e., concentration polarization) at and in the
vicinity of the membrane surface (Fig. 4a) [74-76]. The flow and concentration fields within
membrane channels are complex, but the solute concentration at the membrane surface, Cm,
can
be
estimated,
to
within
a
reasonable
level
of
approximation,
as
Cm / Cb  CP  (1  Ro )  Ro exp(Jw / kc ) , where Cb is the solute concentration in the bulk of the
feed channel, CP is the so-called concentration polarization modulus, RO is the observed solute
rejection (i.e., Ro-1-Cp/Cb) and kc is the feed-side solute mass transfer coefficient.
Salt
concentration in the retentate stream (“brine” or “concentrate”) increases as it flows
downstream along the membrane channel. For example, for seawater RO operation at 40%
recovery, 60% of the processed feed volume will be a brine stream with a salt concentration a
factor of 1.67 above that of the raw feed. At a higher recovery of 80%, which would more
typical for low salinity brackish water, the brine salt concentration would be a factor of five
above that of the raw feed.
14
This article is protected by copyright. All rights reserved.
Feed
Qf ,Cf
Pp
U(x)
Concentrate
Qc ,Cc
Cb
H/2
Cm(x)
Concentration
boundary layer
Permeate
Qp ,Cp
Pp
Jw(x)
(a)
RO membrane
Permeate Collection Holes
Anti-telescoping
device
(b)
Permeate collector
Membrane sheet
Feed channel spacer
Membrane sheet
Element covering
Permeate flow
Figure 4. (a) Cross-flow reverse osmosis in a membrane channel. (Q and C and P denote flow rate, salt
concentration and hydraulic pressure, respectively, and where subscripts f, b, c, p and m denote the feed,
bulk feed-side, concentrate, permeate and membrane surface, respectively; Jw is the water flux through
the membrane; (b) Schematic of a spiral-wound RO membrane element.
Membrane elements are general constructed as multiple membrane sheets arranged (using
spacers) to form flow channels through which pressurized saline feed water flows (Fig. 4b).
Standard commercial spiral-wound RO membrane elements which are housed in pressure
vessels allow operation up to a maximum pressure of ~1200 psi (8.3 MPa) and ~600 psi (4.1
MPa) for seawater and brackish water membranes, respectively. Standard RO (SRO) process
configurations (Fig. 5) typically comprise of one or more pumps and one or more membrane
arrays, coupled with a throttling valve for concentrate depressurization [10, 77, 78]. In order to
avoid wasting the pressure energy of the discharged brine stream, energy recovery devices
(ERDs) can be integrated with the RO system [10, 79]. Typically, single stage RO processes are
integrated with ERDs (Fig. 6) such as a pressure intensifier/booster or parallel feed pumping
[10, 78, 79], which presently are mostly limited to seawater desalination applications.
15
This article is protected by copyright. All rights reserved.
Reduction in energy consumption through the use of multi-stage RO with inter-stage (booster)
pumps is also possible and is most suitable for brackish water desalination in which a high
product level of product recovery is needed [10, 78].
Figure 5. Typical arrangements of pressure vessels for RO/NF membrane elements [16]. (F, C and P
designate the feed, concentrate and permeate streams, respectively).
Conventional RO systems typically operate over a narrow range of water recovery. The
above limitation is because the system feed flow rate is directly coupled with the main feed
pump inlet, the membrane array feed and, for the case in which ERD is utilized, the ERD device
feed- and concentrate-side flow rates. Also, flow-range restrictions, imposed by process
components for effective and energy-optimal operation often restrict the range of water
recovery in conventional systems. The alternative RO configuration of concentrate
recycling/recirculation, at partial or total recycle (i.e., batch operation with permeate
withdrawal and periodic system feed flushing) can improve operational flexibility and increase
RO recovery for a given footprint [80, 81].
16
This article is protected by copyright. All rights reserved.
Figure 6. Conventional RO membrane system configuration integrating energy recovery (ER) devices: (a)
feed pressure booster/intensifier, (b) parallel feed pumping, and (c) inter-stage pressure booster. (MA –
membrane array; P1 and P2: feed and booster pumps, respectively; o, f, c, p and w designate the raw
water source, and RO unit feed, concentrate, permeate and discharge streams, respectively).
17
This article is protected by copyright. All rights reserved.
4
THE COST OF RO DESALINATION
Over the years of RO development, there have been significant technology improvements
with respect to membranes and membrane modules, plant standardization, operational
efficiency, and energy recovery. This has led to a steady decline in the cost of SWRO
desalination (Fig. 7). It is projected that the cost of seawater RO desalination is approaching or
nearly comparable in certain cases with the cost of traditional water sources [82].
Water Price ($/m3)
6
Desalination
Reuse
Freshwater Treatment
Marginal Water Withdrawal
5
4
3
2
1
0
1985
1995
2005
2015
2025
2035
Figure 7. The cost of water seawater RO desalination relative to traditional water sources [82].
Water production cost in a typical RO desalination plant generally consists of the cost of
energy consumption, equipment, membrane replacements, residual concentrate (i.e., brine)
management, labor, maintenance, and financial charges. The cost breakdown for desalination
plants can vary greatly depending on plant size and location, quality and salinity of the water
source, and local electrical energy cost. For a typical large seawater RO desalination plant,
energy and capital costs (Table 3) constitute the major portion of the overall water production
18
This article is protected by copyright. All rights reserved.
cost (e.g., in terms of $/m3 product) with the remainder attributed to operation and
maintenance costs [11, 16, 32].
Table 3. Elements of Capital Expenses (CAPEX) and Operational and Maintenance
(O&M) Costs for Seawater RO Desalination. Source: Adapted from [36].
Cost of RO Desalination(a)
Capital
Energy
Chemicals
Labor
Replacement Parts
Membranes replacement
Overhead
Insurance
Percent of Total Water
Production Cost
34%
38%
5%
3%
9%
5%
5%
1%
Capital Expenses (CAPEX)(a)
Desalination system
Percent of Total CAPEX
31%
Power system
Pre-treatment
Intake & Outfall (Discharge)
Design & permitting
Other
O&M(b)
Fixed Cost
Energy
Operation and Maintenance
26%
12%
11%
7%
13%
Percent of Total O&M
28% - 50%
32 % - 44%
18 % - 28%
(a) Cost for a typical large plant. Actual costs will vary depending on plant design, product water
recovery, level of automation, outfall requirements; (b) O&M costs can vary considerable
depending on plant design, level of automation, intake and outfall and feed pretreatment
requirements.
