Steam electrolysis is a high-efficiency process that generates hydrogen from water using electricity and heat made available from an advanced nuclear power plant.¹  The process depends on multiple solid oxide cells arranged in “stacks” identical to solid oxide fuel cell designs, but operating in reverse polarization.  The stack provides for a manageable voltage/current input as well as a manufacturable cell size.  However, in building stacks several factors are introduced, which lead to degradation during operation.  Most post-test analyses of solid oxide stacks have been performed without the results being made public.  Published reports often rely on in situ current/voltage data taken during stack operation,² and cross-sectional images from scanning electron microscopy (SEM).³  In this work, we developed a procedure to identify regions of degradation in disassembled cells using four-point resistivity measurements, X-ray fluorescence mapping, and Raman spectroscopy, followed by sectioning the cells and analyzing specific regions using SEM.  This presentation will focus on results from examination of the oxygen electrodes and our own efforts to develop alternative electrode materials for high temperature steam electrolysis.

The examined cells were from a Ceramatec, 25-cell solid oxide stack that had operated at ~830°C for 1000 hours, producing hydrogen at a nominal rate of 200 normal liters per hour.¹  Our first experiment mapped the resistivity of the electrode surface using a linear 4-point probe to identify possible degradation regions.  The results were plotted on a 3-D contour map of resistivity versus X-Y position.  These four-point resistivity measurements revealed that the oxygen electrodes were relatively stable, except along the sealing edge where hydrogen exits the cross-flow stack.

To obtain sub-surface information, X-ray fluorescence mapping measurements were performed at The Materials Research Collaborative Access Team beamline of Argonne's Advanced Photon Source (APS).  These measurements showed the presence of chromium deep in the oxygen electrode.  The results indicate that the Cr had permeated from the stainless steel bipolar plate into the electrode towards the electrode-electrolyte interface.  Chromium is known to poison the oxygen electrode reaction in SOFCs by blocking active sites for the oxygen reaction.

Raman micro-spectroscopy was used to identify known constituent phases and unknown impurities within the bond layers and electrodes.  This near-surface technique is useful for analyzing features in a non-destructive manner with a variably focused excitation laser.  Raman spectroscopy identified monoclinic zirconia on the edges of the cells.  Monoclinic zirconia is a crystal phase that has structural instability.  Chromium, in the form of ruby crystals Cr-Al₂O₃, was also identified near the edges of the electrode.

After conducting the non-destructive analyses described above, we were able to identify regions of interest where cell performance degradation most likely occurred and then used cross-sectioning and SEM analysis to further assess the causes.  It was determined that most of the cell degradation occurred along the edges of the electrode and at the interfaces of the various layers.

We are also investigating alternative materials for improvement of the oxygen electrode performance.  Using electrochemical impedance spectroscopy on half-cells of our air electrode materials, we have evaluated the area specific resistance of Pr_0.5Sr_0.5CoO₃ and Pr₂NiO₄ as a function of temperature and time.  We are also currently working to optimize the performance of the cobaltite and nickelate materials by engineering the microstructure using functionally graded compositions and fine-grained interlayers.  We have found that a thin, fine-grained interlayer of Ce_0.8Sm_0.2O₂ improves the performance of the cobaltite and nickelate electrodes.

References:

1.     J. E. O'Brien, C. M. Stoots, J. S. Herring et al., "Performance of Planar High-Temperature Electrolysis Stacks for Hydrogen Production from Nuclear Energy", accepted for publication in Nuclear Technology (2007).

2.     Michele Molinelli, Diego Larrain, Nordahl Autissier et al., "Dynamic behavior of SOFC short stacks", Journal of Power Sources 154, 394 (2006).

3.     S. Benhaddad and I. Protkova, "Failure analysis as an essential component of an SOFC development program", in 9th International Symposium on Solid Oxide Fuel Cells, SOFC IX, edited by S. Singhal and J. Mizusaki (Electrochemical Society Inc., Pennington, NJ, Quebec, Canada, 2005), Vol. PV 2005-07, pp. 923.

Acknowledgments

This work is supported by the U.S. Department of Energy, Office of Nuclear Energy, Science, and Technology, Nuclear Hydrogen Initiative.  The submitted manuscript has been created by the UChicago Argonne, LLC as Operator of Argonne National Laboratory under Contract No. DE-AC02-06CH11357 with the U.S. Department of Energy. The U.S. Government retains for itself, and others acting on its behalf, a paid-up, nonexclusive, irrevocable worldwide license in said article to reproduce, prepare derivative works, distribute copies to the public, and perform publicly and display publicly, by or on behalf of the Government.