Summary

Ion-Exchange Membranes for the Fabrication of Reverse Electrodialysis Device

Published: July 20, 2021
doi:

Summary

We demonstrate the fabrication of a reverse electrodialysis device using a cation-exchange membrane (CEM) and anion-exchange membrane (AEM) for power generation.

Abstract

Reverse electrodialysis (RED) is an effective way to generate power by mixing two different salt concentrations in water using cation-exchange membranes (CEM) and anion-exchange membranes (AEM). The RED stack is composed of an alternating arrangement of the cation-exchange membrane and anion-exchange membrane. The RED device acts as a potential candidate for fulfilling the universal demand for future energy crises. Here, in this article, we demonstrate a procedure to fabricate a reverse electrodialysis device using laboratory-scale CEM and AEM for power production. The active area of the ion-exchange membrane is 49 cm2. In this article, we provide a step-by-step procedure for synthesizing the membrane, followed by the stack's assembly and power measurement. The measurement conditions and net power output calculation have also been explained. Furthermore, we describe the fundamental parameters that are taken into consideration for obtaining a reliable outcome. We also provide a theoretical parameter that affects the overall cell performance relating to the membrane and the feed solution. In short, this experiment describes how to assemble and measure RED cells on the same platform. It also contains the working principle and calculation used for estimating the net power output of the RED stack using CEM and AEM membranes.

Introduction

Energy harvesting from natural resources is an economical method that is environmentally friendly, thereby making our planet green and clean. Several processes have been proposed until now to extract energy, but reverse electrodialysis (RED) has an enormous potential to overcome the energy crisis issue1. Power production from Reverse electrodialysis is a technological breakthrough for the decarbonization of global energy. As the name suggests, RED is a reverse process, where the alternate cell compartment is filled with the high-concentrated salt solution and low-concentrated salt solution2. The chemical potential generated by the salt concentration difference across the ion-exchange membranes, collected from the electrodes at the compartment end.

Since the year 2000, many research articles have been published, providing insight into the RED theoretically and experimentally3,4. Systematic studies on the operation conditions and reliability studies under stress conditions improved the stack architecture and enhanced the overall cell performance. Several research groups have diverted their attention toward RED's hybrid application, such as RED with desalination process5, RED with solar power6, RED with reverse osmosis (RO) process5, RED with the microbial fuel cell7, and RED with the radiative cooling process8. As mentioned earlier, there is a lot of scope in implementing RED's hybrid application to solve the energy and clean water problem.

Several methods have been adopted to enhance the RED cell's performance and the membrane's ion-exchange capacity. Tailoring the cation-exchange membranes with different types of ions using sulfonic acid group (-SO3H), phosphonic acid group (-PO3H2), and carboxylic acid group (-COOH) is one of the effective ways to alter the physicochemical properties of the membrane. Anion-exchange membranes are tailored with ammonium groups (Equation 1)9. The high ionic conductivity of AEM and CEM without deteriorating the membrane's mechanical strength is the essential parameter for selecting an appropriate membrane for device application. The robust membrane under stress conditions provides mechanical stability to the membrane and enhances the device's durability. Here, a unique combination of high-performance free-standing sulfonated poly (ether ether ketone) (sPEEK) as cation-exchange membranes with FAA-3 as anion-exchange membranes are used in the RED application. Figure 1 shows the flow chart of the experimental procedure.

Figure 1
Figure 1: Procedure chart. The flow chart presents the procedure adopted for the preparation of ion-exchange membrane followed by the process for measurement of reverse electrodialysis. Please click here to view a larger version of this figure.