Specific energy consumption (SEC) is strongly affected by water product recovery, energy
recovery (from the high pressure RO concentrate), pumping energy (including feed intake),
operating conditions (e.g., RO feed channel velocity and feed-side pressure), and plant
configuration in terms of membrane modules arrangements [10, 11, 16, 31, 78, 83-85]. The
19
This article is protected by copyright. All rights reserved.
various contributions to the total energy consumption for a seawater RO desalination plant
typically includes the desalination stage and discharge (~67%), RO feed pre-treatment (~13%),
intake (~7%) and post-treatment (~13%) [36]. Because seawater RO is energy intensive, the
costs of seawater RO is often considered high. However, this can be misleading as often
desalination costs are not compared to the costs of locally available (and feasible) alternatives
(Fig. 7,
[31]). It has also been argued that in certain cases, the energy costs of
pumping/conveying water from large distances can be higher than the energy required for
water production by a large desalination plant located in proximity to water consumers and
when such a plant utilizes off-peak electrical energy [11, 27, 31]. Another common concern
regarding seawater desalination is the discharge of high salinity concentrate (up to twice the
salinity of seawater) to the sea. The concentrate stream may contain chemicals (e.g.,
surfactants, antiscalants and acid/base for pH control) used for membrane cleaning and scale
prevention in addition to rejected backwash from feed water pretreatment. Today there are
continuing efforts to introduce “green” water treatment chemicals so as to avoid potential
environmental impacts. At the same time, in order to reduce potential environmental impacts,
modern technologies of concentrate disposal to the sea provide for sufficient concentrate
dispersion through multiple nozzles, upward (subsurface) concentrate discharge, and where
feasible mixing the concentrate with power station cooling water [11].
The energy cost of desalting brackish water is lower than for seawater (given the lower
salinity of brackish water, Table 1). The energy cost is in the range of 20% - 30% of the total
water production cost, but can be higher for heavily brackish water [10, 12, 16]. However,
unlike seawater RO plants, the management of the RO concentrate from inland brackish
20
This article is protected by copyright. All rights reserved.
desalination can represent a major challenge given the often limited options for concentrate
disposal [10, 12, 20, 30]. Inland concentrate management costs can be as high as 0.40–1.78
$/m3 and in certain cases represent more than 50% of the cost of water production [86].
Hence, there is a significant drive to increase RO recovery so as to minimize the volume of
generated RO residual stream [10, 12, 30, 43, 57-59, 86].
5
THE ENERGY COST OF RO DESALINATION
5.1 Minimum energy consumption for a reversible desalination process
The minimum isothermal reversible work, W, for separating salt from water, irrespective of
the separation mechanism, is given as follows [31, 87]:
n2
n2
n2
n1
n1
n1
Wrev   Fdn   RT lnaw dn    sVw dn
(1)
in which F is the change in free energy associating with the initial (1) and final (2) states of a
solution as its salt concentration is altered, R is the ideal gas constant, T is the absolute
temperature, aw is the water activity, n is the number of moles of water,  s is the solution
osmotic pressure, and Vw is the water molar volume. As the feed water salinity increases so
does the osmotic pressure (reasonably approximated for a wide salinity range by  s  CRT ,
where  is the van’t Hoff factor and C is the salt molar concentration [83] ). Following Eq. 1, it
can be shown that the minimum energy of desalination, at the limit of zero product water
recovery, for desalting seawater of salinity of 35,000 mg/L TDS (osmotic pressure is ~26 bar) is
about 0.72 kWh/m3 at 25o C. Increasing the water recovery would elevate osmotic pressure,
21
This article is protected by copyright. All rights reserved.
thereby increasing the required reversible desalination work. The dependence of the minimum
required isothermal reversible work to achieve a given water recovery can be deduced from Eq.
1 [88, 89] and expressed in terms of the SEC normalized with respect to the feed water osmotic
pressure (o):
NSEC rev 
SEC rev
o

ln1  YS 
1  Wrev 

  
 o  Vf  Vc 
YS
(2)
where Vf and Vc are the feed and concentrate volumes, respectively, and YS is the fraction of
product water recovered from the feed. Based on Eq. 2, for example, desalination of seawater
at a water recovery of say 50% would require ~1.0 kWh/m3. However, desalting at the above
level of energy consumption for a reversible process (at infinitesimal permeate flux) is clearly
impractical. It is noted that, at present, the most energy efficient SWRO desalination (under
non-reversible thermodynamic operation) is reported to be nearing 1.58 kWh/m3 (Table 2).
Given the current state of RO efficiency, further RO SEC reduction is likely to require a higher
capital cost for the same level of target permeate productivity.
5.2 RO Energy Consumption and the Thermodynamic Limit
The minimum transmembrane pressure that must be applied to ensure permeate
production along the entire membrane area (in the axial fluid flow direction), for crossflow
membrane channels, must be such that P   exit , where  exit is the osmotic pressure
difference (across the membrane) at the RO channel exit [83]. Using the approximate linear
relation for the osmotic pressure dependence on salt concentration, it has been shown that the
minimum required applied pressure difference is  P min   oRt / (1  YS ) (in which Rt is the
22
This article is protected by copyright. All rights reserved.
membrane salt rejection), which clearly increases with both rising feed osmotic pressure and
water recovery [83]. With the old generation of relatively low permeability ( L p ) cellulose
acetate membranes, it was necessary to operate at a feed pressure that was much higher than
the brine osmotic pressure in order to produce a reasonable level of permeate flux for a viable
desalting operation. However, with the development of highly permeable polyamide based
thin-film composite (TFC) membranes and their improvements over the last two decades,
higher permeate fluxes can be attained at significantly lower feed pressure enabling cross-flow
RO operation near the limit imposed by the thermodynamic restriction [78, 90] .
