Pair your accounts.

Export articles to Mendeley

Get article recommendations from ACS based on references in your Mendeley library.

Pair your accounts.

Export articles to Mendeley

Get article recommendations from ACS based on references in your Mendeley library.

You’ve supercharged your research process with ACS and Mendeley!

STEP 1:
Click to create an ACS ID

Please note: If you switch to a different device, you may be asked to login again with only your ACS ID.

Please note: If you switch to a different device, you may be asked to login again with only your ACS ID.

Please note: If you switch to a different device, you may be asked to login again with only your ACS ID.

MENDELEY PAIRING EXPIRED
Your Mendeley pairing has expired. Please reconnect
ACS Publications. Most Trusted. Most Cited. Most Read
Energy Consumption of Brackish Water Desalination: Identifying the Sweet Spots for Electrodialysis and Reverse Osmosis
My Activity
  • Open Access
Article

Energy Consumption of Brackish Water Desalination: Identifying the Sweet Spots for Electrodialysis and Reverse Osmosis
Click to copy article linkArticle link copied!

  • Sohum K. Patel
    Sohum K. Patel
    Department of Chemical and Environmental Engineering, Yale University, New Haven, Connecticut 06520-8286, United States
    Nanosystems Engineering Research Center for Nanotechnology-Enabled Water Treatment (NEWT), Yale University, New Haven, Connecticut 06520-8286, United States
  • P. Maarten Biesheuvel
    P. Maarten Biesheuvel
    European Centre of Excellence for Sustainable Water Technology, Wetsus, Oostergoweg 9, 8911 MA Leeuwarden, The Netherlands
  • Menachem Elimelech*
    Menachem Elimelech
    Department of Chemical and Environmental Engineering, Yale University, New Haven, Connecticut 06520-8286, United States
    Nanosystems Engineering Research Center for Nanotechnology-Enabled Water Treatment (NEWT), Yale University, New Haven, Connecticut 06520-8286, United States
    *Menachem Elimelech. Email: [email protected]. Phone: (203) 432-2789.
Open PDFSupporting Information (1)

ACS ES&T Engineering

Cite this: ACS EST Engg. 2021, 1, 5, 851–864
Click to copy citationCitation copied!
https://doi.org/10.1021/acsestengg.0c00192
Published January 22, 2021

Copyright © 2021 The Authors. Published by American Chemical Society. This publication is licensed under CC-BY-NC-ND.

Abstract

Click to copy section linkSection link copied!

Though electrodialysis (ED) and reverse osmosis (RO) are both mature, proven technologies for brackish water desalination, RO is currently utilized to desalinate over an order of magnitude more brackish water than ED. This large discrepancy in the adoption of each technology has yet to be thoroughly justified in the literature, particularly from the perspective of energy consumption. Hence, in this study, we performed a direct and systematic comparison of the energy consumption of RO and ED for brackish water desalination, precisely mapping out the ideal operational space of each technology for the first time. Using rigorous system-scale models for RO and ED, we determine the specific energy consumption and energy efficiency of each process over a wide range of brackish water conditions. Specifically, we investigate the effects of varying feed salinity, extent of salt removal, water recovery, and productivity to ultimately identify the operational sweet spots of each technology. By maintaining the same separation parameters (i.e., feed salinity, salt removal, water recovery) and productivity between RO and ED throughout the study, we ensure that our comparison of the technologies is valid and fair. Our results indicate that both RO and ED are capable of operating with high energy efficiency (>30%) for brackish water desalination, though for differing conditions. Particularly, we show that whereas ED excels for low feed salinities (<3 g L–1) and extents of salt removal, RO operates optimally for high salinity feeds (>5 g L–1), which require more extensive desalination. Through our in-depth energetic analysis, we provide guidance for future applications of RO and ED, emphasizing that increased implementation of ED will require significant reduction in the cost of ion-exchange membranes.

Copyright © 2021 The Authors. Published by American Chemical Society

INTRODUCTION

Click to copy section linkSection link copied!

While the planet’s limited freshwater resources incessantly shrink due to pollution and climate change, the global demand for freshwater continues to rapidly increase with population growth. In effect, global water scarcity persists as one of the greatest challenges of our time, already affecting two-thirds of the world population and expanding at an alarming rate. (1−3) Though improved water management and conservation are essential countermeasures, effectively combatting water scarcity will require further intervention, specifically through the harnessing of unconventional water sources to augment dwindling freshwater supplies. Alternative water sources, however, are generally characterized by high salinities, requiring salt to be extracted from the water prior to utilization.
The process of water–salt separation, termed desalination, inherently requires the consumption of energy, making the minimization of energy consumption a long-standing goal in the development of desalination technologies. (4−6) A simple, yet effective approach to reducing the minimum energy requirement of desalination is to target lower salinity waters; (7) hence, as reliance on desalinated water has recently surged, brackish waters (salinity range of approximately 1 g L–1 to 10 g L–1) have increasingly been utilized as feedwater, as opposed to the vastly abundant, but considerably more saline, seawater. Therefore, the identification and implementation of the most energy efficient brackish water desalination technology is critical toward ensuring sustainable water production in the coming decades.
Currently, reverse osmosis (RO) dominates global desalination capacity, accounting for over 80% of the desalinated brackish water. (8) In contrast, the second most utilized technology for brackish water desalination, electrodialysis (ED), commands only 8% of the brackish water desalination market share, (8) despite suggestions that the energy consumption of ED is comparable or even superior to RO for certain brackish water conditions. (9) The highly distinct mechanisms of water–salt separation in RO and ED, however, make the conditions for which each technology is energetically advantageous difficult to discern.
Reverse osmosis, a pressure-driven process, makes use of a semipermeable membrane which allows for the permeation of water molecules while rejecting the passage of solutes. Hence, as pressurized saline feedwater is passed along a membrane module, a freshwater permeate stream and concentrated brine stream are produced. (10,11) The energy consumption of RO is dictated by the pressurization of the feedwater, with the process inherently requiring the application of hydraulic pressures in excess of the osmotic pressure of the exiting brine. Though the first viable cellulose acetate RO membrane was demonstrated in the 1960s, (12) it was not until the development of the polyamide thin-film-composite (TFC) membrane in the late 1970s that RO proliferated in application. (13) Current TFC RO membranes, combined with state-of-the-art energy recovery devices and pumps, facilitate seawater RO to operate within a remarkable 2-fold of the thermodynamic minimum energy requirements, making RO the unrivaled technology for seawater desalination. (10,14) However, at lower feed salinities, such as the brackish water regime, the energy efficiency of RO has been demonstrated to be comparable to other technologies. (7) Furthermore, efforts to reduce the energy consumption of brackish water RO through enhancement of membrane material properties (i.e., water permeability) have been shown to be relatively insignificant, providing alternative technologies the opportunity to compete for brackish water desalination capacity. (15,16)
Electrodialysis, also a relatively mature membrane-based desalination process, is an electrically driven technology that relies on the transmembrane transport of ions, rather than water, to achieve desalination. (17−19) Specifically, in ED an electric field is applied across a stack of ion-exchange membranes to drive ions toward the oppositely charged electrode. As the feedwater is passed through the flow channels, cations and anions selectively pass through alternating cation- and anion-exchange membranes, respectively, thereby forming ion-concentrated and depleted solution streams on each side of the membrane. Since the first demonstration of multicompartment ED in 1940, (20) the technology has gained considerable momentum, being primarily adopted for brackish water desalination, but also applied in the bioprocessing, pharmaceutical, and food industries. (17,21,22) The gradual development of highly selective and conductive ion-exchange membranes (23) has made ED an energy efficient technology for brackish water desalination, as demonstrated by our previous study. (24) Nonetheless, compared to RO, ED’s penetration of the brackish water desalination market is minimal and warrants further assessment.
Previous works have investigated the optimal operation of ED, (25,26) while others have discussed the relative merits of RO and ED (17) and proposed RO-ED hybrid systems. (27,28) Nonetheless, precise and rigorous identification of the operational sweet spots of RO and ED, with respect to one another, has yet to be demonstrated. Hence, there is a critical need for such an analysis to assist in guiding the future direction of the growing brackish water desalination sector.
In this study, we systematically compare the energy consumption of ED and RO over a wide range of brackish water operating conditions to determine the optimal applications of each desalination technology with regard to energy consumption. To ensure a meaningful comparison, the feed salinity, water recovery, and salt removal are fixed across both technologies, thereby setting the theoretical minimum energetic demand of a given water–salt separation. Additionally, the kinetic efficiency of desalination is held constant across ED and RO by fixing the system productivity, defined as the rate of water production normalized by the projected membrane area. Our comparison ultimately provides fundamental insight into the current RO dominated brackish water desalination market and reveals opportunities for the growth of ED in particular applications. We conclude by highlighting practical considerations, outside of energy consumption, which largely influence the employment and outlook of each technology.

ED PROCESS MODEL AND CALCULATION OF ENERGY CONSUMPTION

Click to copy section linkSection link copied!

ED Operation Mode

In ED, numerous cell pairs—each consisting of an anion-exchange membrane (AEM), cation exchange membrane (CEM), concentrate channel, and diluate channel—are placed between a pair of electrodes. Here, we consider the conventional plate-and-frame stack configuration, in which the flow channels and ion-exchange membranes are aligned between the electrodes, as shown in Figure 1. To represent commercial-scale brackish water desalination, we simulate a fifty-cell pair stack operating under steady state continuous flow conditions. (19,29)

Figure 1

Figure 1. Schematic illustration of the variable water recovery electrodialysis (ED) process. A portion of the multicell pair ED stack is shown. Each cell pair contains an anion-exchange membrane (AEM), cation-exchange membrane (CEM), and spacer channels between the membranes. An external voltage is applied across the ED stack as (an equal flow rate of) saline water passes through each of the flow channels. The generated electric field causes ions in the channels to migrate toward the oppositely charged electrode. The AEM and CEM enable selective permeation of anions and cations, respectively. Due to nonideal membrane selectivity, however, the transport of counterions across the membrane (Jct) is also accompanied by some degree of counterproductive co-ion transport (Jco). The overall selective transport of ions results in alternating product and brine channels. Redox reactions at the surface of the electrodes convert the ionic current into an electrical current (i). For a system-scale water recovery greater than 50%, a fraction of the brine effluent (αR) is recycled, thus reducing the overall volume of brackish feedwater and increasing the system-scale water recovery.

Upon polarization of the electrodes, anions and cations in the flow channels migrate toward the oppositely charged electrode, selectively traversing the anion- and cation-exchange membranes, respectively. This counterion flux (Jct) is the underlying principle of ED desalination and results in alternating channels of diluate and concentrate. Practical ion-exchange membranes, however, also permit some degree of co-ion flux (Jco) from the concentrate to diluate channel, detracting from the efficacy of salt removal in ED. Redox reactions at the electrodes enable the conversion of the ionic current in the ED stack to an electrical current, which flows from the cathode to the anode via an external circuit. Here, we assume reduction of hydrogen ions at the cathode, and oxidation of chloride ions at the anode, corresponding to a combined electrode potential (Vel) of 1.4 V throughout the analysis.
Co-current and equal flow is maintained through each of the channels to minimize transmembrane hydrostatic pressure differences, which would otherwise exacerbate water transport across the ion-exchange membranes. (29,30) Hence, the module-scale water recovery is fixed at 50% throughout the analysis. To vary the system-scale water recovery, we employ the commonly practiced “feed-and-bleed” operation mode. (17−19,31) Particularly, a fraction of the produced brine (αR) is recirculated and mixed with the brackish feedwater supplied to the concentrate channels. In effect, the volume of brackish feedwater required to produce a given volume of product water is reduced, thus allowing the system-scale water recovery to exceed the module-scale recovery of 50%. In Figure S1, we provide a schematic illustration of the feed-and-bleed operation mode on a single pair of concentrate and diluate channels to clearly demonstrate this operation mode.

ED Process Model

We utilize a two-dimensional (2-D) Nernst–Planck model to fundamentally capture the underlying ion transport phenomena in the ED stack. The modeling framework utilized was originally demonstrated by Sonin and Probstein (32) in 1968 and was later extended to include co-ion transport in the ion-exchange membranes, Donnan potentials at the membrane–solution interfaces, water transport through the membranes, and the effects of porosity and tortuosity in a spacer-filled channel. (30,33−35) These advancements of the Sonin–Probstein model have facilitated more realistic description of desalination performance, cell pair voltage, and current distribution, as has been confirmed through rigorous validation against several experimental data sets. (30,34) In this 2-D model, the applied cell pair voltage is invariant across the length of the membrane, and the total current in the stack self-consistently distributes over the membrane surface (i.e., more current may go to parts closer to the channel inlets, where the transmembrane concentration differences are lower, than to later parts of the channel). Thus, all positions in the 2-D geometry are numerically coupled together. Further model illustration is provided in Figure S2, where we demonstrate the capability of predicting key ED transport phenomena, such as concentration polarization and the limiting current density.
As demonstrated by previous studies, the model relies on the numerical solution of the extended Nernst–Planck equation, both in the spacer channels and in the ion-exchange membranes. (30,32−34) Two-dimensional Nernst–Planck modeling effectively minimizes the number of independent variables and circumvents the assumption of boundary layer thicknesses commonly used to describe concentration polarization. Specifically, in our 2-D modeling approach we generate concentration profiles for each discretized position down the length of the channel. The concentration at each point in the spacer channels—with the exception of the membrane–solution interfaces, where thin electrical double layers form—is governed by the condition of electroneutrality and the convection–diffusion equation (eq S6). Concentrations at the membrane–solution interfaces are determined according to the continuity of ion flux between the ion-exchange membranes and solution, coupled with the Donnan potentials on each side of the membrane. Thus, the effects of concentration polarization are effectively captured without applying the concept of separate boundary layers next to a thin core “bulk” region, which is not particularly applicable for the narrow spacer-filled channels of an ED membrane stack.
Whereas ion transport across the thickness of the spacer channels (i.e., perpendicular to the membranes) is governed by diffusion and migration, ion transport along the length of the channel (i.e., parallel to the membranes) is solely related to advection. Notably, we neglect water transport across the ion-exchange membranes, as the effects of both osmotic and electro-osmotic transport are insignificant for analysis of brackish water ED. (29,30) A fully developed Hagen–Poiseuille velocity profile is applied for the convective fluid flow through the entirety of the spacer channels. Though the fluid streamlines in a spacer-filled channel may be more accurately described by different flow profiles, it has been shown that the fluid profile applied in the 2-D modeling approach leads to minimal deviation in the prediction of ED process performance. (33) We also confirm the relative lack of importance in selection of the velocity profile in Figure S3, in which we show that changing the Hagen–Poiseuille profile to plug flow has nearly no effect on the predicted energy consumption of ED.
In contrast to the fluid flow profile, we emphasize that the estimation of the porosity (ε) and tortuosity (τ) in the spacer-filled channels plays a more significant role in the ED process modeling. Though a typical mesh-filled channel provides enhanced fluid dispersion, the inclusion of a spacer results in retarded ion mobility from the reduced amount of cross-sectional area (porosity) and the elongation of the transport pathways (tortuosity). Hence, to accurately predict the potential drops in the spacer channels, it is critical to use an effective diffusion coefficient for electromigration (De), determined by scaling the diffusion coefficient in free (bulk) solution (Dd) by a factor of . Here, we use an experimentally verified value of 0.314. (30) We also note that our determination of the full concentration profiles across the thickness of the channel, with the use of an experimentally verified effective diffusion coefficient for electromigration, intrinsically captures the so-called spacer shadow effect. (36,37) Further details regarding the ED process modeling—including specified values for system geometry, mass transport properties, and membrane properties—are provided in the Supporting Information.
Throughout the modeling, feedwater is considered as sodium chloride solution, and the sodium and chloride ions are assumed to possess the same diffusivity, as is common in the modeling of electro-driven desalination processes, such as electrodialysis and capacitive deionization. (27,29,30,33,35,36,38,39) Additionally, the properties of the anion- and cation-exchange membranes (other than the sign of their intrinsic charge) are assumed to be the same. As a result of the assumptions, a plane of symmetry is established in each cell pair, allowing for the computational space to be limited to only half of a cell pair. We note that a more rigorous calculation with unequal diffusion coefficients and ion-exchange membrane properties (34) shows that dropping the assumption of symmetry does not have a considerable effect on the prediction of ED process performance.

