Reverse Electrodialysis: Optimizing Per-formance in Up-Scaled Systems

Reverse Electrodialysis (RED)<!–[if supportFields]> XE “Reverse Electrodialysis (RED)” <![endif]–><!–[if supportFields]><![endif]–> is a salinity<!–[if supportFields]> XE “salinity” <![endif]–><!–[if supportFields]><![endif]–> gradient power (SGP) process whereby electrical power is produced from the reversible mixing of waters that have different salinity concentrations, such as river and sea water. This technology has promise as a future source of clean and sustainable energy, with an estimated global potential of 2.6 TW for all forms of SGP. However, previous research has been done on a relatively small scale. In order to create commercially viable RED power plants, researchers must first determine how to maximize the performance of large cell stacks. When trying to solve this problem, Veerman et al. 2010 found that there exists a tradeoff between the hydrodynamic and electrical requirements of spacers. In a RED system, spacers are open structures which separate alternately stacked cation (CEM) and anion exchange membranes (AEM), provide stack stability, and increase turbulence within the compartments. In attempting to maximize performance, Veerman et al. also considered other parameters such as, flow direction, residence time, flow velocity, and electrode segmentation. Their research is significant because it provides a “[f]irst [s]tep” towards the goal of producing commercial electricity from a RED power plant.—Juliet Archer

J. Veerman and colleagues, from Wetsus (Centre of Excellence for Sustainable Water Technology) and the University of Groningen, compared the performance of small and large laboratory RED stacks. In these stacks, power is generated by the potential difference between sea and fresh water over a membrane and the movement of ions through that membrane. The “large” stacks contained either 25 or 50 cells each, while the “small” stacks consisted of 50 cells each. Cell dimensions of the small stacks, which have been the focus of previous research, measure less than 10 by 10 cm2. These small stacks have a total active membrane area of 1 m2. In this study, Veerman et al. also utilized larger stacks with cell dimensions of 25 by 75 cm2 and total active membrane areas of either 9.4 or 18.75 m2. With these large RED stacks, the researchers also determined the impact of others parameters, when attempting to maximize performance at the lowest possible investment and operational costs. The researchers employed NaCl and hexacyanoferrate electrode rinse solutions, depending on which parameter they were testing. In using these applied electrode solutions, the study drifts from its focus on maximizing the performance of commercial SGP technology because these solutions are only used in laboratories. More advanced systems must be employed in RED power plants. For “sea water” they used a 30 g NaCl/L solution and for “river water” they used a 1 g NaCl/L solution. 
The authors found that hydrodynamic power losses are greatest at high flow rates. On the contrary, they also found that losses from co-ion transport and osmosis were significant at very low flow rates. Therefore, net power density (W/m2) and energy efficiency<!–[if supportFields]> XE “energy efficiency” <![endif]–><!–[if supportFields]><![endif]–> are maximized at optimal rather than maximal or minimal flow rates. In studying the effect of residence time on power density, Veerman et al. used Fumasep and Qianqiu membranes, cross-, co-, and counter-current flow directions, and 25 and 50 cell large stacks. The authors found that the generated electricity of stacks was mostly independent from the aforementioned parameters. Similarly, the number of cells had no effect on power density. This signifies that losses from shortcut currents, which increase as the number of cells increase, are minimal. Co-current and counter-current operations were also tested, to determine which mode is more efficient in RED stacks. In counter-current mode, river water flows downward and sea water flows upward.  However, in co-current mode both river and sea water flow upward. Contrary to evidence from other processes, the authors found that co-current mode resulted in a higher power density within the RED stacks. The researchers explained this surprising discovery based on the competing effects of a high potential difference near the inlet side and a high conductivity near the outlet side. Furthermore, the authors speculated that co-current operations minimize the pressure within the compartments and thus minimize leakage. Lastly, thin, delicate membranes and open spacers, both of which maximize power density, can be used with co-current operations.

Veerman et al. also considered the effect of electrode segmentation on generated power (W). The power of a segmented stack was found to be 11% more than that of an unsegmented stack. However, segmentation may not be practical in actual RED power plants because it is probable that segmentation’s small advantage disappears at high flow rates. Furthermore, segmentation requires the use of complicated and costly electronics, which may reduce its theoretical benefits.  The authors also measured the pressure from fluid resistance in the manifolds, bore holes, and compartments—around the supply and drain holes and where uniform flow exists—in order to calculate hydrodynamic losses. Fluid resistance in the manifolds and bore holes resulted in negligible and very low losses, respectively. Around the outlet and inlet holes, on the other hand, resistance was very high and these areas accounted for the majority of fluid resistance within the system. When graphed against flow velocity (cm/s), uniform spacer resistance for horizontal and vertical operation in co-current mode had approximately equal slopes.  To decrease fluid resistance, the authors recommend that more inlet and outlet places be created and that very open spacers be used around supply and drain holes. The researchers conclude by recommending that future RED designs utilize very open spacers, co-current operation, and very thin membranes.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s