The energy footprint of RO desalination can be expressed in terms of the specific energy
consumption (SEC) normalized with respect to the feed osmotic pressure (  o ):
NSEC 
SEC
o



1  N Wi 
 
 o qp  i 1 p ,i 


(3)
where W i and  p ,i are the rate of work and efficiency of pump i in a system with N pumps, qp is
the total permeate flow rate and  o is the osmotic pressure of the RO feed. In principle, NSEC
of conventional single-stage RO with an ERD, at the limit of cross-flow thermodynamic
restriction (tr), can be described by the following relationship[85]:
NSECtr 
1  
ERD
 1  YS   Rt
p  YS 1  YS 
(4)
where  p and  ERD are the pump and ERD efficiencies, respectively, and YS  q p / qo (where qo
and qp are the raw feed and product water flow rates, respectively) is the product water
recovery and Rt is the salt rejection. In the absence of an ERD (i.e.,ERD  0 )
23
This article is protected by copyright. All rights reserved.
NSECtr  Rt / p  YS 1  YS  . thus, the energy savings of utilizing and ERD is given by
ERD  Rt / p  YS  , which is more pronounced at lower water recovery. A plot of Eq. 4 (Fig. 8)
illustrates that the energy consumption, for RO process operation (for the case of a membrane
of 100% salt rejection) up to the thermodynamic limit, is highly dependent on the product
water recovery, as well as pump and ERD efficiencies, but it is independent of membrane
permeability (Fig. 3).
The global minimum (w.r.t energy consumption) for RO operation, without an ERD and
assuming constant pump efficiency, is at recovery of 50%. For seawater desalination (salinity of
~35,000 mg/L TDS) and ideal pump (i.e., p  1 ) this implies an energy consumption of 2.88
kWh/m3 (NSEC =4). This energy consumption level may be reduced to 1.44 kWh/m3 h (NSEC =2)
with an ideal ERD (i.e., ERD  1 ) and to 1.51 kWh/m3 (NSEC =2.1) for ERD of 95% efficiency
which is consistent with the lowest reported SEC for seawater desalination (Table 2). RO plants
that are designed to operate close to the thermodynamic limit, which can be feasible with
currently available commercial RO membranes, should be able to operate optimally so as to
reduce energy consumption. When the RO plant is incapable of operating up to the
thermodynamic limit (due to design and equipment restrictions), one can drive the RO process
toward the optimal operating condition via suitable model-based control [49, 91]. In such an
approach one must also consider constraints imposed by the target water production capacity
and limitations placed on the operability of plants components (e.g., maximum allowable
operating pressure for pumps and pressure vessels).
24
This article is protected by copyright. All rights reserved.
Figure 8. Variation of the normalized specific energy consumption with product
water recovery for a single stage RO operation at the thermodynamic limit
(excluding frictional pressure losses which ae typically <5-10%; note:  p and
 ERD are the RO pump and ERD efficiencies, respectively; frictional pressure
losses).
It is stressed that for RO process operation up to the thermodynamic limit, the SEC is not
impacted by membrane permeability (Eq. 4). Therefore, utilization of more permeable
membranes will not provide additional reduction in energy consumption, unless the plant is
operating away from the thermodynamic limit [78, 90]. For a plant operating at the
thermodynamic limit, the use of higher permeability membranes will, however, allow reduction
in plant footprint owing to the higher achievable water flux which will enable the
thermodynamic restriction to be reached with a smaller membrane surface area. Membrane
25
This article is protected by copyright. All rights reserved.
fouling, however, is more severe at high flux operation and thus high flux membranes would
have to be more fouling resistant and/or feed pretreatment will need to be more effective.
5.3 Multi-Stage RO Desalination
The energy foot-print of RO desalting is reduced with increasing number of pumping stage.
Unlike a single stage RO configuration that requires pressurizing the entire feed, at a pressure
equal or above the concentrate pressure, in a multi-stage system membrane units are staged
serially with inter-stage booster pumps (Fig. 7). This enables incremental increase in the feed
pressure to each stage and correspondingly an incremental reduction in the required feed
pumping capacity for each successive membrane stage. The NSEC at the limit of
thermodynamic restriction (for each membrane stage, assuming complete salt rejection and
ideal pump efficiency) is given by [83, 92]:
NSECtr   N  (1  YS )1/N  N  1 ERD  / YS
(5)
where N is the number of pumping stages and YS is the overall system water recovery, which is
related to the individual stage water recovery ( Yn ) by the relationship 1  YS   1  Yn  . Note
N
that Eq. 5 reflects the energy-optimal condition at which that the individual stage water
recovery is identical for all stages [83, 92]. In the limit of infinite number of pumping stages and
ideal energy recovery (  ERD =1), Eq. 5 for the multi-stage RO process reduces to Eq. 2 for the
reversible desalination process. Analysis of Eq. 5 reveals that at recovery above about 68%,
infinite-stage RO without an ERD would have lower energy consumption relative to single-stage
RO with an ideal ERD [92]. With a finite number of stages with inter-stage pumps (without
ERD), the advantage over a single-stage RO with an ERD will be shifted to even higher recovery.
26
This article is protected by copyright. All rights reserved.
5.4 Concentrate recirculation
Membrane process systems are composed of highly integrated and interdependent process
components. Optimal operation of each system component is often limited to a narrow flow
range, which in turn restricts the integrated process system operational range. Pumps (and
associated motors) and energy recovery devices, for example, typically have narrow flow ranges
at which energy efficiencies are optimal. Furthermore, membrane modules are inherently
limited with respect to the flow ranges for optimal operation, while maintaining conditions that
protect physical integrity, maximize membrane area utilization, minimize concentration
polarization, and avoid excessive membrane fouling and/or mineral scaling. Such module-level
operational limits can significantly restrict water recovery and permeate productivity ranges of
the integrated system, which is highly dependent on the specific membrane array design.
Existing methods for improving operational flexibility of RO membrane based processes rely
on partial recirculation of RO concentrate (to RO feed); this provides an additional degree of
freedom for steady-state process regulation. The simplest approach is to recirculate a portion
of depressurized RO concentrate to the RO feed pump (Fig. 9a). Although simple, RO with low
pressure concentrate recirculation (LPCR) is highly energy intensive with energy consumption
(for single stage RO) given by:
NSECLPCR1,tr 
1
p  YMA 1  YS 
(6)
Compared to the NSEC for conventional RO without concentrate recirculation (Eq. 6), LPCR has
a higher energy footprint due to the required higher flow capacity of the main feed pump (by a
factor of YMA /YS) for achieving the same productivity. As a consequence, conventional LPCR is
often deemed undesirable and only utilized in limited applications when operational flexibility
27
This article is protected by copyright. All rights reserved.
is significantly more important than the energy footprint (e.g., as proposed certain small-scale
industrial applications [80]).
