Table of contents

The Fuel Cells and Hydrogen Joint Undertaking (FCH JU) was set up in 2008 to accelerate the development of fuel cells and hydrogen technologies in Europe towards commercialization from 2015 onwards. To reach this target the FCH JU intends to bring together resources under a cohesive public-private partnership to ensure commercial focus, to match RTD activities to industry's needs and expectations and to scale-up and intensify links between the Industry Community and the Research Community. The applications are open to all fuel cells technologies, SOFC being mostly developed for the stationary applications, including back-up powers and APUs for transportation (trucks, planes). There are different groups working in Europe and supported partially by the FCH JU on the SOFC technology. By supporting such a project portfolio, FCH JU is going to reach most of the objectives set-up at European level mainly in terms of potential reduction of costs through sufficient number of units demonstrated across Europe in transport and stationary applications. The EU public support will continue for the next period of 2014-2020 for activities with an advanced TRL, moving towards market penetration of the FCH technologies.

The New Energy and Industrial Technology Development Organization (NEDO) has been conducting technology development and demonstrative research programs related to SOFC, as one of the most important and promising technologies for CO₂ reduction and realizing dispersed power sources.  One of their achievements was the-first-in-the-world commercialization of SOFC-based combined heat and power (CHP) system "ENE-FARM type S" in 2011.  

In June 2014, the Ministry of Economy, Trade and Industry (METI) has compiled a Strategic Road Map for Hydrogen and Fuel Cells.  It was established, releasing fuel cells for commercial and industrial use onto the market in 2017 as one of the processes of expanding the use of fuel cell technology.  NEDO's activities on SOFC, including a program "Technology development for promoting SOFC commercialization" , will be overviewed.

Development of electric power generation technology that efficiently and economically utilizes coal and natural gas while meeting environmental requirements is of crucial importance to the United States.  The U.S. Department of Energy (DOE) Office of Fossil Energy (FE), through the National Energy Technology Laboratory (NETL), is leading the research and development of advanced Solid Oxide Fuel Cells (SOFC) as a key enabling technology.  This work is being done in partnership with private industry, academia, and national laboratories. 

Program participants are making significant progress toward commercialization.  Milestones include: the development and validation of multiple generations of engineered seal concepts, cells with low degradation and high power density, development of integrated stack and system modeling tools, and validation of low-cost stainless steel interconnect solutions.  With advances in cost reduction, the program is now focused on research to improve durability and reliability, especially in an integrated realistic system context.  Integrated system tests at larger scales (10-100 kWe) are yielding respectable degradation rates in the 1 to 1.5%/1,000 hrs under cost-effective operating conditions.  

In this presentation, these highlighted accomplishments will be reviewed along with the status of the program and a road map for the future.

Fuel cell systems span a wide range of temperatures, materials, and type of ionic transport. An intermediate temperature range of 200 to 500 degrees Celsius has received less attention than higher and lower temperature fuel cell systems due to the relatively few number of electrolytes that operate in this range. However materials research over the past 15 years has resulted in new electrolytes with sufficiently high ionic conductivity to enable intermediate temperature fuel cells (ITFCs). The focus of the Reliable Electricity Based on ELectrochemical Systems (REBELS) program at the U.S. Department of Energy Advanced Research Projects Agency-Energy (ARPA-E) is to create ITFCs with the potential for drastically reduced system costs, as well as enhanced electrochemical functionality such as rapid response to transients and electrochemical production of liquid fuels. Compared to lower temperature polymer electrolyte membrane (PEM) fuel cells, little or no precious metal catalysts are required for sufficient electrode kinetics. The higher temperatures also increase fuel flexibility and tolerance to impurities. Compared to higher temperature solid oxide fuel cells (SOFCs), a range of less expensive interconnects and seals are available and the system can be operated in more of a load-following manner. In addition to these ITFC benefits, challenges will be identified and discussed particularly with respect to higher temperature SOFCs. These challenges include methane reforming, ohmic resistance of the electrolyte, and electrode overpotentials at reduced temperatures. Potential pathways to overcome these challenges will be proposed and examples of the new concepts under development in the REBELS program will be highlighted.

