The ongoing growth of the demand of electrolytic hydrogen produced from renewable electrical energy sources is, at present, mainly met by low temperature electrolysers (alkaline and PEM). Yet, high temperature (steam) electrolysis using the solid oxide cell (SOC) technology is rapidly developing, with demonstration systems approaching the MW power range. System upscaling and industrialisation is, e.g., subject of the R&D project H2Giga in Germany /1/. Steam electrolysis with SOC yields highest electrical-to-chemical energy-conversion efficiency. That efficiency is further increased if steam generation is feasible without electrical heating but via heat sources of moderate temperature in industrial installations. Moreover, as an often not considered advantage of weight, initial efficiency can be maintained over a large part of the foreseen lifetime (or even the entire lifetime) by temperature adjustments which keep the cell voltage close to the thermal neutral value for a simplified thermal management of the electrolyser system (U th ~ 1.29 V @ 830° C) /2/. Finally, a matching to a fluctuating load or for grid-services is also possible with the SOC technology /3/.SOC system development heavily relies on cell research, in particular on the experimental verification of long-term stability. In that context, as already discussed in earlier meetings, electrolysis with electrolyte supported solid oxide cells (ESC) with Ni/GDC cathodes shows excellent durability /2-5/. Operation with an elevated current density of 0.9 Acm-2 was demonstrated in up to 30,000 h long tests leading to a low voltage-degradation rate of around 5 mV/1000 h, sufficient for practical application /5/. Comparison of intrinsic cell degradation and the, usually larger, stack/system degradation facilitates the identification of degradation sources (from e.g., cells, seals, tubing, and feed media). Cells with Ni/GDC cathodes and the known mechanically robust 3YSZ electrolyte are currently used in pilot installations by the company Sunfire (Germany).This contribution will focus on two follow-up activities in cell testing, (i), a further increase in current density (>1 Acm-2), and (ii), a lowering of the operation temperature, both at maintained cell voltage. The therefore required decrease of the (area specific) cell resistance, ASR, implies a reduction of the ionic conduction losses in the electrolyte. This is achievable with electrolytes of higher specific conductivity (e.g., 10Sc1CeSZ or 6Sc1CeSZ) or with thinner electrolyte layers. Tests are, so far, done with cells based on 3YSZ electrolytes down to 30 µm thickness (standard thickness being 90 µm), with testing times in the range from 4,000 to above 10,000 h. Degradation is determined under constant current operation, and different degradation contributions are separated with impedance spectroscopy at maintained steady-state conditions.Electrolysis operation around the U th value means relatively high cell-voltage losses (~0.4 V), which is in marked difference to the SOFC mode where losses are usually much smaller. A reasonably large electrolysis current density of 0.7 Acm-2 (at U ~ U th) could be demonstrated in a 10,000 h long test with a cell with a 32 µm thick electrolyte at a temperature of 755°C. This test will be discussed in more detail. Temperature is within the typical temperature range of operation of classical electrode (cathode) supported cells, CSC, which facilitates direct cell comparisons (for a comparison of ESC/CSC performance at different temperature together with a post-test analysis of long-term operated cells see ref. /6/). Provided that degradation remains sufficiently low, it also means a large usable temperature window for electrolysis, favourable for power variation and for degradation compensation with temperature adjustments.The increase in current density relies, above all, on sufficient electrode stability (notably of the hydrogen/steam electrode). Rather low electrode overvoltages found in previous work /2-5/ are encouraging results in that context. More open is the evolution of the generally observed predominantly ohmic degradation coming from the electrolyte, the contacting, and the formation of secondary insulating layers.
Read full abstract