RO with high pressure concentrate recirculation (HPCR) (Fig. 9b) is a more energy efficient
process when compared to LPCR. In HPCR, only a portion of the concentrate stream (w) is
depressurized (and discharged) using a throttling valve, while the remaining concentrate (rc) is
recycled to the main pump (P1) outlet without depressurization [80]. A recirculation pump (P2)
is needed in order to match the main pump outlet pressure and make up for the relatively small
axial pressure loss in the membrane array (Pc1-Pf ≳ 0). However, the concentrate pressure
energy remains unrecovered in the concentrate discharge stream (w), making HPCR suboptimal
with respect to energy utilization. For example, assuming that axial pressure drop in the
membrane array (MA) is small such that P2 energy requirement is minimal and the limit of
complete salt rejection, the minimum HPCR1 energy footprint is no better than that of standard
single stage RO (Eq. 4). Another critical drawback of conventional HPCR (Fig. 9b), is that, because the feed pum
increase system energy footprint. Thus, increased operational flexibility in conventional RO with
HPCR may be at the cost of reduced energy efficiency, depending on the dependence of the
pump efficiency on feed water flow rate.
HPCR operation of RO in unsteady-state cyclic operational modes have recently been
introduced commercially (i.e., closed circuit desalination) [81, 93]. In this approach, the RO
system is operated in alternating periods of complete and no concentrate recirculation in a
cyclical manner. In each cycle, salt buildup in the RO system during the period of complete
concentrate recirculation is purged during subsequent period of concentrate discharge. In
addition to enhanced operational flexibility, recent theoretical analysis under idealized plug-
28
This article is protected by copyright. All rights reserved.
flow conditions suggests that for a single stage RO, the approach can potentially be utilized as
an alternative to using an energy recovery device [94, 95]. Nevertheless, existing analysis have
yet to consider the impact of non-ideal flow conditions in real RO systems, especially during the
concentrate withdrawal period. Hence, the practicality of the approach in reducing energy
consumption still needs to be explored.
(a)
(A)
FD
rc
w
c
TV
c1
o
f
rc
(b)
(B)
o
P1
P1
o1
MA
f1
P2
f
rc1
w1
p
TV
w
c
MA
p
Figure 9. Conventional single-stage reverse osmosis (a) with low (LPCR1) and (b) high pressure (HPCR1) concentrate
recirculation. MA-membrane array, FD- flow diverter, P1-main feed pump, P2: concentrate recirculation pump, TVthrottling valve (for back pressure regulation); o, f , p, c, rc, o and w – source water, and RO unit feed, permeate,
concentrate, recycled concentrate and discharge streams, respectively.
6
THE PATH TO IMPROVING RO DESALINATION
In the forthcoming years improvements in the design and operation of RO plants will lead to
further reduction in the cost of RO desalination. For example, more effective process
configurations could allow operational flexibility (i.e., wide recovery range) with efficient
energy recovery [83-85]. Improvements in membrane element design [96, 97] could also be
29
This article is protected by copyright. All rights reserved.
beneficial to reducing pressure losses while possibly enabling reduction in concentration
polarization, thereby allowing higher recovery operation per element and thus decreasing the
plant footprint in and possibly plant components. Operational considerations include optimal
plant control that will ensure energy optimal operation while meeting the target permeate
productivity [49, 50].
In recent years there have also been mounting efforts to develop effective methods of
plant fault detection and isolation [49, 91, 98-100] so as to avoid plant failures and reduce
maintenance costs. Effective self-adaptive RO feed pretreatment [48, 51] that can handle
temporal variability of raw feed water quality and production demand is also likely to lead to
significant reduction in the use of feed treatment chemicals while reducing the frequency of
membrane cleaning and replacement. All of the above would benefit from real time monitoring
of membrane fouling and mineral scaling [44, 51, 77, 101-103]. It is also stressed that real-time
monitoring of feed and product water quality (e.g., w.r.t emerging contaminants) and of
membrane integrity [104, 105] would contribute to establishing effective operational strategies
to mitigate fouling/scaling and toward gaining acceptability of RO for direct potable reuse
applications.
While efforts to increase membrane permeability are continuing with claims of this
approach as being a viable path to achieving significant reduction reducing in RO energy
consumption, studies about a decade ago have shown that this is unlikely to be realized
(Section 5.2). The classical polyamide membranes appear to be well entrenched in the RO
desalination industry [106]. However, the development of promising membranes [107, 108],
such as those based on graphene, carbon nanotubes, aquaporins, self-organizing block
30
This article is protected by copyright. All rights reserved.
copolymers, crystalline polymers, as well as molecularly printed and nanostructured [109, 110]
and surface modified membranes [103], could pave the way for commercial RO elements
capable of high flux (and thus high recovery per element) operation with high membrane
selectivity. These future membranes, if successful, could allow significant reduction in plant
footprint. Such membrane will need to match or be better than current membranes with
respect to their fouling propensity. Moreover, the development of such membranes will also
have to ensure their mechanical and chemical (e.g., due to exposure to oxidants, extreme pH
conditions) stability.
7
SUMMARY
Water scarcity and groundwater contamination around the globe have sparked major
efforts to preserve and diversify regional water portfolios through water desalination and water
reuse.
While various desalination technologies exist, reverse osmosis (RO) membrane
desalination has emerged as the dominant technology for desalting of seawater and brackish
water in applications that range from small- to municipal-scale applications. The market growth
of RO desalination is attributed, in part, to its simplicity and technological maturity. While it
has been often reported that RO desalination is energy intensive such statements are often
misleading as examples already exist where RO desalination of both brackish and seawater are
competitive with traditional water sources. Admittedly, the energy cost of water desalination is
typically the major cost contributor to water production via seawater desalination, but this is
not always the case (e.g. capital cost can be higher in some cases). In the case of brackish water
desalination, energy (while significant contributor) is often not the major cost component.
31
This article is protected by copyright. All rights reserved.
While the overall cost for seawater RO desalination has been reported to be as low as about
$0.53/m3, there is a wider range of reported costs for both seawater and brackish water
desalination. Cost variability is due to numerous factors including, but not limited to,
differences in plant design and location, equipment, operation (e.g., feed salinity and quality,
recovery level, automation, and real time process optimization), and environmental compliance
costs (e.g., regulatory requirements for inflow and outflow safeguards).
