Redox flow desalination system and use thereof

Abstract

Provided is a system for redox flow desalination. In one aspect, the present disclosure relates to a bench- scale 4-channel redox flow desalination system. In one aspect, the present disclosure relates to an investigation of flow rate in symmetric four channel redox flow desalination system, in which higher flow rates were determined to improve ion flux across ion-exchange membranes.

Description

Attorney Docket No.: 206256-0112-00 WOTITLE OF THE INVENTIONRedox Flow Desalination System and Use ThereofCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] The present application claims priority to U.S. Provisional Application No. 63 / 746.011. filed January 16.2025, which is incorporated by reference herein in its entirety.BACKGROUND OF THE INVENTION

[0002] Although water is abundant on Earth, global statistics estimate that up to 66% of the world’s population suffers from water insecurity (Mekonnen, Sci. Adv. 2016, 2, el500323; Alsharhan, (2020). In: Water Resources and Integrated Management of the United Arab Emirates. World Water Resources, 3, 17-61 Springer, Cham.). This challenge is projected to intensify due to the combined effects of climate change and population growth (Jones, Science of The Total Environment 2019, 657, 1343-1356). In regions lacking access to suitable freshwater sources, seawater desalination is critical for providing potable water (Elimelech, Science, 2016, 333, 712-717). However, seawater’s high salinity (~ 35000 ppm) presents a formidable obstacle. Conventional commercial processes, such as multi-stage flash (MSF) and seawater reverse osmosis (S WRO), are frequently limited by high energy demands and operational costs. Additionally, these methods present significant environmental challenges, including the discharge of concentrated brine streams and carbon dioxide and the necessity for intensive chemical pre-treatment to prevent membrane fouling or scaling (Khawaji, Desalination, 2008, 221. 47-69; Qasim. Desalination 2019. 459, 59-104; Al-Karaghouli, Renewable and Sustainable Energy Reviews, 2013, 24. 343-356).

[0003] While these challenges have prompted research into electrochemical methods-such as electrochemical desalination systems like electrodialysis (ED), electrodeionization, and capacitive deionization (Lu, Chemical Engineering Journal 2022, 431, 133917; Knust, ChemElectroChem, 2014, 1, 850-857), achieving low energy consumption, high salt removal rate and efficiency, and continuous desalination of seawater remain a capabilities remains a significant hurdle for such conventional systems (Shi. Environmental Science & Technology Letters, 2018, 5, 692-700).

[0004] Recently, the four-channel redox flow desalination (RFD) system has been developed to address the challenges of electrochemical seawater desalination (Beh. ACS Sustainable Chemistry & Engineering, 2019, 7, 13411-13417; Chen. Journal of The Electrochemical Society, 2020, 167, 083503). This system consists of electrochemical and salt channels partitioned by ion-exchange membranes (lEMs). The operational mechanism is based on the Donnan exclusive effect; desalination of the salt channels is achieved by coupling the electrochemical reactions of soluble redox species in the electrolyte channels with ion transport across lEMs (Donnan. Chemical Reviews, 1924, 1, 73-90; Barragan. Journal of Colloid and Interface Science. 1998, 205. 365-373; Choi, Journal of Colloid and Interface Science, 2001. 238, 188-195). However, such systems often require complex parametric optimization of redox species, conducting salts, feedwater concentrations, and flow rates to determine optimal separation conditions (Kim. Desalination, 2023, 550, 116406).Attorney Docket No.: 206256-0112-00 WO

[0005] Two distinct redox flow desalination (RFD) branches have emerged based on the feedwater concentration, i.e., brackish (< 10000 ppm) and seawater (~ 35000 ppm) (Greenlee, Water Research. 2009. 43, 2317-2348). While RFD systems have demonstrated efficacy for brackish water (Kim, Desalination, 2023, 550, 116406; Mohandass, Environmental Science & Technology, 2022, 56, 4477-4488), the larger market for seawater desalination remains underserved due to significant technical hurdles (Pilat, Desalination. 2001, 139, 385-392). Specifically, reported RFD systems often exhibit performance trade-offs-such as reduced salt removal rates at high redox concentrations (> 50 mM) and high applied voltages (> 0.5V)-typically attributed to mass transport limitations at electrode sites (Pan, Desalination, 2020, 496, 114762; Chen, Chemical Engineering Journal. 2020. 401, 126111; Han, Environmental Science: Water Research & Technology. 2023, 9, 2368-2377; Kim, ACS Energy Letters, 2023, 8, 2097-2105). But the influences of the hydrodynamic conditions on these performance limitations are not fully resolved, since prior art excludes the effects of electrolyte flow rate.

[0006] Accordingly, there is a need in the art for improved systems and methods for water desalination that overcome mass transport limitations and enable efficient seawater processing. The current disclosure provides solutions to these and other needs.SUMMARY OF THE INVENTION

[0007] In one aspect, tire present disclosure relates to A method of water desalination, comprising tire steps of: provideing a redox flow desalination system, said redox flow desalination system comprising: a first chamber comprising a first chamber input and a first chamber output, wherein the first chamber is electrically connected to a first electrode and configured to comprise an electrolyte material, wherein the first electrode undergoes a reversible redox reaction with the electrolyte material; a second chamber comprising a second chamber input and second chamber output, configured to receive water having a first salinity; a third chamber having a third chamber input and third chamber output, configured to receive the saline water; optionally comprising a first reservoir having a first reservoir input and a first reservoir output; wherein the first reservoir input is fluidly connected to the third chamber output; and wherein the first reservoir output is fluidly coimected to the third chamber input: a fourth chamber comprising a fourth chamber input and fourth chamber output, wherein the fourth chamber is electrically coimected a second electrode and configured to comprise die electrolyte material, wherein die second electrode undergoes a reversible redox reacdon widi the electrolyte material; an electrolyte channel having an electrolyte channel input and an electrolyte channel output, wherein the electrolyte chamrel output is fluidly comrected to the first channel input, the first chamrel output is fluidly comrected to dre fourth chamber input, and the fourth chamber output is fluidly comrected to the electrolyte channel input; a first semi-permeable membrane disposed between the first and second chambers and configured to permit ion flow between dre first and second chambers; a second semi-permeable membrane disposed between dre second chamber and third chambers and configured to permit ion flow between the second and third chambers; and a third semi-permeable membrane disposed between the third chamber and the fourth chamber and configured to permit ion flow bet een the third and fourth chambers; adding w ater having a first salinity into the second chamber; adding water having a second sa 1 i iri tv into the third chamber; adding an electrolyte material to the first chamber and the fourth chamber; and applying a voltage across the first electrode and the second electrode; thereby decreasing the salinity of the w ater in the third chamber and increasing the salinity ofAttorney Docket No.: 206256-0112-00 WOthe water in the second chamber; wherein the flow rate of water in the second chamber and in the first chamber is at least 25 mL / min; and wherein the flow rate of electrolyte material in the first chamber and the fourth chamber is at least 25 mL / min.

[0008] In some examples, flow rate of water in tire second chamber and in the first chamber is at least 30 mL / min; and wherein the flow rate of electrolyte material in the first chamber and the fourth chamber is at least 30 mL / min. In some examples, the flow rate of water in the second chamber and in tire first chamber is at least 40 mL / min; and wherein the flow rate of electrolyte material in the first chamber and tire fourth chamber is at least 40 mL / min. In some examples, the flow rate of water in the second chamber and in the first chamber is at least 50 mL / min; and wherein the flow rate of electrolyte material in the first chamber and tire fourth chamber is at least 50 mL / min.

[0009] In one aspect, the present disclosure relates to a system for redox flow desalination, wherein the system comprises: a first reservoir having a first reservoir input and a first reservoir output; a second reservoir having a second reservoir input and a second reservoir output; a third reservoir having a third reservoir input and a third reservoir output; a first channel having a first channel input and a first channel output; a second channel having a second chaimel input and a second channel output; a third channel having a third channel input and a third chamrel output; a fourth channel having a fourth chaimel input and a fourth channel output; a first electrode in electrical contact with the first channel; and a second electrode in electrical contact with the fourth chamrel; wherein the first chamrel is contacted with the second channel via first semi-permeable membrane; tire second channel is contacted with the third chaimel via a second semi-permeable membrane; and the third channel is contacted with the fourth chamrel via third semi-permeable membrane; wherein the first reservoir output is fluidly connected to tire second chamrel input through a first pump, and the first channel output is fluidly connected to tire first reservoir input; the second fluid reservoir output is fluidly connected to the third chamrel output through a second pump, and the third chamrel output is fluidly comrected to the second fluid reservoir input; wherein the third fluid reservoir output is fluidly connected to the first channel input, tire first channel output is fluidly connected the fourth chamrel input through a third pump, and the fourth chamrel output is fluidly comrected to the third fluid reservoir input; wherein the volume of the first reservoir is no more than 500 mL; the volume of tire second reservoir is no more than 500 nrL; and tire volume of the third reservoir is no more than 500 mL.

[0010] In some examples, the volume of the first reservoir is no more than 400 mL; the volume of the second reservoir is no more than 400 mL; and the volume of the third reservoir is no more than 400 mL. In some examples, the volume of the first reservoir is no more than 300 mL; the volume of the second reservoir is no more than 300 mL; and the volume of the third reservoir is no more than 300 mL. In some examples, the volume of the first reservoir is no more than 250 mL; the volume of the second reservoir is no more than 250 mL; and the volume of the third reservoir is no more than 250 mL. In some examples, the volume of the first reservoir is no more than 200 mL; the volume of the second reservoir is no more than 200 mL; and the volume of the third reservoir is no more than 200 mL. In some examples, the volume of the first reservoir is no more than 150 mL; the volume of the second reservoir is no more than 150 mL; and the volume of the third reservoir is no more than 150 mL. In some examples,Attorney Docket No.: 206256-0112-00 WOthe volume of the first reservoir is no more than 100 mL; the volume of the second reservoir is no more than 100 mL; and the volume of the third reservoir is no more than 100 mL.

[0011] In one aspect, the present disclosure relates to a method of redox flow desalination, the method comprising the steps of: providing tire system disclosed herein; adding water having a first salinity into the first reservoir chamber; adding water having a second salinity’ into the second reservoir chamber; adding an electrolyte material to third reservoir; activating the first pump, the second pump, and the third pump; applying a voltage across the first electrode and the second electrode; thereby decreasing the salinity of the water in the second reservoir chamber and increasing the salinity’ of tire water in the first reservoir; wherein tire flow rate of water through the first pump is at least 25 mL / min; the flow of water through the second pump is at least 25 mL / min; and the flow of water through the third pump is at least 25 mL / min.

[0012] In some examples, the flow rate of water through the first pump is at least 30 mL / min; the flow of water through the second pump is at least 30 mL / min; and the flow of water through the third pump is at least 30 mL / min. In some examples, the flow rate of water through the first pump is at least 40 mL / min; the flow of water through the second pump is at least 40 mL / min; and the flow of water through the third pump is at least 40 mL / min. In some examples, the flow rate of water through the first pump is at least 50 mL / min; the flow of water through the second pump is at least 50 mL / min; and the flow of water through the third pump is at least 50 mL / min.BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The following detailed description of embodiments of the invention will be better understood when read in conjunction with the appended drawings. It should be understood that the invention is not limited to the precise arrangements and instrumentalities of the embodiments shown in the drawings.

[0014] FIG. 1 depicts the experimental cycling of a closed-loop symmetric redox flow desalination (RFD) in batch mode. The four channels are labeled CH 1 and CH 4 for the electrolyte channels and CH 2 and CH 3 for the salt channels. Salinity changes are detected by measuring the conductivity of feedwater reservoir.

[0015] FIG. 2 is a schematic of the exploded view of 4-channel RFD system in batch-mode, showing components. CEM and AEM refer to cation-exchange membrane and anion-exchange membrane, respectively.

[0016] FIG. 3 depicts the RFD components and assembled 4-channel RFD cell.

[0017] FIG. 4 provides a series of plots for salt concentration measurements. At left is a calibration curve for converting conductivity measurement to NaCl NaCl concentration in parts-per-million (ppm) for seawater and brackish water (bottom-right). At right is a schematic showing the range of concentration and time values for determining the average salt removal rates (ASRR values), where Co and Ceruirepresent the initial and final concentration points obtained during a voltage-controlled desalination run.

[0018] FIG. 5 depicts the results of the desalination of seawater with bench scale 4-channel RFD in batch-mode. At left is the chronoamperometiy profile measured during -1750 min of operation at 0.5 V. At right is theAttorney Docket No.: 206256-0112-00 WOconcentration profiles of CH 2 (freshwater, bottom line) and CH 3 (brine, top line) measured during -1750 min of operation at 0.5 V.

[0019] FIG. 6 is a plot of the area specific resistance of bench-top RFD during at different feedwater concentration difference of salt channels during desalination.

[0020] FIG. 7 depicts the experimental cycling of a closed-loop symmetric RFD in single-pass mode, where the recycling of salt channels into large reservoirs ensures a continuous influent supply at constant salinity. The four channels are labeled CH 1 and CH 4 for the electrolyte channels and CH 2 and CH 3 for the salt channels. Salinity changes are detected by measuring the conductivity of the effluent of CH 2.

[0021] FIG. 8 depicts the RFD Single-pass mode set-up. At left is a schematic of in-situ module for salt channel analysis (conductivity measurement) in single-pass mode. The probe was placed 1 cm above stirrer. At right is the RFD system in single-pass mode for electrochemical and desalination analyses, where middle channels are recirculated into a 1 L reservoir of salt solution. Magnetic stirrers (700 rpm) are used to improve homogeneous mixing.

[0022] FIG. 9 demonstrates the effects of the salinity of salt channel's effluent on the precision of salt concentration measurements with 2.000 ppm influent feedwater in salt channels for 25 mM FeCN RFD in singlepass mode operated at applied reverse polarity of 0.2.0.4, and 0.6 V and flow rate of 5 mL min-1in all channels.

[0023] FIG. 10 demonstrates the effects of the salinity of salt channel’s effluent on the precision of salt concentration measurements with 35.000 ppm influent feedwater in salt channels for 25 mM FeCN RFD in singlepass mode operated at applied reverse polarity of 0.2, 0.4. and 0.6 V and flow rate of 5 mL min1in all channels.

[0024] FIG. 11 depicts the effects of salt channel's flow rate on the precision of seawater salt concentration measurements (35000 ppm). At left, is the concentration profile of RFD in single-pass mode with 50 mM FeCN dissolved in 35000 ppm NaCl solution circulating at 5 mL min1and salt channel of seawater at flow rate of 1 mL min1(139.9 L nr2h1) and 30-minute OCV and applied voltages from 0.2 to 1.0 V. At right is the ASRR of RFD in single-pass mode with 25 mM FeCN dissolved in 35000 ppm NaCl solution, calculated for 4 cycles at applied voltages of 0.8 V reported as mean ± SD. Salt channel flow rate of 5 mL min'1(699.3 L m:h1) consistently demonstrates lowest standard deviations, when compared to the increasing standard deviation of ASRR observed from 10 to 50 mL min-1(1398.6 - 6993 L nr2h1)

[0025] FIG. 12 depicts the chronoamperometry for an RFD operating with 25 mM FeCN RFD operating at 5mL min-1(699.9 L nr2h1) in the salt channels, flow rates of 5 - 60 mL min1in the electrolyte channels, and applied reverse polarity voltages of 0.2 - 1.0 V.

[0026] FIG. 13 depicts the concentration profile for an RFD operating with 25 mM FeCN RFD operating at 5mL min-1(699.9 L nr2h1) in the salt channels, flow rates of 5 - 60 mL min-1in the electrolyte channels, and applied reverse polarity voltages of 0.2 - 1.0 V.Attorney Docket No.: 206256-0112-00 WO

[0027] FIG. 14 provides a summary of Ereactor versus ASRR for RFD operated in single-pass mode with 25 mM FeCN dissolved in 35.000 ppm NaCl solution. ASRR plots are provided in FIG. 15.

[0028] FIG. 15 provides a series of plots of tire corresponding ASRR (reported as mean GSD), charge efficiency (CE), and Ee.torfor applied voltages of 0.2 V, 0.4 V, 0.6 V, 0.8 V, and 1.0 V.

[0029] FIG. 16 depicts the chronoamperometry (left) and concentration profile (right) for 12.5 mM FeCN concentration for RFD operated using applied voltage of ±0.8V and flow rates of 5 - 50 mL min'1in electrolyte channels.

[0030] FIG. 17 depicts the chronoamperometiy (left) and concentration profile (right) for 25 mM FeCN concentration for RFD operated using applied voltage of ±0.8V and flow rates of 5 - 50 mL min1in electrolyte channels.

[0031] FIG. 18 depicts the chronoamperometry (left) and concentration profile (right) for 50 mM FeCN concentration for RFD operated using applied voltage of ±0.8V and flow rates of 5 - 50 mL min1in electrolyte channels.