ED Performance Metrics

The ED model is used to predict the specific energy consumption (SEC) of ED across a broad range of brackish water desalination conditions. Specific energy consumption, which is the energy consumption normalized by the volume of product water, is calculated by
(1)
where N is the number of cell pairs in the stack, Vcp is the voltage per cell pair, Vel is the combined redox reaction potential of the electrodes (previously defined as a fixed value of 1.4 V), is the average current density in a cell pair, Am is the projected area of the ion-exchange membrane, and QP is the volumetric flow rate of the product water per cell pair. For stacks with a large number of cell pairs, as in any practical application of ED, (17) the term for the redox potential of the electrodes becomes negligible, simplifying the expression. Nonetheless, to ensure the rigor of our analysis, we maintain the use of eq 1 in its presented form throughout our study.
The calculated SEC is compared to the thermodynamic minimum specific energy consumption (SECmin) to calculate the thermodynamic energy efficiency (η) as
(2)
where SECmin represents the theoretical energy required by a thermodynamically reversible process and is a function of only the separation conditions, namely the system-scale water recovery (WR), salt removal (Rs), and feed salinity (cF). Particularly, for a fully dissociated 1:1 salt, SECmin can be expressed as (6)
(3)
where Rg is the universal gas constant, T is the absolute temperature, cB is the salinity of the concentrated brine solution, and cP is the salinity of the desalinated product water.
The specific energy consumption of ED is dependent on several water–salt separation parameters—namely, the salt removal, water recovery, and productivity. Throughout the study we utilize the conventional definition of salt removal, defined in terms of the diluate channel effluent (product water) salinity (cP) and the feed salinity (cF):
(4)
The water recovery of a desalination process is the volume of product water generated per volume of feedwater supplied to the system. Thus, in the feed-and-bleed operation described, the system-scale water recovery is given by
(5)
in which QB,f is the volumetric flow rate of the “makeup” brackish feedwater supplied to the concentrate (brine) channels, which is required in addition to the recirculated brine to equalize the flow rates in the diluate and concentrate channels (a schematic illustration of the feed-and-bleed operation mode is shown in Figure S1).
Lastly, the productivity is a measure of the kinetic efficiency of the separation process and relates the rate of desalinated water production to the system size. For ED, though a cell pair consists of two membranes, the system size is most accurately represented by the projected area of a single ion-exchange membrane (Am). Hence, the productivity is expressed as
(6)

RO PROCESS MODEL AND CALCULATION OF ENERGY CONSUMPTION

Click to copy section linkSection link copied!

RO Operation Mode

In RO, saline feedwater is pressurized and passed along a dense semipermeable membrane, which allows the permeation of water while blocking the passage of virtually all solutes. The hydraulic pressure applied to the feedwater must be greater than the osmotic pressure of the brine (retentate) at the end of the membrane module to ensure use of the entire membrane area, with higher pressures increasing the transmembrane water flux at the expense of greater energy consumption and salt passage. (40) The membrane, being the core of the RO process, largely dictates the overall performance. Current state-of-the-art thin film composite (TFC) polyamide membranes used in brackish water desalination possess operational permeabilities in the range of 3–5 L m–2 h–1 bar–1 and are capable of salt removals in excess of 99%, with some as high as 99.8%. (16)
In order to tune the salt removal of RO for fair comparison with ED, we employ a single-stage bypass system (Figure 2), as introduced in our previous study, (41) in which only a portion of the feedwater is treated by the RO module. The remainder of the feedwater circumvents the high-pressure pump and RO module by passing through either a brine or product bypass stream. The product bypass stream, which is mixed with the permeate from the RO module to form the overall product water, is used to adjust the system-scale salt removal. Alternatively, the brine bypass stream is mixed with the RO retentate and serves to minimize the load on the RO module. The flow rate fractions of each of the three streams are governed by the specified system-scale salt removal and water recovery, with further details provided in the RO Performance Metrics subsection.

Figure 2

Figure 2. Schematic illustration of the variable salt removal reverse osmosis (RO) process. The brackish feedwater is split among three streams (with the following flow rate fractions): RO module feed (αRO), brine bypass (αB), and product bypass (αP). The hydraulic pressure of the feedwater fed to the RO module is increased (by ΔP) via a high-pressure pump, requiring the input of electrical energy. Water permeates through the RO membrane with a flux of Jw, while complete salt rejection is assumed. The permeate and retentate streams from the RO module are mixed with the corresponding bypass streams, generating the overall product and brine streams, respectively. The system-scale salt removal is controlled by tuning the fraction of feedwater sent to the product bypass stream. Alternatively, the brine bypass stream serves to minimize the flow to the RO module, and thus the energy consumption. An energy recovery device (ERD) is applied to the retentate stream exiting the RO module to recover and reuse a significant portion of the mechanical energy provided by the high-pressure pump. However, an ERD is only applied to certain cases throughout the study, as its use in brackish water RO is circumstantial.

A key advantage of RO is that an energy recovery device can be applied to the retentate stream to recover a large fraction of the stored mechanical energy. However, we note that in brackish water RO an energy recovery device (ERD) is often not economically justifiable (especially for lower feed salinities), as the energy savings are not substantial enough to outweigh the incurred capital cost. (42) Thus, in this study we consider scenarios both with and without the application of an energy recovery device, specifically indicating cases in which energy recovery is utilized.

RO Process Model

To describe the mass transfer and concentration polarization phenomena in the RO module, we utilize a classical solution–diffusion-based model coupled with film theory. (43−45) Particularly, we apply one-dimensional finite element analysis to discretize the membrane module and account for the varying flow rate and solution concentrations along the length of the membrane. With the system-scale salt removals of interest in this study (<95%) being considerably lower than the rejection of typical brackish water RO membranes (>99%), salt flux across the membrane is neglected throughout the analysis. This assumption of ideal salt rejection is often applied to simplify RO modeling (46) and is especially valid for the relatively low-salt-removal bypass system utilized, as the system-scale salt removal is primarily controlled by the product bypass stream, rather than the performance of the RO membrane.
With the assumption of complete salt rejection, mass transfer in the RO module is solely described by water transport across the membrane. According to solution–diffusion theory, the water flux in each finite element is given by
(7)
in which A is the membrane permeability coefficient, ΔP is the hydraulic pressure applied to the feedwater, and πF,m is the osmotic pressure corresponding to the salt concentration at the feed-side membrane surface. Throughout the analysis, a constant A value of 5 L m–2 h–1 bar–1 is utilized, representative of a typical brackish water RO membrane. (16)
To determine the water flux, the concentration at the membrane surface must be simultaneously solved. Here, we use film theory to describe concentration polarization effects near the membrane–solution interface. With the assumption of ideal salt rejection, the concentration at the membrane surface (cF,m) can be determined by
(8)
where cF,b is the local feed-side bulk concentration and kF is the feed-side mass transfer coefficient. The mass transfer coefficient is directly proportional to the boundary layer (film) thickness, which is a function of the hydrodynamic conditions in the RO module. However, with the mass transfer coefficient being highly system-dependent, in this study we use a fixed value of 150 L m–2 h–1, as is typical of current RO systems. (16,46)
As the feedwater flows along the length of the membrane, water is transported from the feed stream to the product stream, affecting the volumetric flow rates on each side of the membrane. Accordingly, in each finite element the change in the product-side flow rate (QP) with respect to the membrane surface area (Am) is equal to the water flux:
(9)
In contrast, the change in the flow rate on the feed-side (QF) is the opposite of the water flux:
(10)
By numerically integrating eqs 9 and 10 over the total membrane area, we can calculate the permeate and retentate flow rates at the exit of the module. Thus, the module-scale water recovery (r), the ratio of the permeate and RO feedwater flow rates, can be determined.

RO Performance Metrics

The RO model is used to determine the specific energy consumption (SEC) over various desalination conditions. In the described RO bypass system, the SEC is governed by the fraction of feedwater, which is directed to the RO module (αRO), and hence must be pressurized. Thus, the SEC is given by
(11)
where ΔP is the applied hydraulic pressure, ηERD is the efficiency of the energy recovery device (if applied), WRmod is the module-scale water recovery, WR is the system-scale water recovery, and ηP is the efficiency of the high pressure pump. The efficiency of the energy recovery device and high-pressure pump are both set to 0.8 throughout the analysis. We note that an efficiency of 0.8 may be considered low with respect to the performance of state-of-the-art pressure exchanger devices and high-pressure pumps; (11,47) however, for the comparative purposes of this study, we opt for relatively conservative values. As with ED, the SEC of RO is compared to the SECmin (eq 3) to determine the energy efficiency of desalination.
The flow rate fractions for the RO module feed (αRO), brine bypass (αB), and product bypass (αP) streams are critical parameters toward the system-scale performance as they are directly related to the energy consumption, salt removal, and water recovery. We note that the system-scale salt removal is defined in the same manner as ED (eq 4), while the system-scale water recovery (WR) is defined as the ratio of the volumetric flow rates of the product water (QP) and feedwater (QF):
(12)
The flow rate fraction of the product bypass stream controls the system-scale salt removal and also influences the system-scale water recovery. Accordingly, a system mass balance reveals that the flow rate fraction of the product water bypass stream is given by
(13)
The RO module flow rate fraction is also related to the system-scale salt removal and water recovery, while additionally affecting the module-scale water recovery:
(14)
After solving the product bypass and RO module flow rate fractions, the brine bypass flow rate fraction can be determined by realizing the sum of the flow rate fractions must equal unity.
Though the system-scale salt removal and water recovery are specified parameters for a given separation, the module-scale water recovery remains a degree of freedom. Affecting both the recoverable energy as well as the required hydraulic pressure (as shown by eq 11), the module-scale water recovery directly influences the SEC of the RO bypass system. Hence, for each investigated separation condition, the module-scale water recovery which minimizes the SEC is determined through numerical optimization. For further details regarding the bypass system modeling, we refer the reader to our previous work. (41)
In order to ensure a fair comparison, the productivity (or average water flux) of the RO system must be consistent with that of ED. The productivity, as previously defined, relates the total rate of water production to the system size. Hence, for the RO system, the productivity is determined in the same manner as ED (eq 6), with the membrane area, Am, being the projected surface area of the RO membrane.

COMPARISON OF ENERGY CONSUMPTION IN RO AND ED

Click to copy section linkSection link copied!

Effect of Varying Feed Salinity and Salt Removal

Increasing the feed salinity or the extent of salt removal demands greater energy consumption in both RO and ED, despite their highly distinct mechanisms of water–salt separation. In ED, treating higher salinity feeds or increasing the extent of desalination ultimately requires the transport of more ions across the ion-exchange membranes. Thus, a greater driving force, in the form of amplified electrical potential, must be applied, increasing SEC. For RO, SEC is dictated by the volume of feedwater that must be pressurized and the magnitude of pressure that must be applied by the pump. To make effective use of the membrane area, the hydraulic pressure applied must exceed the osmotic pressure of the retentate stream at the end of the RO module. Thereby, higher salinity feedwaters, which in turn generate retentates of higher concentration and osmotic pressure, require the application of larger hydraulic pressures. The extent of salt removal in RO, unlike ED, cannot readily be adjusted by manipulating the driving force. Rather, we enable variable salt removal through the bypass system described in Figure 2. In such a bypass system, increasing the degree of desalination requires a larger fraction of the feedwater to pass through the RO module, thereby increasing the energy demand of the pump.
In Figure 3, we show the energy consumption, SEC, of RO (without energy recovery) and ED as a function of salt removal for various salinity feedwaters. Feed salinities of 1 g L–1 (orange), 3 g L–1 (red), 5 g L–1 (blue), and 10 g L–1 (purple) are assessed to encompass the range of brackish water salinities, while water recovery and productivity are fixed at 80% and 20 L m–2 h–1, respectively. As expected, the SEC of both RO and ED consistently increases with the extent of salt removal and feed salinity, albeit at varying rates. Overall, Figure 3 highlights that ED and RO outperform one another for particular feed salinities and salt removals. Whereas ED performs best for relatively low feed salinities and salt removals, RO has an energetic advantage over ED for higher feed salinities and salt removals.

Figure 3

Figure 3. Specific energy consumption (SEC) for RO and ED as a function of salt removal (Rs). The water recovery and productivity are fixed at 80% and 20 L m–2 h–1, respectively. The various colored lines show different feed salinities: 1 g L–1 (orange), 3 g L–1 (red), 5 g L–1 (blue), and 10 g L–1(purple). (A) Specific energy consumption of RO with a water permeability and feed-side mass transfer coefficient of 5 L m–2 h–1 bar–1 and 150 L m–2 h–1, respectively. The module-scale water recovery is optimized for each feed salinity and salt removal to minimize SEC. Energy recovery is not applied. (B) Specific energy consumption of ED with the modeling parameters specified in Table S1. The cell pair voltage is tuned to achieve varying degrees of salt removal. Effluent from the brine channels is recycled at a recycle ratio of 3.0 to achieve a water recovery of 80%.

