Degradation Behaviour of a Solid Oxide Electrolysis Stack with Ni-YSZ Based Cells at 700 °C

ECS Meeting Abstracts, vol. MA2025-03, pp. 416

Abstract

Solid Oxide Electrolysis Cells (SOEC) offer high efficiencies to produce hydrogen via steam electrolysis. Commercial introduction of SOEC systems is, however, still limited due to an insufficient lifetime of cells and stacks. Previous investigations on Ni-YSZ-based fuel electrode supported cells show higher degradation rates in steam electrolysis operation in comparison to fuel cell operation [1,2]. The causes of aging are still controversially discussed. Degradation mechanisms in SOEC are often attributed to Ni-migration and agglomeration at the fuel gas electrode [1], however, there are also voices that attribute the performance loss to the air electrode [2]. In the search for more favourable, but nevertheless system-relevant operating conditions for this cell type, a 4-layer short-stack in the Jülich F10 design with commercially available cells (Elcogen, Estonia) with Ni-YSZ cermet support and fuel electrode, was assembled and tested under different operating conditions. The steam electrolysis operation was divided into 4 consecutive operating phases in which the current density and the steam content were varied. An operating time of at least 1000 hours was planned for each phase. The operation started (phase I) at a current density of -0.5 A/cm² and a steam content in the inlet gas of 50 % (balanced with hydrogen). In the second phase, the current density was increased to -0.75 A/cm² and in two further phases the steam content in the inlet gas was raised to 70 and 90 %, respectively, while keeping the current density at -0.75 A/cm². The operating temperature and the steam utilization were kept constant at 700 °C and at 50 %, respectively, throughout all phases of the steam electrolysis operation. The low operating temperature was chosen, because Sassone et al. [3] observed a reduction in aging with lower operating temperature. Since high steam contents in the reactant gas are system-relevant, the main question of this work was what influence these have on degradation. To evaluate the degradation over the four operating phases, the stack was characterized in detail. In addition to a symmetrical current/voltage (IV) curve measurement in SOEC/SOFC mode, impedance spectra were measured at different current densities between -0,0625 A/cm² and -0.75 A/cm². The measurements were carried out approximately in the middle of each operating phase and at the end or before the start of the next phase. Stationary operation was interrupted for the characterization. With the aim of obtaining an indication of possible leaks, the open-circuit voltages (OCV) were repeatedly determined with dry H 2 as feed gas. Figure 1 shows the area-specific resistances (ASR) determined from the IV curves at -0.5 A/cm² as well as from the impedance spectra at a current density of 0.5 A/cm² (R total ) at the end of each operating phase. Besides the total resistances also the high-frequency contributions (R Ω , ‘Ohmic resistance’) were determined. For clarity only data for layer three is shown. The ASR values from the IV-curves showed a good agreement with the total resistances from the EIS spectra. The increase in current density during stationary operation had no noticeable influence on the ASR value determined subsequently during characterization. However, higher steam contents during operation lead to a considerable increase of the ASR afterwards, especially for 90% steam. The ohmic contributions to the total impedances remained constant throughout all operating phases, pointing to an increase only of the polarization contributions of the impedances. Despite the relatively low temperature of 700 °C for steam electrolysis, degradation rates of ~10 mΩcm²/kh and less were observed up to and including operating phase III, i.e. up to a steam content of 70 % in the fuel gas inlet. Further detailed evaluation of the impedance spectra will be presented together with the results of the ongoing post-test investigations and possible interpretations during the conference. References [1] X. Shao et al. Journal of Power Sources 609 (2024) 234651 [2] Y. Wang et al. Journal of Materials Science & Technology 55 (2020) 35 [3] G. Sassone et al. Journal of Power Sources 605 (2024) 234541 Figure 1

Authors 4

1. L.G.J. de Haart

   Forschungszentrum Jülich

   Affiliation as printed

   Forschungszentrum Jülich GmbH, IET-1

2. Dominik Schäfer

   Forschungszentrum Jülich

   Affiliation as printed

   Forschungszentrum Jülich, IET-1: Fundamental Electrochemistry

3. Remzi Can Samsun

   Forschungszentrum Jülich

   Affiliation as printed

   Forschungszentrum Jülich GmbH, IET-1

4. Rüdiger‐A. Eichel Aachen

   Forschungszentrum Jülich · RWTH Aachen University

   Affiliation as printed

   Forschungszentrum Jülich GmbH, IET-1

   RWTH Aachen University