[0032] FIG. 19 depicts the effects of flow rates of electrolyte channels on seawater desalination using different FeCN concentrations / The ASRR (reported as mean GSD), CE, andare provided for RFD operated in singlepass mode at an applied voltage of 0.8 V and FeCN concentrations of 12.5 mM (top), 25.0 mM (bottom left), and 50 mM (bottom right).

[0033] FIG. 20 depicts the chronoamperometry (left) and concentration profiles (right) of RFD in single-pass mode with 50 mM FeCN dissolved in 3000 ppm cycled at flow rates of 5 - 30 mL min1in electrolyte channels, while applying reverse polarity voltages of 0.2 to 1.0 V.

[0034] FIG. 21 depicts a series of plots showing brackish water desalination performance of RFD in single-pass mode at varying electrolyte channel flow rates. At top right is a plot of Ereactoivs ASRR. The remaining plots present the ASRR (reported as mean ± SD). charge efficiency, and Ereactor for applied voltages of 0.2 V, 0.4 V 0.6 V, 0.8 V, and 1.0V.

[0035] FIG. 22 is a Nyquist plot of 12.5 mM FeCN for frequency ranges of 100 kHz to 1 mHz.

[0036] FIG. 23 is a Nyquist plot of 50 mM FeCN for frequency ranges of 100 kHz to 1 mHz.

[0037] FIG. 24 is a Bode plot of 12.5 M FeCN for frequency ranges of 1 MHz to 1 mHz.

[0038] FIG. 25 is a Bode plot of 50 mM FeCN for frequency ranges of 1 MHz to 1 mHz.

[0039] FIG. 26 is a Nyquist plot of RFD of 12.5 mM at 5mL min1at die electrolyte and salt channels showing experimental and fitting parameters using equivalent circuit (RC-Circuit). Nyquist plots of EIS taken prior to desalination runs in FIG. 19 for frequency range of 10 kHz to 10 mHz; where series cell resistance R, (1); charge transfer resistance (Rcr); capacitance of double layer (QDL, ) (2); convective transport resistance (Room) and capacitance (QCOnv, n2) (3); heterogeneous interfacial transport resistance (RHT) and capacitance (QHT, n3) (4).Attorney Docket No.: 206256-0112-00 WO

[0040] FIG. 27 is a series of plots showing the resistances resolved from fitting EIS spectra of desalination runs for different FeCN concentrations. The summary of z fit values resistances forFeCN concentration of 12.5 mM (left) and 50 mM (right) FeCN for EIS spectra taken for frequency range of representing 100 kHz to 10 mHz.

[0041] FIG. 28 is a series of plots showing tire effects of different flow rates of the electrolyte and salt channels (electrolyte|salt) on EIS spectra of RFD operating in single-pass mode. At left is a Nyquist plot for frequency of 100 kHz to 1 mHz. At right is a Bode plot for frequencies of 1 MHz to 1 mHz for RFD with 50 mM FeCN dissolved in 35,000 ppm NaCl.

[0042] FIG. 29 is a pair of plots showing RFD seawater desalination at productivity of 139.9 L nr2h1.Chronoamperometry (left) and concentration profiles (right) are presented for RFD in single-pass mode with 50 mM FeCN dissolved in 35000 ppm cycled at flow rates of 30 mL min-1in electrolyte channels and 1 mL min1(139.9 L nr2h1) in salt channels, while applying reverse polarity voltages of 0.2, 0.4, and 0.8 V for 45 min per cycle.

[0043] FIG. 30 is a pair of plots showing the effect of difference in flow rates between electrolyte and salt channels on salt removal. RFD operated in batch mode with 50 mM FeCN dissolved in 35000 ppm and flow rates of 50 mL min1in electrolyte channels. At left is a concentration profile of a desalination run from 35000ppm to 350 ppm for a 100 mL volume of reservoir, using 5 mL min-1or 50 mL miff1flow rates, corresponding to batch productivity' of 8.8 and 12.03 L nr2h1in salt channel. At right is a Bode plot of RFD EIS taken prior to desalination run.

[0044] FIG. 31 presents a series of images of lEMs after 20 days of cycling and desalination runs in single-pass mode. Fouling of Selemion® AMV (AEM) by FeCN is visible as yellow color after experimental runs yvith 25 mM (left) (related to FIG. 14-18) and 12.5 - 50 mM FeCN (center) (related to FIG. 19), along with unfouled Nafion® 212 (CEMs) (right).

[0045] FIG. 32 presents a series of plots demonstrating the effects of uneven flow rates in salt and electrolyte channels on UV-Vis spectra. The RFD was operated in batch-mode using 50 mM FeCN at 30 mL miff1in electrolyte channels and different flow rates in salt charnels. Aliquots taken from CH 2 (top left) and CH 3 (top right) at 5 mL min1(699.3 L m2h '), while CH 2 (bottom left) CH 3 (bottom right) at 30 mL min1(4195.8 L m2h ’). At the end of operation, the estimated FeCN concentrations in salt channels were 30, 50, 18, and 24 pM.

[0046] FIG. 33 is a pair of plots showing the UV-Vis spectra calibration curve for characterizing crossover using various concentrations of FeCN. At left is a plot of Absorbance vs wavelength spectra. At right that plot is converted to absorbance peaks vs. concentration of FeCN. The absorbance peak wavelength was taken at 216 ± 2 mn.

[0047] FIG. 34 provides a schematic for exemplary redox flow desalination device 100.

[0048] FIG. 35 provides a schematic for exemplary redox flow desalination device 200.Attorney Docket No.: 206256-0112-00 WODETAILED DESCRIPTION

[0049] The figures and descriptions provided herein illustrate elements relevant for a clear understanding of the present disclosure, while omitting certain standard elements found in desalination systems for clarity. Those of ordinary skill in the art will recognize that additional elements or steps may be required to implement the disclosed embodiments. Because such elements are well known in the art and do not facilitate a better understanding of the underlying technology, they are not detailed herein. This disclosure encompasses all such variations and modifications known to those skilled in the art.

[0050] Unless defined otherwise, all technical and scientific temrs used herein have the same meaning as commonly understood by one of ordinary’ skill in the art to which this invention belongs.

[0001] As used herein, each of the following terms has the meaning associated with it in this section.

[0052] The articles “a” and “an” are used herein to refer to one or to more than one (i.e. , to at least one) of the grammatical object of the article. By way of example, “an element” means one element or more than one element.

[0053] “About” as used herein when referring to a measurable value such as an amount, a temporal duration, and the like, is meant to encompass variations of ±20%, ±10%, ±5%, ±1%, or ±0.1% from the specified value, as such variations are appropriate to perform the disclosed methods.

[0054] Ranges: throughout this disclosure, various aspects of the disclosed embodiments my be presented in a range format. The description in range format is provided for convenience and brevity and should not be construed as an inflexible limitation on the scope of the disclosure. Accordingly, the description of a range should be considered to have specifically disclosed all the possible subranges as well as individual numerical values within that range. For example, description of a range such as from 1 to 6 should be considered to have specifically disclosed subranges such as from 1 to 3, from 1 to 4, from 1 to 5, from 2 to 4, from 2 to 6. from 3 to 6 etc., as well as individual numbers within that range, for example, 1, 2, 2.7, 3, 4, 5, 5.3, and 6. This applies regardless of the breadth of the range.Desalination System

[0055] In some examples, as shown in FIG. 34, the present invention relates to a desalination system 100 comprising a first chamber 101 (CHI) comprising a first chamber input and first chamber output; a first set of second 102 (CH2) and third 103 (CH3) chambers, wherein the second chamber comprises a second chamber input and a second chamber output, and the third chamber comprises a third chamber input and a third chamber output; and a fourth chamber 104 (CH4) comprising a fourth chamber input and fourth chamber output. In some examples, the first chamber is electrically7connected to an anode 110 and the fourth chamber is electrically connected to electrode 120. In some examples, a first semi-permeable membrane 130 is disposed between the first and second chambers and configured to permit ion flow from the first chamber through the first semi-permeable membrane and into the second chamber. In some examples, a second semi -permeable membrane 140 is disposed between the second and third chambers and configured to permit ion flow from the third chamber through the second semiAttorney Docket No.: 206256-0112-00 WOpermeable membrane and into the second chamber. In some examples, a third semi-permeable membrane 150 is disposed between the third and fourth chambers and configured to pennit ion flow from the third chamber through the third semi-penneable membrane and into the fourth chamber.

[0056] In some examples, system 100 comprises electrolyte channel 160. In some examples, the first electrolyte channel 160 has a first electrolyte channel input and a first electrolyte channel output, wherein the first electroly te channel output is fluidly connected to the first chamber 101 input and wherein the first chamber 101 output is fluidly connected to the fourth chamber (104) input. In some examples, the fourth chamber 104 output is fluidly connected to the electrolyte channel 160 input.

[0057] In some examples, system 100 comprises a feed channel 180. In some examples, the feed channel comprises a first feed channel input and optionally a second feed channel input, a first feed chamrel output and a second feed channel output. In some examples, the first feed channel output is fluidly comrected to the second chamber input. In some examples, the second feed channel output is fluidly connected to the third chamber input. In some examples, the second chamber output is fluidly connected to the first feed channel input. In some examples, the third chamber output is fluidly connected to the second feed charnel input.

[0058] In some examples, system 100 comprises first pump 115. second pump 125, and third pump 135, said first pump, second pump, and third pump independently configured to control flow rate within tire system.

[0059] In some examples, as shown in FIG. 35. the present invention relates to desalination system 200 comprising a first reservoir 210 having a first reservoir input and a first reservoir output; a second reservoir 220 having a second reservoir input and a second reservoir output; and a third reservoir 230 having a third reservoir input and a third reservoir output; In some examples, desalination system 200 further comprises a first electrode 240 in electrical contact with a first channel 260 and a second electrode 250 in electrical contact with a fourth channel 290. wherein the first channel 260 has a first channel input and a first channel output; and wherein the fourth channel 290 has a fourth channel input and a fourth channel output. First channel 260 is contacted with a second channel 270, said second channel also having a second channel input and a second channel output, via first semi-permeable membrane 265. The second channel 270 is contacted with third channel 280, said third channel further comprising a third channel input and a third channel output, via a second semi-permeable membrane 275. The third channel 280 is contacted with fourth channel 290, said fourth channel having a fourth channel input and a third channel output, via third semi -permeable membrane 285.

[0060] The first reservoir 210 output is fluidly connected to tire second channel input; this fluid connection is optionally controlled via first pump 215. The first channel output is fluidly connected to the input of first reservoir 210, thereby creating a semi-closed fluid system, with output only through semi-permeable membrane 265 and semi-penneable membrane 275, and only of selected chemical species designed to pass through said semi-permeable membranes. In one embodiment, first reservoir 210 is capable of storing any quantity of solution. In one embodiment, first reservoir 210 is only a fluid connector between the second charnel output and the second chamrel input. In one example, first reservoir 210 has a volume of no more than 300 mL, no more than 250 mL, no more than 200 mL, no more than 150 mL, or no more than 100 mL.Attorney Docket No.: 206256-0112-00 WO

[0061] The second fluid reservoir 220 output is fluidly connected to the third channel output: this fluid connection is optionally controlled via a second pump 225. The third channel 280 output is fluidly connected to the second fluid reservoir 220 input, thereby forming a semi-closed fluid system, with output only through semi-permeable membrane 275 and semi-permeable membrane 285, and only of selected chemical species designed to pass through said semi-permeable membranes. In one example, second reservoir 220 is capable of storing any quantity of solution. In one example, second reservoir 220 is only a fluid connector between the third channel output and the third channel input. In one example, second reservoir 220 has a volume of no more than 300 mL, no more than 250 mL, no more than 200 mL. no more than 150 mL, or no more than 100 mL.

[0062] The third fluid reservoir 230 output is fluidly connected to the first channel 260 input. The first channel 260 output is fluidly connected the fourth channel 290 input, optionally via a third pump 235. The fourth channel output is fluidly connected to the third fluid reservoir 230 input, thereby forming a semi-closed fluid system, with output only through semi-permeable membranes 265 and 285, and only of selected chemical species designed to pass through said semi-permeable membranes. In one example, third reservoir 230 is capable of storing any quantity of solution. In one example, third reservoir 230 is only a fluid connector between the third channel output and the third channel input. In one example, third reservoir 230 has a volume of no more than 300 mL, no more than 250 mL, no more than 200 mL, no more than 150 mL, or no more than 100 mL.

[0063] In one example, desalination system 200 is coimected to a power source: in such examples, electrode 240 is a cathode and withdraws electrons from the system, and electrode 250 is an anode and supplies electrons to the system.

[0064] In some examples, first semi-permeable membrane 265, which contacts first channel 260 and second channel 270, is a cation-cxchangc membrane (CEM). In some examples, second scmi-pcrmcablc membrane 275, which contacts second channel 270 and third chamiel 280, is an anion-exchange membrane (AEM). In some examples, third semi-permeable membrane 285, which contacts third channel 280 and fourth chamiel 290, is a cation-exchange membrane (CEM).

[0065] Note that throughout the disclosure, “channel” and “chamber” are used interchangeably to describe the four components of the desalination system sandwiched at distal ends by electrodes and separated by semi-permeable membranes.

[0066] In some examples, li e system comprises a semi-permeable membrane. Examples of semi-permeable membranes include but are not limited to, bipolar membranes, selective ion exchange membranes, cation exchange membranes (CEMs), anion exchange membranes (AEMs), reverse osmosis membranes, forward osmosis membranes, nanofiltration membranes, semi-transparent membranes, ultrafiltration membranes, monovalent-selective ion exchange membranes, divalent-selective ion exchange membranes, alkaline anion exchange membranes, proton exchange membranes, solid-state electrolytes, polyelectrolyte multilayer films, metal organic frameworks, and covalent organic frameworks. In some embodiments, the bipolar membrane comprises a cation exchange layer and an anion exchange layer. Further examples, of semi-permeable membranes include, but are notAttorney Docket No.: 206256-0112-00 WOlimited to those under the trade names Selemion® AMV (AEM), Selemion® CMV (CEM), and Nafion® 212 (CEM).

[0067] In some examples, the system comprises an anode and a cathode. In some examples, the system comprises a power supply and / or a potentiostat with leads. In some embodiments, the anode and cathode comprise electrodes made of or coated with metals including, but not limited to zinc, cobalt, copper, magnesium, silver, iron, platinum, graphite, titanium, brass, lead, steel, nickel, ruthenium, iridium, Ag / AgCl, and the like. In some examples, the electrodes comprise porous graphite felt. In some examples, the system comprises non-metal electrodes such as hydrogen electrodes, chlorine electrodes, oxygen electrodes, and the like. In some examples the anode and / or cathode comprises further polymeric supports, blocks, and / or binders. In some examples, the anode and / or cathode comprise a polyvinylidene fluoride (PVDF) in the form of a block, film, membrane, or coating that supports or encapsulates the electrode components.

[0068] In some examples, the system comprises a sensor positioned in fluid communication with any one of the chambers. In some examples, the system comprises multiple sensors. In some examples, the sensor is capable of detecting changes in conductivity, pH, pressure, humidity, gas concentrations, velocity of fluids and particulates, magnetic fields, temperature, concentration of specific salts, ions, and molecules, and combinations thereof. Suitable sensors include but are not limited to. glass electrode pH sensors, ion-sensitive field-effect transistor sensors, calorimetric sensors, optical sensors (absorbance / transmittance), conductivity sensors, refractometers, near-infrared sensors, resistance temperature detectors, thermocouples, thermistors, infrared sensors, multi-parameter probes, wireless sensors, photoionization detectors, flame ionization detectors, and combinations thereof. In some examples, the conductivity sensor comprises a plurality of electrodes configured to enable multiple-point conductivity measurements. The electrodes in the conductivity sensor may comprise any of the above electrode materials.

[0069] In some examples, the system further comprises spacer materials to promote turbulence and mixing within any one of the chambers. Spacer materials include, but are not limited to meshes, magnetic stirrers, diffusors, gaskets, ion-exchange resins, other porous materials, and the like.Method of Desalination

[0070] In some examples, the present invention relates to a method of desalination of water. In some examples, the method comprises the step of providing the desalination system described elsewhere herein. In some examples, the system functions as a hybrid desalination and energy storage device. The system may be cycled between energy storage and desalination modes or operate in a coupled mode wherein desalination occurs concurrently with energy charging or discharging.

