Peter Halvor Larsen

Solid Oxide Fuel Cell

SOFC cell comprising a metallic support 1 ending in a substantially pure electron conducting oxide, an active anode layer 2 consisting of doped ceria, ScYSZ, Ni—Fe alloy, an electrolyte layer 3 consisting of co-doped zirconia based on oxygen ionic conductor, an active cathode layer 5 and a layer of a mixture of LSM and a ferrite as a transition layer 6 to a cathode current collector 7 of single phase LSM. The use of a metallic support instead of a Ni—YSZ anode support increases the mechanical strength of the support and secures redox stability of the support. The porous ferrite stainless steel ends in a pure electron conducting oxide so as to prevent reactivity between the metals in the active anode which tends to dissolve into the ferrite stainless steel causing a detrimental phase shift from ferrite to austenite structure.

Prospects and problems of dense oxygen permeable membranes

The prospects of using mixed ionic/electronic conducting ceramics for syngas production in a catalytic membrane reactor are analysed. Problems relating to limited thermodynamic stability and poor dimensional stability of candidate materials are addressed. The consequences for these problems, of flux improving measures like minimization of membrane thickness and minimization of the losses due to oxygen exchange over the membrane surfaces, are discussed. The analysis is conducted on two candidate materials: La 0.6 Sr 0.4 Co 0.2 Fe 0.8 O 3- δ and SrFeCo 0.5 O x . Finally, experimental investigations of the dimensional stability of the latter material under reducing conditions are reported.

The mechanism behind redox instability of anodes in high-temperature SOFCs

Bulk expansion of the anode upon oxidation is considered to be responsible for the lack of redox stability in high-temperature solid oxide fuel cells (SOFCs). The bulk expansion of nickel-yttria stabilized zirconia (YSZ) anode materials was measured by dilatometry as a function of sample geometry, ceramic component, temperature, and temperature cycling. The strength of the ceramic network and the degree of Ni redistribution appeared to be key parameters of the redox behavior. A model of the redox mechanism in nickel-YSZ anodes was developed based on the dilatometry data and macro- and microstructural observations.

Sr-Doped LaCrO3 Anode for Solid Oxide Fuel Cells

A number of doped lanthanum chromite perovskites are considered as anode materials for solid oxide fuel cells with an yttria-stabilized zirconia (YSZ) electrolyte operating in hydrogen at 850°C. The polarization resistance is measured by impedance spectroscopy, and shown to depend significantly on the type and amount of doping. In particular, the composition La 0.8 Sr 0.2 Cr 0.97 V 0.03 O 3 (LSCV) is examined in detail. Reactivity studies indicate the presence of secondary phases in LSCV. These phases are reactive toward YSZ, resulting in the formation of SrZrO 3 . The secondary phases may he readsorbed during prolonged calcination under reducing conditions. The polarization resistance is shown to increase severely over a few days, and to be recoverable by temporary oxidation. The time constant of the degradation is shown not to match that of the changes in stoichiometry and conductivity during reduction of the perovskite. Two rate limiting processes arc generally observed. The low frequency process is suggested to relate to adsorption of hydrogen on the LSCV surface or a chemical reaction step. The high frequency process is suggested to relate to the LSCV/YSZ contact interface. LSCV does not exhibit significant catalytic activity toward steam reforming of methane, and shows no sign of direct methane oxidation.

Dimensional Behavior of Ni–YSZ Composites during Redox Cycling

The dimensional behavior of Ni-yttria-stabilized zirconia (YSZ) cermets during redox cycling was tested in dilatometry within the temperature range 600-1000°C. The effect of humidity on redox stability was investigated at intermediate and low temperatures. We show that both the sintering of nickel depending on temperature of the initial reduction and the operating conditions, and the temperature of reoxidation are very important for the size of the dimensional change. Cumulative redox strain (CRS) is shown to be correlated with temperature. Measured maximum CRS after three redox cycles varies within 0.25-3.2% dL/L in dry gas and respective temperature range of 600-1000°C. A high degree of redox reversibility was reached at low temperature, however, reversibility is lost at elevated temperatures. We found that at 850°C, 6% steam and a very high p H2O /p H2 ratio is detrimental for redox stability, whereas at 600°C no negative effect was observed. Pre-reduction at 1100 instead of 800°C more than doubled redox strain on reoxidation at 800°C. For samples similarly pre-reduced at 1000°C, lowering the reoxidation temperature from 1000 to 750°C or below reduces the redox strain to less than half.

The influence of SiO2 addition to 2MgO–Al2O3–3.3P2O5 glass

2MgO–Al2O3–3.3P2O5 glasses with increasing amounts of SiO2 are considered for sealing applications in solid oxide fuel cells (SOFC). The change in chemical durability under SOFC anode conditions and the linear thermal expansion are measured as functions of the SiO2 concentration. Raman spectroscopy analysis of the glasses reveals no sign of important changes in the glass structure upon SiO2 addition. Some increase in glass durability with SiO2 concentration is reported and its cause discussed.