It is questionable if the energy cost of desalination can be significantly reduced in the
coming years given the current availability of high permeability RO membranes which should
enable operation that approaches the cross-flow thermodynamic restriction. Higher
permeability membranes of low fouling propensity and high resistance to disinfectants should,
however, allow reduction in plant footprint and increase in membrane lifetime. Moreover, cost
reduction can also be achieved through advances in RO operational flexibility, energy optimal
plant control and self-adaptive operation, real-time monitoring of membrane fouling and
integrity, smaller plant footprint, membrane element design improvements, and higher
efficiency pumps and energy recovery devices. It is stressed that all of the above are
inextricably linked whereby an improvement in one area is also likely to impact the others.
Although there are various proposed alternatives to RO desalination, to date, none have been
shown in field studies to be superior to RO desalination for large-scale applications of potable
water production. However, it is conceivable that future developments will drive the
competition which will ultimately lead to expansion and broader acceptability of desalination
technologies as an important step toward water sustainability.
32
This article is protected by copyright. All rights reserved.
8
REFERENCES
[1] M.A. Maupin, J.F. Kenny, S.S. Hutson, J.K. Lovelace, N.L. Barber, K.S. Linsey, Estimated use of water in
the United States in 2010, U.S. Geological Survey Circular, 1405 (2014).
[2] Global Water Intelligence, Municipal Water Reuse Markets 2010, in, Oxford, England, 2010.
[3] T.S. Kelly, E.W. Michael, Evaluating the energy consumed for water use in the United States,
Environmental Research Letters, 7 (2012) 034034.
[4] American Society of Civil Engineers, 2009 Report Card for America's Infrastructure,
www.asce.org/reportcard, in, ASCE, Washington, D.C., 2009.
[5] B.I. Cook, T.R. Ault, J.E. Smerdon, Unprecedented 21st century drought risk in the American
Southwest and Central Plains, Science Advances, 1 (2015).
[6] Committee on Future Options for Management in the Nation’s Subsurface Remediation Effort,
National Research Council, Alternatives for Managing the Nation's Complex Contaminated
Groundwater Sites, National Academies Press, Washington, D.C., 2013.
[7] E.M. Biggs, E. Bruce, B. Boruff, J.M.A. Duncan, J. Horsley, N. Pauli, K. McNeill, A. Neef, F. Van Ogtrop,
J. Curnow, B. Haworth, S. Duce, Y. Imanari, Sustainable development and the water–energy–food
nexus: A perspective on livelihoods, Environmental Science & Policy, 54 (2015) 389-397.
[8] D.C. DiGiulio, R.B. Jackson, Impact to Underground Sources of Drinking Water and Domestic Wells
from Production Well Stimulation and Completion Practices in the Pavillion, Wyoming, Field,
Environmental Science & Technology, 50 (2016) 4524-4536.
[9] F. Virgili, T. Pankratz, J. Gasson, IDA Desalination Yearbook 2015-2016, Global Water Intelligence,
2016.
[10] R. Semiat, D. Hasson, Seawater and Brackish-Water Desalination with Membrane Operations, in:
Membrane Technology, Wiley-VCH Verlag GmbH & Co. KGaA, 2010, pp. 147-168.
[11] S. Miller, H. Shemer, R. Semiat, Energy and environmental issues in desalination, Desalination, 366
(2015) 2-8.
[12] Y. Cohen, B. McCool, A. Rahardianto, M.-m. Kim, J. Faria, Membrane Desalination of Agricultural
Drainage Water, in: A.C. Chang, D. Brawer Silva (Eds.) Salinity and Drainage in San Joaquin Valley,
California, Springer Netherlands, 2014, pp. 303-341.
[13] U.S. National Archives and Records Administration, Secondary maximum contaminant levels, Code
of Federal Regulations, Title 40, Section § 143.3 (2016).
[14] I.C. Watson, O.J. Morin Jr., L. Henthorne, Desalting Handbook for Planners, Third Edition, in:
Desalination and Water Purification Research and Development Program Report No. 72, U.S.
Bureau of Reclamation, Denver, CO, 2003.
[15] I. Shiklomanov, World fresh water resources, in: P.H. Gleick (Ed.) Water in Crisis: A Guide to the
World's Fresh Water Resources, Oxford University Press, New York, 1993.
[16] S. Gray, R. Semiat, M. Duke, A. Rahardianto, Y. Cohen, Seawater Use and Desalination Technology,
Treatise on Water Science, Vol 4: Water-Quality Engineering, (2011) 73-109.
[17] R. Semiat, Desalination - Present and Future, Water International, 25 (2000) 54-65.
33
This article is protected by copyright. All rights reserved.
[18] E. Güler, C. Kaya, N. Kabay, M. Arda, Boron removal from seawater: State-of-the-art review,
Desalination, 356 (2015) 85-93.
[19] A.K. Venkatesan, S. Ahmad, W. Johnson, J.R. Batista, Salinity reduction and energy conservation in
direct and indirect potable water reuse, Desalination, 272 (2011) 120-127.
[20] C.J. Gabelich, P. Xu, Y. Cohen, Chapter 10 Concentrate Treatment for Inland Desalting, in: C.E.
Isabel, I.S. Andrea (Eds.) Sustainability Science and Engineering, Elsevier, 2010, pp. 295-326.
[21] S.S. Shenvi, A.M. Isloor, A.F. Ismail, A review on RO membrane technology: Developments and
challenges, Desalination, 368 (2015) 10-26.
[22] G. Amy, N. Ghaffour, Z. Li, L. Francis, R.V. Linares, T. Missimer, S. Lattemann, Membrane-based
seawater desalination: Present and future prospects, Desalination, 401 (2017) 16-21.
[23] N. Ghaffour, T.M. Missimer, G.L. Amy, Technical review and evaluation of the economics of water
desalination: Current and future challenges for better water supply sustainability, Desalination,
309 (2013) 197-207.
[24] L. Henthorne, B. Boysen, State-of-the-art of reverse osmosis desalination pretreatment,
Desalination, 356 (2015) 129-139.
[25] R. Schwantes, A. Cipollina, F. Gross, J. Koschikowski, D. Pfeifle, M. Rolletschek, V. Subiela,
Membrane distillation: Solar and waste heat driven demonstration plants for desalination,
Desalination, 323 (2013) 93-106.