[0071] In some examples, the method comprises the steps of: providing a desalination comprising a first chamber comprising a first chamber input and first chamber output, wherein the first chamber is electrically connected to an anode and configured to comprise an electrolyte material and the anode undergoes a reversible redox reaction with the electrolyte material; a second chamber comprising a second chamber input and second chamber output, configured to receive saline water; a third chamber having a third chamber input and third chamber output,Attorney Docket No.: 206256-0112-00 WOconfigured to receive saline water having a first salinity; a fourth chamber comprising a fourth chamber input and fourth chamber output, wherein the fourth chamber is electrically connected to a cathode and configured to comprise the electrolyte material which undergoes a reversible redox reaction with the cathode; wherein the fourth chamber output is fluidly connected with the first chamber input; and wherein the first chamber output is fluidly connected with the fourth chamber input; a first semi-permeable membrane disposed between the first and the second chambers and configured to permit ion flow from the first chamber to the second chamber; a second semi-permeable membrane disposed between the third chamber and the second chamber and configured to permit ion flow from the third chamber to the second chamber; a third semi-permeable membrane disposed between the third chamber and the fourth chamber and configured to permit ion flow from the third chamber to the fourth chamber; transporting water into the second and third chambers; applying a voltage to the anode and the cathode to provide water having a second salinity in the third chamber, wherein the second salinity is lower than the first salinity; and removing the water having die second salinity from die third chamber.

[0072] In some examples, the method comprises the steps of: providing a desalination system comprising a first chamber comprising a first chamber input and first chamber output, wherein the first chamber is electrically connected to an anode and configured to comprise an electrolyte material and the anode undergoes a reversible redox reaction with the electrolyte material; a second chamber comprising a second chamber input and second chamber output, configured to receive saline water having a first salinity; a third chamber having a third chamber input and third chamber output, configured to receive saline water; a fourth chamber comprising a fourth chamber input and fourth chamber output, wherein the fourth chamber is electrically connected to a cathode and configured to comprise the electrolyte material which undergoes a reversible redox reaction with the second electrolyte material; wherein the first chamber output is fluidly connected to the fourth chamber input, and the fourth chamber output is fluidly connected to the first chamber input; a first semi-permeable membrane disposed between the first and the second chambers and configured to permit ion flow between the first and second chambers; a second semi-permeable membrane disposed between the third chamber and the second chamber and configured to permit ion flow between the second and third chambers; a third semi-permeable membrane disposed betw een the third chamber and the fourth chamber and configured to permit ion flow between the third and fourth chambers; transporting saline water into the second and third chambers; applying a voltage to the anode and the cathode to provide water having a second salinity in the second chamber, wherein the second salinity is lower than the first salinity; and removing the water having the second salinity from the second chamber.

[0073] In some examples, the methods described herein comprise providing an electrochemical cell comprising a first chamber and a second chamber, and a semi-permeable membrane disposed between the first chamber and the second chamber. Steps and embodiments related to tire desalination system tircrcby apply to the electrochemical cell.

[0074] Examples of semi-permeable membranes are described elsew here herein. In some examples, the first and third semi-permeable membranes are anionic exchange membranes and the second semi-permeable membrane is a cationic exchange membrane. In some examples, this exemplary configuration of membranes (ACA, Figure ID) is incorporated when a positive voltage is applied to the anode and cathode. In some examples, applying a positiveAttorney Docket No.: 206256-0112-00 WOvoltage permits negative ions from the saline water to flow from the second chamber to first chamber, permits positive ions from the saline water to flow from the second chamber to the third chamber, and permits negative ions from the electrolyte material to flow from the fourth chamber to the third chamber.

[0075] In some examples, the first and third semi-permeable membranes are cationic exchange membranes and the second semi-permeable is an anionic exchange membrane. In some examples, this exemplary' configuration of membranes (CAC, Figure 1C) is incorporated when a negative voltage is applied to the anode and cathode. In some examples, a negative voltage is applied to the anode and cathode. In some examples, applying a negative voltage pemrits positive ions from the saline water to flow from the second chamber to first chamber, permits negative ions from the saline water to flow from the second chamber to tire third chamber, and permits positive ions from the electrolyte material to flow from the fourth chamber to the third chamber.

[0076] In some embodiments, the saline water may comprise any combination of seawater, oil extraction waste, brine, mining influence water, groundwater, brackish water, freshwater, wastewater, fluids rich in lithium, or recirculated water from any one of the channels in the system. In some examples, the saline water has a pH of about 7. In some examples tire pH of the saline water is between 5-9. In some examples, the pH of the saline water is between 0-5. In some examples, the pH of the saline water is between 9-14.

[0077] The saline water may comprise any ty pe of ion. Exemplary ions include, but are not limited to, calcium, hydrogen, hydroxide, aluminum, magnesium, sodium, potassium, lithium, strontium, barium, ammonia, carbonate, sulfate, chloride, cobalt, copper, neodymium, dysprosium, gallium, iridium, terbium, nitrate, boron, silicon dioxide, and iron. In some embodiments, the saline water comprises a salt including, but not limited to, sodium chloride (NaCl), lithium hydroxide (LiOH), lithium chloride (LiCl), potassium chloride (KC1), magnesium chloride (MgCb), magnesium carbonate (MgCO3), magnesium sulfate (MgSO4), calcium chloride (CaCl2), calcium sulfate (CaSO4), calcium carbonate (CaCO3), potassium acetate (Kac) and calcium magnesium acetate (CaMgAc).

[0078] In some examples, the saline water has sodium chloride (NaCl) in a concentration of at least 0.1 mM. at least 1 mM. at least 10 mM, at least 20 mM, at least 30 mM. at least 40 mM, at least 50 mM. at least 100 mM, at least 250 mM, and at least 500 mM. In some examples, the saline water has sodium chloride (NaCl) in a concentration of between 0.1 mM and 1 M, between 1 mM and 500 mM, between 1 mM and 250 mM, between 1 mM and 100. and between ImM and 75 mM. In some examples, the saline water has sodium chloride (NaCl) in a concentration of less than 10 M, less than 5 M, less than 1 M. less than 500 mM, less than 400 mM, less than 300 mM. less than 200 mM, less than 100 mM, less than 75 mM, less than 60 mM, less than 50 mM. and less than 25 mM.

[0079] In some examples, the step of applying a voltage generates an electric field. In some embodiments, the electric field has a fixed direction during operation. In some embodiments, the electric field is a dynamic electric field, wherein tire magnitude and / or direction of the field is tunable based on the size and polarity' of the applied voltage. In some embodiments, tire electric field has a direction which can be reversed which may help dislodge ions in the semi-permeable membranes and / or electrodes.Attorney Docket No.: 206256-0112-00 WO

[0080] In some examples, the step of apply a voltage comprises applying voltage of at least -10 kV. at least -5 kV. at least -1 kV, at least -500 V, at least -250 V, at least -100 V. at least -10 V, at least -8 V. at least -6 V. at least -4 V. at least -2 V, at least -1 V, at least 0.1 V, at least 1 V. at least 2 V, at least 3 V, at least 5 V, at least 10 V, at least 20 V, at least 50 V, at least 100 V. at least 250 V. at least 500 V, at least 1 kV, at least 5 kV, and at least 10 kV. In some examples, the step of apply a voltage comprises applying voltage of less than 10 kV, less than 5 kV, less than 1 kV, less than 500 V, less than 250 V, less than 100 V, less than 50 V, less than 25 V, less than 10 V, and less than 5 V, less than IV, less than OV, less than -1 V, less than -5 V, less than -10 V, less than - 100 V, less than -500 V, less than -1 kV, less than -5 kV, and less than -10 kV. In some examples, In some examples, the step of applying a voltage comprises applying voltage of between -10 kV and 10 kV, between -1 kV and 1 kV, between -500 V and 500 V, between -100 V and 100 V, and betw een -10 V and 10 V.

[0081] In some examples, the system can be operated in batch mode or in a continuous mode in other embodiments. In batch mode, a volume of saline water to be treated is provided (e.g., pushed) in the system. A voltage is applied to the electrodes, and ions are collected in the two electrodes until the salt concentration in the water drops below a set limit. Then the water is removed from the system. In some examples, the treated water is then provided to a second desalination system that utilizes a different type of desalination process such as reverse osmosis, capacitive deionization, or a thermal-based process. In a continuous flow mode, water flows through the system, and the total residence time for a volume of water in each part of the system is sufficient to achieve a desired reduction in salt concentration. In certain embodiments, separate units can be broken up into different stages and / or components with independently controlled electrodes to accommodate decreasing salinity levels at each successive step during a desalination process. This can also accommodate increasing salinity levels at each successive step during a salination process.

[0082] In some examples, the method further comprises introducing and / or circulating the first or second electrolyte material in the first and / or fourth chambers. In some examples a desired flow rate is set for any saline water and / or electrolyte material. In some examples, the flow rate is at least 0.1 mL / min, at least 0.5 mL / min, at least 1 mL / min, at least 3 mL / min, at least 5 mL / min, at least 10 mL / min, at least 20 mL / min, at least 25 mL / min, at least 30 mL / min, at least 35 mL / min, at least 40 mL / min, at least 45 mL / min. at least 50 mL / min, at least 100 mL / min, at least 500 mL / min, at least 1 L / min, and at least 10 L / min. In some examples, the flow rate is less than 100 L / m, less than 50 L / min, less than 25 L / min, less than 5 L / min, less than 1 L / min, less than 500 mL / min, less than 250 mL / min, less than 100 mL / min, and less than 50 mL / min.

[0083] In some examples, the flow rate is configured to maintain the system within an optimal hydrodynamic regime, transitioning from laboratory-scale volumetric flow to industrial-scale linear velocity. While bench-scale embodiments may employ flow rates of at least 0.1 to 50 mL min1(180 - 270 m3h1). industrial -scale implementations — comprising 20 to 60 cells connected hydraulically in parallel — require significantly higher throughput to support current densities ranging from 75 to 150 mA cm'2For example, a megawatt-scale installation may require a total system flow' rate between 3.000 and 4.500 L min1, fn certain embodiments, the flow rate is managed to maintain a linear velocity between 6 cm s1and 12 cm s1. This velocity is critical to minimize theAttorney Docket No.: 206256-0112-00 WOstagnant diffusion boundary layer, thereby maximizing the limiting current density (LCD) and preventing concentration polarization. Furthermore, achieving a target process flux in the range of 15 to 20 L nr2h'1(LMH) allows the system to balance mass transport against parasitic energy consumption and hydraulic pressure drops, which typically reach approximately 120 kPa (1.2 bar) in standard industrial stacks.

[0084] In some examples, the electrolyte material is redox-active. In some examples, the electrolyte material comprises an anolyte and / or catholyte. In some examples, the anolyte and / or catholyte, may comprise any of the following, but is not limited to, in one or more of their oxidation states, as their ions, oxocations, or oxoanions, and / or complexed to ligand(s): titanium(III), titanium(IV), vanadium(II), vanadium(III), vanadium(IV), vanadium(V), chromium(II), chromium(III), chromium(VI), manganese(II), manganese(III), manganese(VI), manganese(VII), iron(II), iron(III), iron (VI), cobalt(II), cobalt(III), nickel(II), copper(I), copper(II), zinc(II), ruthenium(II), ruthenium(III), tin(II), tin(IV), cerium(III), cerium(IV), tungsten(IV), lungsten(V). osmium(II), osmium(III), lead(II), zincate, aluminate, chlorine, chloride, bromine, bromide, tribromide, iodine, iodide, triiodide, polyhalide, halide oxyanion, sulfide, polysulfide, sulfur oxyanion, ferrocyanide, ferricyanide, a quinone derivative, an alloxazine derivative, a flavin derivative, a viologen derivative, a ferrocene derivative, any other metallocene derivative, a nitroxide radical derivative, a N,N-dialkyl-N-oxoammonium derivative, a nitronyl nitroxide radical derivative, and / or polymers incorporating complexed or covalently bound components of any of the aforementioned materials, and combinations thereof. In some embodiments, the electrolyte material comprises sodium ferrocyanide, Na|Fe(CN)s, potassium ferrocyanide, K / |Fe(CN)6, sodium ferricyanide Na3Fe(CN)6, or potassium ferricyanide K3Fe(CN)6. In some embodiments, the electrolyte material comprises a mixture of ferrocyanides and ferricyanides.

[0085] In some examples, the electrolyte material comprises an aqueous solution of an anolyte or catholyte described herein, wherein the anolyte and / or catholyte is in a concentration of at least 0.1 mM. at least 5 mM, at least 10 mM, at least 25 mM, at least 50 mM, at least 100 mM, at least 250 mM, at least 500 mM, at least 750 mM, at least 1 M, and at least 2 M. In some examples, the anolyte and / or catholyte is in a concentration of less than 10 M, less than 5 M, less than 2 M, less than 1 M, less than 750 mM, less than 500 mM, less than 250 mM, less than 100 mM, and less than 50 mM. In some examples, the anolyte or catholyte is in a concentration of betw een 0.1 mM and 3 M, between 1 mM and 1 M, between 1 mM and 750 mM, between 1 mM and 500 mM, and between 1 m and 100 mM.EXPERIMENTAL EXAMPLES

[0086] The invention is further described in detail by reference to the following experimental examples. These examples are provided for purposes of illustration only and are not intended to be limiting unless otherwise specified. Thus, the invention should in no way be construed as being limited to the following examples, but rather, should be construed to encompass any and all variations which become evident as a result of the teaching provided herein.

[0087] Without further description, it is believed that one of ordinary skill in the art can, using the preceding description and the following illustrative examples, make and utilize the present invention and practice the claimedAttorney Docket No.: 206256-0112-00 WOmethods. The following working examples therefore are not to be construed as limiting in any way the remainder of the disclosure.

[0088] Example 1 : Developing a bench-scale 4-channel redox flow desalination system

[0089] The closed-loop, symmetric four-channel RFD cell for batch and continuous desalination is the most frequently demonstrated experimental architecture in the literature due to its ability to continuously dilute and concentrate influent feedwater (brackish water to brine) in two separate channels (Cheng, 2022, Desalination 534, 115783; Lu. 2022 Chem. Eng. J. 431, 133917; Pan, 2020 Desalination 496. 114762; Chen, 2020. Chem. Eng. J. 401, 126111; Kim. 2023. Desalination 550, 116406; Zhang, 2020, Mater. Today Commun. 23, 100921). In a typical batch-mode cell (FIG. 1). the redox couples — equal moles of the oxidized and reduced forms of a redox species (e.g., Fe2+ / Fe3+, V2+ / V3+) — are dissolved in a conducting salt solution (e.g., NaCl) and pumped through the outer channels (CH 1 and 4). where they exchange electrons at the surface of a porous carbon electrode due to an applied electric field across the cell The redox electrolyte is maintained at a 50% state of charge (SOC). such that the mirrored redox reactions at the anode and cathode facilitate continuous ion removal without a net change in the electrolyte's chemical potential. Thus, the generated redox potential difference is equilibrated by the selective transport of counterions across the lEMs — Na+through the CEM and Cl“ through the AEM — due to the Donnan potential (Wang, 2021, Desalination 504, 114964; Porada, 2013, Prog. Mater. Sci. 58. 1388-1442; Barragan, 1998, J. Colloid Interface Sci. 205. 365-373; Choi, 2001, J. Colloid Interface Sci. 238, 188-195) and other potential driving forces, such as hydraulic, temperature (Dai, 2021, Chem. Eng. J. 416, 127716) and diffusion differences across the channels, leading to dilution and concentration in CH 2 and CH 3, respectively (Darling. 2016. J.Electrochem. Soc. 163, A5029-A5040). This makes the symmetric four-channel RFD effective at producing freshwater from feedwater of all salinities, including brine with over 200,000 ppm (Beh, 2019. ACS Sustain. Chem. Eng. 7. 13411-13417). Additionally, it requires a low amount of redox species (< 200 mM) and consumes little energy.

[0090] Building a four-channel RFD system for closed-loop batch and continuous desalination requires the following key active components: redox electrolytes (redox species and conducting salt solutions), lEMs, porous electrodes, and current collectors, as well as gaskets, spacers, tubing, fittings, and electric pumps (FIG. 2). The redox species and lEMs not only play critical roles in this electrochemically driven ion separation but also contribute to over 40% of the capital cost of the RFD (Darling, 2014, Energy Environ. Sci. 7, 3459-3477; Dmello, 2016 J. Power Sources 330, 261-272). Therefore, developing cost-effective RFDs hinges on maintaining the sustained performance of this coupled process, emphasizing compatible, high-performing, and robust components.

