Effect of Electrode Structure and Electrolyte Flow on Performance of Redox Flow Battery

As a storage battery suitable for leveling output of renewable energy, a redox flow battery with features such as long life and high safety has attracted attention. Since a redox flow battery uses a pump for the electrolyte circulation, it is necessary to supply the electrolyte with keeping the pressure drop as small as possible. Mench et al. reported that the interdigitated design (IDD) of electrolyte flow fields can maintain lower pressure drop than the conventionally-used flow through design (FTD) (1). Some studies have reported that carbon paper electrodes show better cell performance with the serpentine design of electrolyte flow fields than carbon felt electrodes that have been widely used until now (2). In this study, by using the carbon felt electrode, I-V performance and pressure loss with two kinds of flow designs, the IDD and the FTD, were compared, and the influence of the flow designs on the cell performance was examined. In addition, the performance for the IDD was also compared with that using carbon paper electrode, and the influence of type of porous carbon electrode was investigated in the IDD. Furthermore, we tried to improve the cell performance by applying the combination of carbon felt and carbon paper electrodes with the IDD. Figure 1 shows schematic views of the FTD and the IDD. In the FTD, the electrolyte flows uniformly through the porous electrode, as shown in Figure 1(a), and reacts on the surface of carbon electrode. In the IDD, the electrolyte flows from the inlet flow path of the channel, dives into the porous electrode, and exits to the outlet flow path, as in Figures 1(b) and (c). In this study, two kinds of electrodes, carbon felt and carbon paper, were used, and the reaction area was 21.6cm 2 (4.6cm × 4.7cm). The carbon felt with a thickness of 3.1mm was compressed to 2.0 mm, and four carbon papers with the thickness of 1.1mm (0.28mm × 4) were compressed to 0.85mm. Figure 2 shows the I-V characteristics with the FTD and the IDD. The SOCs are 40 and 80%, and the flow rate is 30mL/min. It was confirmed that the performance with the FTD is higher than that with the IDD with the same flow rate. Figure 3 shows the relationship between the pressure drop and the flow rate of the electrolyte using various flow fields. In the case of the FTD, the pressure loss for the carbon paper electrode with lower porosity becomes larger than that of the carbon felt electrode. In the case of the IDD, small pressure loss can be maintained regardless of the type of electrode. The IDD is suitable for upsizing because of small pressure loss, but the I-V characteristics with the IDD is inferior to that with the FTD with the same flow rate of the electrolyte, as described above. Therefore, it is necessary to improve the I-V performance of the IDD with keeping the pressure loss low. In this study, we tried to optimize the electrode structure by combining the two types of electrodes having opposite properties. The carbon paper electrode having a low porosity has an advantage that the surface area can be secured even if it is thin. On the other hand, the carbon felt electrode having a small pressure loss has an advantage that it can suppress the pump power. Figure 4 shows results of the I-V measurement with the IDD using four types of electrode structures. In addition to the previous structures, the combination of carbon paper and felt, and two 4.0mm thick two carbon felts are compared. In the combination, one carbon paper was stacked on one carbon felt electrode at the current collector side. In the case with the combination structure, the I-V performance is improved compared to that with the carbon felt electrode. The pressure loss can be also maintained small. This suggests that combination of various types of carbon electrode is effective to improve the cell performance with the ID. Detailed investigations on the mechanism and further improvements of the cell performance with the IDD will be conducted. Reference (1) M. M. Mench, et al. , Journal of Power Sources , 302, 369 (2016). (2) T. J. Schmidt, et al. , ECS Trans ., 57, 535 (2018). Figure 1