Specifically, for feed salinities of 1 and 3 g L–1, ED has lower energy consumption than RO throughout the entire range of salt removals (20% to 90%), though the disparity between the SEC decreases as the salt removal and feed salinity are increased. For example, for a 1 g L–1 feed and 30% salt removal, ED utilizes 0.013 kWh m–3 while RO requires more than 3-fold the energy (0.042 kWh m–3). When salt removal is increased to 80%, however, this factor decreases to nearly 2-fold. Similarly, the energy consumption of RO and ED converges when assessing a 3 g L–1 feed salinity. For a salt removal of 80%, for instance, the SEC of ED is 0.29 kWh m–3, while that of RO is only 24% higher at 0.36 kWh m–3.
Though ED requires less energy for the treatment of relatively low feed salinities, RO becomes the energetically superior technology as feed salinity is increased, particularly at high salt removals. When the feed salinity is increased to 5 g L–1, for example, ED maintains lower SEC than RO up to 80% salt removal. However, further increasing the feed salinity to 10 g L–1 shifts this transition point to only 65% salt removal. Notably, for high feed salinities and salt removals, the magnitude of the SEC considerably grows, thereby also making the difference in the performance of RO and ED more substantial. For instance, in the case of 90% salt removal of a 10 g L–1 feed, the SEC of ED is 2.99 kWh m–3 while that of RO is only 1.51 kWh m–3, a sizable difference of 1.48 kWh m–3. In contrast, for a 3 g L–1 feed, the maximum difference in the SEC of ED and RO is only 0.093 kWh m–3.
The superior performance of RO at higher feed salinities and salt removals can readily be discerned by noting the relatively linear trend of its SEC curves with respect to those of ED. The near linear relation observed for the RO curves is a result of the bypass system utilized, by which increasing the salt removal generally requires the direction of more feedwater through the RO module. By contrast, the effects of increasing the salt removal in ED are more complex, leading to deviation from linearity. Particularly, as greater extents of desalination are achieved, the diluate solution becomes depleted of ions, thereby increasing the solution resistance and the magnitude of the associated potential drop. Greater salt removal (for a fixed productivity) also necessitates an increase of the counterion flux, leading to sharper concentration gradients at the membrane–solution interface and, thus, intensified driving force for the passage of co-ions. Additionally, higher degrees of salt removal generate more severe transmembrane concentration gradients, increasing the rate of back-diffusion of counterions from the concentrate channel to the diluate channel. (17) The compromised selectivity of the ion-exchange membranes and the larger rate of back-diffusion that result from an increase in salt removal ultimately deteriorate the current efficiency of the ED process, thus requiring more energy consumption. Notably, these detrimental effects are exacerbated at higher feed salinities, as is apparent by the increasingly exponential behavior of the SEC curves in Figure 3B. For example, for a 3 g L–1 feed, increasing the salt removal from 30% to 80% results in a decline of current efficiency from 88% to 76%. However, for a 10 g L–1 feed, the current efficiencies for 30% and 80% salt removal are reduced to 70% and 49%, respectively.

Effect of Varying Water Recovery

Maximizing water recovery, the volume of product water generated per volume of feedwater, is a primary objective of desalination processes. However, with an increase in water recovery corresponding to a greater degree of water–salt separation, additional energy input is typically required. In RO, increasing the system-scale water recovery necessitates that either more water be directed through the RO module or the module-scale recovery be increased, both of which increase SEC. We note that for each unique set of operating conditions, the module-scale recovery, and thus the flow rate fraction to the RO module, are numerically optimized to minimize SEC.
In ED, the volumetric flow rates through the diluate and concentrate channels are held equal to minimize transmembrane water flux, thus fixing the module-scale water recovery at 50%. Therefore, the system-scale water recovery is increased by recirculating a fraction of the produced brine to the feed of the concentrate channels, in effect reducing the volume of fresh feedwater required to generate a given volume of product water. However, the implementation of such brine recirculation further increases the salinity in the concentrate channels, intensifying the effects of performance deteriorating phenomena (i.e., co-ion transport and back-diffusion). The resulting decline in current efficiency thus requires a larger SEC for a given salt removal.
As determined in the previous section, the energetic performance of RO and ED begins to notably diverge for a feed salinity of 5 g L–1. Hence, we continue our analysis of a 5 g L–1 feed salinity in Figure 4, where we show the thermodynamic energy efficiency of RO and ED for varying water recovery and salt removal. The productivity, as before, is maintained at 20 L m–2 h–1 throughout. In each color map, the energy efficiency for a given water recovery and salt removal corresponds to the provided scale bars. Contour lines, representing fixed SEC values, are also shown to illustrate the energetic penalty of increasing either salt removal or water recovery.

Figure 4

Figure 4. Energy efficiency color maps of RO and ED for varying salt removal (Rs) and system-scale water recovery (WR). The energy efficiency (η) is defined as the ratio of the thermodynamic minimum specific energy consumption of separation (SECmin) and the actual specific energy consumption (SEC). The values for energy efficiency correspond to the color legends shown on the right side of each plot. The productivity is set to 20 L m–2 h–1 throughout. (A) Energy efficiency of RO for 5 g L–1 feed salinity. The water permeability coefficient and the feed-side mass transfer coefficient are set to 5 L m–2 h–1 bar–1 and 150 L m–2 h–1, respectively. The module-scale water recovery is optimized for each salt removal and system-scale water recovery to maximize the energy efficiency. Energy recovery is not applied. (B) Energy efficiency of ED for 5 g L–1 feed salinity. The salt removal and water recovery are varied by tuning the cell pair voltage and brine effluent recycle ratios, respectively. The black contour lines shown in (A) and (B) represent fixed SEC values.

The scale bars in Figure 4 reveal that RO and ED perform over a similar range of energy efficiency values. Specifically, the energy efficiency of RO over the conditions shown ranges from 9.8% to 27.0%, while ED operates at 5.5% to 29.5% energy efficiency. Nonetheless, the salt removal conditions over which each technology operates with maximum energy efficiency are in stark contrast. Whereas RO attains optimal energy efficiency when salt removal is maximized, ED displays its best performance for moderate salt removals, in agreement with the findings of Figure 3. Notably, both RO and ED display their highest energy efficiency for relatively high-water recoveries, ranging from approximately 70% to 80%. Moving away from these optimal regions leads to gradual decline in energy efficiency, as the rate of growth of the SECmin overtakes that of the SEC. However, ED, unlike RO, maintains high energy efficiency for most conditions outside of its optimal operating zone, showing particularly low energy efficiencies only when both the water recovery and salt removal are increased beyond 90%.

Effect of Varying Productivity

In addition to energy consumption, the productivity, which relates a system’s throughput to its size, is a key performance metric for assessing desalination process performance. Whereas the energy consumption determines operational costs, the productivity is indicative of the required capital, with higher productivities corresponding to more compact and cost-efficient systems. However, in membrane separation processes, increased productivities generally come at the expense of energy efficiency. (48−52) For RO, an increase in productivity requires a faster rate of water transport across the membrane, for which larger hydraulic pressures must be applied. To increase the productivity in ED, in contrast, the volumetric flow rate through each of the spacer channels is increased, decreasing the hydraulic retention time. Thus, for a fixed salt removal, a larger current density must be realized by increasing the applied voltage.
In Figure 5, we show the specific energy consumption, SEC, and energy efficiency, η, of RO and ED for productivities of 20, 40, and 60 L m–2 h–1, covering the typical range of values reported in brackish water desalination. (36,46,50,52−56) For each productivity, we show salt removals of 50% (red), 70% (blue), and 90% (orange), while fixing the feed salinity and water recovery at 5 g L–1 and 80%, respectively. As expected, for a fixed salt removal, the SEC of both RO and ED increase with productivity, with ED showing slightly greater sensitivity to changes in productivity than RO. For example, for a salt removal of 70%, increasing the productivity from 20 to 60 L m–2 h–1 increases the SEC of ED by 33.9%, whereas RO increases by 27.3%. Nonetheless, the magnitude of the SEC of ED remains smaller than RO, except for in the case of 90% salt removal.

Figure 5

Figure 5. (A) Specific energy consumption and (B) corresponding energy efficiency of ED (open bars) and RO with no energy recovery (hatched bars) for varying productivity. The energy efficiency is defined as the ratio of the thermodynamic minimum specific energy consumption of separation and the actual specific energy consumption. Salt removals of 50% (red bars), 70% (blue bars), and 90% (orange bars) are shown. The water recovery and feed salinity are set to 80% and 5 g L–1, respectively. Productivity is defined as the rate of water production normalized by the projected membrane area (eq 6). For ED, the productivity is the flow rate through an individual spacer channel divided by the projected area of an ion-exchange membrane. In RO, the productivity is the product water flow rate divided by the membrane area of the module.

Though increasing the productivity results in a higher SEC by RO and ED, it is important to note that it has no effect on the SECmin. Thereby, the energy efficiency of RO and ED (for a fixed salt removal) declines with increasing productivity, as is shown by Figure 5B. As previously discussed, the energy efficiency of ED is reduced with increasing salt removal, whereas the opposite effect is observed for RO. It is notable, however, that as the productivity is increased, the energetic advantage of ED for lower salt removals begins to narrow. For instance, for 70% salt removal and 20 L m–2 h–1 productivity, the energy efficiency of ED is 5.3% higher than that of RO. However, when the productivity is increased to 60 L m–2 h–1 the difference in the energy efficiency of ED and RO is reduced to 3.1%. Hence, in cases where very high throughputs are required, ED’s superior performance for low salt removal may potentially be overturned by RO.

Identifying the Operational Sweet Spots of RO and ED

Thus far, we have demonstrated the general trends in the energy consumption of RO and ED for varying separation and operating conditions. Specifically, we showed that ED outperforms RO for lower feed salinities and extents of salt removal and that both technologies are capable of performing efficiently at similarly high water recoveries. Nonetheless, the particular conditions under which each process exhibits superior performance have yet to be identified. Hence, in this section we focus on locating the operational sweet spots of RO and ED.
Notably, in the following analysis we investigate only the effects of varying separation conditions (i.e., feed salinity, salt removal, and water recovery), while fixing the productivity at 20 L m–2 h–1, a typical value for brackish water desalination. (29,46,53,55) We choose to exclude productivity as a variable on the basis of our results from the previous section, which showed both technologies demonstrate remarkable stability in their energy efficiency, even when the productivity is increased up to 3-fold. Instead, we now consider the application of energy recovery, a feature exclusive to RO. The energy recovery device (ERD) is assumed to operate at 80% efficiency, representative of the performance of a turbocharger or Pelton wheel type ERD. (47) Though such ERDs operate at considerably lower efficiencies than modern pressure exchangers, they continue to serve as economical alternatives for energy recovery in brackish water RO. (11,42,57)
To make the optimal operating conditions for each technology apparent, in Figure 6 we show color maps of the difference between the energy efficiency of ED and RO. The sign of the values signifies the energetically superior technology, with positive and negative values corresponding to higher energy efficiency in ED and RO, respectively. The magnitude of the values, in contrast, indicates the extent of the advantage the technology provides. White contour lines are provided on each of the color maps to clearly show the conditions at which RO and ED operate with equal energy efficiency. Hence, these transition lines effectively demarcate the operational sweet spots of each technology. In Figure 6A,C,E, energy recovery is not used, whereas in Figure 6B,D,F the application of energy recovery in RO is included. The difference in ED and RO energy efficiency is shown for varying water recovery and salt removal, while the feed salinity is specified on the individual plots. We continue with the evaluation of 3, 5, and 10 g L–1 feed salinities, but no longer consider a feed salinity of 1 g L–1 since ED was shown to considerably outperform RO across the entire range of salt removals. Furthermore, at 1 g L–1 feed salinities and below, the values for SEC are very small, making energy recovery in RO impractical and the comparison of energy consumption among ED and RO of minimal significance.

Figure 6

Figure 6. Energy efficiency phase diagrams of ED and RO for varying salt removal (Rs) and system-scale water recovery (WR). The difference between the ED and RO energy efficiencies is plotted, with values corresponding to the color legends shown to the right of each panel. The white contour lines show the transition between ED and RO being more energy efficient: above the line, RO is more energy efficient, while below the line, ED is more energy efficient. The energy efficiency (η) is defined as the ratio of the thermodynamic minimum specific energy consumption of separation and the actual specific energy consumption. The productivity is set to 20 L m–2 h–1 throughout. The water permeability coefficient and the feed-side mass transfer coefficient for RO are set to 5 L m–2 h–1 bar–1 and 150 L m–2 h–1, respectively. The RO module-scale water recovery is optimized for each salt removal and system-scale water recovery to maximize the energy efficiency. In ED, the salt removal and water recovery are varied by tuning the cell pair voltage and brine effluent recycle ratios, respectively. (A), (C), (E) Energy efficiency phase-diagram of ED-RO with no energy recovery in RO for 3, 5, and 10 g L–1 feed salinity, respectively. (B), (D), (F) Energy efficiency phase-diagram of ED-RO with an RO energy recovery device operating at 80% efficiency (ηERD = 0.8) for 3, 5, and 10 g L–1 feed salinity, respectively.

In Figure 6A,B, color maps for a feed salinity of 3 g L–1 are shown. For the case of no energy recovery in RO (Figure 6A), the transition line extends across the top right corner of the plot, indicative of ED being the dominant technology up to very high salt removals and water recoveries. Specifically, until a water recovery of 73%, the energy efficiency of ED remains higher than RO across all the shown salt removals (up to 95%). As the water recovery is further increased, the transition line trends toward lower salt removal values, demonstrating that when high recoveries and salt removals are needed, RO is the more favorable technology. For example, at 90% water recovery and 95% salt removal, the energy efficiency of RO is 8.4% higher than that of ED. Nonetheless, it is important to note that desalination of a 3 g L–1 feed to drinking water standards requires only 83% salt removal, and that further extending the depth of desalination incurs greater energetic and operational costs. Hence, for the case of the more practical 83% salt removal, the energy efficiency of ED is superior to RO across all water recoveries. When energy recovery is applied to RO (Figure 6B), the transition line shifts considerably downward, providing RO with a much larger region of favorable conditions. However, even with energy recovery, at 83% salt removal ED retains an advantage over RO, albeit marginal.
Upon increasing the feed salinity to 5 g L–1 in Figure 6C,D, the transition lines shift further downward from the 3 g L–1 counterparts, demonstrating the increasing advantage of RO. Nonetheless, when energy recovery is not applied, ED remains competitive with RO. Particularly, in the case of 90% salt removal to achieve a 0.5 g L–1 product water salinity, ED remains energetically superior up to a water recovery of 67%. However, in the case of energy recovery and 90% salt removal, RO overtakes ED with a notable difference in energy efficiency of up to 11%. When the feed salinity is further increased to 10 g L–1, any practical advantage of ED is eliminated, even in the case of no energy recovery in RO. With a 10 g L–1 feed salinity, 95% salt removal is required to achieve drinking water quality. For such high salt removals, RO becomes the superior technology, showing over a 17% difference in energy efficiency in the case of no energy recovery and 80% water recovery. The application of energy recovery further widens the gap in energy efficiency to 22%, highlighting RO’s dominance for high salinity and salt removal applications.

IMPLICATIONS AND OUTLOOK

Click to copy section linkSection link copied!