[26] B.D. Coday, P. Xu, E.G. Beaudry, J. Herron, K. Lampi, N.T. Hancock, T.Y. Cath, The sweet spot of
forward osmosis: Treatment of produced water, drilling wastewater, and other complex and
difficult liquid streams, Desalination, 333 (2014) 23-35.
[27] R. Semiat, J. Sapoznik, D. Hasson, Energy Aspects in Osmotic Processes, Desalination and Water
Treatment, 15 (2010) 228-235.
[28] A. Sagiv, A. Zhu, P.D. Christofides, Y. Cohen, R. Semiat, Analysis of forward osmosis desalination via
two-dimensional FEM model, Journal of Membrane Science, 464 (2014) 161-172.
[29] J.R. Ziolkowska, R. Reyes, Impact of socio-economic growth on desalination in the US, Journal of
Environmental Management, 167 (2016) 15-22.
[30] M.J. Mickley, J. Jordahl, A. Arakel, Development of a Knowledge Base for Desalination Concentrate
and Salt Management, WateReuse Research Foundation, 2011.
[31] R. Semiat, Energy Issues in Desalination Processes, Environmental Science & Technology, 42 (2008)
8193-8201.
[32] N. Voutchkov, R. Semiat, Seawater Desalination, in: Advanced Membrane Technology and
Applications, John Wiley & Sons, Inc., 2008, pp. 47-86.
[33] L.M. Camacho, L. Dumée, J. Zhang, J. Li, M. Duke, J. Gomez, S.R. Gray, Advances in Membrane
Distillation for Water Desalination and Purification Applications, Water, 5 (2013) 94-196.
[34] S. Tavakkoli, O.R. Lokare, R.D. Vidic, V. Khanna, Systems-Level Analysis of Waste Heat Recovery
Opportunities from Natural Gas Compressor Stations in the United States, ACS Sustainable
Chemistry & Engineering, 4 (2016) 3618-3626.
[35] D.B. Gingerich, M.S. Mauter, Quantity, Quality, and Availability of Waste Heat from United States
Thermal Power Generation, Environmental Science & Technology, 49 (2015) 8297-8306.
[36] P. Rao, A. Aghajanzadeh, P. Sheaffer;, I. William R. Morrow, S. Brueske, C. Dollinger, K. Price, P.
Sarker, N. Ward, J. Cresko, Volume 1: Survey of Available Information in Support of the Energy34
This article is protected by copyright. All rights reserved.
Water Bandwidth Study of Desalination Systems, Energy Technology Area, LBNL-1006424, in,
2016.
[37] A. Rahardianto, W.-Y. Shih, R.-W. Lee, Y. Cohen, Diagnostic characterization of gypsum scale
formation and control in RO membrane desalination of brackish water, Journal of Membrane
Science, 279 (2006) 655-668.
[38] W.-Y. Shih, A. Rahardianto, R.-W. Lee, Y. Cohen, Morphometric characterization of calcium sulfate
dihydrate (gypsum) scale on reverse osmosis membranes, Journal of Membrane Science, 252
(2005) 253-263.
[39] G. Greenberg, D. Hasson, R. Semiat, Limits of RO recovery imposed by calcium phosphate
precipitation, Desalination, 183 (2005) 273-288.
[40] D. Hasson, A. Drak, R. Semiat, Inception of CaSO4 scaling on RO membranes at various water
recovery levels, Desalination, 139 (2001) 73-81.
[41] D. Hasson, R. Semiat, D. Bramson, M. Busch, B. Limoni-Relis, Suppression of CaCO3 scale deposition
by anti-scalants, Desalination, 118 (1998) 285-296.
[42] R. Semiat, I. Sutzkover, D. Hasson, Technique for evaluating silica scaling and its inhibition in RO
desalting, Desalination, 140 (2001) 181-193.
[43] J. Thompson, A. Rahardianto, H. Gu, M. Uchymiak, A. Bartman, M. Hedrick, D. Lara, J. Cooper, J.
Faria, P.D. Christofides, Rapid field assessment of RO desalination of brackish agricultural drainage
water, Water research, 47 (2013) 2649-2660.
[44] J. Thompson, A. Rahardianto, S. Kim, M. Bilal, R. Breckenridge, Y. Cohen, Real-time direct detection
of silica scaling on RO membranes, Journal of Membrane Science, 528 (2017) 346-358.
[45] J. Thompson, N. Lin, E. Lyster, R. Arbel, T. Knoell, J. Gilron, Y. Cohen, RO membrane mineral scaling
in the presence of a biofilm, Journal of Membrane Science, 415–416 (2012) 181-191.
[46] H. Shemer, D. Hasson, R. Semiat, Review of the State of the Art of Antiscalant Selection, in: Mineral
Scales in Biological and Industrial Systems, CRC Press, 2013, pp. 227-256.
[47] OCWD, OCSD, Groundwater Replenishment System Technical Brochure, Orange County Water
District and Orange County Sanitation District, in, Fountain Valley, California, 2016.
[48] L.X. Gao, A. Rahardianto, H. Gu, P.D. Christofides, Y. Cohen, Novel design and operational control of
integrated ultrafiltration — Reverse osmosis system with RO concentrate backwash, Desalination,
382 (2016) 43-52.
[49] L. Gao, A. Rahardianto, H. Gu, P.D. Christofides, Y. Cohen, Energy-Optimal Control of RO
Desalination, Industrial & Engineering Chemistry Research, 53 (2014) 7409-7420.
[50] L.X. Gao, H. Gu, A. Rahardianto, P.D. Christofides, Y. Cohen, Self-adaptive cycle-to-cycle control of
in-line coagulant dosing in ultrafiltration for pre-treatment of reverse osmosis feed water,
Desalination, 401 (2017) 22-31.
[51] H. Gu, A. Rahardianto, L.X. Gao, P.D. Christofides, Y. Cohen, Ultrafiltration with self-generated RO
concentrate pulse backwash in a novel integrated seawater desalination UF-RO system, Journal of
Membrane Science, 520 (2016) 111-119.
[52] S. Surawanvijit, A. Rahardianto, Y. Cohen, An Integrated approach for characterization of polyamide
reverse osmosis membrane degradation due to exposure to free chlorine, Journal of Membrane
Science, 510 (2016) 164-173.
35
This article is protected by copyright. All rights reserved.
[53] J.-E. Gu, B.-M. Jun, Y.-N. Kwon, Effect of chlorination condition and permeability of chlorine species
on the chlorination of a polyamide membrane, Water Research, 46 (2012) 5389-5400.