Protocol

1. Experimental requirement

  1. Purchase ion-exchange ionomer polymer, E-550 sulfonated-PEEK polymer fiber to prepare CEM and FAA-3 to prepare AEM. Ensure that all ionomer polymers are stored in a clean, dry, and dust-free environment before use.
  2. Use high purity (>99%) solvents, including N-Methyl-2- pyrrolidone with molecular weight 99.13 g mol-1 and N, N-Dimethylacetamide with molecular weight 87.12, for preparing homogeneous ionomer solution. Ensure all analytical grade chemicals and solvents are used for membrane preparation as received without any further purification.
  3. After the membranes' activation process, immediately immerse all membranes in a 0.5 M NaCl solution for better performance. After activation of both membranes, drying is not required. Water with resistivity is 18.2 MΩ at room temperature was used throughout the synthesis of the membrane.
  4. Characterize membrane properties using a dry membrane. The detailed description of the characterization techniques and their physicochemical properties such as ion-exchange capacity, ion-conductivity, thickness, thermal analysis, and surface morphology, are as presented in the literatures10,11.
  5. Use a cutter to shape the membrane for CEM and AEM to the RED stack size with an active area of 49 cm2, as displayed in Figure 2.
  6. For the RED stack fabrication, make an alternate CEM and AEM arrangement, separated by spacer and gasket; a real picture of the working RED stack is presented in Figure 3a, and its schematic diagram of each layer is illustrated in Figure 3b.
    1. First, place the PMMA plate facing electrode upside; now, place the rubber gasket and spacer on it, then place the CEM. After that, place the silicone gasket with the spacer on the CEM then place the AEM on it. Similarly, add the silicon gasket and spacer on the top of AEM followed by CEM. Now, place the end PMMA plate, rubber gasket, and spacer followed with tightening using screw and nut bolts.
  7. After assembling the RED stack, check the free flow of the high-concentration (HC), low-concentration (LC), and rinse the solutions one by one. Any crossflow or leakage is required to be eliminated before the measurement.
  8. Prior to the current and voltage measurement, monitor the flow rate of salt solutions and pressure gauge reading and make sure it gets stabilized. Make sure all the connections are in the exact place before the measurement starts. Avoid touching the RED stack and its connecting tubes while the measurement is running.
    ​NOTE: HC and LC solution flow from their compartments to discard compartment through a peristaltic pump, pressure gauge, and RED stack, respectively.
  9. Use galvanostat method for the measurement of current and voltage, the source meter instrument connected to the RED stack through crocodile clips.

Figure 2
Figure 2: Size and shape of the prepared membrane, gasket, and spacer for the fabrication of reverse electrodialysis. (a) outer silicone gasket, (b) outer spacer and inner spacer, (c) inner silicone gasket, (d) cation-exchange membrane, (e) anion-exchange membrane, and (f) gasket and membrane assembly. Please click here to view a larger version of this figure.

Figure 3
Figure 3: Reverse electrodialysis stack. (a) setup of reverse electrodialysis stack with connecting tubes, and (b) schematic illustration of different layers, including PMMA endplates, electrodes, gasket, spacer, CEM, and AEM. Please click here to view a larger version of this figure.

2. Ion-exchange membrane preparation

NOTE: The amount of precursor material was optimized for obtaining a membrane with 18 cm diameter and ~50 µm thickness.

  1. Cation-exchange membrane
    1. Take 5 wt% of sulfonated-PEEK fibers in a 250 mL round bottom flask and dissolve the fibers in Dimethylacetamide (DMAc) as a solvent having molecular weight 87.12 g mol-1. Shake the flask for 10 min so that all ionomer polymers settle down.
    2. Place a magnetic bar in the flask and then keep the mixture in the silicon oil bath, followed by vigorously stirring at 500 rpm for 24 h at 80 °C to obtain a homogenous solution.
    3. Filter the sulfonated-PEEK solution through a 0.45 µm pore size Polytetrafluoroethylene (PTFE) filter.
    4. After that, pour the filtered solution onto a circular glass dish with a diameter of 18 cm. Ensure that all air bubbles are removed using an air blower before placing the Petri dish in the oven.
    5. Place the Petri dish inside an oven for drying out the solution at 90 °C for 24 h, resulting in ~50 µm thick free-standing membrane. Do this for extracting free-standing membrane: To peel off the membrane from the Petri dish, fill the Petri dish with warm distill water (~60 °C) and let it stand for 10 min untouched. The free-standing membrane will automatically come out.
    6. For membrane activation, immerse the prepared free-standing membrane in 1 M sulfuric acid (H2SO4) aqueous solution, i.e., 98.08 g, in 1 L of distilled water, and incubate for 2 h at 80 °C.
      NOTE: This step will ensure the removal of foreign particles and other chemicals such as solvents that will reduce the possibility of membrane from fouling.
    7. Wash the soaked membrane with 1 L of distilled water for 10 min, at least three times at room temperature.
  2. Anion-exchange membrane
    1. Dissolve FAA-3 ionomer solution 10 wt.% in N-Methyl-2-pyrrolidone (NMP) solvent.
    2. Keep the solution for stirring at room temperature for 2 h at ~500 rpm.
    3. After that, filter the solution using the mesh with 100 µm pore size.
    4. Pour ~30 mL filtered solution into a circular glass Petri dish with a diameter of 18 cm. Ensure that all air bubbles were removed using an air blower before placing the glass Petri dish in the oven. The drying process takes place at 100 °C for 24 h.
    5. To obtain a free-standing membrane, pour hot distilled water into the glass Petri dish and keep it for at least 10 min. Now peel off the membranes and place in 1 liter of sodium hydroxide (NaOH) solution (concentration 1M and molecular weight 40 g mol-1) for 2 h.
    6. Then, wash the membrane thoroughly with 1 L of distilled water for 10 min, at least three times in ambient condition.
      ​NOTE: All prepared membranes were stored in the 0.5 M NaCl solution overnight before using it in the RED stack. So that the membrane conductivity gets enhanced and can achieve stabilized output performance during the measurement of the RED stack. Table 1 describes the membrane properties10,11.
Specification Unit CEM AEM
Swelling degree  % 5±1 1±0.5
Charge density or Ion exchange capacity  meq/g 1.8 ~1.6
Mechanical properties
(Tensile strength)
MPa >40 40-50
Elongation to Break  % ~42 30-50
Young Modulus (MPa) 1500±100 1000-1500
Conductivity at room temperature S/cm ~0.03 ~0.025
Permselectivity % 98-99 94-96
Thickness μm 50±2 50±3
Solvent Dimethylacetamide (DMAc) N-methyl-2-pyrrolidone (NMP)

