0
Research Papers

Impact of Materials and Design on Solid Oxide Fuel Cell Stack Operation

[+] Author and Article Information
Stefan Diethelm, Michele Molinelli

 HTceramix S.A., 26 Avenue des Sports, CH-1400—Yverdon-les-Bains, Switzerland

Jan Van herle1

Laboratory of Industrial Energy Systems (LENI),  Ecole Polytechnique Fédérale de Lausanne, CH-1015 Lausanne, Switzerlandjan.vanherle@epfl.ch

Zacharie Wuillemin, Arata Nakajo, Nordahl Autissier

Laboratory of Industrial Energy Systems (LENI),  Ecole Polytechnique Fédérale de Lausanne, CH-1015 Lausanne, Switzerland

1

Corresponding author.

J. Fuel Cell Sci. Technol 5(3), 031003 (May 22, 2008) (6 pages) doi:10.1115/1.2889025 History: Received November 30, 2005; Revised October 16, 2007; Published May 22, 2008

Planar SOFC stack technology based on a unique concept (SOFConnex™) uses structured gas distribution layers between unprofiled metal sheet interconnects and thin Ni-YSZ anode supported electrolyte cells. The layers are flexible both in material and design and allow to implement new configurations relatively simply; manifolding can be internal, external, or combined. Together with thin stack components, independent of the supplier, the SOFConnex™ stacking approach allows compact planar assembly with low cost potential and adequate power density. Different cell and flow designs have been realized. With a basic flow configuration, short stacks (50cm2 cell active area) were assembled and tested, power density at 800°C reaching 0.5Wcm2 at 0.7V average cell voltage (1.5kWeL, 0.36Ωcm2 area specific resistance), for 65% fuel utilization and 35% lower heating value electrical efficiency. Short stacks were thermally cycled and operated with both hydrogen and syngas. Degradation was essentially Ohmic (confirmed from impedance spectroscopy on stacks) and at first mainly due to the cathode-electrolyte interfacial reaction, performance loss was subsequently strongly reduced after cathode replacement. Using multiple voltage probes with additional interconnects allowed to separately monitor current collection losses during polarization. With an improved design in terms of sealing, postcombustion control and flow field, stacks up to 1kWe have been operated.

Copyright © 2008 by American Society of Mechanical Engineers
Your Session has timed out. Please sign back in to continue.

References

Figures

Grahic Jump Location
Figure 1

Different SOFC cell (anode supported) and stack designs using the adaptable SOFConnex™ stack concept. Left: R cells (base design), 80×80mm2, internally manifolded counterflow. Right: S cells (improved design), 150×75mm2, externally (air)+internally (fuel) manifolded coflow.

Grahic Jump Location
Figure 2

Current density—average repeat element voltage output of five-cell SOFC stack (R design), in the temperature range of 700–800°C, for 5ml∕mincm2 humidified H2 flow, λair=1.6. OCV in mV, and asr in Ωcm2

Grahic Jump Location
Figure 3

Current density—average repeat element voltage output of five-cell SOFC stack (R design), in the temperature range of 700–800°C, for 8ml∕mincm2 humidified H2 flow, λair=1.6. OCV in mV, and asr in Ωcm2

Grahic Jump Location
Figure 4

Current density—individual repeat element voltage output of five-cell SOFC stack (R design), at 800°C (Fig. 3), for 8ml∕mincm2 humidified H2 flow, λair=1.6. OCV in mV, and asr in Ωcm2

Grahic Jump Location
Figure 5

Degradation of four-cell stack, as average repeat element voltage decay (given in −mV∕h) for different load periods, 6ml∕mincm2H2 fuel flow, λair=1.6

Grahic Jump Location
Figure 6

Sketch of single-cell stack with extra pair of metal interconnects (MICs) and SOFConnex™ (GDL or gas diffusion layer) on each cell side (cathode and anode), with separate voltage probes for monitoring current collection losses in situ

Grahic Jump Location
Figure 7

Degradation of one-cell stack, equipped with an extra current collection pair on anode and cathode sides (metalsheet+SOFConnex™) to separate out losses (see Fig. 6). Total RE voltage drop at constant current (0.51Acm−2, 800°C) is plotted left; anode and cathode sided current collection losses are plotted right (mV)

Grahic Jump Location
Figure 8

Degradation of one-cell stack, fed alternatingly with H2 and syngas (750°C, 5ml∕mincm2 fuel flow). No change in behavior is seen with CO addition to the anode stream.

Grahic Jump Location
Figure 9

EIS measurement on single-cell stack of Fig. 8, at times indicated in the legend (H2 fuel). Numbers −2 to 3 correspond to applied ac frequency in powers of 10Hz.

Grahic Jump Location
Figure 10

Temperature cycling profile on a five-cell stack (R design). Five temperature levels are indicated: fuel and air inlet, as well as the temperatures of bottom, middle, and top MICs. Heating rate was varied between 180K∕h and 400K∕h.

Grahic Jump Location
Figure 11

Average repeat element voltage of a five-cell stack versus current density (800°C, 6ml∕mincm2H2, λair=1.6) before and after three consecutive complete thermal cycles (800-30-800°C), see Fig. 1. Degradation of performance is seen to occur during constant current load at 800°C, and not as a consequence of thermal cycling itself.

Grahic Jump Location
Figure 12

SEM cross section image of Cell 1 of a five-cell stack after 500h operation. Delamination is observed to occur between both cathode layers (LSF in contact with the electrolyte, LSC top layer for current collection).

Grahic Jump Location
Figure 13

SEM cross section image of a cell from a five-cell stack showing very severe degradation (−20%) in 500h operation at 800°C. An interfacial layer of SrZrO3 formed at the LSF cathode ∕ YSZ electrolyte interface.

Grahic Jump Location
Figure 14

Repeat element voltage output with time of five-cell SOFC stack (R design) with LSM-YSZ instead of LSF active cathodes, 800°C, 6ml∕mincm2 humidified H2 flow, λair=2, 60% fuel conversion.

Grahic Jump Location
Figure 15

Durability of five-cell stack (800°C, 0.46A∕cm2, uF 40%, λ=1.6), integrating cells from another supplier (ECN (NL)), 0.55mm thick ASE cells with LSM-YSZ cathodes, with SOFConnex™ stacking. Total average degradation was −4%∕1000h, with −3%∕1000h loss estimated to current collection (derived from Fig. 7) and −1%∕1000h attributed to the cells.

Tables

Errata

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In