As mature desalination technologies, ED and RO have both been proven effective and reliable for full-scale brackish water desalination. Nonetheless, RO has established dominance over the brackish water desalination market share in recent years, with little discussion in the literature regarding the limited adoption of electrodialysis. In this study, we presented the first direct and thorough comparison of the energy consumption of RO and ED for brackish water desalination, providing insight and guidance toward the application of each technology. The operational sweet spots for ED and RO were identified, revealing that ED and RO are both capable of operating with similarly high energy efficiencies, though under different conditions. Here, we summarize the determined energetic advantages of each technology and extend our discussion to include additional practical factors that may influence which process is preferred. To make the advantages of each technology apparent, we rank RO and ED on a five-point scale for each of the considered metrics, as shown in Table 1.
Table 1. Rating of ED and RO across Several Key Brackish Water Desalination Performance Metrics
As demonstrated throughout our analysis, ED and RO operate with high energy efficiency in brackish water desalination, exceeding values of 30% for operation within each technology’s respective sweet spot. When compared to efficiencies of emerging alternatives such as capacitive deionization (CDI) or thermal distillation technologies, which typically operate with single-digit thermodynamic energy efficiency, the energy efficiencies of ED and RO are remarkable. (7) However, it is important to note that whereas ED operates most efficiently for low feed salinity separations, for which the magnitude of SEC is small, RO provides energy efficient operation for energy intensive separations of higher feed salinities and salt removals, for which energy consumption is a major concern. Hence, we discern that RO is the overall more energy efficient technology compared to ED, as shown by RO and ED’s five- and four-point ratings in Table 1, respectively.
We note that in this study we considered only single-pass systems for RO and ED. Alternative process designs, particularly for RO, offer the potential to significantly improve energy efficiency, though at greater capital cost. For example, it is well established that significant energy savings can be realized in brackish water RO by implementing brine staging, in which the brine produced by one stage serves as the feed for the following stage. (10,11,58) With such a multistage configuration, the applied hydraulic pressure in each stage can be varied according to the osmotic pressure of the respective feed, bringing the process closer to the ideal case of thermodynamic reversibility. (6) Splitting the RO process into just two stages, for example, has been shown to decrease the specific energy consumption for brackish water desalination by up to 40%. (59) Similarly, closed-circuit RO, in which the hydraulic pressure is temporally varied as the brine is recirculated and mixed with the feed, has been shown to provide up to 45% reduction in energy consumption as compared to single-stage BWRO. (59,60) Though the energy efficiency of RO can effectively be enhanced by such system configurations, the energy consumption of ED is relatively insensitive to similar approaches due to the large current densities (and resulting high rates of entropy generation) characteristic of cost-effective, compact ED stacks. (61,62) For instance, implementing two-stage ED for brackish water desalination was shown to result in less than a 5% decrease in SEC. (63) Accordingly, we further emphasize RO’s superior ranking to ED in terms of energy efficiency.
A key outcome of our analysis is the identification of precise feed salinity ranges that are most suitable for efficient operation of ED and RO. As shown in Figures 1 and 6, ED outperforms RO across all practically relevant water recoveries and salt removals for feed salinities up to 3 g L–1, even when energy recovery is applied in RO. With the treatment of such low feed salinities constituting nearly 30% of brackish water desalination, (8) expanded application of ED is warranted from the perspective of energy efficiency. However, we reiterate that the specific energy consumption for the desalination of low feed salinities is relatively small, often making capital investment a greater consideration than operational costs.
In our analysis, we further showed that the energetic performances of ED and RO begin to notably diverge at a feed salinity of 5 g L–1, with RO gaining a considerable advantage for practically relevant salt removals beyond this point. It is important to note that though RO only outperforms ED for feed salinities of 5 g L–1 and greater, it retains relatively strong performance even in the treatment of lower feed salinity applications. In contrast, the energy efficiency of ED deteriorates rapidly outside its optimal salinity range (i.e., high feed salinities), restricting its practical applicability across a wide range of feed salinities. For example, whereas RO still operates with 13% energy efficiency for 50% salt removal of a 1 g L–1 feed, the energy efficiency of ED for 95% salt removal of a 10 g L–1 feed is less than 7% (with productivity and water recovery in both scenarios fixed at 20 L m–2 h–1 and 80%, respectively). Thus, RO provides greater versatility for applications in which the feed salinity considerably varies.
Though high feed salinities inherently require large degrees of salt removal to achieve drinking water quality, applying such extensive desalination to lower feed salinities unnecessarily expends additional energy. Additionally, due to its corrosivity and low alkalinity, completely deionized water is unsuitable for widespread distribution and consumption, thus requiring remineralization. (13) Desalination technologies that readily facilitate variable degrees of salt removal are therefore essential toward efficient brackish water desalination. In ED, the applied voltage to the stack can easily be adjusted, thereby allowing for precise tunability in the amount of salt removal. (18,26,64) In contrast, the degree of salt removal in RO is difficult to control. Though in this study we utilized an RO bypass system that employs both brine and product water bypass streams to control the extent of salt removal, we note that such a process has yet to be implemented in practice and would require complex control systems to optimize and manipulate the feed flow rate distribution in real-time. Hence, for relatively low feed salinity applications, ED holds a significant advantage over RO by avoiding inefficient overtreatment and remineralization.
Another important distinction between ED and RO worth noting is their ability to remove various types of solutes, specifically charged and uncharged species. Whereas RO relies on a combination of charge and size exclusion to reject both charged and uncharged solutes in the feedwater, ED’s reliance on an electric field driving force inherently limits its applicability to the removal of only charged species. Therefore, RO, unlike ED, is capable of effectively removing many uncharged micropollutants, viruses, and micro-organisms to simultaneously meet multiple water quality objectives. (65) Nonetheless, it is worth noting that RO membranes still struggle with the rejection of smaller uncharged species, such as 1,4-dioxane and N-nitrosodimethylamine (NDMA)—contaminants of growing concern. (66,67) The selection of RO or ED for brackish water desalination should therefore be source-water specific, with complex or polluted feedwaters opting for the use of RO due to its rejection of a broad spectrum of contaminants.
Through our analysis, we revealed that RO and ED are both capable of operating efficiently at high water recoveries (∼80%). We note, however, that our study considered water recoveries only up to 90%, as further increasing the recovery results in severe membrane fouling and scaling. (68) Particularly, brackish groundwaters typically contain high concentrations of scale-forming inorganic species, such as calcium, sulfate, and silica, which precipitate on the membrane surface when concentrated at high water recoveries. (69) Similarly, the propensity for colloidal and organic fouling on membrane surfaces increases as the concentration of foulants is elevated with increasing water recovery. In ED, fouling and scaling of the ion-exchange membranes results in increased electrical resistance and reduced ion selectivity, increasing SEC for a given separation. (17,70) Fouling and scaling in RO also results in increased process energy consumption, as higher hydraulic pressures are required to maintain the necessary water flux. (40)
Though fouling and scaling remain a challenge in both ED and RO, severity and effectiveness of mitigation strategies for each technology vary. In RO, typical approaches to control membrane fouling and scaling include pretreatment, mechanical or chemical cleaning, and membrane surface modification. (71) Such methods are also applicable to ED, in addition to exclusive strategies that exploit the use of the electric field driving force. Specifically, periodic reversal or pulsation of the electric field have been shown to be highly effective methods to prevent fouling and scaling of the ion-exchange membranes. (72) Though pulsed current operation is still under investigation at the lab-scale, current reversal is commonly practiced at the industrial scale and has been shown to minimize the frequency of membrane cleaning and replacement. (73) An additional aspect of ED that reduces the severity of fouling and scaling with respect to RO is the tendency of only charged foulants to be drawn to the ion-exchange membranes via electromigration. Conversely, in RO, the bulk convective permeate flow transports all foulants, including uncharged species, to the membrane surface, upon which permeation drag further exacerbates fouling rates. (74) Thus, neutral foulants that continue to plague RO operation, such as silica, (75,76) are less of a concern in ED.
The primary focus of this study was to distinguish ED and RO in terms of energy consumption. However, an equally important consideration in the determination of a preferred technology is capital cost. Here, we estimate the capital cost of each technology on the basis of the respective membranes utilized. Estimates from industry reveal that the area normalized cost of current ion-exchange membranes (>200 $ m–2) (28) is more than an order of magnitude greater than thin-film composite RO membranes (<10 $ m–2), (77) severely hampering the practical implementation of ED. By fixing the system productivity between RO and ED throughout our analysis, we ensured the technologies were fairly compared with regard to system size. However, it should be noted that the productivity in ED is normalized by the projected area of only one of the ion-exchange membranes in a cell pair, which actually consists of both an anion- and cation-exchange membrane. Thus, for a given productivity, ED requires double the membrane area compared to RO, inherently putting ED at an even larger disadvantage in terms of cost. Furthermore, capital cost becomes increasingly important as the magnitude of the energy consumption, and thus the operational costs, decrease. It is therefore sensible why ED, despite being more energy efficient for low-feed salinity desalination, has been dominated by the more cost-effective RO. Though the extremely high cost of ion-exchange membranes currently hampers the application of ED, economies of scale and renewed interest in the research and development of ion-exchange membranes—largely due to their application in energy conversion and storage technologies—could optimiztically drive down the future cost of ion-exchange membranes to be more economically competitive with RO membranes.

Supporting Information

Click to copy section linkSection link copied!

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsestengg.0c00192.

  • Description of ED process model; parameters utilized for ED process modeling (Table S1); schematic illustration of feed-and-bleed ED operation mode (Figure S1); ED model validation (Figure S2); effect of changing velocity profile on prediction of ED process performance (Figure S3); ED model-predicted ion-exchange membrane resistances as a function of feedwater salinity (Figure S4) (PDF)

Terms & Conditions

Most electronic Supporting Information files are available without a subscription to ACS Web Editions. Such files may be downloaded by article for research use (if there is a public use license linked to the relevant article, that license may permit other uses). Permission may be obtained from ACS for other uses through requests via the RightsLink permission system: http://pubs.acs.org/page/copyright/permissions.html.

Author Information

Click to copy section linkSection link copied!

  • Corresponding Author
    • Menachem Elimelech - Department of Chemical and Environmental Engineering, Yale University, New Haven, Connecticut 06520-8286, United StatesNanosystems Engineering Research Center for Nanotechnology-Enabled Water Treatment (NEWT), Yale University, New Haven, Connecticut 06520-8286, United StatesOrcidhttp://orcid.org/0000-0003-4186-1563 Email: [email protected]
  • Authors
    • Sohum K. Patel - Department of Chemical and Environmental Engineering, Yale University, New Haven, Connecticut 06520-8286, United StatesNanosystems Engineering Research Center for Nanotechnology-Enabled Water Treatment (NEWT), Yale University, New Haven, Connecticut 06520-8286, United StatesOrcidhttp://orcid.org/0000-0001-5228-9449
    • P. Maarten Biesheuvel - European Centre of Excellence for Sustainable Water Technology, Wetsus, Oostergoweg 9, 8911 MA Leeuwarden, The NetherlandsOrcidhttp://orcid.org/0000-0002-5468-559X
  • Notes
    The authors declare no competing financial interest.

Acknowledgments

Click to copy section linkSection link copied!

This work was supported by the NSF Nanosystems Engineering Research Center for Nanotechnology-Enabled Water Treatment (EEC-1449500).

References

Click to copy section linkSection link copied!

This article references 77 other publications.