Table 1: Membranes properties. Summary of both cation-exchange and anion-exchange membrane properties.

3. Fabrication of reverse electrodialysis

  1. Assembly of RED stack
    1. Prepare a model solution using 0.6 M NaCl for high concentration (HC) and 0.01 M NaCl for low concentration (LC) compartments12.
      NOTE: Here, river water is considered a low concentration salt solution, and seawater is represented as a high concentration salt solution.
    2. Prepare 5 L of high concentration and low concentration solution in a large container connected with the tubes. Keep the solutions stirring at ambient conditions (room temperature) for at least 2 h before it is used in the RED stack.
    3. Prepare the mixture of 0.05 M of [Fe (CN)6]-3/ [Fe (CN)6]-4 and 0.3 M NaCl in 500 mL water as a rinse solution for RED.
    4. Connect all three solution containers with RED stack using rubber tubes through the peristaltic pump and pressure gauges. Use the tube of size L/S 16 for rinse solution, and use the tube of size L/S 25 for HC and LC solution.
    5. To make a RED stack, take two endplates made-up of polymethyl methacrylate (PMMA). Connect both endplates horizontally face to face with nuts, bolts, and washers using 25 Nm force using a digital wrench driver. The thickness of PMMA endplates 3 cm, and the path of the flow channels was designed in plates for HC, LC, and rinse solution by a driller2.
    6. Place two mesh electrodes made from metal Titanium (Ti) coated with a mixture of Iridium (Ir) and Ruthenium (Ru) in a 1:1 ratio and place at the end of the PMMA plates. Both end electrodes are connected with the crocodile clip of the source meter.
      ​NOTE: Both PMMA end plates are equipped with mesh electrodes, both electrodes were layered with a square shape spacer, and the PMMA endplate covered with a rubber gasket facing inside. After that, CEM and AEM are placed alternatively, separated by silicone gasket and spacer, as shown in Figure 3.
    7. Install silicon gaskets, polymer spacers, and ion-exchange membranes (CEM and AEM) layer by layer, as presented in the schematic diagram Figure 4 and Figure 5. Ensure the active area of electrodes, both membranes, outer and inner spacer, outer and inner gasket is 7 x 7 = 49 cm2.
    8. Pass high-concentration and low-concentration solutions from respective compartments by peristaltic pumps, as displayed in the schematic diagram in Figure 4.
    9. Circulate the rinse solution in the outer electrode and membrane compartments in recirculation mode using peristaltic pumps. The flow rate used for the rinse solution is 50 mL min-1.
    10. Fixed flow rate is used for analyzing the performance of each membrane. In this experiment, we have used 100 mL min-1 through a peristaltic pump.