[54] A. Rahardianto, B.C. McCool, Y. Cohen, Reverse osmosis desalting of inland brackish water of high
gypsum scaling propensity: kinetics and mitigation of membrane mineral scaling, Environmental
science & technology, 42 (2008) 4292-4297.
[55] H. Gu, A.R. Bartman, M. Uchymiak, P.D. Christofides, Y. Cohen, Self-adaptive feed flow reversal
operation of reverse osmosis desalination, Desalination, 308 (2013) 63-72.
[56] A. Sagiv, R. Semiat, Backwash of RO spiral wound membranes, Desalination, 179 (2005) 1-9.
[57] R. Segev, D. Hasson, R. Semiat, Improved high recovery brackish water desalination process based
on fluidized bed air stripping, Desalination, 281 (2011) 75-79.
[58] C.J. Gabelich, A. Rahardianto, C.R. Northrup, T.I. Yun, Y. Cohen, Process evaluation of intermediate
chemical demineralization for water recovery enhancement in production-scale brackish water
desalting, Desalination, 272 (2011) 36-45
[59] A. Rahardianto, B.C. McCool, Y. Cohen, Accelerated desupersaturation of reverse osmosis
concentrate by chemically-enhanced seeded precipitation, Desalination, 264 (2010) 256-267.
[60] V. Freger, H. Shemer, A. Sagiv, R. Semiat, Chapter 8 - Boron Removal Using Membranes A2 - Kabay,
Nalan, in: M. Bryjak, N. Hilal (Eds.) Boron Separation Processes, Elsevier, Amsterdam, 2015, pp.
199-217.
[61] R. Semiat, I. Sutzkover, D. Hasson, Scaling of RO membranes from silica supersaturated solutions,
Desalination, 157 (2003) 169-191.
[62] A. Sagiv, R. Semiat, Analysis of parameters affecting boron permeation through reverse osmosis
membranes, Journal of Membrane Science, 243 (2004) 79-87.
[63] D. Lisitsin, D. Hasson, R. Semiat, The potential of CO2 stripping for pretreating brackish and
wastewater desalination feeds, Desalination, 222 (2008) 50-58.
[64] H. Shemer, D. Hasson, R. Semiat, State-of-the-art review on post-treatment technologies,
Desalination, 356 (2015) 285-293.
[65] P. Xu, T.Y. Cath, A.P. Robertson, M. Reinhard, J.O. Leckie, J.E. Drewes, Critical Review of Desalination
Concentrate Management, Treatment and Beneficial Use, Environmental Engineering Science, 30
(2013) 502-514.
[66] A. Pérez-González, A.M. Urtiaga, R. Ibáñez, I. Ortiz, State of the art and review on the treatment
technologies of water reverse osmosis concentrates, Water Research, 46 (2012) 267-283.
[67] A. Giwa, V. Dufour, F. Al Marzooqi, M. Al Kaabi, S.W. Hasan, Brine management methods: Recent
innovations and current status, Desalination, 407 (2017) 1-23.
[68] J. Morillo, J. Usero, D. Rosado, H. El Bakouri, A. Riaza, F.-J. Bernaola, Comparative study of brine
management technologies for desalination plants, Desalination, 336 (2014) 32-49.
[69] J. Leong, J. Tan, J. Charrois, B.P. Ladewig, Review of high recovery concentrate management
options, Desalination and Water Treatment, 52 (2014) 7609-7627.
[70] R. Semiat, D. Hasson, G. Zelmanov, I. Hemom, Threshold scaling limits of RO concentrates flowing in
a long waste disposal pipeline, Water Sci Technol., 49 (2004) 211-219.
[71] M.C. Mickley, Membrane Concentrate Disposal: Practices and Regulation, in: Desalination and
Water Purification Research and Development Program Report No. 123, U.S. Department of the
Interior, Bureau of Reclamation, Denver, CO, 2006.
36
This article is protected by copyright. All rights reserved.
[72] R. Segev, D. Hasson, R. Semiat, Modeling CaCO3 precipitation in fluidized bed CO2 stripping
desalination process, Desalination, 311 (2013) 192-197.
[73] K.S. Spiegler, O. Kedem, Thermodynamics of hyperfiltration (reverse osmosis): criteria for efficient
membranes, Desalination, 1 (1966) 311-326.
[74] E. Lyster, J. Au, R. Rallo, F. Giralt, Y. Cohen, Coupled 3-D hydrodynamics and mass transfer analysis
of mineral scaling-induced flux decline in a laboratory plate-and-frame reverse osmosis membrane
module, Journal of Membrane Science, 339 (2009) 39-48.
[75] E. Lyster, Y. Cohen, Numerical study of concentration polarization in a rectangular reverse osmosis
membrane channel: Permeate flux variation and hydrodynamic end effects, Journal of Membrane
Science, 303 (2007) 140-153.
[76] I. Sutzkover, D. Hasson, R. Semiat, Simple technique for measuring the concentration polarization
level in a reverse osmosis system, Desalination, 131 (2000) 117-127.
[77] A. Antony, J.H. Low, S. Gray, A.E. Childress, P. Le-Clech, G. Leslie, Scale formation and control in high
pressure membrane water treatment systems: A review, Journal of Membrane Science, 383
(2011) 1-16.
[78] A. Zhu, A. Rahardianto, P.D. Christofides, Y. Cohen, Reverse osmosis desalination with high
permeability membranes—cost optimization and research needs, Desalination and Water
Treatment, 15 (2010) 256-266.
[79] R. Huehmer, S. Alt, J. Lozier, A. Kupp, T. Nading, K. Egrican, B. Emerson, Evaluation and Optimization
of Emerging and Existing Energy Recovery Devices for Desalination and Wastewater Membrane
Treatment Plants, WateReuse Research Foundation Project No. WRF-08-14, (2014).
[80] Dow Water & Process Solutions, FILMTEC Reverse Osmosis Membranes Technical Manual, Form No.
609-00071-1009, in: http://dowac.custhelp.com/app/answers/detail/a_id/3428, 2014.
[81] A. Efraty, Closed circuit desalination series no-3: high recovery low energy desalination of brackish
water by a new two-mode consecutive sequential method, Desalination and Water Treatment, 42
(2012) 256-261.
[82] A. Stikker, Market Profile, Desalination Mark., in, Global Water Intelligence, 2006.