Mitsubishi Hitachi Power Systems, Ltd. (MHPS) is one of the first companies to focus on the potential of the Solid Oxide Fuel Cells (SOFC) as a component of large-scale power plant and has promoted segmented-in-series tubular type cell, the module and the system development since the 1980s. We are developing for commercialization of an SOFC-micro gas turbine (MGT) system from several hundred to several MW class. Recently prototype of a 250 kw class SOFC-MGT hybrid system was operated at Tokyo Gas Senju Techno Station. This system was very stable without voltage degradation at the rated for 4,100 hours. At the same time, we have started the development of fundamental technologies for realization of SOFC-Gas Turbine (GT)-Steam Turbine (ST) triple combined cycle system that aims to achieve more than 70% of power generation efficiency. We report the recent progress of the SOFC-micro gas turbine (MGT) hybrid system and development of triple combined cycle system.

Versa Power Systems (VPS) is a developer of solid oxide fuel cells (SOFCs) for clean power generation and is a wholly owned subsidiary of  FuelCell Energy (FCE).  FCE is a global leader in the design, manufacture and distribution of Molten Carbonate Fuel Cell (MCFC) power plants.  From an economic perspective, MCFCs scale-up very well and as a result FCE's MCFC products are in the multi-megawatt size range.  SOFCs are complementary because they scale-down well and hence are well suited to sub-megawatt applications.  This paper highlights the status of VPS and FCE's SOFC technology in the areas of cell, stack and system development.

Ceres Power is continuing to make excellent progress in the development of its low-temperature metal supported SOFC design (the 'Steel Cell') based predominantly around the use of ceria. This unique design architecture allows for a robust, low cost, subsidy free fuel cell product, whilst retaining the advantages of fuel flexibility, high efficiency and low degradation.

Over the last year, extensive development and verification of the technology has been undertaken on the latest generation of Steel Cell technology, which offers further enhancements in technology readiness level and manufacturability.  Over the same period, third parties have deepened their testing and development engagements with Ceres Power for a broadening range of market applications at cell, stack, fuel cell module and product levels. This paper provides an update on Ceres' development, verification approach and latest results.

Solid oxide fuel cell (SOFC) technology requires further maturations to become commercialized broadly. Many existing problems are expected to be solved by intermediate temperature SOFC (IT-SOFC). On the other hand, electrochemical resistances of cathode materials increase rapidly at lower temperature. So it is necessary to understand the underlying mechanisms of oxygen reduction reaction and develop new materials. Although many experimental and theoretical researches for this purpose are in progress, traditional trial-and-error materials search must be inefficient and take too long time.

 Considering the above situation, we explore critical factors that can explain the oxygen exchange rate of perovskite oxide cathode in SOFCs. It has been said that the bulk ionic conductivity is strongly related to the oxygen exchange rate, while the electronic conductivity is not as far as it is high enough. However, we found that the combination of ionic and electronic conductivities shows much stronger correlation than ionic conductivity alone while analyzing reported experimental data on 18 materials in figure 1. Materials which have high oxygen exchange rate can be explained by the multiplication of bulk ionic conductivity and electronic conductivity while others explained by the ionic conductivity.

Fig. 1 Relationship between oxygen exchange rate and bulk electronic conductivity, ionic conductivity, and the model function with both conductivities.

Figure 1

Chromium deposition and poisoning of cathodes is one of the most important degradation mechanisms of solid oxide fuel cell (SOFC). In this presentation, we will report the development of a BaO modified La_0.6Sr_0.4Co_0.2Fe_0.8O_3-δ (LSCF) cathode with substantial chromium tolerance, using EIS, SEM-EDS, XRD, XPS and ToF-SIMS. The BaO infiltration dramatically enhances the performance stability of LSCF in the presence of Fe-Cr interconnect. The infiltrated BaO nanoparticles inhibit the formation of SrCrO₄ by forming thermodynamically stable and conducive BaCrO₄, preventing the extraction of Sr at the A-site of LSCF perovskite structure and thus successfully mitigating the poisoning effect of Cr on the electrocatalytic activity and electrical conductivity of LSCF. This study offers a new and effective strategy to develop highly tolerant and resistant cathodes towards Cr contaminant for SOFCs.