Figure 4
Figure 4: Schematic representation of the tube connection with reverse electrodialysis stack. Connection of reverse electrodialysis with peristaltic pumps, high-concentration solution container, low-concentration solution container, rinse solution container, and discard solution container. It also shows the spacer's alignment with both an anion exchange membrane (AEM) and cation exchange membrane (CEM). Please click here to view a larger version of this figure.

Figure 5
Figure 5: Schematic diagram of different layers in the reverse electrodialysis setup. (a) Cross-section view of a schematic illustration of reverse electrodialysis shows the flow direction of the high-concentration solution, low-concentration solution, and electrode rinse solution. Other components such as electrodes, outer and inner gaskets, outer and inner spacers, cation-exchange membrane, and anion-exchange membrane. (b) Front view of the stack, which shows the flow direction of a solution. Please click here to view a larger version of this figure.

4. Measurement of reverse electrodialysis

  1. Power calculation
    1. Let the high concentration, low concentration, and rinse solution, run through the stack at least for 5 min. Measure the RED output performance by a source meter, which is connected to both electrodes of the RED stack13.
    2. Calculate the RED stack's current-voltage characteristics in terms of power density using the galvanostat method.
      NOTE: In the galvanostat method, a constant current is applied across electrodes and measures the resulting current. The resulting current is the current generated due to the electrochemical reaction in the stack. The measurement is carried out under 0.05 V static voltage with a fixed sweep current that is 10 mA.
    3. The maximum power density for the RED stack is measured with the help of the following equation 1.
      Equation 2 (1)
      Here, Pmax is the maximum power density of the RED stack (Wm-2), Ustack is the voltage (V) produced by the membrane in the stack, Istack is the recorded current (A), and Amem is the active area of the membranes (m2).

Representative Results

Net power output
RED cell generally generates electrical energy from the salinity gradient of the salt solution, i.e., ions' movement in the opposite direction through the membrane. To assemble the RED stack correctly, one needs to align all the layers, including electrodes, gaskets, membranes, and spacers in the stack carefully, as demonstrated in the schematic diagram in Figure 4 and Figure 5. If the stack is not perfectly aligned, two problems may arise: (i) HC and LC solution crossflow may occur in the stack and (ii) leakage of the solution in the stack may occur. It is necessary to eliminate both the problems before starting the actual measurement of power output. Other parameters need to be fixed, including the HC and LC solution's flow rate, pumping pressure, and applied voltage, to obtain efficient power output. To estimate the RED stack's net power, one needs to deduct the hydrodynamic power loss from obtained net power10. The maximum power output is obtained from the RED stack by multiplying the obtained voltage and current. In contrast, the active area and number of the membrane pairs must be divided to obtain the stack's actual power density, as given by equation 114,15. The total power obtained from the RED stack is subtracted by a hydrodynamic power loss or pumping power loss generated by the pump and given by the following equation 2.

Equation 3(2)

Here, Ploss is a hydrodynamic pumping power loss (W m-2) produced in the RED stack by internal loss. Pmax is the maximum power (W m-2) obtained from the experiment. The highest net power output reported for RED is 1.2 W m-2 using river water and seawater by Vermaas16. Power loss is represented as a difference of pressure at inlet and outlet of HC and LC solution at the stack and given by pressure drop (ΔP), flow rate (Q), and pump efficiency (ηpump)17,18.

Equation 4 (3)

Here, QH and QL are the flow rate (mL mim-1) of a high-concentration solution and low concentration solution in mL min-1 and ΔPH and ΔPL is the pressure drop at the high -concentration side and low concentration compartment in Pa. Here, the measured pressure drop from the pressure gauge for the HC compartment is 11,790 Pa and LC compartment is 11,180 Pa. The calculated pumping power loss (Ploss) is 0.038 W m-2.

Theoretical parameter estimation
Basically, A RED system is made up of two different types of ion-exchange membranes, gasket, pump, spacers, and electrode. The pressure drop across the RED stack is estimated theoretically using the Darcy-Weisbach equation11,19. In an ideal RED system, a laminar flow of solution in an infinite wide uniform channel is used for calculating the pressure drop.

Equation 5 (4)

Here, dh (m) is the channel's hydraulic diameter, whereas the hydraulic diameter for an infinite wide channel is 2h. Other parameters is the viscosity of water (Pa·s), tres is the residence time (s), L is the length of the membrane (cm). In RED stack, sPEEK as CEM and FAA-3 as AEM is used, and the distance between both membranes is given by the term b, which is directly proportional to the hydraulic diameter's value in the case of the profiled membrane, and "h" is the intermembrane distance (m), is given by equation 520.