[0091] The redox species selection for RFDs: Redox species (couples) are redox-active solutes that undergo reversible oxidation by releasing electrons and reduction by accepting electrons in solution (Chen, Journal of Energy Chemistry, 27, 1304-1325; Kwabi, 2020, Chem. Rev. 120, 6467-6489). This allows for energy storage in this solution, unlike conventional batteries, where energy is stored in solid electrochemical materials. The total energy capacity is directly linked to the amount of dissolved redox species, making them highly scalable.Attorney Docket No.: 206256-0112-00 WO

[0092] Inorganic and organic / organometallic redox species have been primarily developed for large-scale energy storage inRFBs (Bartolozzi. 1989, J. Power Sources 27, 219-234; Kwabi, 2020, Chem. Rev. 120, 6467-6489; Pan, 2015, Molecules 20. 20499-20517). To meet new cost-effective standards of desalination, properties have been proposed for cost-effective redox species: i) high electrochemical stability at or near neutral pH and ambient conditions; ii) fast electron transfer redox kinetics and long cycle life, ideally without proton-electron coupled transfer; iii) nontoxicity and water solubility; and iv) low permeability coefficients. Despite their commercialization potential in RFBs. inorganic redox species such as sodium iodide (Nal), vanadium chloride, zinc chloride, and ferric chloride (FeCty) may present challenges in RFDs due to the high crossover rates (i.e ., high permeability coefficients. cm2s'1) in commercial IEMS (Winsberg, 2017, Angew. Chemie Int. Ed. 56, 686-711; Kim, 2020, Int. J. Heat Mass Transf. 148, 119040). Crossover is undesirable because it leads to electrolyte loss, capacity degradation, and contamination of salt channels, posing risks to water quality, health, and the environment.

[0093] Thus, most RFD researchers have focused on identifying highly tunable organic / organometallic redox species that meet all the criteria for RFD development. Among these, ferri- / ferrocyanide (FeCN). a well-known organometallic compound, serves as the benchmark redox species in RFD systems due to its excellent electrochemical properties, low capacity' fade, chemical stability (Luo, 2019, Joule 3, 149-163), and affordability (~$10 / kg) (Wang. 2022, Mater. Today Energy' 28, 101061).

[0094] Ion exchange membrane selection for RFDs: IEMs remain the preferred separation membranes for high-performing RFDs due to their superior ion conductivity, mechanical strength, chemical stability, and permselectivity (Nagarale, 2006, Adv. Colloid Interface Sci. 119, 97-130; Pintauro, 2015, Polym. Rev. 55, 201-207). Ion transport is directly correlated with water uptake percentage and ion conductivity (the inverse of area-specific resistance) and indirectly related to thickness. Permselectivity — the IBM’s efficiency at blocking species other than counterions — is inversely related to bulk properties and operating conditions that increase ion transport and aging. Inherently, these bulk properties are intrinsically linked to the nanoscale morphology of the IEMs (Le, 2009, J. Memb. Sci. 340, 133— 140; Kusoglu, 2017. Chem. Rev. 117. 987-1104).

[0095] When an IEM is fully hydrated, a characteristic phase-separated morphology forms at tire nanoscale (Kusoglu, 2017, Chem. Rev. 117, 987-1104). This morphology' consists of the hydrophobic region of the hydrocarbon backbone and the hydrophilic region, which includes the ionomer, bound water, and free water, along with other solutes present in the bulk electrolyte (Verbrugge, 1990, J. Electrochem. Soc. 137, 886-893; Kamcev, 2018. J. Memb. Sci. 547. 123-133). This phase-separated morphology' dictates the IEM’S chemical and mechanical equilibrium and the ion and water transport properties. The sorption behavior at the ionomer site, which is highly sensitive to the type of adsorbed species, can provide insight into this phase-separated morphology and various stressors that impact overall stability. This sensitivity manifests as changes to water and counterion permeability', leading to the strong positive correlation between the IEM’s ion transport and water content. Therefore, changes that reduce water content (or solvent uptake percentage) typically reduce bulk IEM conductivity.Attorney Docket No.: 206256-0112-00 WO

[0096] Several commercial IEMS (Table 1) have been used for electrochemical desalination and energy storage systems. Selemion® (with a hydrocarbon backbone) and Nafion® (with a perfluorocarbon backbone) are commonly employed in RFDs and possess high stability for electrochemistry and desalination.

[0097] Table 1 : Representative commercial ion exchange membranes’ information provided by manufacturers. *Retail cost quote obtained from Bellex InternationalJEM name Structure Thickness Ion Area Water Cost Gum) Exchange Resistance Uptake ($ m Capacity fl cm2(wt %)2) (meq g1)Selemion® CMV Sulfonated Styrene- 150 2.4 2.9 25 320* (CEM. divinylbenzeneHomogeneous)Selemion® AMV Aminated 140 1.9 2.0 - 4.5 19 (AEM, Styrene / ethylene- Homogeneous) butadiene / styrenetriblock polymerNafion® 212 Fluorinated 50.8 0.92 0.92 50 50083(CEM,Homogeneous)Nafion® 117 Fluorinated 183 0.9 2.9 25 (CEM,Homogeneous)

[0098] Electrodes and current collectors’ selection for RFD system: These are the electrical conductors of the RFD systems and distribute and collect current from the redox species. Because electrochemical reactions occur at the surface of the electrodes, in addition to high electrical conductivity, they must also be inert to redox species and corrosion-resistant to handle high conducting salt concentrations. For electrodes, RFD systems use materials such as carbon or graphite felts, paper, and cloths. These carbon electrodes are low-cost, highly porous, and have a large surface area. PAN graphite felts have the advantage of three-dimensional porosity' and have been shown in several studies to be electrochemically superiorto carbon paper (Zhang, 2012, J. Hydrogen Energy 37, 16935-16942; Liu, 2022, Chemistry Select 7, e202201217). For current collectors, graphite bipolar plates and titanium plates have been demonstrated. However, as the cheaper option, graphite bipolar plates are more cost-effective for RFD systems.

[0099] Separation conditions and analysis of RFD system: The operation of the RFD system is complex. First, RFD systems allow independent control of several separation conditions. These include: a) the applied current or voltage; b) the composition of the redox electrolyte (i.e., the amount or concentration of redox species and conducting salt); c) the feedwater concentrations; and d) the flow rates of individual channels. Second, analysis of RFD systems depends on decoupled data collection of the electrical signals from the reactor and compositional analysis of the salt channel, typically using a conductivity meter. Finally, the analysis of the effluent channel also depends on the cycling mode (i.e., batch or single-pass cycling modes). Because batch mode uses two fixed-volumeAttorney Docket No.: 206256-0112-00 WOreservoirs, one for each salt channel of the RFD system, the salinity change of the reservoir’s solution can be detected by a conductivity probe.

[0100] The present disclosure relates in part to the development of a symmetric, closed-loop, four-channel RFD. As critical components, Nafion® 117, Selemion® AEM, and ferri- / ferrocyanide (FeCN) are used. The fundamental electrochemical properties of sodium ferrocyanide are confirmed, as well as the structural and ion transport properties of Selemion® AEM. To demonstrate the RFD's desalination capability, the RFD was operated in batch mode at a constant applied voltage, observing salt changes in both channels.

[0101] Materials and Methods

[0102] Chemicals and Materials: Sodium ferrocyanide decahydrate (Na4[Fe(CN)6]) (98%). potassium ferricyanide (K3[Fe(CN)6]) (99%), nitric acid (HNO3), hydrogen peroxide (H2O2), and sodium chloride (NaCl) were purchased from Sigma-Aldrich. Selemion® AMV (AEM) and Nafion® 117 (CEM) were obtained from Mihama Corporation and Ion-Power, respectively. Graphite felts were purchased from Fuel Cell Store. 316 Steel plates, continuous flex santoprene tubes, and fluorosilicone gasket sheets, along with nylon and propylene tube fittings, were obtained from McMaster Carr. Impervious graphite plates were purchased from The Graphite Store. Redox species chemicals were stored in glove box at less than 0.5 ppm of O2 and H2O prior to use.

[0103] Assembly of 4-channel redox flow desalination system: The RFD cell (FIG. 3) was assembled by adapting zero gap architecture with parts manufactured from the NYU MakerSpace and materials purchased from commercial distributors. The RFD cell consisted of two end frames made of corrosion resistant stainless steel with CNC-milled holes for #10-24 bolts and polypropylene tube fitting adapters. Each porous graphite felt electrode, 0.3cm thick with no compression, had an active area of 4.29 cm2(1.3 cm x 3.3 cm) and was supported by a polyvinylidene fluoride block. Additionally, two impervious graphite plates connected to titanium wires were used as current collectors. Fluorosilicone sheets were used as insulating gaskets between components. The middle chambers were 3-D printed using a Form 3 and had a spacer channel height of 4.06 mm. The lEMs. Nafion® 117 (CEM) and Selemion® AMV (AEM). with exposed areas of 5.64 cm2, were used. The CEM was prepared by chemical conditioning through sequential boiling in 3% H2O2, deionized water, and 0.5 M HNO3for one hour each and then equilibrating in 2 M NaCl solution for 48 hours (Berezina, 2022, J. Memb. Sci. 209, 509-518). Both the CEM and AEM w ere prepared by equilibrating in 0.5 M NaCl solutions for at least 24 hours before experimental runs. The CEMs were used to separate the electrolyte channels from the salt channels, w hile the AEM divided the two salt channels. A 100 mM FeCN electrolyte was prepared with a 1:1 molar ratio of Na Fe(CN)6] and K3[Fe(CN)6] in 50 mL of 35,000 ppm NaCl solution and circulated in batch mode under constant N2purge. Three NE-9000B syringe pumps were used to circulate the electrolyte and salt solutions.

[0104] Analysis of salt channels: The salt solution’s conductivity changes (in pS cm-1) were detected by inserting an Orion™ DuraProbe™ four-cell conductivity probe into the salt solutions of CH 2 and CH 3. The conductivity of tire feedwater was measured by dipping the probe into the salt solutions of tire reservoirs at various times during desalination. To determine the salinity of the chamrels, a calibration curve of conductivity against salt concentration was used, based on measured conductivities of NaCl standard solutions in brackish water and seawater regimesAttorney Docket No.: 206256-0112-00 WO(FIG. 4, right). This curve was fitted using the measured conductivities of standard solutions of 1,413 pS cm ‘, 12.9 mS cm-1, and 111.9 mS cm-1. During voltage holds, the salinity changes were recorded, as shown in FIG. 4, left.

[0105] Electrochemical measurements: All electrochemical tests were performed using a BioLogic VMP-3 potentiostat. Cliroiioamperometr of tire RFD, conducted at a constant voltage of 0.5 V, was used to cany out the desalination run, resulting in simultaneous desalination and salination in CH 2 and CH 3, respectively (Luo, 2017, Nano Energy 42, 215-221). Cyclic voltammetry (CV) measurements were performed with a three-electrode setup. The setup included a glassy carbon electrode (2 mm diameter) and a Pt wire, which served as the working and counter electrodes, respectively. Ag / AgCl was used as the reference electrode. Rotating disk electrode linear sweep voltammetry’ (RDE-LSV) tests were performed with a Pine Research Instrumentation system using a three-electrode configuration: a glassy carbon rotating electrode (5 mm diameter), a glassy carbon counter electrode (2 nun), and an Ag / AgCl reference electrode. The current-voltage data of the electrolytes were collected at rotating speeds of 300 to 2,400 rotations per minute (rpm) and a scan rate of 5 mV s'1. The Levich equation was used to calculate the redox species diffusion coefficient (D) in the aqueous solution from the current-voltage curves of the RDE experiment.where the bulk concentration, Cbuik. solution kinematic viscosity, v, number of electrons partaking in charge transfer, n, area of glassy carbon RDE electrode, A, and Faraday’s constant, F, were known values, and D calculated by resolving the slope of the plot of the limiting current, ilim, against the squared-root of the angular velocity, co1 / 2. The electron exchanges current, i0, and reaction rate constant, k0, were determined with extrapolations of the K-L and Tafel plots with using the Koutecky-Levich and Butler- Volmer equations, respectively as follows:

[0106] Chronopotentiometry of the AEM was performed to characterize the ion transport properties in a four-electrode H-cell setup at 25 °C under N2purge, as previously reported (Freijanes, 2016, J. Memb. Sci. 510, 79-90). Each half-cell included an Ag / AgCl reference electrode and an electroplated Ag / AgCl coil wire (d = 0.64 mm) immersed in 2,000 ppm salt solution (100 mL), ensuring ideal transport properties. The tips of the Luggin capillaries of the reference electrodes were maintained at 0.5 mm to measure tire potential difference across the AEM (surface area of 1.98 x 10'4nF), which was sandwiched between the half-cells. Constant currents (7-25 mA), ranging from below to above the limiting currents, were applied for 5-minute intervals, followed by 5-minute open-circuit voltages (OCVs), and then reversed to maintain the Ag / AgCl coil electrodes. Consequently, the applied constant currents resulted in membrane voltage-current curves with a characteristic S-shape. The V-I plots were analyzed to determine the difference in AEM’s ion transport number of Cl’ between membrane and solution, Atcl, and its effective fraction of conducting region, e*, as follows:<Attorney Docket No.: 206256-0112-00 WO<where Vstis the quasi-stable voltage plateau of tire S curve, representing the measured voltage difference between two fixed Luggin capillaries in the bulk solution placed on two sides of the membrane. Ro is the ohmic resistance, R is the universal gas constant, T is tire temperature of tire system, IL is tire limiting current density, T is the transition time, and 6 is the effective polarization layer thickness.

[0107] Results and Discussion

[0108] Fundamental electrochemical properties of sodium ferrocyanide: In solution, sodium ferrocyanide undergoes a highly reversible, single electron redox reaction as shown below.e-+ [Fe(CN)6]3’^ [Fe(CN)6]4’

[0109] To confirm the half-electrode potential of sodium ferrocyanide, performed CV was performed in a three-electrode set-up containing 1 mM of the redox species in 0.5 M NaCl. The CV peaks were well-defined, indicating the stable and fast electrochemical characteristics of sodium ferrocyanide between -0.2 and 0.6 V. To investigate the reaction kinetics of sodium ferrocyanide, RDE-LSV was performed at a 5 mV s-1scan rate and 300-2,400 rpm (rotations per minute). The D and ko constants were calculated using the Butler-Vomer and Koutecky-Levich (K-L) equations and the Tafel plot extrapolation. These resulted in a Doand ko of 4.45xl0'6cm2s-1and 1.47xl0'2cm s1, respectively, verifying the diffusion rate and reactions kinetics of FeCN were consistent with literature (see Table 2) (Tan. 2019, Nat. Mater. 2019 192 19, 195-202).

[0110] Table 2: Fundamental electrochemical properties of Na4[Fe(CN)6] at pH-neutral conditions.0111] Ionic transport properties of pristine Selemion® AMV: Chronopotentiometiy is a well-established method for the precise characterization of bulk and interfacial transport properties of IEMS, such as the difference in chloride ion transport number between the solution and membrane (Afcl) and the effective fraction of conducting region (c‘)-also related to fraction of homogeneous and microheterogeneity regions (Freijanes, 2016. J. Memb. Sci.510. 79-90; Tian. 2022. Int. J. Thermophys. 43, 1-19). Therefore, the pristine AEMs were tested under isothermal conditions (25 °C) using a 100 mL double-layer H-cell with a four-electrode setup. Constant currents were applied for 5-minute intervals, followed by 5-minute open-circuit voltage measurements. These steps were reversed to maintain the coating of the Ag / AgCl reversible coils.Attorney Docket No.: 206256-0112-00 WO

[0112] The ion transport properties of pristine Seleinion® AMV in are summarized in Table 3. Vst-I curves were converted to a Cowen plot to detennine the limiting current, IL of 14.83 mA. by taking the inverse current value at the intersections of the two slopes. The experiment was repeated with applied current (20-25 mA), i.e., >1.51L, to generate voltammograms and the transition times (r) plots. Using the slope (B) of the transition times and inverse of squared current densities, values of Aty]- of 0.5, 6 of 444 pm, and e* of 0.95 by were obtained by non-linear minimization of Eq. 1-4. equality of Eq. 1-5, and linear fit of Eq. 1-6, confirming the ideal properties of the pristine AEM is consistent with reports.

[0113] Table 3 : Summary' of ion transport properties of pristine Selemion® AMV from chronopotcntiomctn

[0114] Demonstration of 4-channel seawater RFD in batch mode: To demonstrate the seawater desalination using the custom system, the RFD was operated in batch-mode with 100 mM FeCN dissolved in simulated seawater of 35000 ppm NaCl solution. At the start of the experiment, each of the feedwater reservoirs contained 50 mL of seawater. The flow rates in all channels and applied voltage across the RFD cell were maintained at 30 mL min1and at 0.5 V, respectively.

[0115] As observed in the chronoampcromctry results in FIG. 5, right, a constant applied voltage led to a continuous decrease in current density over time. By measuring die conductivity of the feedwater in the beakers at various time intervals, a continuous decrease was observed in one channel from approximately 35,000 ppm to 805 ppm and an increase in the other channel within 1,750 minutes, corresponding to more than 97% salt removal from the desalination channel. This indicated the first successful continuous desalination and salination by the custom four-channel RFD, matching similarly reported trends for batch-mode RFDs. However, the manual conductivity data acquisition method was unsuitable for continuous measurement, leading to a gap in data acquisition, as exemplified by the data gap between approximately 300 and approximately 1, 100 minutes in FIG. 5, left.