La_0.6Sr_0.4Co_0.2Fe_0.8O_3-δ (LSCF) perovskite oxide is one of the most important cathode materials being developed for intermediate-temperature solid oxide fuel cells (IT-SOFCs); however, it is prone to surface segregation leading to the formation of undesirable secondary phases and reduction of active surface sites. In this study, dense LSCF thin films having controlled surfaces were epitaxially grown using pulsed laser deposition (PLD) on gadolinia-doped ceria (GDC) films with (100), (110), and (111) orientations, and then subjected to long-term annealing at 800°C and 900°C in air for 1-3 months. Detailed microstructural characterizations revealed preferential surface segregation for LSCF thin films which are predominantly (110)-oriented, conversely this appears to be mostly inhibited for (100)-oriented (pseudocubic) LSCF films, which exhibited extensive pore formation on the surface. Compositional and structural analyses revealed that the surface-segregated crystallites are SrSO₄ in origin, which is attributed to the reaction of the segregated Sr cations from LSCF with SO₂ present in the air. Furthermore, SrSO₄ formation on LSCF surfaces was accompanied by drastic morphological changes as well as decomposition of the perovskite phase. These results have implications toward tailoring the performance of cathode surfaces by understanding the dependence of cation segregation on driving forces such as surface chemistry and microstructure.

The exchange of oxygen between the gas phase and a mixed conducting ceramic, and its subsequent solid state diffusion, is key to the operation of electrochemical energy conversion devices such as Solid Oxide Fuel Cells (SOFCs) and Solid Oxide Electrolysers (SOECs), as well as dense membranes for oxygen separation. Indeed, it is often the surface reaction that determines the overall performance in such applications. However, whilst the mechanisms of oxygen diffusion in the bulk of these materials are generally well understood in terms of the defect chemistry and crystal structure, such an atomistic description of the surface exchange reaction still eludes us.

Whilst ¹⁸O isotopic tracer diffusion experiments coupled with SIMS analysis have been used for more than 30 years to study the kinetics of oxygen exchange and transport in candidate mixed conductors, relatively little attention has been paid to the characterization of the realistic surface composition of such materials; despite the great attention which is paid to the bulk chemistry, it is often assumed that the surface is a simple termination of the bulk. More recently, it is becoming apparent that the surface and near-surface composition after high temperature treatment is often radically different from the bulk.

Some of this recent work has been facilitated by the availability of new tools to probe the surface composition. Our group in particular has been applying Low Energy Ion Scattering (LEIS) spectroscopy, which is uniquely capable of determining the elemental composition of the single outer atomic layer of a material (i.e. the very same atomic surface that interacts with oxygen in the gas phase).

In this contribution we will use the commonly used perovksite La_0.6Sr_0.4Co_0.2Fe_0.8O_3-d (LSCF) to illustrate how LEIS and SIMS can contribute to our understanding of the relevant surface composition, complementing the existing understanding of the bulk properties of this commonly used electrode material. Starting from studies of surface segregation and restructuring in dense ceramic pellets after heat treatments similar to sintering conditions, we will then show how compositional changes can also occur at targeted device operating temperatures (around 500 ºC). Finally, we will show how laterally resolved LEIS analysis can be applied to understand compositional changes and interdiffusion occurring in the very outer atomic surfaces of LSCF microelectrode – YSZ electrolyte electrochemical half cells.

The present study is focused on alternative oxygen electrodes for Intermediate Temperature Solid Oxide Fuel Cells (IT-SOFCs) using Metal Supported Cells (MSCs) conditions. To prevent detrimental oxidation of the metal support, sintering of the cell components at high temperature under low pO₂ is required. Ln₂NiO_4+δ (Ln = La, Pr) compounds with the K₂NiF₄-type structure act as cathode materials for IT-SOFC due to their mixed ionic and electronic conductivity (i.e. MIEC properties). Pr₂NiO_4+δ shows excellent electrochemical properties at intermediate temperature (i.e. low polarization resistance R_p value, R_p = 0.03 Ω.cm² at 700 °C), while La₂NiO_4+δ exhibits higher chemical stability. Thus, the properties of La_2-xPr_xNiO_4+δ mixed nickelates were investigated with the aim to find best compromise between chemical stability and electrochemical performances.