Equation 6 (5)

For an infinite wide channel, the value calculated from equation 6 is usually much lower than the finite wide channel's value. The values obtained are low in magnitude, which is due to the non-uniformity of inlet and outlet of feed solutions. The spacer mesh restricts the flow of aqueous salt solutions due to the spacer shadow effect, resulting in an increment in pumping power. Placing the value obtained from the ratio of surface to volume (Ssp / Vsp) of spacer mesh in the formula, ε is the porosity, one can estimate the thickness of spacer-filled channels from equation 621,22.

Equation 7 (6)

The spacer thickness and the other parameters, including open ratio, mesh opening, and wire diameters, are kept constant in all the compartments. Both HC and LC compartments used the same solution (NaCl) with different concentrations. Therefore, it is easy to initialize the parameters, and theoretical pumping loss can be given by equation 723.

Equation 8 (7)

Where, A is the active membrane area in m2 and Q feed solution flow rate in m3 s-1. Here, μ is the viscosity of water measured in Pa·s, L is the length of membrane given by cm, and tres is a residence time in second.

The performance of the RED stack
The RED stack's output performance was investigated using one cell pair at a fixed flow rate of 100 mL min-1. The feed solution's concentration was also kept fixed for a higher concentration (0.6 M), and a lower concentration (0.01 M) prepared from NaCl salt. It is observed that the maximum power density is 0.69 W m-2 at 100 mL min-1, and the net power density is 0.66 W m-2 as presented in Figure 6. Higher flow rate and high ion-exchange capacity play a significant role in obtaining better cell performance because ions' transport is more active at a higher flow rate. On the other hand, it decreases the diffusion-boundary-layer resistance at the interface. The difference in the salinity gradient of the salt concentration gives rise to the open-circuit voltage, as illustrated in Figure 6. This voltage depends on the internal resistance of the RED stack and other parameters. It is noted that as the current density increases, the voltage start decreases whereas, initially the power density of the cell increases obtaining maxima at a certain current density value and then drops down. This decrease in the power density is due to an increase in the stack's internal resistance, as shown in Figure 6.

Figure 6
Figure 6: Output performance of the reverse electrodialysis device: (a) variation of output voltage with varying current, and (b) net power density with a varying current density of the RED stack. Please click here to view a larger version of this figure.

Discussion

The RED's working principle is mainly dominated by the membrane's physicochemical properties, which is a crucial part of the RED system, as illustrated in Figure 3. Here, we describe the fundamental characteristics of the membrane for delivering a high-performance RED system. Membrane's specific ion permeability makes it pass one type of ions through their polymer nanochannel. As the name suggests, CEM can pass cation from one side to another and restricts anion, whereas AEM can pass anion and restricts cation. As shown in Figure 2, all membranes were shaped into a RED stack size containing inlet and outlet passage for flow solution. The amount of ion exchanged through the membrane is directly proportional to the membrane's conductivity and, therefore, the power output of the stack24. The movement of ions in the ion-exchange membrane works on the Donnan exclusion principle25. The charge group attached with the polymer backbone repels the same charge present in the solution. Thus, higher the charge density greater will be the repulsion, which usually depends on the perm selectivity. Generally, in RED cells, ions' movement takes place through the membrane from higher concentration to lower concentration of the solution. This ion transport from one compartment to another through the membrane gives an open circuit voltage and current values, which is used to calculate the net power output of the cell26.

The RED stack's performance mainly depends on the ion exchange capacity and swelling density of CEM- and AEM-based membranes27. It is observed that the higher the ion-exchange capacity of the CEM and AEM, the better is the conductivity. However, the higher ion-exchange capacity of the membrane leads to high swelling, easily deteriorating the membrane's mechanical strength. Thus, it is essential to optimize swelling density and the membranes' conductivity for better and reliable cell performance. On the other hand, it is also crucial to optimize the stack resistance with the function of the feed solution's flow in both the compartments. As the flow rate increases, the stack resistance decreases, and the output cell performance increases. Theoretically, RED stack resistance is given by equation 8.

