Water electrolysis is a key technology in future sustainable energy scenarios, since hydrogen as a universal energy carrier could promote the decarbonization of the energy economy, or even become its backbone in the context of a ‘hydrogen economy’. Polymer electrolyte water electrolysis (PEWE) is a key technology in energy applications, where the possibilities of rapid start-up and dynamic operation with large differential pressures in compact units are of importance. The most prominent challenge for electrolyzers with improved performance and lower cost so far are the durability targets. (1, 2)

In this work we compare the effect of dynamic operating conditions (start-stop vs. steady-state) on the component aging in 250 h experiments using Nafion 115-based CCMs. The main mechanism of performance decay in 250 h experiments, independent of the type of galvanostatic cycle, is the cation poisoning of the ionomer. It is, however, reversible by CCM re-protonation in sulfuric acid (3). More interestingly, on-line monitoring of the anodic gas purity indicates an increase in the hydrogen crossover (HC) during both experiments. Post test SEM analysis suggests that the cause is the creep of the membrane material into the voids of the titanium porous transport layer (PTL), with concomitant local thinning of the membrane. This effect could represent a bottleneck for implementing thinner membranes towards more efficient PEWE.

The increase of HC over time is alleviated by changing to the less-expensive cathodic carbon PTLs. Being more compressible, the carbon based cathode PTL deforms under the cell clamping pressure, minimizing the creep of the polymer. The mechanical deformation of the CCM and the material creep into the PTL pores is much less pronounced according to post test SEM analysis. Moreover, carbon cathode PTLs lead to improvements in overall cell performance and lower HC (Figure 1). Both benefits are postulated to stem from a more uniform contact between the cathode catalyst layer (CL) and the PTL structure. The lower cell potential is related to the decrease in the contact resistance. Lower HC is thought to be a result of the more uniform current distribution over the CL and PTL interface. Hydrogen evolution reaction is thought not to take place on the catalytic sites situated in the voids of the PTL (nominally 30 - 40 µm). Due to the limited electrical in-plane CL conductivity, areas of the CL which are not in direct electronic contact to the PTL material remain inactive. Thus, the active regions need to sustain a higher local current density, which might result in local hot-spots and higher gas pressures in the CL. To test this hypothesis, we introduced a cathodic micro-porous layer (MPL) to further improve the current distribution. The presence of the MPL leads to further improvements in both HC and cell performance (Figure 1). We conducted X-ray tomography of PEWE PTL materials to characterize the interface and further investigate the correlations between the PTL morphology and the hydrogen crossover.

References

1. M. Carmo, D. L. Fritz, J. Mergel and D. Stolten, International Journal of Hydrogen Energy, 38, 4901 (2013).

2. U. Babic, M. Suermann, F. N. Büchi, L. Gubler and T. J. Schmidt, Journal of The Electrochemical Society, 164, F387 (2017).

3. X. Wang, L. Zhang, G. Li, G. Zhang, Z.-G. Shao and B. Yi, Electrochimica Acta, 158, 253 (2015).