[0116] The observed attenuation in current density serves as a real-time proxy for the deceleration of ion transport. This phenomenon is primarily driven by tire non-linear rise in polarization area-specific resistance as the system works against an increasing concentration gradient and cell resistance, which arises from the reduced availability of ions in CH 2 and the high concentration in CH 3, as observed in FIG. 6. Specifically, as the desalination channel (CH 2) reaches a state of high dilution (e.g., < 5.000 ppm), the paucity of available charge carriers at the membranesolution interface results in a drastic resistance surge (from -165 Q cm2to -780 Q cm2). Thus, energy consumption increases proportionally with the increasing salt channel salinity difference during desalination, an inevitable tradeoff in batch-mode operation that necessitates the optimized flow regimes described in Example 2.

[0117] In conclusion, a custom four-channel redox flow desalination system was designed and built using 3D printing and CNC milling techniques, as well as commercially available components. The fundamental electrochemical properties of pristine sodium ferrocyanide and Selemion® AMV were consistent with literatureAttorney Docket No.: 206256-0112-00 WOvalues. Using 100 mM FeCN as the redox species, as well as Nation® 117 (CEMs) and Selemion® AMV (AEM), the simultaneous desalination and salination of CH 2 and CH 3 was demonstrated, respectively, using a custom-built four-channel RFD in batch mode at an applied voltage of 0.5 V. A 50 mL solution of seawater was desalinated from approximately 35.000 ppm to 805 ppm within 1.750 minutes.

[0118] Example 2: Investigation of flow rate in symmetric four channel redox flow desalination system

[0119] It has been previously demonstrated that flow rate is a critical parameter for improving the performance of flow-based electrochemical and desalination systems by altering the mass transport properties of redox-active materials and ions at the fluid-membrane interfaces. The present disclosure investigates the impact of electrolyte channel flow rates on a redox flow desalination (RFD) system utilizing a ferricyanide / ferrocyanide (FeCN) redox couple. In addition to flow dynamics, other operational conditions-including redox species concentration and applied voltage-are examined to evaluate their influence on key performance metrics: average salt removal rates (ASRR), energy consumption Ereactor, and charge efficiency at 50% water recovery. These experiments are performed in both single-pass and batch modes to accentuate the operational differences and performance shifts driven by varying flow regimes. In-situ electrochemical impedance spectroscopy (EIS) is employed to demonstrate that flow rates strongly influence desalination performance by modulating the mass transport properties of redox species and ions within the system architecture. The disclosure highlights that increasing flow rates of electrolyte channels predominantly diminished electrolyte-membrane interfacial resistances and promoted ionic flux across ionexchange membranes. Consequently, this improved the RFD system’s desalination performance by increasing ASRR and charge efficiency and reducing Ereactor, especially at high voltages. Additionally, decoupling the flow rates of electrolyte and salt channels improved the single-pass mode analysis of RFDs but also slightly increased interfacial resistances at the membrane surface and promoted FeCN crossover into salt channels. These findings thus provide insight into the influence of fluid dynamics in RFD systems, enabling their facile and comprehensive evaluation for high-performing seawater desalination units.

[0120] The state of commercial and emerging desalination technologies: The three ty pes desalination technologies thermal, membrane, and electrochemical are semi-specialized based on the feedwater type and their performance decreases as the salinity of the feedwater increases (summary’ of commercial and emerging desalination technologies are depicted in Table 4.) (Alsharhan Overview on Global Water Resources. In (Springer, Cham), pp.17-61; Donnan, Chemical Reviews. 1924, 7, 73-90; Barragan, Journal of Colloid and Interface Science, 1998, 205, 365-373; Choi, Journal of Colloid and Interface Science, 2001, 238, 188-195; Wang, Desalination. 2021, 504, 114964; Kim, Desalination, 2023. 550, 116406; Jones, Sci. Total Environ., 2019, 657, 1343-1356; Al-Karaghouli. Renew. Sustain., 2013. Energy Rev. 24, 343-356; Elimelech, Science, 2011, 333, 712-717; Lin, Desalination. 2015, 366, 9-14; Lee, ACS Omega. 20172, 1653-1659 ). Hence, their application is often limited by three main factors of cost, energy consumption, and water recovery ratios or efficiency. Currently, multi-stage-flash (MSF), multi-effect-desalination (MED), both thermal, and membrane reverse osmosis (RO) are the major producers of drinkable water at a commercially scale, comprising of about 84% of all operating plants and contributing to about 93% volume of all worldwide desalinated water. RO has been the leading technology for seawater (SW) and brackish (B W)Attorney Docket No.: 206256-0112-00 WOdesalination since 1980s because of its lower energy requirement and environmental impact than MSF and MED. But the limited SW and B W desalination recovery ratio of these major technologies leads to adverse environmental implications, resulting in 142 million cubic meters of brine produced daily. This is about 1.5 times as much brine as produced desalinated water. Emerging membrane technologies have also been employed to address these issues of energy consumption and low recovery ratios, including processes like ultrafiltration and nanofiltration (NF); however, these have high capital costs that scale with feedwater salt concentrations, making them challenging for seawater desalination (Shi. Environmental Science & Technology Letters, 2018, 5, 692-700).

[0121] Table 4: Summary of the state-of-the-art seawater and brackish water desalination technologies.>< & &

[0122] A promising strategy to address the limitations of cost, energy consumption, and recovery ratio is to use electrochemical desalination systems that transport ions instead of water molecules, and thus consuming less energy for desalination that membrane or thermal processes (Knust, ChemElectroChem, 2014, 7, 850-857; Lee, ACS Appl. Mater. Interfaces, 2016, 8, 11154-11161). Out of these, electrodialysis (ED) and electrodeionization (EDI) technologies are capable of desalinating at high efficiency and low energy consumption, however they have relatively high cost. Capacitive deionization (CDI) and faradaic electrochemical processes (FEC) require high surface area electrodes, limiting their application for highly concentrated feedwater like seawater (Shi, Environ. Sci. Technol. Lett., 2018. 5, 692-700; Desai. ACS Energy Lett., 2018, .?. 375-379). As such, there is still a need for systems that can address these three limitations, as well as effectively utilize renewable energy sources, driving research into redox flow desalination (RFD) systems that have the potential to address these issues.

[0123] The state of 4-channel RFD research in single-pass and batch modes: Since the introduction cost-effective symmetric 4-channel RFDs for desalination of highly saline feedwater by Beh et al., who established their capabilities with BTMAP-Fc and FeCN, along with Nafion® R212 and Fumasep Fas-15 lEMs, numerous systematic studies have been published to understand their electrochemical and salt removal performance with various redox species (Al-Karaghouli, Renewable and Sustainable Energy Reviews, 2013, 24, 343-356; Knust, ChemElectroChem, 2014, 7, 850-857; Shi, Environmental Science & Technology Letters. 2018, 5. 692-700; Panagopoulos, Energy Conversion and Management, 2021, 235, 113957; Deb ruler, Advanced Functional Materials,Attorney Docket No.: 206256-0112-00 WO2020, 30. 2000385; Cheng, Desalination, 2022, 534, 115783; Chen, Desalination. 2023. 558, 116616; Pan, Desalination, 2020, 496. 114762). These reports have provided insight into how concentration of redox species, applied voltage, and salt concentration influence the average salt removal rates (ASRR) and energy consumption of RFDs (Chen, Journal of The Electrochemical Society, 2020, 167, 083503; Chen, Chemical Engineering Journal. 2020, 401, 126111).

[0124] Previous parametric investigations in batch or single-pass mode have been crucial for optimizing operational conditions of RFD systems (see Table 5) (Wang, J. Mater. Chem. A, 2019, 7, 13941-13947; Chen, Chem. Eng. J., 2020, 401, 126111; Zhang, Mater. Today Commun., 2020, 23, 100921; Beh, ACS Sustain. Chem. Eng.. 2019, 7. 13411-13417; Lu, Chem. Eng. J., 2022, 431, 133917; Mohandass, Environ. Sei. Technol., 2022, 56, 4477-4488; Cheng, Desalination, 2022, 534, 115783). In batch mode (see FIG. 1), tw o fixed-volume reservoirs, one for each salt channel, supply saltwater to the RFD system to produce either freshwater or brine after a specified period. The salinity change of the reservoir’s solution is detected by a conductivity probe. In single-pass mode (see FIG. 7), a large reservoir supplies a constant concentration of saltwater to the RFD’s salt channels. The salinity change of the salt channel’s effluent is detected in real-time by a conductivity probe. Most reported parametric investigations are batch mode (Han, Environmental Science: Water Research & Technology. 2023, 9. 2368-2377). From these investigations several research teams have observed that the energy consumption increased w ith increments of applied current density or applied voltage. Var ing current density and concentrations of redox species, studies have show n that increasing the concentration of redox species from 2 mM to 10 mM reduces overpotentials, which simultaneously enhances salt removal rates and reduces energy consumption for brackish water (3000 ppm) desalination. Studies confirm the reduction of overpotential due to increased concentration with current-voltage overpotential curves. Recently. Cheng et al. further observed that 10 mM still showed limited brackish water desalination performance of ~30 pg cm'2min1and ~ 80 kJ mol1, even at a high applied voltage of 0.8V. Rather, redox species concentration above 100 mM showed promising salt removal of about 91 pg cm'2min1, comparable to SWRO. Yet, their report demonstrated that differences in ASRR values between the RFDs with 50 mM and 100 mM FeCN for brackish water desalination were marginal above 0.4 V. which they attribute to the saturation of electrochemical reaction sites (Chen, Journal of The Electrochemical Society, 2020, 167, 083503). The highlight of their study was the increase of desalination performance due to increase in salinity of salt channels . Using lOOmM FeCN and voltage of 0.2V. they observed up to 5-fold gain in salt removal rates with feedwater salinity of 35000 ppm and flow rates of 5 mL min1, achieving desalination performance of 96.2 pg cm'2min-1and 20.6 kJ mol1that were comparable to SWRO. However, their study lacked further flow rate studies to understand the influence it has on seawater desalination. These and other batch-mode parametric investigations have been crucial to identifying optimal and limiting operation parameters for RFDs.

[0125] Table 5: Summary' of Sy mmetric 4-channel RFDs demonstrated in literature.&& Attorney Docket No.: 206256-0112-00 WO&&&&&Attorney Docket No.: 206256-0112-00 WO&&&Attorney Docket No.: 206256-0112-00 WOCeiec: electrolyte concentration ; CC: constant Current; CV: constant Voltage ; Qeiec : electrolyte flow rate ; CM : feed water cycling mode; Cfeeii: feedwater initial concentration; Vres: feedwater reservoir size; Qfeeci: feedwater flow rate; Atcycle: cycle duration; WR: water recovery; P: desalination productivity; (Ac): average concentration reduction.

[0126] Previously, Lu et. al, achieved remarkable brackish desalination performance of RFD in single-pass using 4-sulfonatooxy-2.2.6,6-tetramethyl-piperidine-l-oxyl (NaSO4-TEMPO) (see Table 6) dissolved in 5000 ppm NaCl. They reported that a 2000 ppm NaCl solution could be desalted to 762 ppm in a single pass. Their report also included detailed single pass performance analysis of various redox species concentration (0 — > 100 mM), electrolyte channel salt concentration ( 0 — > 5,000 ppm), salt channel salinity (500 — > 10000 ppm), applied voltage (0.2 —> 1.4 V), and flow rates of 10 mL min1and 5 mL min1in the electrolyte and salt channels, respectively. But their study did not include seawater concentration (-35000 ppm) and flow rate operational comparison to understand their effects on single-pass RFD desalination performance.

[0127] Tabic 6: Summary' of RFD desalination performance for seawater (35000 ppm).>Attorney Docket No.: 206256-0112-00 WO

[0128] Chemicals and Materials: Sodium ferrocyanide decahydrate (Na4[Fe(CN)6]) (98%). potassium ferricyanide (K3[Fe(CN)6]) (99%), and sodium chloride (NaCl) were purchased from Sigma-Aldrich. Selemion® AMV and Nafion® 212 were obtained from Mihama Corporation and Ion-Power, respectively. Graphite felts were purchased from Fuel Cell Store. 316 Steel plates, continuous flex santoprene tubes, and fluorosilicone gasket sheets, along with nylon and propylene tube fittings, were obtained from McMaster Carr. Impervious graphite plates were purchased from the Graphite Store. Redox species chemicals were stored in glove box at <0.5 ppm of O2and H2O prior to use.

[0129] Assembly of 4-channel redox flow desalination system: The RFD cell was assembled by adapting zero gap architecture (FIG. 2) with parts manufactured from the NYU MakerSpace and materials purchased from commercial distributors. The RFD cell consisted of two end frames made of corrosion resistant stainless steel w ith CNC-milled holes for #10-24 bolts and polypropylene tube fitting adapters. Each porous graphite felt electrode, 0.3cm thick with no compression, had an active area of 4.29 cm2(1.3 cm x 3.3 cm) and was supported by a polyvinylidene fluoride block. Additionally, two impervious graphite plates connected to titanium wires were used as current collectors. Fluorosilicone sheets were used as insulating gaskets between components. The middle chambers were 3-D printed using a Form 3 and had a spacer channel height of 4.06 mm. The IEMS. Nafion® 212 and Selemion® AMV membranes, with active areas of 5.64 cm2, were prepared by equilibrating in 0.5 M NaCl solutions for at least 24 hours before experimental runs.11Nafion® 212 were used to separate the electrolyte channels from the salt channels, while the Selemion® AMV divided the tw o salt channels. FeCN electrolytes of 12.5, 25.0, and 50 mM were prepared with 1 : 1 molar ratio of Na4[Fe(CN)6] and K3[Fe(CN)6] in 50 mL of NaCl solutions and circulated in the electrolyte channels, under constant N2purge Three NE-9000B Syringe pumps were used to circulate electrolyte and salt channels.

[0130] Analysis of salt channels: The salt solution’s conductivity changes in pS cmJwere detected in-situ at the effluent of CH 2 of the RFD operated in single-pass mode using an Orion™ DuraProbe™ 4-cell conductivity probe inserted into a custom 3-D printed module. The in-situ module for housing the conductivity probe was 3-D printed using Formlab’s Form 3 SLA 3D-printer (see FIG. 9, left). To reduce errors in salt conductivity measurement, design considerations were made to limit fringe effect and maintain fluid homogeneity. To measure salinity of salt channels, a calibration curve of conductivity against salt concentration was employed using measured conductivities of NaCl standard solutions in brackish w ater and seawater regimes (FIG. 33). Conductivity readings were collected at 3-second intervals in all experiments. During voltage holds, the salinity changes were noted as shown in FIG. 32 (top right). FeCN crossover was detected ex-situ for CH 2 and CH 3 with a Cary Series UV-Vis-NIRAttorney Docket No.: 206256-0112-00 WOSpectrophotometer (Agilent Technologies). The concentration of FeCN crossover was determined using a calibration curve of absorbance peak at wavelength taken at 216 ± 2 mn. against concentration of FeCN (See FIG.33).

[0131] Electrochemical measurements: All electrochemical tests were completed with a two-electrode setup using a Biologic VMP-3 potentiostat connected to the RFD’s titanium wire-graphite plate current coimectors at room temperature conditions. Chronoampcromctry was used to carry' out desalination runs by reversing the applied voltage difference for four cycles. Each of these reverse polarity cycles consisted of four periods, including an open circuit voltage (OCV), a negative voltage hold for salination in CH 2, another OCV, and a positive voltage hold for desalination in CH 2. The applied voltages difference ranged from 0.2 to 1.0 V. Specifically, each cycle in Section 2.2 included 30 min voltage holds and 30 min rests at OCV. while that of Section 2.3 included 15 min voltage holds and 20 min rests at OCV. These voltages were chosen because 0.2 V could sufficiently drive ion-coupled electrochemical reactions of FeCN, while 1.0 V was adequately high but lower than the electrolytic window of 1.23 V for water splitting at pH neutral conditions, and thus minimized current density contributions from parasitic reactions (Tang. Journal of Power Sources, 2014, 248, 154-162). EIS measurements were collected within a frequency range of 1 MHz to 1 mHz at an amplitude of 10 mV.

[0132] Calculation of desalination performance parameters: The desalination performance parameters were calculated as follows: The throughput productivity, P, in L nr3h1. defined as the volume of desalinated water per total cycle time per face area, and average concentration reduction, (Ac), over tire cycle in mM were calculated as previously reported (Hawks, Water Systems, 2019. 152, 126-137).Single-pass mode:(Eq. 2-la)Batch-mode: PBM= ~ (Eq. 2-lb)(Ac)=;°(Cf°QC) Qdtiddt (R| 2‘2)where n (1) is the number of cells used in desalination. Co is the influent feedwater concentration from the reservoir (ppm), C is the effluent concentration from the reactor (ppm), t is the time (min), T is the duration of voltage hold (min), Qfeed is the flow rate of middle channel (mL min1), and A is the electrode area (4.29 cm2).