Equation 9 (8)

N is the number of cell pairs (alternate arrangement of anion- and cation-exchange membranes), A is the effective area of both the membranes (m2), RA is the anion exchange membrane resistance (Ω m2), RC is the cation exchange membrane resistance (Ω m2), dc is the thickness of the compartment with the concentrated solution (m), kc is its ionic conductivity (S m-1 ), dd is the thickness of the compartment with the diluted solution (m), kd is its ionic conductivity (S m-1), and Re is the electrode resistance (Ω). Reducing the stack resistance is an essential factor for enhancing the net output power, but other factors also influence the cell performance28, which also need to be considered. The spacer shadow effect, the flow of feed solution, compartment width, and concentration of feed solution, the schematic illustration of the RED cell are presented in Figure 5.

In RED cells, the membrane acted as a limiting factor and required a stable high conducting membrane. Apart from that, both CEM and AEM are required to have comparable ion-conducting properties so that the cell can produce an efficient and optimized power output. Degradation of ion-exchange capacity and salt accumulation also need to be taken into account for reliable RED performance. Novel membrane material and state-of-the-art device architecture may further improve cell performance in the coming future and will pave a path for future research direction.

Offenlegungen

The authors have nothing to disclose.

Acknowledgements

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MEST) (No. NRF-2017R1A2A2A05001329). The authors of the manuscript are grateful to the Sogang University, Seoul, Republic of Korea.

Materials

AEM based membrane Fumion P1810-194 Ionomer
CEM based membrane Fumion E550 Ionomer
Digital torque wrench Torqueworld WP2-030-09000251 wrench
Labview software Natiaonal Instrument Software
Laptop LG PC
Magnetic stirrer Lab Companion MS-17BB
N, N-Dimethylacetamide Sigma aldrich 271012 Chemical
N-Methyl-2- pyrrolidone Daejung 872-50-4 Chemical
Peristaltic pump EMS tech Inc EMP 2000W
Potassium hexacyanoferrate(II) trihydrate Sigma aldrich P3289 Chemical
Potassium hexacyanoferrate(III) Sigma aldrich 244023 Chemical
Pressure Gauge Swagelok Guage
Reverse electrodialysis setup fabricated in lab Device
RO system pure water KOTITI Water
Rotary evaporator Hitachi YEFO-KTPM Induction motor
Sodium Chloride Sigma aldrich S9888 Chemical
Sodium Hydroxide Merk 1310-73-2 Chemical
Source meter Keithley 2410
Spacer Nitex, SEFAR 06-250/34 Spacer
Sulfuric acid Daejung 7664-93-9 Chemical
Tube Masterflex tube 96410-25 Rubber tube