  1. 1
    Mekonnen, M. M.; Hoekstra, A. Y. Four billion people facing severe water scarcity. Sci. Adv. 2016, 2 (2), e1500323,  DOI: 10.1126/sciadv.1500323
  2. 2
    Greve, P.; Kahil, T.; Mochizuki, J.; Schinko, T.; Satoh, Y.; Burek, P.; Fischer, G.; Tramberend, S.; Burtscher, R.; Langan, S. Global assessment of water challenges under uncertainty in water scarcity projections. Nature Sustainability 2018, 1 (9), 486,  DOI: 10.1038/s41893-018-0134-9
  3. 3
    Shannon, M. A.; Bohn, P. W.; Elimelech, M.; Georgiadis, J. G.; Marinas, B. J.; Mayes, A. M. Science and technology for water purification in the coming decades. In Nanoscience and technology: a collection of reviews from nature Journals; World Scientific: 2010; pp 337346.
  4. 4
    Plappally, A. K.; Lienhard, J. H. Energy requirements for water production, treatment, end use, reclamation, and disposal. Renewable Sustainable Energy Rev. 2012, 16 (7), 48184848,  DOI: 10.1016/j.rser.2012.05.022
  5. 5
    Semiat, R. Energy Issues in Desalination Processes. Environ. Sci. Technol. 2008, 42 (22), 81938201,  DOI: 10.1021/es801330u
  6. 6
    Wang, L.; Violet, C.; DuChanois, R. M.; Elimelech, M. Derivation of the Theoretical Minimum Energy of Separation of Desalination Processes. J. Chem. Educ. 2020, 97, 4361,  DOI: 10.1021/acs.jchemed.0c01194
  7. 7
    Lin, S. H. Energy Efficiency of Desalination: Fundamental Insights from Intuitive Interpretation. Environ. Sci. Technol. 2019, 54 (1), 7684,  DOI: 10.1021/acs.est.9b04788
  8. 8
    Jones, E.; Qadir, M.; van Vliet, M. T. H.; Smakhtin, V.; Kang, S. M. The state of desalination and brine production: A global outlook. Sci. Total Environ. 2019, 657, 13431356,  DOI: 10.1016/j.scitotenv.2018.12.076
  9. 9
    Tanaka, Y. A computer simulation of batch ion exchange membrane electrodialysis for desalination of saline water. Desalination 2009, 249 (3), 10391047,  DOI: 10.1016/j.desal.2009.06.055
  10. 10
    Elimelech, M.; Phillip, W. A. The future of seawater desalination: energy, technology, and the environment. Science 2011, 333 (6043), 712717,  DOI: 10.1126/science.1200488
  11. 11
    Greenlee, L. F.; Lawler, D. F.; Freeman, B. D.; Marrot, B.; Moulin, P. Reverse osmosis desalination: Water sources, technology, and today’s challenges. Water Res. 2009, 43 (9), 23172348,  DOI: 10.1016/j.watres.2009.03.010
  12. 12
    Loeb, S. The Loeb-Sourirajan Membrane - How It Came About. ACS Symp. Ser. 1981, 153, 1,  DOI: 10.1021/bk-1981-0153.ch001
  13. 13
    Kucera, J. Reverse Osmosis: Industrial Processes and Applications; John Wiley & Sons, 2015.
  14. 14
    Porter, C. J.; Werber, J. R.; Zhong, M.; Wilson, C. J.; Elimelech, M. Pathways and Challenges for Biomimetic Desalination Membranes with Sub-Nanometer Channels. ACS Nano 2020, 14 (9), 1089410916,  DOI: 10.1021/acsnano.0c05753
  15. 15
    Patel, S. K.; Ritt, C. L.; Deshmukh, A.; Wang, Z. X.; Qin, M. H.; Epsztein, R.; Elimelech, M. The relative insignificance of advanced materials in enhancing the energy efficiency of desalination technologies. Energy Environ. Sci. 2020, 13 (6), 16941710,  DOI: 10.1039/D0EE00341G
  16. 16
    Werber, J. R.; Deshmukh, A.; Elimelech, M. The Critical Need for Increased Selectivity, Not Increased Water Permeability, for Desalination Membranes. Environ. Sci. Technol. Lett. 2016, 3 (4), 112120,  DOI: 10.1021/acs.estlett.6b00050
  17. 17
    Strathmann, H. Ion-exchange membrane separation processes; Elsevier, 2004; Vol. 9.
  18. 18
    Strathmann, H. Electrodialysis, a mature technology with a multitude of new applications. Desalination 2010, 264 (3), 268288,  DOI: 10.1016/j.desal.2010.04.069
  19. 19
    Campione, A.; Gurreri, L.; Ciofalo, M.; Micale, G.; Tamburini, A.; Cipollina, A. Electrodialysis for water desalination: A critical assessment of recent developments on process fundamentals, models and applications. Desalination 2018, 434, 121160,  DOI: 10.1016/j.desal.2017.12.044
  20. 20
    Meyer, K. H.; Straus, W. La perméabilité des membranes VI. Sur le passage du courant electrique a travers des membranes sélectives. Helv. Chim. Acta 1940, 23 (1), 795800,  DOI: 10.1002/hlca.19400230199
  21. 21
    Gurreri, L.; Tamburini, A.; Cipollina, A.; Micale, G. Electrodialysis Applications in Wastewater Treatment for Environmental Protection and Resources Recovery: A Systematic Review on Progress and Perspectives. Membranes 2020, 10 (7), 146,  DOI: 10.3390/membranes10070146
  22. 22
    Xu, T.; Huang, C. Electrodialysis-based separation technologies: a critical review. AIChE J. 2008, 54 (12), 31473159,  DOI: 10.1002/aic.11643
  23. 23
    Xu, T. W. Ion exchange membranes: State of their development and perspective. J. Membr. Sci. 2005, 263 (1–2), 129,  DOI: 10.1016/j.memsci.2005.05.002
  24. 24
    Patel, S. K.; Qin, M.; Walker, W. S.; Elimelech, M. Energy Efficiency of Electro-Driven Brackish Water Desalination: Electrodialysis Significantly Outperforms Membrane Capacitive Deionization. Environ. Sci. Technol. 2020, 54 (6), 36633677,  DOI: 10.1021/acs.est.9b07482
  25. 25
    McGovern, R. K.; Zubair, S. M.; Lienhard, J. H. The cost effectiveness of electrodialysis for diverse salinity applications. Desalination 2014, 348, 5765,  DOI: 10.1016/j.desal.2014.06.010
  26. 26
    Chehayeb, K. M.; Farhat, D. M.; Nayar, K. G.; Lienhard, J. H. Optimal design and operation of electrodialysis for brackish-water desalination and for high-salinity brine concentration. Desalination 2017, 420, 167182,  DOI: 10.1016/j.desal.2017.07.003
  27. 27
    McGovern, R. K.; Zubair, S. M.; Lienhard, V. J. H. The benefits of hybridising electrodialysis with reverse osmosis. J. Membr. Sci. 2014, 469, 326335,  DOI: 10.1016/j.memsci.2014.06.040
  28. 28
    Nayar, K. G.; Fernandes, J.; McGovern, R. K.; Dominguez, K. P.; McCance, A.; Al-Anzi, B. S.; Lienhard, V. J. H. Cost and energy requirements of hybrid RO and ED brine concentration systems for salt production. Desalination 2019, 456, 97120,  DOI: 10.1016/j.desal.2018.11.018
  29. 29
    Wright, N. C.; Shah, S. R.; Amrose, S. E.; Winter, A. G. A robust model of brackish water electrodialysis desalination with experimental comparison at different size scales. Desalination 2018, 443, 2743,  DOI: 10.1016/j.desal.2018.04.018
  30. 30
    Tedesco, M.; Hamelers, H. V. M.; Biesheuvel, P. M. Nernst-Planck transport theory for (reverse) electrodialysis: II. Effect of water transport through ion-exchange membranes. J. Membr. Sci. 2017, 531, 172182,  DOI: 10.1016/j.memsci.2017.02.031
  31. 31
    Tanaka, Y. A computer simulation of feed and bleed ion exchange membrane electrodialysis for desalination of saline water. Desalination 2010, 254 (1–3), 99107,  DOI: 10.1016/j.desal.2009.12.008
  32. 32
    Sonin, A. A.; Probstein, R. F. A Hydrodynamic Theory of Desalination by Electrodialysis. Desalination 1968, 5 (3), 293,  DOI: 10.1016/S0011-9164(00)80105-8
  33. 33
    Tedesco, M.; Hamelers, H. V. M.; Biesheuvel, P. M. Nernst-Planck transport theory for (reverse) electrodialysis: I. Effect of co-ion transport through the membranes. J. Membr. Sci. 2016, 510, 370381,  DOI: 10.1016/j.memsci.2016.03.012
  34. 34
    Tedesco, M.; Hamelers, H. V. M.; Biesheuvel, P. M. Nernst-Planck transport theory for (reverse) electrodialysis: III. Optimal membrane thickness for enhanced process performance. J. Membr. Sci. 2018, 565, 480487,  DOI: 10.1016/j.memsci.2018.07.090
  35. 35
    Biesheuvel, P. M.; Dykstra, J. E. Physics of Electrochemical Processes ; 2020; http://www.physicsofelectrochemicalprocesses.com/.
  36. 36
    Lee, H. J.; Sarfert, F.; Strathmann, H.; Moon, S. H. Designing of an electrodialysis desalination plant. Desalination 2002, 142 (3), 267286,  DOI: 10.1016/S0011-9164(02)00208-4
  37. 37
    Fidaleo, M.; Moresi, M. Optimal strategy to model the electrodialytic recovery of a strong electrolyte. J. Membr. Sci. 2005, 260 (1–2), 90111,  DOI: 10.1016/j.memsci.2005.01.048
  38. 38
    Wang, L.; Dykstra, J. E.; Lin, S. H. Energy Efficiency of Capacitive Deionization. Environ. Sci. Technol. 2019, 53 (7), 33663378,  DOI: 10.1021/acs.est.8b04858
  39. 39
    Wang, L.; Lin, S. Membrane capacitive deionization with constant current vs constant voltage charging: Which is better?. Environ. Sci. Technol. 2018, 52 (7), 40514060,  DOI: 10.1021/acs.est.7b06064
  40. 40
    Baker, R. W. Membrane technology and applications; John Wiley & Sons, 2012.
  41. 41
    Qin, M.; Deshmukh, A.; Epsztein, R.; Patel, S. K.; Owoseni, O. M.; Walker, W. S.; Elimelech, M. Comparison of energy consumption in desalination by capacitive deionization and reverse osmosis. Desalination 2019, 455, 100114,  DOI: 10.1016/j.desal.2019.01.003
  42. 42
    Drak, A.; Adato, M. Energy recovery consideration in brackish water desalination. Desalination 2014, 339, 3439,  DOI: 10.1016/j.desal.2014.02.008
  43. 43
    Geise, G. M.; Paul, D. R.; Freeman, B. D. Fundamental water and salt transport properties of polymeric materials. Prog. Polym. Sci. 2014, 39 (1), 142,  DOI: 10.1016/j.progpolymsci.2013.07.001
  44. 44
    Wijmans, J. G.; Baker, R. W. The Solution-Diffusion Model - a Review. J. Membr. Sci. 1995, 107 (1–2), 121,  DOI: 10.1016/0376-7388(95)00102-I
  45. 45
    Porter, M. C. Concentration Polarization with Membrane Ultrafiltration. Ind. Eng. Chem. Prod. Res. Dev. 1972, 11 (3), 234,  DOI: 10.1021/i360043a002
  46. 46
    Bartholomew, T. V.; Mauter, M. S. Computational framework for modeling membrane processes without process and solution property simplifications. J. Membr. Sci. 2019, 573, 682693,  DOI: 10.1016/j.memsci.2018.11.067
  47. 47
    Manth, T.; Gabor, M.; Oklejas, E. Minimizing RO energy consumption under variable conditions of operation. Desalination 2003, 157 (1–3), 921,  DOI: 10.1016/S0011-9164(03)00377-1
  48. 48
    Lin, S. H.; Elimelech, M. Kinetics and energetics trade-off in reverse osmosis desalination with different configurations. Desalination 2017, 401, 4252,  DOI: 10.1016/j.desal.2016.09.008
  49. 49
    Wang, L.; Lin, S. Intrinsic tradeoff between kinetic and energetic efficiencies in membrane capacitive deionization. Water Res. 2018, 129, 394401,  DOI: 10.1016/j.watres.2017.11.027
  50. 50
    Hawks, S. A.; Ramachandran, A.; Porada, S.; Campbell, P. G.; Suss, M. E.; Biesheuvel, P. M.; Santiago, J. G.; Stadermann, M. Performance metrics for the objective assessment of capacitive deionization systems. Water Res. 2019, 152, 126137,  DOI: 10.1016/j.watres.2018.10.074
  51. 51
    Chehayeb, K. M.; Lienhard, J. H. On the electrical operation of batch electrodialysis for reduced energy consumption. Environ. Sci-Wat Res. 2019, 5 (6), 11721182,  DOI: 10.1039/C9EW00097F
  52. 52
    Hand, S.; Shang, X.; Guest, J. S.; Smith, K. C.; Cusick, R. D. Global Sensitivity Analysis To Characterize Operational Limits and Prioritize Performance Goals of Capacitive Deionization Technologies. Environ. Sci. Technol. 2019, 53 (7), 37483756,  DOI: 10.1021/acs.est.8b06709
  53. 53
    Wang, L.; Lin, S. H. Theoretical framework for designing a desalination plant based on membrane capacitive deionization. Water Res. 2019, 158, 359369,  DOI: 10.1016/j.watres.2019.03.076
  54. 54
    Hand, S.; Guest, J. S.; Cusick, R. D. Technoeconomic Analysis of Brackish Water Capacitive Deionization: Navigating Tradeoffs between Performance, Lifetime, and Material Costs. Environ. Sci. Technol. 2019, 53 (22), 1335313363,  DOI: 10.1021/acs.est.9b04347
  55. 55
    Lopez, A. M.; Williams, M.; Paiva, M.; Demydov, D.; Do, T. D.; Fairey, J. L.; Lin, Y. J.; Hestekin, J. A. Potential of electrodialytic techniques in brackish desalination and recovery of industrial process water for reuse. Desalination 2017, 409, 108114,  DOI: 10.1016/j.desal.2017.01.010
  56. 56
    Hand, S.; Cusick, R. D. Emerging investigator series: capacitive deionization for selective removal of nitrate and perchlorate: impacts of ion selectivity and operating constraints on treatment costs. Environ. Sci-Wat Res. 2020, 6 (4), 925934,  DOI: 10.1039/C9EW01105F
  57. 57
    Andrews, W. T.; Pergande, W. F.; McTaggart, G. S. Energy performance enhancements of a 950 m(3)/d seawater reverse osmosis unit in Grand Cayman. Desalination 2001, 135 (1–3), 195204,  DOI: 10.1016/S0011-9164(01)00150-3
  58. 58
    Zhu, A. Z.; Christofides, P. D.; Cohen, Y. Effect of Thermodynamic Restriction on Energy Cost Optimization of RO Membrane Water Desalination. Ind. Eng. Chem. Res. 2009, 48 (13), 60106021,  DOI: 10.1021/ie800735q
  59. 59
    Lin, S. H.; Elimelech, M. Staged reverse osmosis operation: Configurations, energy efficiency, and application potential. Desalination 2015, 366, 914,  DOI: 10.1016/j.desal.2015.02.043
  60. 60
    Werber, J. R.; Deshmukh, A.; Elimelech, M. Can batch or semi-batch processes save energy in reverse-osmosis desalination?. Desalination 2017, 402, 109122,  DOI: 10.1016/j.desal.2016.09.028
  61. 61
    Chehayeb, K. M.; Lienhard, J. H. Entropy generation analysis of electrodialysis. Desalination 2017, 413, 184198,  DOI: 10.1016/j.desal.2017.03.001
  62. 62
    Mistry, K. H.; McGovern, R. K.; Thiel, G. P.; Summers, E. K.; Zubair, S. M.; Lienhard, J. H. Entropy Generation Analysis of Desalination Technologies. Entropy 2011, 13 (10), 18291864,  DOI: 10.3390/e13101829
  63. 63
    Chehayeb, K. M.; Nayar, K. G.; Lienhard, J. H. On the merits of using multi-stage and counterflow electrodialysis for reduced energy consumption. Desalination 2018, 439, 116,  DOI: 10.1016/j.desal.2018.03.026
  64. 64
    Parulekar, S. J. Optimal current and voltage trajectories for minimum energy consumption in batch electrodialysis. J. Membr. Sci. 1998, 148 (1), 91103,  DOI: 10.1016/S0376-7388(98)00148-3
  65. 65
    Werber, J. R.; Osuji, C. O.; Elimelech, M. Materials for next-generation desalination and water purification membranes. Nat. Rev. Mater. 2016, 1 (5), 16018,  DOI: 10.1038/natrevmats.2016.18
  66. 66
    Li, W.; Patton, S.; Gleason, J. M.; Mezyk, S. P.; Ishida, K. P.; Liu, H. Z. UV Photolysis of Chloramine and Persulfate for 1,4-Dioxane Removal in Reverse-Osmosis Permeate for Potable Water Reuse. Environ. Sci. Technol. 2018, 52 (11), 64176425,  DOI: 10.1021/acs.est.7b06042
  67. 67
    Plumlee, M. H.; Lopez-Mesas, M.; Heidlberger, A.; Ishida, K. P.; Reinhard, M. N-nitrosodimethylamine (NDMA) removal by reverse osmosis and UV treatment and analysis via LC-MS/MS. Water Res. 2008, 42 (1–2), 347355,  DOI: 10.1016/j.watres.2007.07.022
  68. 68
    Stanford, B. D.; Leising, J. F.; Bond, R. G.; Snyder, S. A. Inland desalination: Current practices, environmental implications, and case studies in Las Vegas, NV. Sustainability Science and Engineering 2010, 2, 327350,  DOI: 10.1016/S1871-2711(09)00211-6
  69. 69
    Brady, P. V.; J, K. R.; M, M. T.; Hightower, M. M. Inland desalination: Challenges and research needs. Journal of Contemporary Water Research & Education 2005, 132, 4651,  DOI: 10.1111/j.1936-704X.2005.mp132001007.x
  70. 70
    Thompson, D. W.; Tremblay, A. Y. Fouling in Steady and Unsteady State Electrodialysis. Desalination 1983, 47 (May), 181188,  DOI: 10.1016/0011-9164(83)87071-4
  71. 71
    Jiang, S. X.; Li, Y. N.; Ladewig, B. P. A review of reverse osmosis membrane fouling and control strategies. Sci. Total Environ. 2017, 595, 567583,  DOI: 10.1016/j.scitotenv.2017.03.235
  72. 72
    Mikhaylin, S.; Bazinet, L. Fouling on ion-exchange membranes: Classification, characterization and strategies of prevention and control. Adv. Colloid Interface Sci. 2016, 229, 3456,  DOI: 10.1016/j.cis.2015.12.006
  73. 73
    Katz, W. E. The electrodialysis reversal (EDR) process. Desalination 1979, 28 (1), 3140,  DOI: 10.1016/S0011-9164(00)88124-2
  74. 74
    Song, L. F.; Elimelech, M. Particle Deposition onto a Permeable Surface in Laminar-Flow. J. Colloid Interface Sci. 1995, 173 (1), 165180,  DOI: 10.1006/jcis.1995.1310
  75. 75
    Koo, T.; Lee, Y. J.; Sheikholeslami, R. Silica fouling and cleaning of reverse osmosis membranes. Desalination 2001, 139 (1–3), 4356,  DOI: 10.1016/S0011-9164(01)00293-4
  76. 76
    Lisitsin, D.; Hasson, D.; Semiat, R. Critical flux detection in a silica scaling RO system. Desalination 2005, 186 (1–3), 311318,  DOI: 10.1016/j.desal.2005.06.007
  77. 77
    Shi, L.; Rossi, R.; Son, M.; Hall, D. M.; Hickner, M. A.; Gorski, C. A.; Logan, B. E. Using reverse osmosis membranes to control ion transport during water electrolysis. Energy Environ. Sci. 2020, 13 (9), 31383148,  DOI: 10.1039/D0EE02173C

Cited By

Click to copy section linkSection link copied!

This article is cited by 56 publications.