[83] A. Zhu, P.D. Christofides, Y. Cohen, Effect of Thermodynamic Restriction on Energy Cost
Optimization of RO Membrane Water Desalination, Industrial & Engineering Chemistry Research,
In Press (2009).
[84] A. Zhu, P.D. Christofides, Y. Cohen, Minimization of energy consumption for a two-pass membrane
desalination: Effect of energy recovery, membrane rejection and retentate recycling, Journal of
Membrane Science, 339 (2009) 126-137.
[85] A. Zhu, P.D. Christofides, Y. Cohen, Effect of Stream Mixing on RO Energy Cost Minimization,
Desalination, 261 (2010) 232-239.
[86] B.C. McCool, A. Rahardianto, J.I. Faria, Y. Cohen, Evaluation of chemically-enhanced seeded
precipitation of RO concentrate for high recovery desalting of high salinity brackish water,
Desalination, 317 (2013) 116-126.
[87] K.S. Spiegler, Y.M. El-Sayed, The energetics of desalination processes, Desalination, 134 (2001) 109128.
[88] K.S. Spiegler, Y.M. El-Sayed, A desalination primer : introductory book for students and newcomers
to desalination, Balaban Desalination Publications, Santa Maria Imbaro, Italy, 1994.
37
This article is protected by copyright. All rights reserved.
[89] Water Science and Technology Board, Review of the Desalination and Water Purification
Technology Roadmap, National Academies Press, 2004.
[90] A. Zhu, P.D. Christofides, Y. Cohen, On RO membrane and energy costs and associated incentives for
future enhancements of membrane permeability, Journal of Membrane Science, 344 (2009) 1-5.
[91] A.R. Bartman, A. Zhu, P.D. Christofides, Y. Cohen, Minimizing energy consumption in reverse
osmosis membrane desalination using optimization-based control, Journal of Process Control, 20
(2010) 1261-1269.
[92] M. Li, Energy Consumption in Spiral-Wound Seawater Reverse Osmosis at the Thermodynamic
Limit, Industrial & Engineering Chemistry Research, 53 (2014) 3293-3299.
[93] A. Efraty, R.N. Barak, Z. Gal, Closed circuit desalination — A new low energy high recovery
technology without energy recovery, Desalination and Water Treatment, 31 (2011) 95-101.
[94] T. Qiu, P.A. Davies, Comparison of Configurations for High-Recovery Inland Desalination Systems,
Water, 4 (2012) 690.
[95] D.M. Warsinger, E.W. Tow, K.G. Nayar, L.A. Maswadeh, J.H. Lienhard V, Energy efficiency of batch
and semi-batch (CCRO) reverse osmosis desalination, Water Research, 106 (2016) 272-282.
[96] A.J. Karabelas, M. Kostoglou, C.P. Koutsou, Modeling of spiral wound membrane desalination
modules and plants – review and research priorities, Desalination, 356 (2015) 165-186.
[97] A.J. Karabelas, C.P. Koutsou, M. Kostoglou, The effect of spiral wound membrane element design
characteristics on its performance in steady state desalination — A parametric study, Desalination,
332 (2014) 76-90.
[98] X. Pascual, H. Gu, A. Bartman, A. Zhu, A. Rahardianto, J. Giralt, R. Rallo, P.D. Christofides, Y. Cohen,
Fault Detection and Isolation in a Spiral-Wound Reverse Osmosis (RO) Desalination Plant,
Industrial & Engineering Chemistry Research, 53 (2014) 3257-3271.
[99] D. Garcia-Alvarez, M.J. Fuente, A UPCA-based monitoring and fault detection approach for reverse
osmosis desalination plants, Desalination and Water Treatment, 52 (2014) 1272-1286.
[100] C.W. McFall, A. Bartman, P.D. Christofides, Y. Cohen, Control and Monitoring of a High Recovery
Reverse Osmosis Desalination Process, Industrial & Engineering Chemistry Research, 47 (2008)
6698-6710.
[101] F. Ahmed, B.S. Lalia, V. Kochkodan, N. Hilal, R. Hashaikeh, Electrically conductive polymeric
membranes for fouling prevention and detection: A review, Desalination, 391 (2016) 1-15.
[102] M. Uchymiak, A. Rahardianto, E. Lyster, J. Glater, Y. Cohen, A novel RO ex situ scale observation
detector (EXSOD) for mineral scale characterization and early detection, Journal of Membrane
Science, 291 (2007) 86-95.
[103] D.J. Miller, D.R. Dreyer, C.W. Bielawski, D.R. Paul, B.D. Freeman, Surface Modification of Water
Purification Membranes, Angewandte Chemie International Edition, (2017) n/a-n/a.
[104] M.-L. Pype, M.G. Lawrence, J. Keller, W. Gernjak, Reverse osmosis integrity monitoring in water
reuse: The challenge to verify virus removal – A review, Water Research, 98 (2016) 384-395.
[105] S. Surawanvijit, J. Thompson, A. Rahardianto, V. Frenkel, Y. Cohen, Pulsed marker method for realtime detection of reverse osmosis membrane integrity loss, Desalination, 370 (2015) 25-32.
[106] J.M. Gohil, P. Ray, A review on semi-aromatic polyamide TFC membranes prepared by interfacial
polymerization: Potential for water treatment and desalination, Separation and Purification
Technology, 181 (2017) 159-182.
38
This article is protected by copyright. All rights reserved.
[107] J.R. Werber, C.O. Osuji, M. Elimelech, Materials for next-generation desalination and water
purification membranes, Nature Reviews Materials, 1 (2016) 16018.
[108] Y. Jiang, P. Biswas, J.D. Fortner, A review of recent developments in graphene-enabled membranes
for water treatment, Environmental Science: Water Research & Technology, 2 (2016) 915-922.
[109] Y. Cohen, N. Lin, K.J. Varin, D. Chien, R.F. Hicks, Membrane Surface Nanostructuring with
Terminally Anchored Polymer Chains, in: Functional Nanostructured Materials and Membranes
for Water Treatment, Wiley-VCH Verlag GmbH & Co. KGaA, 2013, pp. 85-124.
[110] J.H. Jhaveri, Z.V.P. Murthy, A comprehensive review on anti-fouling nanocomposite membranes
for pressure driven membrane separation processes, Desalination, 379 (2016) 137-154.
39
This article is protected by copyright. All rights reserved.