Referenzen

  1. Dlugolecki, P., Gambier, A., Nijmeijer, K., Wessling, M. Practical potential of reverse electrodialysis as process for sustainable energy generation. Environmental Science & Technology. 43, 6888-6894 (2009).
  2. Kim, D., Kwon, K., Kim, D. H., Li, L. . Energy Generation Using Reverse Electrodialysis: Principles, Implementation, and Applications. , (2019).
  3. Mei, Y., Tang, C. Y. Recent developments and future perspectives of reverse electrodialysis technology: A review. Desalination. 425, 156-174 (2018).
  4. Yip, N. Y., Brogioli, D., Hamelers, H. V. M., Nijmeijer, K. Salinity gradients for sustainable energy: primer, progress, and prospects. Environmental Science & Technology. 50, 12072-12094 (2016).
  5. Li, W., et al. A novel hybrid process of reverse electrodialysis and reverse osmosis for low energy seawater desalination and brine management. Applied Energy. 104, 592-602 (2013).
  6. Brauns, E. Salinity gradient power by reverse electrodialysis: effect of model parameters on electrical power output. Desalination. 237, 378-391 (2009).
  7. Cusick, R. D., Kim, Y., Logan, B. E. Energy capture from thermolytic solutions in microbial reverse-electrodialysis cells. Science. 335, 1474-1477 (2012).
  8. Kim, D. H., Park, B. H., Kwon, K., Li, L., Kim, D. Modeling of power generation with thermolytic reverse electrodialysis for low-grade waste heat recovery. Applied Energy. 189, 201-210 (2017).
  9. Hong, J. G., et al. Potential ion exchange membranes and system performance in reverse electrodialysis for power generation: A review. Journal of Membrane Science. 486, 71-88 (2015).
  10. Choi, S. -. Y., et al. Controlling fuel crossover in open electrochemical cells by tuning the water nanochannel for power generation. ACS Sustainable Chemistry & Engineering. 8, 8613-8623 (2020).
  11. Shah, S. A., et al. Modified single-wall carbon nanotube for reducing fouling in perfluorinated membrane-based reverse electrodialysis. International Journal of Hydrogen Energy. 45, 30703-30719 (2020).
  12. Kwon, K., Han, J., Park, B. H., Shin, Y., Kim, D. Brine recovery using reverse electrodialysis in membrane-based desalination processes. Desalination. 362, 1-10 (2015).
  13. Kwon, K., Park, B. H., Kim, D. H., Kim, D. Parametric study of reverse electrodialysis using ammonium bicarbonate solution for low-grade waste heat recovery. Energy Conversion and Management. 103, 104-110 (2015).
  14. Hatzell, M. C., Ivanov, I., Cusick, R. D., Zhu, X., Logan, B. E. Comparison of hydrogen production and electrical power generation for energy capture in closed-loop ammonium bicarbonate reverse electrodialysis systems. Physical Chemistry Chemical Physics. 16, 1632-1638 (2014).
  15. Zhu, X. P., He, W. H., Logan, B. E. Reducing pumping energy by using different flow rates of high and low concentration solutions in reverse electrodialysis cells. Journal of Membrane Science. 486, 215-221 (2015).
  16. Vermaas, D. A., Saakes, M., Nijmeijer, K. Doubled power density from salinity gradients at reduced intermembrane distance. Environmental Science & Technology. 45, 7089-7095 (2011).
  17. Veerman, J., Saakes, M., Metz, S. J., Harmsen, G. J. Reverse electrodialysis: Performance of a stack with 50 cells on the mixing of sea and river water. Journal of Membrane Science. 327, 136-144 (2009).
  18. Veerman, J., Saakes, M., Metz, S. J., Harmsen, G. J. Electrical power from sea and river water by reverse electrodialysis: a first step from the laboratory to a real power plant. Environmental Science & Technology. 44, 9207-9212 (2010).
  19. Batchelor, C. K., Batchelor, G. K. . An Introduction to Fluid Dynamics. , (2000).
  20. Schock, G., Miquel, A. Mass transfer and pressure loss in spiral wound modules. Desalination. 64, 339-352 (1987).
  21. Da Costa, A. R., Fane, A. G., Wiley, D. E. Spacer characterization and pressure drop modelling in spacer-filled channels for ultrafiltration. Journal of Membrane Science. 87, 79-98 (1994).
  22. Vermaas, D. A., Veerman, J., Saakes, M., Nijmeijer, K. Influence of multivalent ions on renewable energy generation in reverse electrodialysis. Energy & Environmental Science. 7, 1434-1445 (2014).
  23. Vermaas, D. A., Saakes, M., Nijmeijer, K. Enhanced mixing in the diffusive boundary layer for energy generation in reverse electrodialysis. Journal of Membrane Science. 453, 312-319 (2014).
  24. Moreno, J., Grasman, S., van Engelen, R., Nijmeijer, K. Upscaling reverse electrodialysis. Environmental Science & Technology. 52, 10856-10863 (2018).
  25. Sarkar, S., SenGupta, A. K., Prakash, P. The donnan membrane principle: opportunities for sustainable engineered processes and materials. Environmental Science & Technology. 44, 1161-1166 (2010).
  26. Kim, H. -. K., et al. High power density of reverse electrodialysis with pore-filling ion exchange membranes and a high-open-area spacer. Journal of Materials Chemistry A. 3, 16302-16306 (2015).
  27. Długołęcki, P., Nymeijer, K., Metz, S., Wessling, M. Current status of ion exchange membranes for power generation from salinity gradients. Journal of Membrane Science. 319, 214-222 (2008).
  28. Geise, G. M., Curtis, A. J., Hatzell, M. C., Hickner, M. A., Logan, B. E. Salt concentration differences alter membrane resistance in reverse electrodialysis stacks. Environmental Science & Technology Letters. 1, 36-39 (2014).

Play Video

Diesen Artikel zitieren
Singh, R., Hong, S. H., Kim, D. Ion-Exchange Membranes for the Fabrication of Reverse Electrodialysis Device. J. Vis. Exp. (173), e62309, doi:10.3791/62309 (2021).

View Video