  1. Hanqing Fan, Yuxuan Huang, Peter Cruz-Grace, Ngai Yin Yip. Hypersaline Electrodialysis Desalination: Intrinsic Membrane and Module Performance Trade-offs. ACS ES&T Engineering 2024, 4 (9) , 2294-2305. https://doi.org/10.1021/acsestengg.4c00246
  2. Madeline Tuai, Pooria Karami, Aria Khalili, Muhammad Amirul Isalm, Jae-Young Cho, Michael J. Serpe, Mohtada Sadrzadeh. Sustainable Osmotic-Driven Clean Water Recovery through Thermoresponsive Polymer Hydrogels with a Selective Polyamide Coating. ACS Applied Polymer Materials 2024, 6 (16) , 9440-9452. https://doi.org/10.1021/acsapm.4c00958
  3. Sarah M. Dischinger, Daniel J. Miller, David A. Vermaas, Ryan S. Kingsbury. Unifying the Conversation: Membrane Separation Performance in Energy, Water, and Industrial Applications. ACS ES&T Engineering 2024, 4 (2) , 277-289. https://doi.org/10.1021/acsestengg.3c00475
  4. Yu-I Lin, Shu-Yuan Pan, Po-Chih Tseng, Chin-Hao Yeh. Electrodialysis-Based Recovery of Water and Chemicals in Recycled Polyethylene Terephthalate Industries: Energy, Environmental, and Economic Performance. ACS Sustainable Chemistry & Engineering 2024, 12 (4) , 1584-1594. https://doi.org/10.1021/acssuschemeng.3c06873
  5. Le Shi, Xuechen Zhou, Rachel F. Taylor, Chenghan Xie, Bin Bian, Derek M. Hall, Ruggero Rossi, Michael A. Hickner, Christopher A. Gorski, Bruce E. Logan. Thin-Film Composite Membranes for Hydrogen Evolution with a Saline Catholyte Water Feed. Environmental Science & Technology 2024, 58 (2) , 1131-1141. https://doi.org/10.1021/acs.est.3c07957
  6. Weifan Liu, Ruoyu Wang, Anthony P. Straub, Shihong Lin. Membrane Design Criteria and Practical Viability of Pressure-Driven Distillation. Environmental Science & Technology 2023, 57 (5) , 2129-2137. https://doi.org/10.1021/acs.est.2c07765
  7. Sabyasachi Das, Margaret G. O’Connell, Hui Xu, Roy Bernstein, Jae-Hong Kim, Kirti Sankhala, Tamar Segal-Peretz, Rahul Shevate, Wei Zhang, Xuechen Zhou, Seth B. Darling, Jennifer B. Dunn. Assessing Advances in Anti-fouling Membranes to Improve Process Economics and Sustainability of Water Treatment. ACS ES&T Engineering 2022, 2 (11) , 2159-2173. https://doi.org/10.1021/acsestengg.2c00184
  8. Yoav Green. Electrical Conductance of Charged Nanopores. ACS Omega 2022, 7 (41) , 36150-36156. https://doi.org/10.1021/acsomega.2c02266
  9. Soumodip Sur, Ravikumar Thimmappa, Zahid Manzoor Bhat, Neethu Christudas Dargily, Sanchayita Mukhopadhyay, Xi Liu, Pingwei Cai, Zhenhai Wen, Musthafa Ottakam Thotiyl. Hybrid Alkali–Salt–Acid Electrochemical Device for Electricity-Efficient Desalination and H2 Generation. ACS Sustainable Chemistry & Engineering 2022, 10 (33) , 10781-10788. https://doi.org/10.1021/acssuschemeng.2c01497
  10. Gowri Mohandass, Weikun Chen, Sitaraman Krishnan, Taeyoung Kim. Asymmetric and Symmetric Redox Flow Batteries for Energy-Efficient, High-Recovery Water Desalination. Environmental Science & Technology 2022, 56 (7) , 4477-4488. https://doi.org/10.1021/acs.est.1c08609
  11. Zhe Yang, Li Long, Chenyue Wu, Chuyang Y. Tang. High Permeance or High Selectivity? Optimization of System-Scale Nanofiltration Performance Constrained by the Upper Bound. ACS ES&T Engineering 2022, 2 (3) , 377-390. https://doi.org/10.1021/acsestengg.1c00237
  12. Gowri Mohandass, Taeyoung Kim, Sitaraman Krishnan. Continuous Solar Desalination of Brackish Water via a Monolithically Integrated Redox Flow Device. ACS ES&T Engineering 2021, 1 (12) , 1678-1687. https://doi.org/10.1021/acsestengg.1c00266
  13. Ramadan Abu-Rjal, Yoav Green. Bipolar Nanochannels: A Systematic Approach to Asymmetric Problems. ACS Applied Materials & Interfaces 2021, 13 (23) , 27622-27634. https://doi.org/10.1021/acsami.1c05643
  14. Peter Ozaveshe Oviroh, Tien-Chien Jen, Jianwei Ren, Lesego M. Mohlala, Robert Warmbier, Sina Karimzadeh. Nanoporous MoS2 Membrane for Water Desalination: A Molecular Dynamics Study. Langmuir 2021, 37 (23) , 7127-7137. https://doi.org/10.1021/acs.langmuir.1c00708
  15. Haoqi Zhao, Zhangxin Wang, Yuanmiaoliang Chen. A theoretical analysis on upgrading desalination plants with low-salt-rejection reverse osmosis. Desalination 2023, 565 , 116827. https://doi.org/10.1016/j.desal.2023.116827
  16. Leila Ghadiri, Leila Naji, Mehran Javanbakht. Highly permselective polyvinylidene fluoride-based cation-exchange membranes containing phosphorated graphene oxide for electrodialysis. Journal of Environmental Chemical Engineering 2023, 11 (5) , 110629. https://doi.org/10.1016/j.jece.2023.110629
  17. Ruoyu Wang, Ravindra Duddu, Shihong Lin. Extended Donnan-Manning theory for selective ion partition and transport in ion exchange membrane. Journal of Membrane Science 2023, 681 , 121782. https://doi.org/10.1016/j.memsci.2023.121782
  18. Iliya Krastev Iliev, Andrey Alexandrovich Chichirov, Antonina Andreevna Filimonova, Natalia Dmitrievna Chichirova, Alexander Vadimovich Pechenkin, Ivan Hristov Beloev. Development of Hybrid Membrane Systems for Highly Mineralized Waste Utilization in the Power Industry. Energies 2023, 16 (17) , 6166. https://doi.org/10.3390/en16176166
  19. Mitra Ebrahimi Gardeshi, Hamed Arab, Alae Benguit, Patrick Drogui. Runoff water loaded with road de‐icing salts: Occurrence, environmental impact and treatment processes. Water and Environment Journal 2023, 2 https://doi.org/10.1111/wej.12897
  20. Chirag G. Patel, Jaichander Swaminathan. Energy savings by implementing a serpentine flow path in electrodialysis. Science Talks 2023, 7 , 100244. https://doi.org/10.1016/j.sctalk.2023.100244
  21. John Sebastian, Yoav Green. Electrical Circuit Modeling of Nanofluidic Systems. Advanced Physics Research 2023, https://doi.org/10.1002/apxr.202300044
  22. Agnes Maria Mani, Ashwani Kumar, Sanhita Chaudhury. Decoding transport selectivity of ions in polymer membranes by In-situ impedance spectroscopy. Separation Science and Technology 2023, , 1-11. https://doi.org/10.1080/01496395.2023.2219377
  23. Suhaib M. Alawad, Ridha Ben Mansour, Fahad A. Al-Sulaiman, Shafiqur Rehman. Renewable energy systems for water desalination applications: A comprehensive review. Energy Conversion and Management 2023, 286 , 117035. https://doi.org/10.1016/j.enconman.2023.117035
  24. Neda Seyedhassantehrani, James W. Palko. Outflow geometry for electrochemical desalination cells. Electrochimica Acta 2023, 449 , 142180. https://doi.org/10.1016/j.electacta.2023.142180
  25. Zitao Wu, Haibo Zhai, Eric J. Grol, Chad M. Able, Nicholas S. Siefert. Treatment of brackish water for fossil power plant cooling. Nature Water 2023, 1 (5) , 471-483. https://doi.org/10.1038/s44221-023-00072-x
  26. Soraya Honarparvar, Rashed Al-Rashed, Amos G. Winter V.. Investigation of pulsed electric field operation as a chemical-free anti-scaling approach for electrodialysis desalination of brackish water. Desalination 2023, 551 , 116386. https://doi.org/10.1016/j.desal.2023.116386
  27. Muhammad Faheem, Rahat Alam, Yao He, Linda Zou. Hybrid pseudocapacitive sodium titanate/rGO and MXene/rGO nanocomposite electrodes in capacitive deionization. Desalination 2023, 551 , 116435. https://doi.org/10.1016/j.desal.2023.116435
  28. Na Xue, Jiaqi Lu, Dungang Gu, Yuhang Lou, Yuan Yuan, Guanghui Li, Shogo Kumagai, Yuko Saito, Toshiaki Yoshioka, Nan Zhang. Carbon footprint analysis and carbon neutrality potential of desalination by electrodialysis for different applications. Water Research 2023, 232 , 119716. https://doi.org/10.1016/j.watres.2023.119716
  29. Anas Azzam, Ghada Samy, Mona A. Hagras, Rasha ElKholy. Geographic information systems-based framework for water–energy–food nexus assessments. Ain Shams Engineering Journal 2023, 36 , 102224. https://doi.org/10.1016/j.asej.2023.102224
  30. Nguyen Thi Thu, Santanu Patra, Antika Pranudta, Trung Thanh Nguyen, Medhat Mohamed El-Moselhy, Surapol Padungthon. Desalination of brackish groundwater using self-regeneration hybrid ion exchange and reverse osmosis system (HSIX-RO). Desalination 2023, 550 , 116378. https://doi.org/10.1016/j.desal.2023.116378
  31. Cassidy Carey, José Carlos Díaz, David Kitto, Carolina Espinoza, Eungjin Ahn, Jovan Kamcev. Interfacial interactions between polymers and selective adsorbents influence ion transport properties of boron scavenging ion-exchange membranes. Journal of Membrane Science 2023, 669 , 121301. https://doi.org/10.1016/j.memsci.2022.121301
  32. Mourin Jarin, Zeou Dou, Haiping Gao, Yongsheng Chen, Xing Xie. Salinity exchange between seawater/brackish water and domestic wastewater through electrodialysis for potable water. Frontiers of Environmental Science & Engineering 2023, 17 (2) https://doi.org/10.1007/s11783-023-1616-1
  33. Zaheen Ullah Khan, Maku Moronshing, Marina Shestakova, Ahmed Al-Othman, Mika Sillanpää, Zhengshuo Zhan, Bingnan Song, Yang Lei. Electro-deionization (EDI) technology for enhanced water treatment and desalination: A review. Desalination 2023, 548 , 116254. https://doi.org/10.1016/j.desal.2022.116254
  34. Viorel Ionescu. Numerical Modeling of a Desalination Process Through the Ion-Exchange Membranes of a Electrodialysis Cell. 2023, 803-814. https://doi.org/10.1007/978-3-031-22375-4_65
  35. Guangzhong Cao, Md Mofasserul Alam, Ajkia Zaman Juthi, Zirui Zhang, Yaoming Wang, Chenxiao Jiang, Tongwen Xu. Electro-desalination: State-of-the-art and prospective. Advanced Membranes 2023, 3 , 100058. https://doi.org/10.1016/j.advmem.2022.100058
  36. Drew Lilley, Ravi Prasher. Ionocaloric refrigeration cycle. Science 2022, 378 (6626) , 1344-1348. https://doi.org/10.1126/science.ade1696
  37. Arianna Baldinelli, Giovanni Cinti, Linda Barelli, Gianni Bidini. Hydrogen production from low-quality water: challenges and perspectives. Journal of Physics: Conference Series 2022, 2385 (1) , 012048. https://doi.org/10.1088/1742-6596/2385/1/012048
  38. Qing-Bai Chen, Yong Xu, Peng-Fei Li, Jin Wang, Lin Dong, Jinli Zhao, Jianyou Wang. An emerging pilot-scale electrodialysis system for desalination of SWNF permeate: Evaluating the role of typical factors. Desalination 2022, 542 , 116064. https://doi.org/10.1016/j.desal.2022.116064
  39. Peng Lin, Tao Yang, Zhengtong Li, Wei Xia, Xiaoxu Xuan, Xun Sun, Saad M. Alshehri, Tansir Ahamad, Yusuke Yamauchi, Xingtao Xu, Yoshio Bando. Ion transport channels in redox flow deionization enable ultra-high desalination performance. Nano Energy 2022, 102 , 107652. https://doi.org/10.1016/j.nanoen.2022.107652
  40. Vineeth Pothanamkandathil, Christopher A. Gorski. Comparing energy demands and longevities of membrane-based capacitive deionization architectures. Environmental Science: Water Research & Technology 2022, 8 (7) , 1489-1496. https://doi.org/10.1039/D2EW00188H
  41. Gowri Mohandass, Sitaraman Krishnan, Taeyoung Kim. Experimental Analysis and Modeling of Closed-Loop Redox Flow Desalination. Journal of The Electrochemical Society 2022, 169 (6) , 063521. https://doi.org/10.1149/1945-7111/ac790b
  42. Chirag G. Patel, Digvijaysinh Barad, Jaichander Swaminathan. Desalination using pressure or electric field? A fundamental comparison of RO and electrodialysis. Desalination 2022, 530 , 115620. https://doi.org/10.1016/j.desal.2022.115620
  43. Indah Prihatiningtyas, Al-Hasan Ammar Hussien Al-Kebsi, Yusak Hartanto, Tsegahun Mekonnen Zewdie, Bart Van der Bruggen. Techno-economic assessment of pervaporation desalination of hypersaline water. Desalination 2022, 527 , 115538. https://doi.org/10.1016/j.desal.2021.115538
  44. Yiming Yin, Sifat Kalam, Joshua L. Livingston, Ronny Minjarez, Jongho Lee, Shihong Lin, Tiezheng Tong. The use of anti-scalants in gypsum scaling mitigation: Comparison with membrane surface modification and efficiency in combined reverse osmosis and membrane distillation. Journal of Membrane Science 2022, 643 , 120077. https://doi.org/10.1016/j.memsci.2021.120077
  45. Rui Gao, Suanny Mosquera-Romero, Eleftheria Ntagia, Xiaofei Wang, Korneel Rabaey, Luiza Bonin. Review—Electrochemical Separation of Organic and Inorganic Contaminants in Wastewater. Journal of The Electrochemical Society 2022, 169 (3) , 033505. https://doi.org/10.1149/1945-7111/ac51f9
  46. Tsai-Hsuan Chen, Kuang-Hao Yeh, Cheng-Fang Lin, Mengshan Lee, Chia-Hung Hou. Technological and economic perspectives of membrane capacitive deionization (MCDI) systems in high-tech industries: From tap water purification to wastewater reclamation for water sustainability. Resources, Conservation and Recycling 2022, 177 , 106012. https://doi.org/10.1016/j.resconrec.2021.106012
  47. Marcello Di Martino, Styliani Avraamidou, Efstratios N. Pistikopoulos. A Neural Network Based Superstructure Optimization Approach to Reverse Osmosis Desalination Plants. Membranes 2022, 12 (2) , 199. https://doi.org/10.3390/membranes12020199
  48. Nur Syahirah Suhalim, Norherdawati Kasim, Ebrahim Mahmoudi, Intan Juliana Shamsudin, Abdul Wahab Mohammad, Fathiah Mohamed Zuki, Nor Laili-Azua Jamari. Rejection Mechanism of Ionic Solute Removal by Nanofiltration Membranes: An Overview. Nanomaterials 2022, 12 (3) , 437. https://doi.org/10.3390/nano12030437
  49. Tahir Muhammad, Lei Li, Yang Xiao, Yunpeng Zhou, Zeyuan Liu, Xin He, Nazir Ahmed Bazai, Yunkai Li. Multiple fouling dynamics, interactions and synergistic effects in brackish surface water distribution systems. Chemosphere 2022, 287 , 132268. https://doi.org/10.1016/j.chemosphere.2021.132268
  50. Yoav Green. Effects of surface-charge regulation, convection, and slip lengths on the electrical conductance of charged nanopores. Physical Review Fluids 2022, 7 (1) https://doi.org/10.1103/PhysRevFluids.7.013702
  51. Jacob Easley. Feasibility and Design of Solar-Powered Electrodialysis Reversal Desalination Systems for Agricultural Applications in The Middle East and North Africa. SSRN Electronic Journal 2022, IFPRI https://doi.org/10.2139/ssrn.4184456
  52. Ahmed Abarkan, Nabil Grimi, Hubert Métayer, Tarik Sqalli Houssaïni, Cécile Legallais. Electrodialysis Can Lower the Environmental Impact of Hemodialysis. Membranes 2022, 12 (1) , 45. https://doi.org/10.3390/membranes12010045
  53. Ihab Shigidi, Ali E. Anqi, Abubakar Elkhaleefa, Azam Mohamed, Ismat H. Ali, Eid I. Brima. Temperature Impact on Reverse Osmosis Permeate Flux in the Remediation of Hexavalent Chromium. Water 2022, 14 (1) , 44. https://doi.org/10.3390/w14010044
  54. A.V. Kovalenko, M. Wessling, V.V. Nikonenko, S.A. Mareev, I.A. Moroz, E. Evdochenko, M. Kh. Urtenov. Space-Charge breakdown phenomenon and spatio-temporal ion concentration and fluid flow patterns in overlimiting current electrodialysis. Journal of Membrane Science 2021, 636 , 119583. https://doi.org/10.1016/j.memsci.2021.119583
  55. Moon Son, Sanghun Park, Namhyeok Kim, Anne Therese Angeles, Youngsik Kim, Kyung Hwa Cho. Simultaneous Energy Storage and Seawater Desalination using Rechargeable Seawater Battery: Feasibility and Future Directions. Advanced Science 2021, 8 (18) https://doi.org/10.1002/advs.202101289
  56. Eric Lichtfouse, Nadia Morin-Crini, Corina Bradu, Youssef-Amine Boussouga, Mehran Aliaskari, Andrea Iris Schäfer, Soumya Das, Lee D. Wilson, Michihiko Ike, Daisuke Inoue, Masashi Kuroda, Sébastien Déon, Patrick Fievet, Grégorio Crini. Technologies to Remove Selenium from Water and Wastewater. 2021, 207-304. https://doi.org/10.1007/978-3-030-69090-8_3
Open PDF