Solid oxide fuel cell development at Topsoe Fuel Cell and Risø

The consortium of Topsoe Fuel Cell A/S and Riso National Laboratory has scaled up its production capacity of anode-supported cells to about 1100 per week. The consortium has an extended program to develop the SOFC technology all the way to a marketable product. Standard stacks have been tested for more than 13 000 h. Post-mortem analysis has revealed the dominating degradation mechanisms. Recently, the degradation rate has been reduced to below 0.5% per 1000 h by introduction of improved stack component materials, including improved metallic interconnects. Several 50- or 75-cell stacks in the 1 kWe and above power range have been tested successfully on methane-rich reformate gas at a fuel utilization up to 92%. Stack and system modeling – including cost optimization analysis – is used to develop 5–25 kWe stack modules for operation in the 700–850°C temperature range. A special effort is focused on manufacturing and testing of larger anode-supported cells and stacks with a footprint of 18 × 18 cm 2 . The SOFC program comprises the development of next-generation cells with metallic supports for operation at lower temperature with increased durability, lower cost, and high mechanical robustness. A range of fuels have been studied including natural gas, LPG, methanol, DME, diesel and ammonia. A multi-stack design study for a 24-stack module prototype has been finalized, and stack construction is under way for a 20 kWe prototype running on natural gas. The studies predict system electrical efficiencies from 50% to 56% (AC out/LHV fuel in) depending on the fuel used and the size of the system.

Solid oxide fuel cell development at Topsoe Fuel Cell A/S and Risø National Laboratory

The consortium of Topsoe Fuel Cell A/S and Risoe National Laboratory has up-scaled its production capacity. Stacks are based on a compact thin plate multilayer design with metallic interconnects and 12x12 cm{sup 2} or 18x18 cm{sup 2} foot print. Larger (500 cm{sup 2}) cells are currently under evaluation. Stacks have been tested successfully for more than 13000 hours. Several 50 or 75 cell stacks in the 1+ kW power range have been tested successfully at a fuel utilisation of up to 92%. Multi stack modules consisting of four 75 cell stacks have been tested for more than 4000 hours with pre-reformed natural gas and modules consisting of twelve stacks are under development. Our SOFC program comprises development of next generation cells with porous ferritic steel is used as a cheap, ductile, robust cell support and the electrolyte is based on scandia-doped zirconia with improved durability. In collaboration with Waertsilae, a 24-stack prototype based on natural gas is being tested. The range of fuels have further been extended to include ethanol and coal syn-gas by development of a new coke resistant catalyst suitable for future SOFC technology.

Solid Oxide Fuel Cell (SOFC) Development in Denmark

The SOFC technology under development at Risø National Laboratory (RISØ) and Topsoe Fuel Cell A/S (TOFC) is based on an integrated approach ranging from basic materials research on single component level over development of cell and stack manufacturing technology to system studies and modelling. The effort also comprises an extensive cell and stack testing program. Systems design, development and test is pursued by TOFC in collaboration with various partners. The standard cells are thin and robust with dimensions of 12 x 12 cm2 and cell stacks are based on internal manifolding. Production of cells is being up-scaled continuously. The durability of the standard stack design with standard cells has been tested for more than 13000 hours including nine full thermal cycles with an overall voltage degradation rate of about 1% per 1000 hours. Recently, the degradation rate has been significantly reduced by introduction of improved stack component materials. 75-cell stacks in the 1+ kW power range have been tested successfully. Stacks have been delivered in a pre-reduced state to partners and tested successfully in test systems with natural gas as fuel. The consortium of TOFC and RISØ has an extended program to develop the SOFC technology all the way to a marketable product. Stack and system modelling including cost optimisation analysis is used to develop multi kW stack modules for operation in the temperature range 700-850oC. To ensure the emergence of cost-competitive solutions, a special effort is focused on larger anode-supported cells as well as a new generation of SOFCs based on porous metal supports and new electrode and electrolyte materials. The SOFC program comprises development of next generation of cells and multi stack modules for operation at lower temperature with increased durability and mechanical robustness in order to ensure long-term competitiveness.

Processing of Ce1-xGdxO2-δ (GDC) Thin Films from Precursors for Application in Solid Oxide Fuel Cells

Extensive interfacial reactions are known to occur between Fe-Co based perovskite cathode materials and the standard solid oxide fuel cell (SOFC) yttria stabilized zirconia (YSZ) electrolyte. Thin films of gadolinia doped ceria (GDC) could be used as a diffusion barrier between the cathode and the electrolyte. The present work investigates spin coating thin diffusion reaction inhibiting films onto SOFC electrolytes. The chemical and structural evolution of ethylene glycol based precursor solution is studied by means of rheology, x-ray diffraction (XRD), high temperature XRD (HT-XRD), Fourier-transformed infrared spectroscopy (FTIR) and differential thermal analysis (DTA). The studies show that cerium formate is formed as an intermediate resin. Thin films, up to 500 nm thick, of gadolinia doped ceria (GDC) are successfully produced by multiple spin coating of polymerized ethylene glycol derived solutions on 200 1m thick YSZ tapes. The GDC and YSZ interfacial surface morphology and film thickness are studied by scanning electron microscopy (SEM) and atomic force microscopy (AFM). These films are shown to successfully prevent the creation of non-conducting reaction phases at the cathode-electrolyte interface by blocking interdiffusion.