[0133] The average salt removal rate (ASRR) in pg cm-2min1, representing the average amount of salt removed by RFD system over a period in the single-pass and batch modes were calculated as reported in previous literature (Porada, Progress in Materials Science, 2013, 58. 1388-1442). The ASRR was calculated as follows:Single-pass mode: ASRRSPM=' ' ''C)ATFEEDPNIDT(Eq. 2-3a)Batch-mode: ASRRI; I=(C1'Cf^n|Vres(Eq. 2-3b)Attorney Docket No.: 206256-0112-00 WOwhere pmis the density of water (1 g mL1), Ci and Cf are tire starting and final concentration of the desalinated reservoir (ppm), Vresis the volume of desalinated reservoir (mL).

[0134] The charge efficiency in % was calculated using chronopotentiometry and salt concentration measurements as reported.Charge Efficiency (Eq. 2-4)where Irand F are the measured current (A) and Faraday’s constant (96,485 C mol1), respectively, while Mw,Naci is the molecular weight of NaCl (58.44 g mol1).

[0135] The energy consumption (Ereactor) in kJ mol'1of the RFD reactor was calculated using chronoamperometry data of the reactor.8 17-24where V is the applied voltage (V). Additionally, the energy consumption in Wh nr1is the product of (Ac) (Eq. 2) and E^tor (Eq. 5).

[0136] Effects of salt channel flow rate on RFD performance: The flow rate of salt channels directly affects throughput productivity in L nr2h-1(Eq 2-1), and inversely influences tire average concentration reduction in mM (Eq. 2-2) due to the increased residence time allowed for ion flux across the IEMS. Since the conductivity probe is placed en route to the tank (see FIG. 7 and FIG. 8), the precision of effluents salt concentration measurements in single-pass mode also depends on the channel's salinity properties and flow rate (Mo handass, Environmental Science & Technology-, 2022, 56, 4477-4488). As such, the first objective was to determine a flow rate of salt channels that allowed facile and precise concentration measurements for a wide range of feedwater concentrations and electrolyte flow rates. To achieve this objective, the single-pass RFD was operated at low to medium reverse polarity applied voltages (0.2 to 0.6 V) to induce minute salt changes. Similar to other reports, the salt and electrolyte channels were fed with 2000 ppm NaCl solution and 25 mM FeCN mixed in 2000 ppm NaCl solution, respectively, at a flow rate of 5 mL min1. This separation condition equated to a high throughput productivity of 699.3 L nr2h'1and water recovery of 50%. The same concentrations of NaCl solutions were used to reduce concentration gradient across the IEMs that could introduce overpotential.

[0137] Effects of salt channels salinity on the precision of desalination measurement in single-pass mode:Owing to the sensitivity limits of the conductivity probe and the complex flow dynamics in the in-situ module, precise measurement of minute salt concentration changes at high salinity can be challenging.

[0138] To understand how the salinity of CH 2’s effluent affects salt concentration measurements in the in-situ module, the RFD was operated in single-pass mode with 25 mM FeCN at 5 mL min'1all channels, and the applied polarity- voltages varied from 0.2 to 0.6 V. FIG. 9 and FIG. 10 show the concentration profiles for feedwater with influent concentration of 2000 ppm and 35000 ppm, respectively. At low salinity, precise measurement of CH 2’s effluent’s concentration was confirmed by the low standard deviations, as shown in FIG. 9. After 4 cycles, theAttorney Docket No.: 206256-0112-00 WOaverage salt removal rates (ASRR values) of 22.90 ± 2.45, 42.98 ± 1.10. 61.47 ± 1.34 pg cm'2min1, which corresponded to average concentration reductions of 0.34 ± 0.01, 0.63 ± 0.02, and 0.96 ± 0.02 mM, for 0.2, 0.4, and 0.6 V, respectively. On the other hand, the precision of the concentration profiles for feedwater of 35000 ppm influent salinity showed high inconsistency in similar operating conditions, as observed in FIG. 10. This was attributed to the sensitivity limits of the Orion™ DuraProbe™ 4-cell conductivity probe to detect minute conductivity changes in highly saline salt solutions.

[0139] As observed in FIG. 9 that shows a cycle of applied voltage, the salt concentration profile of tire effluent of CH 2 was highly consistent. After four cycles, the ASRR values (see Eq. 2-3a) were 22.90 ± 2.45, 42.98 ± 1.10, 61.47 ± 1.34 pg cm'2min1, which corresponded to average concentration reductions of 0.34 ± 0.01, 0.63 ± 0.02. and 0.96 ± 0.02 mM, for applied voltages of 0.2, 0.4, and 0.6 V, similar to previous reports (Cheng, Desalination, 2022, 534, 115783). The low ASRR standard deviations indicate that adequate precision for single-pass salt analysis of brackish water was achieved. However, determining minute salt changes of single-pass RFD was more challenging for effluent seawater at 5 mL min-1in all channels. When the seawater was used for mixing 25 mM FeCN in the electrolyte channels and for the salt channels’ influent, the accuracy significantly deteriorated (see FIG. 10). The concentration profile for seawater was extremely inconsistent at the same salt channel flow rate and voltage operation conditions. Tthese inconsistencies can be attributed to the reduced accuracy of the conductivity probe to detect minute salt changes at high effluent salinity. This emphasized a need for higher average salt concentration reductions, achieved through increased ionic flux or reduced residence time in single-pass analysis for seawater.

[0140] Due to the direct relationship of precision and average concentration reduction, higher productivity would decrease the precision of salinity measurement. To study this, the single-pass RFD was either operated at the same flow rates across all four charnels or decoupled to maintain a constant flow rate of 5 mL min1(699.3 L nr2h1) in the salt channels while the electrolyte channels were varied. To study the precision of salinity changes of the effluent seawater, the effects of flow rate of the RFD system’s salt channels on ASRR were studied at different flow conditions with a high applied voltage of 0.8 V (see FIG. 11) to induce high ionic flux. As shown in FfG. 11, 5 mL min'1was observed to minimize errors in ASRR. while increasing flow rates of the salt channels significantly increased the performance parameter's standard deviation. For example, die ASRR values when all the channels were run at 10 mL min'1and 50 mL min'1(1398.6 and 6993 L nr2h'1) were 326.28 ± 43.73 and 787.28 ± 206.4 pg cm-2min1, respectively. But when the salt channels were maintained at 5 mL min'1and the electrolyte channels varied at 10 mL min'1and 50 mL min1, die ASRR of 336.56 ± 27.73 and 756.28 ± 35.8 pg cm'2min1, clearly indicating higher precision of salt chamiel analysis. As such, decoupling die flow rates of the single-pass RFD ensured greater precision of salt channel analysis.

[0141] Effects of salt channels flow rates on the precision of desalination measurement in single-pass mode:Owing to the sensitivity limits of the conductivity probe and the complex flow dynamics in the in-situ module, precise measurement of minute salt concentration changes at high flow rates can be challenging.

[0142] The flow rates of salt channels influenced the conductivity probe’s precision to detect salinity changes in single-pass mode. FIG. 11 compares the ASRR values of the RFD operated in single-pass mode for seawaterAttorney Docket No.: 206256-0112-00 WOdesalination at different flow rate conditions and 25 mM FeCN dissolved in 35000 ppm. The ASRR values were calculated for 4 cycles of applied reverse polarity of ± 0.8 V. As observed in FIG. 11, the standard deviations of the ASRR values increased significantly as flow rates of salt channels were incremented from 10 mL min1to 50 mL min1. This was due to the increasing noise, as similarly observed in FIG. 10. possibly due to the conductivity probe sensitivity to salt changes limited average salt concentration changes in the in-situ module. On the other hand, the standard deviation remained minimal for flow rates of 5 mL min1, indicating high precision of salt measurement in the salt channel. For example, the ASRR values when all the channels were run at 10 mL min1and 50 mL min-1(1398.6 - 6993 L m'2h1) were 326.28 ± 43.73 and 787.28 ± 206.4 pg cm'2min1, respectively. But when the salt channels were maintained at 5 mL min'1(699.3 L nr2h1) and the electrolyte channels varied from 10 mL min1to 50 mL min1, the ASRR values of 336.56 ± 27.73 and 756.28 ± 35.8 pg cm2min1, were recorded. Based on these results, 5 mL min1in the salt channels exhibited the adequate precision, regardless of the flow rates of the electrolyte channels, and was thus used to evaluate the RFD in single-pass mode.

[0143] It is worth noting that 5 mL min1(699.3 L m'2h'1) in the salt channels is still a relatively high productivity when compared to practical seawater desalination that typically operate at 12-15 L nr2h-1productivity (Greenlee, Water Research. 2009, 43, 2317-2348). Reducing the flow rates below 5 mL min'1would not only allow the RFD to meet practical comparison, but also further increase the precision of the salinity measurement. But in this RFD analysis, reducing the productivity further came at the cost of the response time to reach a quasi-stable concentration plateau. The duration needed to reach a quasi-stable concentration plateau also increased beyond 30 minutes, evident from the concentration profile in FIG. 11 that lacked plateaus. As such, the salt channels were maintained at 5 mL min-1to expedite data acquisition at sufficient precision and emphasize the role of the electrolyte channels’ flow rate on RFD performance.

[0144] Effects of electrolyte flow rate on RFD performance: Thus, to extensively investigate tire impact of the electrolyte channel flow rates on the seawater desalination, the single-pass RFD was operated with 25 mM FeCN dissolved in seawater in the electrolyte channels. The flow rates in the electrolyte channels varied from 5 to 60 mL min1, while that of the salt charnels were constant at 5 mL min1. Applied voltages difference ranged from 0.2 to 1.0 V.

[0145] The chronopotentiometry and concentration profiles are presented in FIG. 12 and FIG. 13. respectively. •Generally, increasing flow rates strongly entranced the performance, particularly at high voltages. For instance, when operating with 0.2 V (FIG. 12), the current density increased from 2.84 at 5 mL min'1to 6.67 mA cm'2at 60 mL min1. Likewise, the current density increased from 4.34 to 33.13 mA cm'2when 1.0 V was applied.Additionally, increasing the applied voltages at low flow rates (e g., 5 mL min1) increased ideal polarization effects, exemplified by the ‘thickening’ of the current density profile. The phenomena of visible polarization effects were similarly observed in the chronoamperometry profiles reported by Lu et al. their single-pass RFD, which was operated at flow' rates of 10 mL min'1in the electrolyte channels and 5 mL min1in the salt channels, showed increasing polarization effects at higher voltages (Panagopoulos. Energy Conversion and Management, 2021, 235,Attorney Docket No.: 206256-0112-00 WO113957). These polarization effects have been attributed to the saturation of the electrochemical reaction sites in other reports (Milshtein. Journal of The Electrochemical Society, 2017, 164, E3265-E3275).

[0146] Conversely, increasing the flow rates from 5 to 60 mL min1led to higher current densities and decreased the observed ideal polarization effects of the RFD in FIG. 12. These findings were consistent with changes in salinity (FIG. 13), indicating that increasing tire flow rates may have contributed to the desaturated reaction sites that were limiting tire electrochemical reactions and the improved ion transport for enhanced salinity changes at high voltages. Overall, the chronoamperometry and concentration profiles suggest that minimizing the ideal polarization effects by increasing flow rates from 5 to 60 mL min1contributes to the strong positive correlation of increased seawater desalination performance changes to flow rates as voltage increases (Tang, Journal of Power Sources, 2014, 248, 154-162).

[0147] This is supported by the analysis of the calculated ASRR. (Ac), E]eactor(Eq. 2-5), and charge efficiency results (Eq. 2-4). which exhibited comparable results for the salination and desalination steps (refer to Table 6). Increasing flow rates had a significant positive impact on ASRR and resulted in decreasedvalues, particularly at high voltages. This effect is clearly illustrated in FIG. 14. which compares the different operating conditions and shows a reduced slope in tire linear correlation between EreactOr and ASRR values. The decrease in slope indicates a more efficient utilization of molar energy consumed for the removal of NaCl ions from the CH 2 solution.

[0148] A comprehensive analysis of tire desalination performance at each voltage level follows. (FIG. 15). At a low flow rate of 5 mL min1and applied voltage of 0.2 V, tire RFD exhibit an ASRR of 81.27 ± 37.51 pg cm'2min'1(1.17 ± 0.54 inM), an Ereactor of 29.36 ± 16.38 kJ mol'1, and a charge efficiency of ~78.8 %. Increasing the applied voltage while maintaining the same flow rate led to higher Ereactor and reduced charge efficiency, with minimal impact on the ASRR. At 5 mL min1and 1 V, the observedand charge efficiency values were 153.25 kJ mol'1and 64.99 %, respectively. It can be inferred from the restricted ASRR that the applied voltage was poorly utilized for desalination at low flow rate, resulting in energy losses and voltage inefficiency that increasedand decreased charge efficiency (Zhou, Journal of Power Sources, 2017, 339, 1-12).

[0149] Conversely, above 5 mL min1, the salt removal rates were positively correlated with flow rate and applied voltage. For example, at 0.2 V, the ASRR increased by more than two-fold from 85.67± 19.48 pg cm'2min1at 5 mL min-1to 235.15 ± 48.84 pg cm'2min1(1.26 ± 0.29 — > 3.45 ± 0.72 mM) at 60 mL min1, whileshowed marginal changes from 25.67 ± 7.52 to 21.02 ± 4.43 kJ mol'1. But at 60 mL min1and 1.0 V. the ASRR exhibited approximately 9.9-fold increase from 102.64 ± 21.14 pg cm'2min1at 5 mL min1to a value of 994.03 ± 24.97 pg cm'2min1. Additionally, thereduced from 155.00 kJ mol'1to about 116.71 kJ mol1, while the charge efficiency increased to about 82.58%. These observations provide further support to tire conjecture that low electrolyte channel flow rates combined with high voltages can lead to overpotentials, as evident by the ideal polarization curves in the chronoamperometry results in FIG. 12. These overpotentials adversely affect desalination performance by limiting the average salt removal rates, increasing energy consumption, and reducing charge efficiency.Attorney Docket No.: 206256-0112-00 WO

[0150] Effects of flow rates for different electrolyte concentrations on RED performance: To investigate the impact of different concentrations of redox species and flow rates in the electrolyte channels on the desalination performance of seawater, the RFD was operated at applied voltage of ± 0.8 V and flow rates ranging from 5 to 50 mL min1. The redox electrolytes in electrolyte channels were sequentially replaced with FeCN of 50 mM, 25 mM, and 12.5 mM, achieved through dilution with 35000 ppm NaCl solutions (Luo, Nano Energy, 2017, 42. 215-221). Similarly, higher flow rates resulted in increased ASRR values; however, the observed trend varied depending on the concentrations of redox species.

[0151] The chronoamperometry results and the corresponding concentration profiles for 12.5 mM, 25 mM, and 50 mM FeCN, are presented in FIG. 16, FIG. 17, and FIG. 18 respectively. Generally, increasing the concentration of the redox species resulted in higher current densities and salinity changes, which were further amplified with increasing flow rates. However, the ideal polarization effects observed in FIG. 16 to FIG. 18, exhibited different trends depending on tire redox species concentration and flow rates. For a flow rate of 5 mL min1, the ideal polarization increased with redox species concentration, indicating an increase in the overpotential at low flow rate. In contrast, from 20 mL min1to 50 mL min-1the ideal polarization effects decreased with increasing redox species concentration and flow rates. At 30 mL min1, note the lack of ideal polarization effects in FIG. 18 (left), which visibly decline in FIG. 16 and FIG. 17. Furthermore, increasing salinity changes were observed alongside higher current densities and reduced polarization effects as flow rates increased, as evident in the left-hand plots of FIG. 16 to FIG. 18.

[0152] Accordingly, provided are insights into the influence of flow rates and concentration of redox electrolytes on the desalination performance (FIG. 19). Consistent with previous reports, the ASRR increased with higher concentrations of redox species. The ASRR displayed a nearly linear increase for 12.5 mM, and a non-linear increase for 25 and 50 mM FeCN, respectively, within flow rate range of 5 and 50 mL min1. Specifically, at a flow rate of 5 mL min1, the ASRR for 12.5 mM, 25 mM, and 50 mM FeCN were 48.48 ± 9.55. 116.65 ± 9.41, and 209.65 ± 40.46 pg cm2min1(0.71 ± 0.14, 1.71 ± 0.13. and 3.08 mM), respectively, representing the lowest values observed. Accompanying these ASRR values were Ereactor values of 137.96 ± 26.27, 104.31 ± 24.10, and 108.61 ± 8.28 kJ mol1. These observations can be attributed to the increase in electrochemical activity offered by the increased redox species amount at the electrochemical active sites that simultaneously increase salt removal rates and decrease energy consumption by reducing overpotentials, as reported by other researchers.

