Introduction: Water electrolysis is an essential technology for efficiently combining renewable energy with hydrogen energy. Long-term durability against voltage fluctuations is necessary to produce hydrogen by water electrolysis directly using renewable energy. In our previous study, potential fluctuation protocols have been developed based on actual wind power voltage fluctuations (1). In our recent work, the durability of water electrolysis cells has been evaluated for 320 days (2). Reversible and irreversible performance losses were present. As an irreversible performance loss, delamination of IrO2 anode layer was found, which was most likely due to increase in the internal pressure derived by stagnation of generated gases, as similarly seen in the durability test up to 160 days (1). However, regarding to changes in pore structures, the trend was different from our previous study. That was assumed that the change in initial pore structures of anode layers derived by different ionomer/catalyst ratio, rather than the change in the length of the durability test, made different trends.Therefore, in the first part of this study, the durability test condition was aligned with that of the previous study (1), equivalent to 160 days, and the effect of the initial pore structure on durability was examined. In the second part, for the purpose of developing more durable anode pore structure, addition of a hydrophobic material was considered since partially hydrophobic surface probably promotes releasing of gas bubbles to eliminate gas stagnation. Experimental: A water electrolysis cell was made by spray printing 0.3 mg Pt/cm2 using 46.3% Pt/KB for the cathode and 0.5 mg IrO2/cm2 using commercial IrO2 (Tokuriki Honten) for the anode on Nafion117 membrane. The amount of Nafion ionomer in the anode was kept to 15% in this study. The resulting membrane electrode assembly (MEA) was placed in a holder which can flow water in the both anode and cathode with porous transport layers of titanium mesh and carbon paper for the anode and cathode, respectively. After initial electrochemical measurements of AC impedance and I-V performance were taken, a durability test of 80 sets corresponding to 160 days was performed using a potential fluctuation protocol developed in our laboratory (1). The potential was applied to the anode while the cathode was maintained at 0 V. Electrochemical measurements were similarly taken at the end of each set. In addition to electrochemical properties, anode structural properties were evaluated from 3D reconstructed images made by hundreds of FIB-SEM images. For the second part of this study, MEAs with an increased hydrophobic property were prepared by adding 10-30 wt% of polypyrrole against the mass of IrO2 to the anode. Results and discussion: The current density at 2.0 V was monitored during the durability test equivalent to 160 days. Both MEAs with 33% (1) and the 15% Nafion ionomer in the anode showed a similar trend of the current decrease and recovery even though initial performance was higher for MEA with 15% Nafion ionomer. However, the irreversible current loss was larger for MEA in this study. This irreversible current loss is most likely caused by the stagnation of the generated gases based on our previous studies (1), (2). Since gas stagnation is deeply related to the pore structure of the anode layer, 3D reconstruction of the anode was performed using several hundred FIB-SEM images, and comparison to our previous study (1) was done. As a result, when pore size distributions based on 3D reconstructed images were analyzed as shown in Figure 1, a denser anode structure was found with 15% Nafion ionomer. In other words, MEA with 15% Nafion ionomer has high initial IV performance but low durability, while 33% Nafion has low initial IV performance but high durability, indicating that there is a trade-off between the two.Optimizing the amount of Nafion ionomer and the mixing method is one way to achieve both high performance and high durability. However, in the second part of this study, addition of a hydrophobic material was rather considered since partially hydrophobic surface promotes releasing of gas bubbles to eliminate gas stagnation. Polypyrrole was chosen because of its conductivity as well as it hydrophobicity. Unlike the addition of an insulator, PTFE, there was almost no change in initial IV performance up to 30% addition, but a significant improvement in hydrophobicity was observed. The durability of MEA with the anode containing 30% polypyrrole has been evaluated.
Read full abstract