Herein, the chemical stability of the nickelates under air at operating temperatures as well as the evolution of the polarization resistances during ageing (recorded under air at i_dc = 0 and i_dc ≠ 0 conditions) were studied for duration up to one month. La₂NiO_4+δ is chemically stable whereas Pr₂NiO_4+δ dissociates after 1 month at 600, 700 and 800 °C. At i_dc = 0 condition, the ageing of the half cells during 1 month under air shows no change in R_p, despite their various degrees of chemical stability. A different behavior is observed under i_dc ≠ 0 conditions with a large increase in R_p for Pr-rich phases in SOFC mode while interestingly, R_p remain stable in SOEC mode. In this presentation, the electrochemical properties of La_2-xPr_xNiO_4+δ under ageing and postmortem analysis will be discussed in details.

To increase Solid Oxide Fuel Cells performances, the optimization of the oxygen electrode is mandatory as its activation overpotential is higher than the one of the hydrogen electrode. One way to increase the cathode electrocatalytic properties and to decrease its polarization resistance is to increase the Triple Phase Boundaries zone, where the oxygen is reduced. A promising solution to achieve this goal is to disperse an oxygen reduction catalyst on the surface of a porous ionic conductor. The lanthanum nickelate La₂NiO_4+δ (LNO) has remarkable surface exchange reaction rate and oxygen diffusion coefficient. However, when the electrode is shaped by classical screen printing technique, it displays higher polarization resistance (≈ 1 Ω.cm²) than the standard lanthanum cobaltite-based materials (≈ 0.1 Ω.cm² for La_0.6Sr_0.4Co_0.2Fe_0.8O_3-δ).

In this study, cathodes were prepared by infiltration of lanthanum nickelate into a Gd-doped ceria (GDC) backbone sintered on 8YSZ electrolyte pellets. The influence of the preparation parameters of the composite electrodes on their electrochemical activity will be presented. The optimization of the preparation parameters led to a large decrease of the polarization resistance, down to 0.1 Ω·cm² at 600 °C. Using the Adler-Lane-Steele model, the impedance diagrams were fitted, allowing the determination of the surface exchange rate and ionic conductivity of the electrode: the obtained values well agree with those previously reported in the literature. Using this infiltration process, single cells were made starting from commercial half cells (Ni cermet/8YSZ), the cathode being the composite LNO/GDC. Voltammetry measurements showed power density higher than 1 W.cm^-2 at 800°C.

Solid oxide fuel cells (SOFCs) rely on long term, consistent operation of non-stoichiometric oxide components under elevated temperatures and oxygen partial pressure gradients.  Materials that exhibit the desired ion transport and reactivity properties generally also exhibit electro-chemo-mechanical coupling that can affect mechanical stability in operando.  The importance of chemomechanical coupling such as chemical expansion of SOFC electrodes is increasingly relevant to the goal of reduced operating temperatures enabled by thin film electrodes. Here, we explore this coupling experimentally and computationally, for the model SOFC cathode material, (Pr, Ce)O_2-δ. We quantified the Young's elastic modulus E of (Pr, Ce)O_2-δ thin films at temperatures up to 600^ºC and oxygen partial pressures pO₂ below 10^-3 atm via environmentally controlled nanoindentation. The observed decrease in E with increased temperature or decreased pO₂ correlated with the changes in oxygen vacancy concentration and lattice parameter expected due to chemical expansion. We compared these results with those we calculated for several compositions of bulk (Pr, Ce)O_2-δ via density functional theory, and found that the experimentally observed reduction in mechanical stiffness of reduced (Pr, Ce)O_2-δ thin films exceeded that predicted computationally for bulk counterparts. These results demonstrate that the in operando mechanical properties of non-stoichiometric oxides used in thin film SOFCs can differ significantly from those expected from characterization of bulk forms or at standard temperature and pressure. Thus, this chemomechanical coupling must be considered when designing mechanically robust SOFCs with thin film electrode components.