ACS ES&T Engineering

Cite this: ACS EST Engg. 2021, 1, 5, 851–864
Click to copy citationCitation copied!
https://doi.org/10.1021/acsestengg.0c00192
Published January 22, 2021

Copyright © 2021 The Authors. Published by American Chemical Society. This publication is licensed under CC-BY-NC-ND.

Article Views

16k

Altmetric

-

Citations

Learn about these metrics

Article Views are the COUNTER-compliant sum of full text article downloads since November 2008 (both PDF and HTML) across all institutions and individuals. These metrics are regularly updated to reflect usage leading up to the last few days.

Citations are the number of other articles citing this article, calculated by Crossref and updated daily. Find more information about Crossref citation counts.

The Altmetric Attention Score is a quantitative measure of the attention that a research article has received online. Clicking on the donut icon will load a page at altmetric.com with additional details about the score and the social media presence for the given article. Find more information on the Altmetric Attention Score and how the score is calculated.

  • Abstract

    Figure 1

    Figure 1. Schematic illustration of the variable water recovery electrodialysis (ED) process. A portion of the multicell pair ED stack is shown. Each cell pair contains an anion-exchange membrane (AEM), cation-exchange membrane (CEM), and spacer channels between the membranes. An external voltage is applied across the ED stack as (an equal flow rate of) saline water passes through each of the flow channels. The generated electric field causes ions in the channels to migrate toward the oppositely charged electrode. The AEM and CEM enable selective permeation of anions and cations, respectively. Due to nonideal membrane selectivity, however, the transport of counterions across the membrane (Jct) is also accompanied by some degree of counterproductive co-ion transport (Jco). The overall selective transport of ions results in alternating product and brine channels. Redox reactions at the surface of the electrodes convert the ionic current into an electrical current (i). For a system-scale water recovery greater than 50%, a fraction of the brine effluent (αR) is recycled, thus reducing the overall volume of brackish feedwater and increasing the system-scale water recovery.

    Figure 2

    Figure 2. Schematic illustration of the variable salt removal reverse osmosis (RO) process. The brackish feedwater is split among three streams (with the following flow rate fractions): RO module feed (αRO), brine bypass (αB), and product bypass (αP). The hydraulic pressure of the feedwater fed to the RO module is increased (by ΔP) via a high-pressure pump, requiring the input of electrical energy. Water permeates through the RO membrane with a flux of Jw, while complete salt rejection is assumed. The permeate and retentate streams from the RO module are mixed with the corresponding bypass streams, generating the overall product and brine streams, respectively. The system-scale salt removal is controlled by tuning the fraction of feedwater sent to the product bypass stream. Alternatively, the brine bypass stream serves to minimize the flow to the RO module, and thus the energy consumption. An energy recovery device (ERD) is applied to the retentate stream exiting the RO module to recover and reuse a significant portion of the mechanical energy provided by the high-pressure pump. However, an ERD is only applied to certain cases throughout the study, as its use in brackish water RO is circumstantial.

    Figure 3

    Figure 3. Specific energy consumption (SEC) for RO and ED as a function of salt removal (Rs). The water recovery and productivity are fixed at 80% and 20 L m–2 h–1, respectively. The various colored lines show different feed salinities: 1 g L–1 (orange), 3 g L–1 (red), 5 g L–1 (blue), and 10 g L–1(purple). (A) Specific energy consumption of RO with a water permeability and feed-side mass transfer coefficient of 5 L m–2 h–1 bar–1 and 150 L m–2 h–1, respectively. The module-scale water recovery is optimized for each feed salinity and salt removal to minimize SEC. Energy recovery is not applied. (B) Specific energy consumption of ED with the modeling parameters specified in Table S1. The cell pair voltage is tuned to achieve varying degrees of salt removal. Effluent from the brine channels is recycled at a recycle ratio of 3.0 to achieve a water recovery of 80%.

    Figure 4

    Figure 4. Energy efficiency color maps of RO and ED for varying salt removal (Rs) and system-scale water recovery (WR). The energy efficiency (η) is defined as the ratio of the thermodynamic minimum specific energy consumption of separation (SECmin) and the actual specific energy consumption (SEC). The values for energy efficiency correspond to the color legends shown on the right side of each plot. The productivity is set to 20 L m–2 h–1 throughout. (A) Energy efficiency of RO for 5 g L–1 feed salinity. The water permeability coefficient and the feed-side mass transfer coefficient are set to 5 L m–2 h–1 bar–1 and 150 L m–2 h–1, respectively. The module-scale water recovery is optimized for each salt removal and system-scale water recovery to maximize the energy efficiency. Energy recovery is not applied. (B) Energy efficiency of ED for 5 g L–1 feed salinity. The salt removal and water recovery are varied by tuning the cell pair voltage and brine effluent recycle ratios, respectively. The black contour lines shown in (A) and (B) represent fixed SEC values.

    Figure 5

    Figure 5. (A) Specific energy consumption and (B) corresponding energy efficiency of ED (open bars) and RO with no energy recovery (hatched bars) for varying productivity. The energy efficiency is defined as the ratio of the thermodynamic minimum specific energy consumption of separation and the actual specific energy consumption. Salt removals of 50% (red bars), 70% (blue bars), and 90% (orange bars) are shown. The water recovery and feed salinity are set to 80% and 5 g L–1, respectively. Productivity is defined as the rate of water production normalized by the projected membrane area (eq 6). For ED, the productivity is the flow rate through an individual spacer channel divided by the projected area of an ion-exchange membrane. In RO, the productivity is the product water flow rate divided by the membrane area of the module.

    Figure 6

    Figure 6. Energy efficiency phase diagrams of ED and RO for varying salt removal (Rs) and system-scale water recovery (WR). The difference between the ED and RO energy efficiencies is plotted, with values corresponding to the color legends shown to the right of each panel. The white contour lines show the transition between ED and RO being more energy efficient: above the line, RO is more energy efficient, while below the line, ED is more energy efficient. The energy efficiency (η) is defined as the ratio of the thermodynamic minimum specific energy consumption of separation and the actual specific energy consumption. The productivity is set to 20 L m–2 h–1 throughout. The water permeability coefficient and the feed-side mass transfer coefficient for RO are set to 5 L m–2 h–1 bar–1 and 150 L m–2 h–1, respectively. The RO module-scale water recovery is optimized for each salt removal and system-scale water recovery to maximize the energy efficiency. In ED, the salt removal and water recovery are varied by tuning the cell pair voltage and brine effluent recycle ratios, respectively. (A), (C), (E) Energy efficiency phase-diagram of ED-RO with no energy recovery in RO for 3, 5, and 10 g L–1 feed salinity, respectively. (B), (D), (F) Energy efficiency phase-diagram of ED-RO with an RO energy recovery device operating at 80% efficiency (ηERD = 0.8) for 3, 5, and 10 g L–1 feed salinity, respectively.

  • References


    This article references 77 other publications.