[0153] Furthermore, as the flow rates increased, the RFD’s ASRR exhibited a corresponding increase. Notably, the highest ASRR values of 811.55 ± 39.22. 985.18 ± 47.04, and 1058.85 ± 20.59 pg cm'2min1(11.91 ± 0.56, 14.46 ± 0.69, and 15.54 ± 0.30 mM) were recorded when cycling the electrolyte channels at 50 mL min1for the same sequence of redox species, representing approximately 16.7-, 8.4-, and 5-fold increases, respectively when compared to the values for 5 mL min1. Additionally, increasing flow rates led to a decrease in the lieactor for all FeCN concentrations, reaching an approximate values between 92 and 95 kJ mol1, comparable to other reports of RFD operating at 0.8 V. For 50 mM FeCN, the maximum value was only slightly greater than the value of 1036.98 ± 63.73 pg cm'2min1(15.22 ± 0.94 mM) observed at 30 mL min1, explaining the non-linearity of the trend. ThisAttorney Docket No.: 206256-0112-00 WOplateau may suggest saturation of electrochemical active sites in the absence of mass transport limitation, impeding salt removal rates and increasing overpotentials, as evidenced by the absence of ideal polarization effects at 30 mL min-1in FIG. 18.

[0154] It is worth noting that the ASRR for 25 mM FeCN in this section were mostly higher than those in above section. However, this difference could be attributed to the slight differences in average series resistance between the two independent studies, i.c., ~5 and ~4 Q for this and previous sections, respectively, as further optimization of the RFD architecture was not a definite objective of these studies. For 12.5 and 25 mM FeCN, there were no clear saturation limits of ASRR observed up till 50 mL min1. It is also worth noting that the highest salt removal rates were recorded when using 50 m FeCN at 50 mL min1, but the highest enhancement across flow rates is observed for 12.5 mM FeCN from 5 to 50 mL min1(See Table 7). This finding highlights the potential for achieving significant desalination of seawater using relatively low redox species concentrations, simply by adjusting flow rates at sufficiently high voltages. Based on these patterns, it can be inferred that increasing flow rates led to a reduction in Ercactolat high voltages by lowering energy losses caused by overpotentials, particularly for low redox species concentrations.

[0155] Table 7 : Summary of RFD desalination performance for brackish water (3000 ppm).

[0156] Additionally, the charge efficiency of desalination for each concentration in the RFD improved by roughly 10% with increasing flow rates and then plateaued at ~ 84%. The lack of high efficiency (> 90 %) may be attributed to the reduced selectivity of the ion exchange membranes (IEMS) at high salt concentrations (Kamcev, Physical Chemistry Chemical Physics, 2016. 18, 6021-6031). According to the Manning Model, the concentration of mobileAttorney Docket No.: 206256-0112-00 WOco-ions in the IEMS is expected to rise with higher salt concentrations (Maiming, The Journal of Chemical Physics, 1969, 57, 924-933). As such, tire selectivity to counter ion transport e.g.. Na+across CEM reduces as concentration of salt in the bulk solutions increases. This decrease in selectivity and increase in migration of co-ions in the IEMs are attributed to a decrease in tire electrical potential for repulsion by Doiman equilibrium due to charge screening. In this phenomenon, fixed charged counter-ions are replaced by co-ions that are absorbed at the electrolytemembrane interface, leading to a decrease in selectivity (Chen, Water Research, 2020, 175, 115681; Veerman, Journal of Membrane Science, 2009, 343, 7-15; Tedesco, Journal of Membrane Science, 2016, 510, 370-381). The observations supports that increasing flow rates at high voltages is relevant for increasing the charge efficiencies for seawater desalination.

[0157] In contrast, the present study demonstrated that increasing the flow rates had a marginal effect on the brackish water (3,000 ppm) desalination performance of the single-pass RFD when using 50 mM FeCN mixed with 3000 ppm NaCl (see FIG. 20 and FIG. 21; Table 7). Increasing flow rates resulted in improvement in ASRR values and charge efficiencies, while minimizing Ereactor, at each voltage. Briefly, increasing flow rate from 5 - 20 mL min-1merely enhanced ASRR by 1.28-fold (121.38 ± 1.98 — > 155.91 ± 2.27 pg cm'2min1) and reduced Exactor from 108.58 ± 1.09 to 104.00 ± 1.88 kJ mol1, while charge efficiency increased from 88.87 ± 0.893 to 92.803 ± 1.69%, even at an applied voltage of 1.0 V (see FIG. 21). Additionally, the charge efficiencies of the RFD were expectedly higher for brackish water desalination than for seawater desalination (refer to Table 7), as explained by Manning Model. For instance, charge efficiency values were above 88% for brackish water RFD. Regardless, the desalination performances did not improve after 20 mL min1, suggesting a plateau in performance enhancement offered by flow rates for brackish water desalination by RFDs.

[0158] Effects of electrolyte channel flow rates on the desalination of brackish water by RFD in single-pass mode: To understand the effects of electrolyte channel flow rates on brackish water desalination (3000 ppm NaCl solution), the single-pass RFD was operated with 50 mM FeCN. The redox species amount was expected to sufficiently drive ion-coupled electrochemical reactions for desalinating brackish water .The concentrations of salt in electrolyte and salt channels were matched to reduce concentration gradient across the IEMs (Luo, Nano Energy.2017, 42, 215-221). For precise conductivity measurements, the flow rate of the salt channels was maintained at 5 mL min-1(699.3 L m'2h'1). The flow rate of electrolyte channels varied from 5 to 30 mL min1. The applied reverse polarity voltages ranged from 0.2 to 1.0 V, chosen within the electrolytic voltage window for water splitting near pH neutral conditions(1.23 V). and thus minimizes current density contributions due to parasitic reactions.

[0159] the chronoamperometry and the salt concentration profiles are presented in FIG. 20. which demonstrated very high stability' for consecutive cycles. Generally, current density and salinity change strongly responded to increment applied voltages. Additionally, the profiles were comparable for salination and desalination steps. A negative applied voltage resulted in a negative current density (FIG. 20, left), resulting in a positive salinity change in CH 2 (FIG. 20, right). The lowest current densities and salinity changes were observed at 0.2 V and increased proportionally with increment voltages. Increasing flow rates of electrolyte channels resulted in increased current density and salt changes, particularly at higher voltages. At low voltage of 0.2 V, the current density andAttorney Docket No.: 206256-0112-00 WOconcentration profiles were nearly identical, suggesting that flow rate had a low effect on ion transport across the lEMs. However, above 0.4 V, an increasing response to flow rates was observed until 20 mL min-2. Particularly at 1.0 V, the disparities amongst concentration profiles were significant from 5 to 10 mL min1, marginal from 10 to 20 mL min1, and indistinguishable 20 mL-min1and 30 mL min1. This implies that the most drastic increase in salinity changes for brackish water was achieved by increasing flow rates of electrolyte channels from 5 to 10 mL min1.

[0160] Correspondingly, the calculated ASRR values and Ereactor increased strongly, while the charge efficiency decreased, with the incrementing applied voltages. Increasing flow rates resulted in improvement in ASRR values and charge efficiencies, while minimizingat each voltage (Table 7). Interestingly, increasing flow rates improved the linear correlation, along with a decreased slope, betweenand ASRR values (FIG. 21). This is the first reported case of this observation for RFD of brackish water. For example, when comparing performance of batch RFD of brackish using different concentration of FeCN and flow rates of 5 mL min1in all channels. A nonlinear relationships betweenand ASRR between applied voltage of 0.2 and 0.8 V. particularly with 50 and 100 mM FeCN had previously been observed, which was attributed to the saturation of electrochemical sites, which limited the performance of 100 mM FeCN. But as observed in FIG. 21, top left, increasing flow rate from 5 to 20 mL min-1resulted in a linear relationship, suggesting that low flow rate could contribute to the non-linear relationship and may be a consequence of mass transport limitations. As such, increasing flow rates resulted in desaturation of electrochemical sites.

[0161] A detailed analysis of the desalination performance at each voltage level is provided in FIG. 21. At 0.2 V and increasing flow rate from 5 mL min1to 20 mL min1, the ASRR values increased by 1.17-fold from 1. 29.51 ± 1.20 to 36.00 ± 1.59 pg cm'2min1(0.46 ± 0.02 — > 0.50 ± 0.05 mM). Accordingly, there was a slight decrease in energy consumption from 21.71 ± 0.73 to 20.11 ± 0.76 kJ mol1(2.58 ± 0.04 — > 2.82 ± 0.02 Wh nr3), while the charge efficiency substantially increased from 88.74 ± 2.92 to 97.04 ± 3.58 %. Likewise at 1.0 V, a 1.28-fold increase was observed in ASRR values (121.38 ± 1.98 — > 155.91 ± 2.27 pg cm'2min1) or ACR values of (1.78 ± 0.03 — > 2.31 ± 0.03 mM). These results imply a relevant need for increasing flow rate of electrolyte channels to improve performance of RFD of brackish water, especially at high voltages. Nevertheless, this improvement to desalination performance was limited after 10 mL min1and optimal performance can be achieved at 20 mL min1.

[0162] Effects of electrolyte channel flow rate on RFD mass transport limitations for seawater: Given the complex hydrodynamic effects of the observed results, it is imperative to gain a deeper understanding of the underlying mechanisms governing the response of RFDs to different electrolyte flow rate conditions. The variable effects of flow rate, voltage, and concentration on the desalination performance are intricately linked to the interplay of faradaic reactions and mass transports processes involving redox species and ions in the carbon felt electrodes and across the lEMs (Zhang, Batteries, 2009, 9, 17; Darling, Journal of The Electrochemical Society, 2014, 767. A1381-A1387). The flow rate of electrolytes and salt solutions induces changes in the mass transport characteristics of redox couple molecules and ions in the bulk media and at the interfaces. Consequently, impedance changes occur within the RFD (Joy. The Journal of Physical Chemistry C, 2021,725, 27556-27565). In order to gain deeper insights into the sources of overpotentials leading to desalination performance limitations, electrochemicalAttorney Docket No.: 206256-0112-00 WOimpedance spectroscopy was used, a powerful technique that provides spatial resolution of complex transport phenomena (Park. Journal of Colloid and Interface Science. 2006. 300, 655-662).

[0163] Nyquist plots illustrate how different flow rates affect the EIS spectra of 12.5 M and 50 mM FeCN within frequencies of 100 kHz and 1 mHz (FIG. 22 and FIG. 23). Notable decrease in impedance occur in the low-frequency region (< 1 Hz) of the impedance curve with increasing flow rates from 5 to 50 mL min1, as indicated by the reduction of the capacitive tail. Similar decrease to capacitive tail have been previously reported for cases of enhanced forced convection or increased electrolyte concentration at the membrane surface (Park, Journal of Colloid and Interface Science, 2006, 294, 129-138; Cazot, Electrochimica Acta, 2019, 321, 134705). The Bode plots (FIG.24 and FIG. 25) reveal three distinct frequency regions of phase shifts, also observable as the semi-circles in the Nyquist plots (refer to FIG. 26). The first phase shift region corresponds to the first semi-circle tire high-frequency regime ( ~100 kHz - 100 Hz), followed by the two later semi-circles at the medium-frequency (~100Hz - 2Hz) and the low-frequency regimes (~ 1Hz - ImHz), respectively (Leuaa, The Journal of Physical Chemistry C, 2019, 123, 21440-21447).

[0164] The series cell resistance ( 1 in FIG. 26), Rs. accounted for the resistance of the salt solutions, electrode, and membranes components (Trovb, Journal of Power Sources. 2021, 493. 229703). The high-frequency semi-circle (2 in FIG. 26), typically observed in electrochemical systems, represents the charge transfer resistance (RCT) and constant phase element CPE from the double layer (QDL). The medium-frequency (3 in FIG. 27). resembling a Warburg impedance in the Nyquist plot, is associated with the resistance and capacitance of convective mass transport (Rconv and Qconv) in electrolyte and membrane media. This region matches the frequency characteristics of flow -dependent impedance in porous electrodes reported previously. At the low-frequency region, the diffusion boundary layer, typically prevalent in low electrolyte concentrations, becomes negligible due to the high NaCl salt concentrations (-35000 ppm) in the electrolyte and salt channels. Consequendy, the low-frequency region (4 in FIG.26) is solely detennined by the resistance and capacitance of heterogenous transport at interface of electrolyte and membrane, resolved as, RHT and QHT. These parameters account for the finite mass transport of electrochemical species paired with ions transport in response to hydrodynamic conditions in the RFD resulting from flow rates. Therefore, increasing flow rates likely lead to significant reduction of the impedance at the membrane surface, improving heterogeneous transport and microfluidic motion of ions across the membrane interface.

[0165] The electrochemical impedance spectra were quantitatively evaluated prior to each desalination run in different electrolyte concentrations (refer to FIG. 19) using Z-fit-Biologic over frequency range of 100 kHz to lOmHz and summarized in Table 8. Fitting parameters for resistances obtained from the desalination runs are summarized in FIG. 27. Expectedly, the interfacial heterogenous resistances, RHT exhibited a significant dependence on flow rate increase, particularly at the low-frequency range (< 1 Hz). This is observed as the reduced impedance that was more pronounced at lower redox species concentration. The highest interfacial resistances observed were 24.98 and 5.57 at flow rate of 5 mL min-1for 12.5 (FIG. 16) and 50 mM FeCN (FIG. 17), respectively. These values reduced to their lowest values of 1.34 and 0.79 at 50 mL min1. The substantial drop in interfacial resistance for 12.5 mM (24.98 — > 1.3 ) from 5 to 50 mL min1provides compelling evidence of a significantAttorney Docket No.: 206256-0112-00 WOreduction in overpotential. This reduction correlates with the notable increase in ASRR values (48.48 — > 811.55 pg cm-2min1) and reduction in Ereactor (—145 — >~95 kJ mol'1) observed in FIG. 14. On the other hand, slight changes were noticed for the RCT and Rconv, which slightly decreased as concentration increased but showed limited response to flow rates increase. Therefore, the interfacial resistances to ion transfer at the electrolyte-membrane interface was the dominant contributor to increased impedance, resulting in overpotential. However, this can be reduced by increasing the flow rate of electrolyte channels.

[0166] Table 8: Equivalent circuit fitting parameters for single-pass RFD with different electrolyte concentrations, related to FIG. 19

[0167] Effects of flow rate disparity between electrolyte and salt channels on RFD performance: While maintaining an influent flow rate of 5 mL min1in CH 2 allowed facile salt concentration measurement, the effect of the flow rate disparity across tire electrolyte and salt membranes on Hie RFD system’s perfonnance is unclear. To further investigate this, EIS studies were performed on an RFD with 50 inM FeCN at different flow rates in salt and electrolyte channels. Nyquist and Bode plots of 50 mM FeCN for are provided different flow rates in salt and electrolyte channels (FIG. 28). Qualitatively analyzing these plots reveal that increasing the flow rates of the salt channels to match those of the electrolyte channels further reduces the impedance observed at low-frequency region (< 1 Hz). As discussed earlier, this is predominantly attributed to reduction of the clcctrolytc-mcmbranc intcrfacial resistance, i.c., RHT. Thus, a flow rate disparity between salt and electrolyte chamrcls could decrease the desalination performance of RFDs in single-pass mode. However, this decrease of impedance is minimal, when compared to the scenario where electrolyte channels were increased from 5 mL min1to 50 mL min'1and the salt channels maintained at 5 mL min1.

[0168] Since single-pass and batch RFDs have different desalination goals, it is also important to understand the effects that flow rate disparity has on their practical separation conditions for seawater. In a single pass, the desalination goal should be achieved after the effluent leaves the reactor, requiring a high average concentrationAttorney Docket No.: 206256-0112-00 WOreduction at a selected productivity. As such, increasing membrane active area and limiting the productivity to achieve greater average salt reduction is necessary for practical applications of single-pass mode for seawater desalination. For example, after two 45-minute cycles using 50 inM FeCN in the single-pass RFD at 30 mL min1, salt channels containing seawater at 1 mL min1(139.9 L m'2Ir1), and an applied voltage of 0.8 V, an average salt concentration reduction value of 55.24 ± 0.14 mM was observed, 3 ,6-fold greater than the value obtained at 5 mL miir1(699.3 L nr2h1) as well as an ASRR value of 752.49 ± 1.84 pg cnr2min1, and Ereactor value of 100.46 ± 0.54 kJ mol1(refer FIG. 29). Aside from possible inefficiencies resulting from the lack of reactor optimization, the minor drop in ASRR may be due to interfacial resistance at the lower productivity. This suggests that the trade-off of increased precision and average concentration reduction achieved with unequal flow rates in single-pass mode outweighs tire further improvement offered by reducing tire interfacial resistance achieved with equal flow rates, possibly due its symmetric nature. In this mode, there is a low potential difference across the reactor due to the constant concentration of influent from the large reservoirs that ensures a low concentration gradient across tire channels (Goulet, Journal of The Electrochemical Society, 2018, 165, A1466-A1477).