Intermediate Temperature-Solid Oxide Fuel Cell (IT-SOFC) electrodes manufactured through electrospinning are at the cutting edge of research since they offer the possibility of coupling good mechanical properties to specific geometric characteristics, such as a high surface/volume ratio [1]. This preparation technique offers advantages in terms of simplicity, efficiency, low cost and high degree of reproducibility of the obtained materials. Furthermore, suitable adjustment of the processing parameters allows to manufacture complex nanostructures, usually fiber-made, whose properties can be further tuned by the addition of nanoparticles, for instance through the infiltration method.

We consider infiltrated fiber-made electrodes from a modelling point of view, and in particular we address our attention towards MIEC (Mixed Ionic-Electronic Conductor) fibers. We consider that each MIEC fiber features two unconnected charge conduction paths, one for electrons and another one for oxygen-ions. Infiltrated dopant particles, adherent to the MIEC fibers, create contact points between the ionic and the electronic conductive paths, among which, otherwise, the charge transfer reaction would be negligible. Based on this picture of the doped MIEC fibers, a model is developed. The model includes the evaluation of i) electron and oxygen-ion conduction along the MIEC fiber, and ii) charge transfer reaction occurring at the doping particles and, possibly, at the electrode/electrolyte interface. The model relies on a detailed estimation of the model parameters, i.e. exchange current density and geometrical features, and includes an evaluation of the ionic and electronic electrode effective conductivities, accounting for percolation within the network of infiltrated particles.

We apply our model to different types of fibrous electrodes, anodes or cathodes [2-3], doped or undoped, manufactured through electrospinning but also through different techniques.

For example, we report here results about infiltrated cathodes based on fibrous LSCF scaffolds with different internal compositions. Simulation results are compared to literature experimental data, demonstrating good agreement (Fig. 1). In particular, the model captures very well the improvement of performance of the doped electrodes over the undoped ones, which can be five to ten fold or even more in some cases, and can bring the 1/R_p values to the order of magnitude of 10 S cm^-2 at 1000 K, which makes them good candidates for intermediate temperature solid oxide fuel cell (IT-SOFC) applications. The model allows to investigate, in detail, the effect of morphological and geometrical parameters on the various sources of losses, which is the first step for an optimized electrode design. This sensitivity analysis shows that, when increasing the doping level, the simulated 1/R_pincreases up to a plateau, and this is confirmed by literature experimental data (Fig. 1). In addition, the model allows to identify the extent of the electrode thickness where the electrochemical reaction effectively occurs, which is 10-20 µm close to the electrode/electrolyte boundary, depending on the operating conditions and the doping level.

In parallel, we are developing an experimental campaign devoted to investigate electro-spun fibrous LSCF cathodes. Electrospinning is based on the application of an electric field to a drop of fluid polymer on the tip of a spinneret. When the applied electric field reaches a critical value, then a charged jet of the solution is then ejected with evaporation of the solvent and simultaneous formation of solidified, continuous, ultra-thin fibers. For electrospinning experiments, we use a custom design equipment (Spinbow s.r.l., BO, Italy) which contains a high electric voltage supplier connected to the stainless steel needle and a sample collector. A syringe pump connected to the needle controls the flow rate of the polymer solution. The instrument is also equipped with coaxial spinneret in order to made core-shell nanofibers as well as hollow nanofibers. As an example, we report in Fig. 2 SEM images of one of our electro-spun nanofibrous polymeric scaffolds.

By coupling the experimental and modelling approaches we aim at (i) additional model validation; (ii) further understanding of the mechanism of electrochemical promotion of dopants; and (iii) optimization of the electrochemical performance.

• Cavaliere S., Subianto S., Savych I., Jones D. J. , Rozière J., 2011, Electrospinning: designed architectures for energy conversion and storage devices. Energy Environmental Science, 4, 4761-4785.

• Enrico A., Costamagna P., 2014, Model of an infiltrated La_1-xSr_xCo_1-yFe_yO_3-δ cathode for intermediate temperature solid oxide fuel cells. Journal of Power Sources, 272, 1106-1121.

