In the wake of the growing environmental challenges, hydrogen produced from renewable energy is a good candidate as a substitute for conventional fuels. Indeed, fuel cells can produce electricity thanks to air and hydrogen, and only reject water. The use of this technology is currently emerging in various sectors, such as the maritime industry. However, sea salt aerosols, mostly composed of sodium chloride (NaCl), can damage fuel cells. So far, only a few studies have been dedicated to the impact of NaCl on Proton Exchange Membrane Fuel Cells (PEMFCs) though showing that salty air can induce a significant loss of performance without clearly identifying the underlying physicochemical mechanisms [1][2][3].

Thus, the present experimental work investigates the effects of a salt spray injected into the airflow of single PEMFC cells and stacks at various time scales and contamination levels, with the final aim of identifying the appropriate PEMFC stack protection needed for maritime applications. A dedicated experimental setup generating salt spray in the cathode airflow of a single cell and 5-cells stack has been designed, realized and characterized. The NaCl concentration range used is from 10 mg/m³ to 5 g/m³, to reproduce different types of salty environments. Cells and stacks are operated in constant current density mode.

The study is conducted under the conditions of accelerated stress tests. When polluted at high concentrations, single cells degradation becomes significant after only a few minutes of operation (Figure 1) and after a few hours for stacks. Even under low concentrated environments, the single cell voltage decreases faster than the baseline (ie reference voltage with pure water in air stream), leading to PEMFC lifetime shortening. It has been observed that cathode rinsing with nitrogen flow or operation under pure water results in partial cell performance recovery. Moreover, starts-up and shuts-down operated during tests can have a positive or negative influence on cell performance, depending on the break duration.

Even when the tests revealed only weak decrease in power, Membrane Electrode Assemblies (MEA) and bipolar plates (BP) post-mortem characterizations highlight degradations. Scanning Electron Microscopy (SEM) studies coupled with Energy Dispersive X-ray Spectroscopy (EDS) have contributed to the identification of components in MEA and BP. As expected, chlorine reaches not only the cathode gas diffusion layer (GDL) and catalyst layer (CL) but also the anode CL by crossing the membrane [4]. According to the literature, chloride adsorption on Pt surface or Pt dissolution reduces the active area [3] whereas sodium ion contaminates the membrane [1]. Moreover, corrosion products from metallic components have been found. Corrosion marks identified as stainless steel from BP are present in the stack effluents and MEAs. As for single cell, compounds from current collectors (gold, nickel and copper) are present in the GDL and CL. In the near future, for single cell tests, non-metallic current collectors will be used in order to discriminate the effects of the direct degradation of salt spray on MEA and indirect degradation coming from corrosion products of metallic cell components.

Overall, the present study reports that NaCl effects on PEMFCs not only depend on the concentration of the NaCl mist but also on the MEA type, current collectors/BP material, and startup/shutdown procedure. In the full paper and oral presentation, tests conducted with less concentrated NaCl droplets will also be presented, with the aim of determining a degradation threshold for single cells.

References

[1] M. S. Mikkola et al., Fuel Cells, vol. 7, p. 153‑158, 2007.

[2] B. V. Sasank et al., J. Mar. Sci. Technol., vol. 21, p. 471‑478, 2016.

[3] S. Uemura et al., ECS Trans., vol. 80, p. 651‑655, 2017.

[4] H. Li et al., J. Power Sources, vol. 196, p. 6249‑6255, 2011.