[0169] In a batch-mode configuration, on the other hand, the desalination goal is to reduce the concentration of a fixed volume of water to drinkability (~500 ppm). Additionally, it is possible to achieve precise salt measurement for a wide range of flow rates in batch mode due to the infinite rate of salt changes in the reservoirs. As such, reducing the salt chamrel flow rate to attain a high average salt concentration and improve analysis is not as important in batch as it is in single-pass. Also, the impact of flow rate disparity across the channels on desalination performance can significantly exacerbate the increasing specific area polarization from the growing difference in concentration across salt channels. The increasing the depth of salt removal during batch RFD operation leads to higher specific area polarization, resulting from salinity' disparities across the salt channels. This result was confirmed in a complementary batch-mode analysis of a 25 mM FeCN RFD (see FIG. 30). A significant drop in the ASRR (Eq. 2-lb) was observed for seawater desalination (35000 — > 350 ppm). For equivalent flow rates of 50|50 mL min1, the ASRR was 826.89 pg cm2min1(1000 min) at productivity of 12.03 L m2h1(Eq. 2-lb). Whereas for unequal flow rates of 50|5 mL min1, it was 591.2 pg cm-2min1(1500 min) at productivity of 8.01 L nr2h1. The actual water recovery was -43% in both separation conditions, possibly due to osmotic driven water transport across the salt and electrolyte channels (Berezina, Journal of Membrane Science, 2002, 209, 509-518).

[0170] Effects of uneven flow rates across electrolyte and salt channels on RFD performance in batch-mode:To demonstrate tire consequences of uneven flow rates in batch-mode, RFD was operated in batch-mode with 25 mM FeCN mixed in 50 mL seawater solution and two fixed-volume reservoirs of 100 mL containing seawater. The flow rates of the electrolyte channels were maintained at 50 mL min1, while that of salt channels were cy cled at either 5 or 50 mL min1(699.3 or 6993 L nr2Ir1). When an applying a voltage of 0.8 V to desalinate seawater (35000 350 ppm) in FIG. 29, the duration of desalination for uneven flow rate (50| 5 mL min1) was about 1.5-fold greater than for even flow rates. The ASRR values were 530.75 (-1500 min) and 826.89 pg cm-2min-1(-1000 min), respectively, for flow rates of 5 and 50 mL min’1in the salt channels. As noted, the latter ASRR value was comparable to that of single-pass RFD at 60 mL min1in electrolyte channels and 5 mL min1(see Table 6), suggesting that flow rate disparity in single-pass mode had limited impact on the ASRR. Comparatively, theAttorney Docket No.: 206256-0112-00 WOdesalination time observed in FIG. 29, left, were drastically impacted by the flow disparity between electrolyte and salt channels. As such, although flow rate disparity between channels can be useful in analysis of single-pass RFD, it can significantly decrease desalination performance in batch RFD by introducing interfacial resistances to iontransport. which is exacerbated by specific polarization resistances due to salt concentration difference across the RFD channels.

[0171] Other major consequences of uneven flow rates in RFD arc electrolyte crossover and fouling of AEMs. FeCN cannot only degrade to toxic acidic cyanide due to photodegradation or when in acidic media but can also foul the AEM. Notably, FeCN fouls AEMs due to the strong electrostatic attraction betw een the negatively charged surface of FeCN and the positively charged ionogenic groups present in Selemion® AMV. As such, AEMs in direct contact with FeCN solution are highly prone to fouling. For example, Kim et al observed significant fouling of Selemion® AMV by FeCN after 20 hours of batch RFD (Kim, Desalination, 2023, 550, 116406). In the RFD configuration described herein, where CEMs are in contact with electrolyte channels, FeCN crossover through Nafion® 212 is a prerequisite to salt channel contamination and fouling of the AEM. According to Porcellinis et al., 0.3M K3[Fe(CN)6] in IM KOH has a diffusion permeability co-efficient of 4.5xl0-9cm2s1through Nafion® 212 (De Porcellinis, J. Electrochem. Soc., 2018, 165, Al 137-Al 139). Fouling is expected to be influenced by redox species amount operational conditions (Gao, J. Power Sources, 2022. 527, 231180). For instance, the IEM images after 20 days of desalination runs (FIG. 30). Selemion® AMV in after desalination runs completed for FIG. 19 shows clearer visible fouling by FeCN than the AEM forthose in FIGs. 14 and 15. This indicates fouling was higher when using 50 mM FeCN than w hen using 25 mM FeCN. despite the higher flow rates and voltages applied in the latter.

[0172] It was initially unclear how uneven flow rates between electrolyte and salt channels influenced crossover, which may lead to fouling. Thus, to further understand this impact, a batch-RFD was operated with 50 mM FeCN mixed in 50 mL seawater and 160 rnL seawater reservoirs. The flow rates of the electrolyte channels were maintained at 30 mL min1electrolyte channels, while those of the salt channels were run at either 5 or 30 mL min1(699.3 or 4195.8 L nr2h1). After 5 days of operation at no applied voltage, higher FeCN crossover was observed for uneven flow rates than that for even flow rates (see FIG. 31). This is evident from the distinct absorbance peak of Fe[CN6]4' at about 215 nm as previously reported. When using flow rate of 5 mL min1in tire salt channels, tire estimated amount of FeCN were approximately 30 pM (CH 2) and 50 pM (CH 3). In contrast, at a flow rate of 30 mL min1, the estimated amounts were approximately 18 pM (CH 2) and 24 pM (CH 3). Clearly, flow rate disparity between the middle and electrolyte streams promoted electrolyte crossover. Surprisingly, the CH 3 showed a greater amount of FeCN than CH 2. In the absence of concentration gradient or applied potential, the observed variation in the quantity of FeCN present in the salt channels may be attributed to unaccountcd-for pressure drops across the IEMS, which could be influenced by the reactor’s flow-through design (see FIG. 2 and FIG. 8). There w as a negative pressure applied at the inlet of CH 1 from the pump (pull), while there was a positive pressure applied at the inlet of CH 4 (push). Despite IEMs being resistant to hydraulic pressure differences, such drops might impact the rates of FeCN crossover (Darling, I. Electrochem. Soc., 2016, 163, A5029-A5040; Darling, J. Electrochem. Soc., 2014, 161, A1381-A1387). As such, further research is required to understand this flow conditions and reactor design affectAttorney Docket No.: 206256-0112-00 WOcrossover and fouling phenomena in RFDs. Nevertheless, equivalent increase of flow rates in all channels can provide the optimal desalination performance of RFD, regardless of the cycling mode, and further research is required to develop practical methods to precisely measure salinity of salt channels in single-pass mode without sacrificing flow rates.

[0173] Additionally, another concern about uneven flow rates was its effects on electrolyte crossover and fouling. Aside concern of its toxicity due to the release of hydrogen cyanide from tire photodegradation in water, FcCN is also an AEM foulant (Kang, Environmental Engineering Research, 2022, 27, 210308-0; Gao, Journal of Power Sources, 2022, 527, 231180). From the observed visible signs for fouling after RFD experiments, fouling was strongly dependent on the redox species concentration in electrolyte channels (see FIG. 31). Differences with the visible signs of fouling between experimental runs were observed, indicating an inverse relationship of FeCN concentration to fouling, i.e., tire crossover and fouling was lower for 25 mM (FIG. 31, left) than for 50 mM (FIG.31. center). Further permeation investigations of the RFD in batch-mode using 50 mM FeCN, revealed that the flow rate disparity across the IEMS also promoted redox species crossover across Nafiori® 212, which could have eventually increased fouling of the Selemion® AMV (AEM). For example, after about 5 days of operating the RFD at flow rates of 30 and 5 mL min1in the electrolyte and salt channels (see FIG. 32), correspondingly, the estimated amount of FeCN observed in CH 2 was approximately 30 pM. Likewise, operating the RFD at equal flow rate of 30 mL min1in all channels, the estimated amount was 18 pM CH 2.

[0174] Based on these results, uneven flow rate across electrolyte and salt channels would be critical for precise experimental analysis and practical performance of the single-pass RFD but at the cost of increased electrolytemembrane interfacial resistance and the redox species crossover. This can be mitigated by using low redox species concentration, which also benefits from reduced interfacial resistance at higher flow rate of electrolyte channels. Furthermore, methods that adjust flow rates ratio of salt channels to further improve water recovery' are necessary for managing brine production, but should be carefully selected in order to minimize the interfacial resistances and redox species crossover characteristics resulting from flow conditions. Thus, these findings highlight the relevance of flow rate parametric investigations to improve performance of RFD, mitigating the need for costly methods of improving ionic flux such as increasing concentration of redox species and active area of electrode and IEMs.

[0175] In conclusion, reported is a parametric investigation of RFD in single-pass mode to show how the system’s analysis and performance for seawater feedwater are affected by different flow rate conditions, when varying ferri- / ferrocyanide (FeCN) concentrations and applied voltages. Increasing flow rates diminishes electrolyte-membrane interfacial resistances, promoting ionic flux across ion-exchange membranes. Interestingly, the RFD achieved a record-breaking improvement to the average salt removal rate of 16.7-fold (48.48 —> 811.55 pg cm-2min1) and energy consumption of (-145 — > -95 kJ mol1) with merely 12.5 mM FeCN when flow rates were increased (5 — > 50 mL min1) at 0.8 V and high desalinated water production rate (699.3 L nr2h1) and 50% water recovery. This remarkable performance is consequently due to the electrolyte-membrane interfacial resistance drop (-25.0 —> 1.3 ). which was verified with in-situ electrochemical impedance spectroscopy in single-pass mode. These findingsAttorney Docket No.: 206256-0112-00 WOprovide new prospects for practical operation of RFD systems to achieve high desalination performance with low material costs.

[0176] The disclosures of each and every patent, patent application, and publication cited herein are hereby incorporated herein by reference in dreir entirety. While this invention has been disclosed widr reference to specific embodiments, it is apparent that odier embodiments and variations of this invention may be devised by others skilled in the art without departing from the true spirit and scope of the invention. The appended claims arc intended to be construed to include all such embodiments and equivalent variations.

Claims

Attorney Docket No.: 206256-0112-00 WOCLAIMSWhat is claimed is1. A method of water desalination, comprising the steps of:providing a redox flow desalination system, said redox flow desalination system comprising:a first chamber comprising a first chamber input and a first chamber output, wherein the first chamber is electrically connected to a first electrode and configured to comprise an electrolyte material, wherein the first electrode undergoes a reversible redox reaction with the electrolyte material:a second chamber comprising a second chamber input and second chamber output, configured to receive water having a first salinity;a third chamber having a third chamber input and third chamber output, configured to receive the saline water;optionally comprising a first reservoir having a first reservoir input and a first reservoir output; wherein the first reservoir input is fluidly connected to the third chamber output; and wherein the first reservoir output is fluidly connected to the third chamber input;a fourth chamber comprising a fourth chamber input and fourth chamber output, wherein the fourth chamber is electrically connected a second electrode and configured to comprise the electrolyte material, wherein tire second electrode undergoes a reversible redox reaction with the electrolyte material;an electrolyte channel having an electrolyte channel input and an electrolyte channel output, wherein the electrolyte channel output is fluidly connected to the first channel input, the first channel output is fluidly connected to the fourth chamber input, and the fourth chamber output is fluidly connected to tire electrolyte channel input;a first semi-permeable membrane disposed between the first and second chambers and configured to permit ion flow betw een the first and second chambers;a second semi-permeable membrane disposed between tire second chamber and third chambers and configured to permit ion flow between the second and third chambers; anda third semi-permeable membrane disposed between the third chamber and the fourth chamber and configured to permit ion flow between the third and fourth chambers; adding water having a first salinity into the second chamber;adding water having a second salinity into the third chamber;adding an electrolyte material to the first chamber and the fourth chamber; andapplying a voltage across the first electrode and the second electrode; thereby decreasing the salinity of the water in the third chamber and increasing the salinity of the water in the second chamber;wherein the flow rate of water in the second chamber and in the first chamber is at least 25 mL / min; and wherein the flow rate of electrolyte material in the first chamber and the fourth chamber is at least 25 mL / min.Attorney Docket No.: 206256-0112-00 WO2. The method of claim 1, wherein the flow rate of water in the second chamber and in the first chamber is at least 30 mL / min; and wherein the flow rate of electrolyte material in the first chamber and the fourth chamber is at least 30 mL / min.

3. The method of claim 1, wherein the flow rate of water in the second chamber and in tire first chamber is at least 40 mL / min; and wherein the flow rate of electrolyte material in the first chamber and the fourth chamber is at least 40 mL / min.

4. The method of claim 1, wherein tire flow rate of water in the second chamber and in tire first chamber is at least 50 mL / min; and wherein tire flow rate of electrolyte material in the first chamber and the fourth chamber is at least 50 mL / min.

5. A system for redox flow desalination, wherein tire system comprises:a first reservoir having a first reservoir input and a first reservoir output;a second reservoir having a second reservoir input and a second reservoir output;a third reservoir having a third reservoir input and a third reservoir output;a first channel having a first charmci input and a first channel output;a second channel having a second channel input and a second channel output;a third channel having a third channel input and a third channel output;a fourth channel having a fourth channel input and a fourth channel output;a first electrode in electrical contact with the first channel; anda second electrode in electrical contact with the fourth channel;wherein the first channel is contacted with the second channel via first semi-permeable membrane; the second channel is contacted with the third channel via a second semi -permeable membrane; and the third channel is contacted with the fourth channel via third semi-permeable membrane; wherein the first reservoir output is fluidly connected to the second channel input through a first pump, and the first channel output is fluidly connected to the first reservoir input;the second fluid reservoir output is fluidly connected to the third channel output through a second pump, and the third channel output is fluidly connected to the second fluid reservoir input;wherein the third fluid reservoir output is fluidly connected to the first channel input, the first channel output is fluidly connected the fourth channel input through a third pump, and the fourth channel output is fluidly connected to the third fluid reservoir input;wherein the volume of the first reservoir is no more than 500 mL; the volume of the second reservoir is no more than 500 mL; and the volume of the third reservoir is no more than 500 mL.Attorney Docket No.: 206256-0112-00 WO6. The system of claim 5, wherein the volume of the first reservoir is no more than 400 mL; the volume of the second reservoir is no more than 400 mL; and the volume of the third reservoir is no more than 400 mL.

7. The system of claim 5, wherein the volume of the first reservoir is no more than 300 mL; the volume of the second reservoir is no more than 300 mL; and the volume of the third reservoir is no more than 300 mL.

8. The system of claim 5, wherein the volume of the first reservoir is no more than 250 mL; the volume of the second reservoir is no more than 250 mL; and the volume of the third reservoir is no more than 250 mL.

9. The system of claim 5, wherein the volume of the first reservoir is no more than 200 mL; the volume of the second reservoir is no more than 200 mL; and the volume of tire third reservoir is no more than 200 mL.

10. The system of claim 5, wherein tire volume of the first reservoir is no more than 150 mL; the volume of the second reservoir is no more than 150 mL; and tire volume of the third reservoir is no more than 150 mL.

11. The system of claim 5, wherein the volume of the first reservoir is no more than 100 mL; the volume of the second reservoir is no more than 100 mL; and the volume of the third reservoir is no more than 100 mL.

12. A method of redox flow desalination, tire method comprising the steps of:providing the system of claim 5;adding water having a first salinity into the first reservoir chamber;adding water having a second salinity into the second resen oir chamber;adding an electrolyte material to third reservoir;activating the first pump, the second pump, and the third pump;applying a voltage across the first electrode and the second electrode; thereby decreasing the salinity of the water in the second reservoir chamber and increasing the salinity of the water in the first reservoir;wherein the flow rate of water through the first pump is at least 25 mL / min; the flow of water through the second pump is at least 25 mL / min; and the flow of water through the third pump is at least 25 mL / min.

13. The method of claim 12, wherein the flow rate of water through the first pump is at least 30 mL / min; the flow of water through the second pump is at least 30 mL / min; and the flow of water through the third pump is at least 30 mL / min.

14. The method of claim 12, wherein the flow rate of water through the first pump is at least 40 mL / min; the flow of water through the second pump is at least 40 mL / min; and the flow of water through the third pump is at least 40 mL / min.Attorney Docket No.: 206256-0112-00 WO15. The method of claim 12, wherein the flow rate of water through the first pump is at least 50 mL / mim the flow of water through the second pump is at least 50 mL / mim and the flow of water through the third pump is at least 50 mL / min.

Images