Iron Contamination and Local Distribution within a 50 Cm Single-Channel Proton Exchange Membrane Water Electrolysis Cell

Abstract

Proton exchange membrane water electrolysis (PEMWE) will play a major role in the clean and versatile energy conversion in the near future.[1] Offering scalable solutions of PEMWE are of high industrial interest and have to provide a sufficient durability and reliability.[2] To achieve this, a better understanding of the impact of contamination on the PEMWE performance is very crucial. In this context, the membrane electrode assembly (MEA) is susceptible to several degradation processes like metallic cations (Fe3+/2+ stemmed from tubing). In particular, soluble and reduced iron species forming by Fenton reaction might attack via radicals the MEA.[3,4] Hence, analyzing impurities and their resulting impacts on e.g. the proton transport resistance and degradation mechanisms at different length scales and local distributions are of large interest.In this work, the influence of spatially resolved iron contamination in MEAs on the cell performance has been investigated using a 50 cm single-channel PEMWE cell equipped with local current density measurement setup.[5] Adding defined amounts of FeSO4 via the anode inlet feed, the changes on the three main regimes of overpotentials, i.e. kinetic, ohmic and mass transport, is correlated with the current density mapping as well as local and global electrochemical impedance spectroscopy (EIS) measurements. Ex-situ scanning electron microscopy (SEM) and energy-dispersive X-ray spectroscopy (EDX) are used to detect the distribution and agglomeration of iron species within the MEA.With this combined approach of electrochemical and ex-situ physical methods, the contamination pathways and their effects of (spatially resolved) iron contamination on the overall performance of MEA are investigated as a function of degree of contamination and operating conditions. Especially, the (spatially resolved) overpotential-specific degradation processes along the 0.5 m channel, which mimic a technically relevant stack size, allow us to conclude on the optimum design of MEA, targeting use of mitigation strategies and development of accelerated stress test protocols.[1] M.A. Pellow, C.J.M. Emmott, C.J. Barnhart, S.M. Benson, Energy Environ. Sci. 8 (2015) 1938-1952.[2] P.K. Shen, C.-Y. Wang, S.P. Jiang, X. Sun, J. Zhang, Electrochemical Energy: Advanced Materials and Technologies, CRC Press, 2015.[3] X. Wang, L. Zhang, G. Li, G. Zhang, Z.-G. Shao, B. Yi, Electrochim. Acta 158(2015) 253-257.[4] C. Rozain, P. Millet, Electrochim. Acta 131 (2014) 160e167.[5] C. Immerz, B. Bensmann, P. Trinke, M. Suermann, R. Hanke-Rauschenbach, JECS 165 (16) (2018)