    1. 1
      Mekonnen, M. M.; Hoekstra, A. Y. Four billion people facing severe water scarcity. Sci. Adv. 2016, 2 (2), e1500323,  DOI: 10.1126/sciadv.1500323
    2. 2
      Greve, P.; Kahil, T.; Mochizuki, J.; Schinko, T.; Satoh, Y.; Burek, P.; Fischer, G.; Tramberend, S.; Burtscher, R.; Langan, S. Global assessment of water challenges under uncertainty in water scarcity projections. Nature Sustainability 2018, 1 (9), 486,  DOI: 10.1038/s41893-018-0134-9
    3. 3
      Shannon, M. A.; Bohn, P. W.; Elimelech, M.; Georgiadis, J. G.; Marinas, B. J.; Mayes, A. M. Science and technology for water purification in the coming decades. In Nanoscience and technology: a collection of reviews from nature Journals; World Scientific: 2010; pp 337346.
    4. 4
      Plappally, A. K.; Lienhard, J. H. Energy requirements for water production, treatment, end use, reclamation, and disposal. Renewable Sustainable Energy Rev. 2012, 16 (7), 48184848,  DOI: 10.1016/j.rser.2012.05.022
    5. 5
      Semiat, R. Energy Issues in Desalination Processes. Environ. Sci. Technol. 2008, 42 (22), 81938201,  DOI: 10.1021/es801330u
    6. 6
      Wang, L.; Violet, C.; DuChanois, R. M.; Elimelech, M. Derivation of the Theoretical Minimum Energy of Separation of Desalination Processes. J. Chem. Educ. 2020, 97, 4361,  DOI: 10.1021/acs.jchemed.0c01194
    7. 7
      Lin, S. H. Energy Efficiency of Desalination: Fundamental Insights from Intuitive Interpretation. Environ. Sci. Technol. 2019, 54 (1), 7684,  DOI: 10.1021/acs.est.9b04788
    8. 8
      Jones, E.; Qadir, M.; van Vliet, M. T. H.; Smakhtin, V.; Kang, S. M. The state of desalination and brine production: A global outlook. Sci. Total Environ. 2019, 657, 13431356,  DOI: 10.1016/j.scitotenv.2018.12.076
    9. 9
      Tanaka, Y. A computer simulation of batch ion exchange membrane electrodialysis for desalination of saline water. Desalination 2009, 249 (3), 10391047,  DOI: 10.1016/j.desal.2009.06.055
    10. 10
      Elimelech, M.; Phillip, W. A. The future of seawater desalination: energy, technology, and the environment. Science 2011, 333 (6043), 712717,  DOI: 10.1126/science.1200488
    11. 11
      Greenlee, L. F.; Lawler, D. F.; Freeman, B. D.; Marrot, B.; Moulin, P. Reverse osmosis desalination: Water sources, technology, and today’s challenges. Water Res. 2009, 43 (9), 23172348,  DOI: 10.1016/j.watres.2009.03.010
    12. 12
      Loeb, S. The Loeb-Sourirajan Membrane - How It Came About. ACS Symp. Ser. 1981, 153, 1,  DOI: 10.1021/bk-1981-0153.ch001
    13. 13
      Kucera, J. Reverse Osmosis: Industrial Processes and Applications; John Wiley & Sons, 2015.
    14. 14
      Porter, C. J.; Werber, J. R.; Zhong, M.; Wilson, C. J.; Elimelech, M. Pathways and Challenges for Biomimetic Desalination Membranes with Sub-Nanometer Channels. ACS Nano 2020, 14 (9), 1089410916,  DOI: 10.1021/acsnano.0c05753
    15. 15
      Patel, S. K.; Ritt, C. L.; Deshmukh, A.; Wang, Z. X.; Qin, M. H.; Epsztein, R.; Elimelech, M. The relative insignificance of advanced materials in enhancing the energy efficiency of desalination technologies. Energy Environ. Sci. 2020, 13 (6), 16941710,  DOI: 10.1039/D0EE00341G
    16. 16
      Werber, J. R.; Deshmukh, A.; Elimelech, M. The Critical Need for Increased Selectivity, Not Increased Water Permeability, for Desalination Membranes. Environ. Sci. Technol. Lett. 2016, 3 (4), 112120,  DOI: 10.1021/acs.estlett.6b00050
    17. 17
      Strathmann, H. Ion-exchange membrane separation processes; Elsevier, 2004; Vol. 9.
    18. 18
      Strathmann, H. Electrodialysis, a mature technology with a multitude of new applications. Desalination 2010, 264 (3), 268288,  DOI: 10.1016/j.desal.2010.04.069
    19. 19
      Campione, A.; Gurreri, L.; Ciofalo, M.; Micale, G.; Tamburini, A.; Cipollina, A. Electrodialysis for water desalination: A critical assessment of recent developments on process fundamentals, models and applications. Desalination 2018, 434, 121160,  DOI: 10.1016/j.desal.2017.12.044
    20. 20
      Meyer, K. H.; Straus, W. La perméabilité des membranes VI. Sur le passage du courant electrique a travers des membranes sélectives. Helv. Chim. Acta 1940, 23 (1), 795800,  DOI: 10.1002/hlca.19400230199
    21. 21
      Gurreri, L.; Tamburini, A.; Cipollina, A.; Micale, G. Electrodialysis Applications in Wastewater Treatment for Environmental Protection and Resources Recovery: A Systematic Review on Progress and Perspectives. Membranes 2020, 10 (7), 146,  DOI: 10.3390/membranes10070146
    22. 22
      Xu, T.; Huang, C. Electrodialysis-based separation technologies: a critical review. AIChE J. 2008, 54 (12), 31473159,  DOI: 10.1002/aic.11643
    23. 23
      Xu, T. W. Ion exchange membranes: State of their development and perspective. J. Membr. Sci. 2005, 263 (1–2), 129,  DOI: 10.1016/j.memsci.2005.05.002
    24. 24
      Patel, S. K.; Qin, M.; Walker, W. S.; Elimelech, M. Energy Efficiency of Electro-Driven Brackish Water Desalination: Electrodialysis Significantly Outperforms Membrane Capacitive Deionization. Environ. Sci. Technol. 2020, 54 (6), 36633677,  DOI: 10.1021/acs.est.9b07482
    25. 25
      McGovern, R. K.; Zubair, S. M.; Lienhard, J. H. The cost effectiveness of electrodialysis for diverse salinity applications. Desalination 2014, 348, 5765,  DOI: 10.1016/j.desal.2014.06.010
    26. 26
      Chehayeb, K. M.; Farhat, D. M.; Nayar, K. G.; Lienhard, J. H. Optimal design and operation of electrodialysis for brackish-water desalination and for high-salinity brine concentration. Desalination 2017, 420, 167182,  DOI: 10.1016/j.desal.2017.07.003
    27. 27
      McGovern, R. K.; Zubair, S. M.; Lienhard, V. J. H. The benefits of hybridising electrodialysis with reverse osmosis. J. Membr. Sci. 2014, 469, 326335,  DOI: 10.1016/j.memsci.2014.06.040
    28. 28
      Nayar, K. G.; Fernandes, J.; McGovern, R. K.; Dominguez, K. P.; McCance, A.; Al-Anzi, B. S.; Lienhard, V. J. H. Cost and energy requirements of hybrid RO and ED brine concentration systems for salt production. Desalination 2019, 456, 97120,  DOI: 10.1016/j.desal.2018.11.018
    29. 29
      Wright, N. C.; Shah, S. R.; Amrose, S. E.; Winter, A. G. A robust model of brackish water electrodialysis desalination with experimental comparison at different size scales. Desalination 2018, 443, 2743,  DOI: 10.1016/j.desal.2018.04.018
    30. 30
      Tedesco, M.; Hamelers, H. V. M.; Biesheuvel, P. M. Nernst-Planck transport theory for (reverse) electrodialysis: II. Effect of water transport through ion-exchange membranes. J. Membr. Sci. 2017, 531, 172182,  DOI: 10.1016/j.memsci.2017.02.031
    31. 31
      Tanaka, Y. A computer simulation of feed and bleed ion exchange membrane electrodialysis for desalination of saline water. Desalination 2010, 254 (1–3), 99107,  DOI: 10.1016/j.desal.2009.12.008
    32. 32
      Sonin, A. A.; Probstein, R. F. A Hydrodynamic Theory of Desalination by Electrodialysis. Desalination 1968, 5 (3), 293,  DOI: 10.1016/S0011-9164(00)80105-8
    33. 33
      Tedesco, M.; Hamelers, H. V. M.; Biesheuvel, P. M. Nernst-Planck transport theory for (reverse) electrodialysis: I. Effect of co-ion transport through the membranes. J. Membr. Sci. 2016, 510, 370381,  DOI: 10.1016/j.memsci.2016.03.012
    34. 34
      Tedesco, M.; Hamelers, H. V. M.; Biesheuvel, P. M. Nernst-Planck transport theory for (reverse) electrodialysis: III. Optimal membrane thickness for enhanced process performance. J. Membr. Sci. 2018, 565, 480487,  DOI: 10.1016/j.memsci.2018.07.090
    35. 35
      Biesheuvel, P. M.; Dykstra, J. E. Physics of Electrochemical Processes ; 2020; http://www.physicsofelectrochemicalprocesses.com/.
    36. 36
      Lee, H. J.; Sarfert, F.; Strathmann, H.; Moon, S. H. Designing of an electrodialysis desalination plant. Desalination 2002, 142 (3), 267286,  DOI: 10.1016/S0011-9164(02)00208-4
    37. 37
      Fidaleo, M.; Moresi, M. Optimal strategy to model the electrodialytic recovery of a strong electrolyte. J. Membr. Sci. 2005, 260 (1–2), 90111,  DOI: 10.1016/j.memsci.2005.01.048
    38. 38
      Wang, L.; Dykstra, J. E.; Lin, S. H. Energy Efficiency of Capacitive Deionization. Environ. Sci. Technol. 2019, 53 (7), 33663378,  DOI: 10.1021/acs.est.8b04858
    39. 39
      Wang, L.; Lin, S. Membrane capacitive deionization with constant current vs constant voltage charging: Which is better?. Environ. Sci. Technol. 2018, 52 (7), 40514060,  DOI: 10.1021/acs.est.7b06064
    40. 40
      Baker, R. W. Membrane technology and applications; John Wiley & Sons, 2012.
    41. 41
      Qin, M.; Deshmukh, A.; Epsztein, R.; Patel, S. K.; Owoseni, O. M.; Walker, W. S.; Elimelech, M. Comparison of energy consumption in desalination by capacitive deionization and reverse osmosis. Desalination 2019, 455, 100114,  DOI: 10.1016/j.desal.2019.01.003
    42. 42
      Drak, A.; Adato, M. Energy recovery consideration in brackish water desalination. Desalination 2014, 339, 3439,  DOI: 10.1016/j.desal.2014.02.008
    43. 43
      Geise, G. M.; Paul, D. R.; Freeman, B. D. Fundamental water and salt transport properties of polymeric materials. Prog. Polym. Sci. 2014, 39 (1), 142,  DOI: 10.1016/j.progpolymsci.2013.07.001
    44. 44
      Wijmans, J. G.; Baker, R. W. The Solution-Diffusion Model - a Review. J. Membr. Sci. 1995, 107 (1–2), 121,  DOI: 10.1016/0376-7388(95)00102-I
    45. 45
      Porter, M. C. Concentration Polarization with Membrane Ultrafiltration. Ind. Eng. Chem. Prod. Res. Dev. 1972, 11 (3), 234,  DOI: 10.1021/i360043a002
    46. 46
      Bartholomew, T. V.; Mauter, M. S. Computational framework for modeling membrane processes without process and solution property simplifications. J. Membr. Sci. 2019, 573, 682693,  DOI: 10.1016/j.memsci.2018.11.067
    47. 47
      Manth, T.; Gabor, M.; Oklejas, E. Minimizing RO energy consumption under variable conditions of operation. Desalination 2003, 157 (1–3), 921,  DOI: 10.1016/S0011-9164(03)00377-1
    48. 48
      Lin, S. H.; Elimelech, M. Kinetics and energetics trade-off in reverse osmosis desalination with different configurations. Desalination 2017, 401, 4252,  DOI: 10.1016/j.desal.2016.09.008
    49. 49
      Wang, L.; Lin, S. Intrinsic tradeoff between kinetic and energetic efficiencies in membrane capacitive deionization. Water Res. 2018, 129, 394401,  DOI: 10.1016/j.watres.2017.11.027
    50. 50
      Hawks, S. A.; Ramachandran, A.; Porada, S.; Campbell, P. G.; Suss, M. E.; Biesheuvel, P. M.; Santiago, J. G.; Stadermann, M. Performance metrics for the objective assessment of capacitive deionization systems. Water Res. 2019, 152, 126137,  DOI: 10.1016/j.watres.2018.10.074
    51. 51
      Chehayeb, K. M.; Lienhard, J. H. On the electrical operation of batch electrodialysis for reduced energy consumption. Environ. Sci-Wat Res. 2019, 5 (6), 11721182,  DOI: 10.1039/C9EW00097F
    52. 52
      Hand, S.; Shang, X.; Guest, J. S.; Smith, K. C.; Cusick, R. D. Global Sensitivity Analysis To Characterize Operational Limits and Prioritize Performance Goals of Capacitive Deionization Technologies. Environ. Sci. Technol. 2019, 53 (7), 37483756,  DOI: 10.1021/acs.est.8b06709
    53. 53
      Wang, L.; Lin, S. H. Theoretical framework for designing a desalination plant based on membrane capacitive deionization. Water Res. 2019, 158, 359369,  DOI: 10.1016/j.watres.2019.03.076
    54. 54
      Hand, S.; Guest, J. S.; Cusick, R. D. Technoeconomic Analysis of Brackish Water Capacitive Deionization: Navigating Tradeoffs between Performance, Lifetime, and Material Costs. Environ. Sci. Technol. 2019, 53 (22), 1335313363,  DOI: 10.1021/acs.est.9b04347
    55. 55
      Lopez, A. M.; Williams, M.; Paiva, M.; Demydov, D.; Do, T. D.; Fairey, J. L.; Lin, Y. J.; Hestekin, J. A. Potential of electrodialytic techniques in brackish desalination and recovery of industrial process water for reuse. Desalination 2017, 409, 108114,  DOI: 10.1016/j.desal.2017.01.010
    56. 56
      Hand, S.; Cusick, R. D. Emerging investigator series: capacitive deionization for selective removal of nitrate and perchlorate: impacts of ion selectivity and operating constraints on treatment costs. Environ. Sci-Wat Res. 2020, 6 (4), 925934,  DOI: 10.1039/C9EW01105F
    57. 57
      Andrews, W. T.; Pergande, W. F.; McTaggart, G. S. Energy performance enhancements of a 950 m(3)/d seawater reverse osmosis unit in Grand Cayman. Desalination 2001, 135 (1–3), 195204,  DOI: 10.1016/S0011-9164(01)00150-3
    58. 58
      Zhu, A. Z.; Christofides, P. D.; Cohen, Y. Effect of Thermodynamic Restriction on Energy Cost Optimization of RO Membrane Water Desalination. Ind. Eng. Chem. Res. 2009, 48 (13), 60106021,  DOI: 10.1021/ie800735q
    59. 59
      Lin, S. H.; Elimelech, M. Staged reverse osmosis operation: Configurations, energy efficiency, and application potential. Desalination 2015, 366, 914,  DOI: 10.1016/j.desal.2015.02.043
    60. 60
      Werber, J. R.; Deshmukh, A.; Elimelech, M. Can batch or semi-batch processes save energy in reverse-osmosis desalination?. Desalination 2017, 402, 109122,  DOI: 10.1016/j.desal.2016.09.028
    61. 61
      Chehayeb, K. M.; Lienhard, J. H. Entropy generation analysis of electrodialysis. Desalination 2017, 413, 184198,  DOI: 10.1016/j.desal.2017.03.001
    62. 62
      Mistry, K. H.; McGovern, R. K.; Thiel, G. P.; Summers, E. K.; Zubair, S. M.; Lienhard, J. H. Entropy Generation Analysis of Desalination Technologies. Entropy 2011, 13 (10), 18291864,  DOI: 10.3390/e13101829
    63. 63
      Chehayeb, K. M.; Nayar, K. G.; Lienhard, J. H. On the merits of using multi-stage and counterflow electrodialysis for reduced energy consumption. Desalination 2018, 439, 116,  DOI: 10.1016/j.desal.2018.03.026
    64. 64
      Parulekar, S. J. Optimal current and voltage trajectories for minimum energy consumption in batch electrodialysis. J. Membr. Sci. 1998, 148 (1), 91103,  DOI: 10.1016/S0376-7388(98)00148-3
    65. 65
      Werber, J. R.; Osuji, C. O.; Elimelech, M. Materials for next-generation desalination and water purification membranes. Nat. Rev. Mater. 2016, 1 (5), 16018,  DOI: 10.1038/natrevmats.2016.18
    66. 66
      Li, W.; Patton, S.; Gleason, J. M.; Mezyk, S. P.; Ishida, K. P.; Liu, H. Z. UV Photolysis of Chloramine and Persulfate for 1,4-Dioxane Removal in Reverse-Osmosis Permeate for Potable Water Reuse. Environ. Sci. Technol. 2018, 52 (11), 64176425,  DOI: 10.1021/acs.est.7b06042
    67. 67
      Plumlee, M. H.; Lopez-Mesas, M.; Heidlberger, A.; Ishida, K. P.; Reinhard, M. N-nitrosodimethylamine (NDMA) removal by reverse osmosis and UV treatment and analysis via LC-MS/MS. Water Res. 2008, 42 (1–2), 347355,  DOI: 10.1016/j.watres.2007.07.022
    68. 68
      Stanford, B. D.; Leising, J. F.; Bond, R. G.; Snyder, S. A. Inland desalination: Current practices, environmental implications, and case studies in Las Vegas, NV. Sustainability Science and Engineering 2010, 2, 327350,  DOI: 10.1016/S1871-2711(09)00211-6
    69. 69
      Brady, P. V.; J, K. R.; M, M. T.; Hightower, M. M. Inland desalination: Challenges and research needs. Journal of Contemporary Water Research & Education 2005, 132, 4651,  DOI: 10.1111/j.1936-704X.2005.mp132001007.x
    70. 70
      Thompson, D. W.; Tremblay, A. Y. Fouling in Steady and Unsteady State Electrodialysis. Desalination 1983, 47 (May), 181188,  DOI: 10.1016/0011-9164(83)87071-4
    71. 71
      Jiang, S. X.; Li, Y. N.; Ladewig, B. P. A review of reverse osmosis membrane fouling and control strategies. Sci. Total Environ. 2017, 595, 567583,  DOI: 10.1016/j.scitotenv.2017.03.235
    72. 72
      Mikhaylin, S.; Bazinet, L. Fouling on ion-exchange membranes: Classification, characterization and strategies of prevention and control. Adv. Colloid Interface Sci. 2016, 229, 3456,  DOI: 10.1016/j.cis.2015.12.006
    73. 73
      Katz, W. E. The electrodialysis reversal (EDR) process. Desalination 1979, 28 (1), 3140,  DOI: 10.1016/S0011-9164(00)88124-2
    74. 74
      Song, L. F.; Elimelech, M. Particle Deposition onto a Permeable Surface in Laminar-Flow. J. Colloid Interface Sci. 1995, 173 (1), 165180,  DOI: 10.1006/jcis.1995.1310
    75. 75
      Koo, T.; Lee, Y. J.; Sheikholeslami, R. Silica fouling and cleaning of reverse osmosis membranes. Desalination 2001, 139 (1–3), 4356,  DOI: 10.1016/S0011-9164(01)00293-4
    76. 76
      Lisitsin, D.; Hasson, D.; Semiat, R. Critical flux detection in a silica scaling RO system. Desalination 2005, 186 (1–3), 311318,  DOI: 10.1016/j.desal.2005.06.007
    77. 77
      Shi, L.; Rossi, R.; Son, M.; Hall, D. M.; Hickner, M. A.; Gorski, C. A.; Logan, B. E. Using reverse osmosis membranes to control ion transport during water electrolysis. Energy Environ. Sci. 2020, 13 (9), 31383148,  DOI: 10.1039/D0EE02173C
  • Supporting Information

    Supporting Information


    The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsestengg.0c00192.

    • Description of ED process model; parameters utilized for ED process modeling (Table S1); schematic illustration of feed-and-bleed ED operation mode (Figure S1); ED model validation (Figure S2); effect of changing velocity profile on prediction of ED process performance (Figure S3); ED model-predicted ion-exchange membrane resistances as a function of feedwater salinity (Figure S4) (PDF)


    Terms & Conditions

    Most electronic Supporting Information files are available without a subscription to ACS Web Editions. Such files may be downloaded by article for research use (if there is a public use license linked to the relevant article, that license may permit other uses). Permission may be obtained from ACS for other uses through requests via the RightsLink permission system: http://pubs.acs.org/page/copyright/permissions.html.