• Enrico A., Costamagna P., 2015, Theoretical analysis of the electrochemical promotion of infiltrations in MIEC based electrodes for IT-SOFCs. Chemical Engineering Transaction, 43, in press.

• Lou X., Wang S., Liu Z., Yang L., Liu M., 2009, Improving La_0.6Sr_0.4Co_0.2Fe_0.8O_3−δ cathode performance by infiltration of a Sm0.5Sr0.5CoO3−δ coating. Solid State Ionics, 180, 1285-1289.

Figure 1

In recent years, solid oxide fuel cells(SOFCs) has been considered as desired devices for generating electricity by electrochemical combination of a fuel with an oxidant. Oxygen ion conductors are usually used as the electolytes for SOFCs. Among several types of oxygen ion conductors such as CeO₂, stabilized Bi₂O₃ and ZrO₂, Bi₂O₃with stabilizers exhibit the highest oxygen ion conductivity. Although Bi₂O₃ tends to be reduced to Bi metal, Bi₂O₃ incorporated with proper dopants will show enhanced stability against hydrogen. The highest OCV for SOFC using Bi₂O₃-based electrolyte is around 0.5V which is still too low to be practical. On the other hand, Bi₂O₃-based system may make a good ionic component for a composite cathode.

In this study, we choose yttria-stabilized bismuth oxides (YSB) as ionic conductor and mixed with strontium-doped lanthanum manganite (LSM) for composite cathode. Size effect of LSM on the polarization resistance was investigated by using convertional solid-state reacted and nano-sized LSM.

The electrochemical performances of composite cathodes with different LSM:YSB ratio have been investigated at temperature ranging from 500℃to 650℃ using AC impedance spectroscopy. The polarization resistance measured from a symmetric cell consisting of nano-sized LSM-YSB electrodes on a YSB electrolyte is 50% lower than that of submicron-sized LSM-YSB cathodes  by SSR (0.21 ohmcm²) at 650℃.

    The polarization of cathode is reduced because of the mixing of high surface area electric conducting materials and ionic conducting materials. The nanostructure of composite cathode is able to extend triple phase boundary(electronic conductor, ionic conductor,and oxygen gas) and provide numerous reaction sites for electrochemical reaction to occur.

The 2R-Cell, developed by Fiaxell Sàrl, provides robustness and reliability upon multiple thermic and redox cycles. In this article we present the results of 2R-Cell tested under thermic and redox cycles. Different cathodes were studied, composite LSC-GDC and LSCF, in terms of electrical performances and resistance to thermic cycles. Degradation due to multiple thermic and redox cycles is reported and discussed. Three composite LSC-GDC cathodes were tested (40, 60 and 75 % GDC). Surprisingly the highest power density was achieved with a cathode containing 75% of GDC in the composite layer. The cell provided a current density of 1060 [mA/cm²] at 0.8 V and 780°C, before thermic cycling. This cell underwent 40 thermic cycles with an average power loss of 0.7 % per cycle and providing a power of 621 [mW/cm²] at the end. A LSCF cathode equipped 2R-Cell, tested by EPFL our partner in an EU project (Roxsolidcell), was subjected to 20 thermic cycles and 20 redox cycles. With this cathode material the average ASR increase was 0.3% per thermic cycle and 0.5 % per redox cycle. SEM pictures of the tested cells after thermic and redox cycles are presented and discussed.

This paper reviews work carried out at Edinburgh Napier University on the manufacture of anodes and cathodes for solid oxide fuel cells (SOFCs) using electroless co-deposition of nickel and ceramic powders. Most of the work has been focused on anodes where the co-deposition of nickel and yttria stabilised zirconia (YSZ) negates the need for traditional manufacturing methods such as screen printing and removes the need for expensive and time consuming high temperature consolidation of the ceramic part of the composite.

Work has also been carried out manufacturing cathodes by electroless co-deposition replacing YSZ with lanthanum strontium manganite and lanthanum nickel ferrite. Anodes and cathodes of both planar and tubular designs of SOFCs have been successfully manufactured using this technique and process variables have included ceramic particle size, bath pH, ceramic loading, the inclusion of pore-formers and tubular orientation in the electroless bath.

Testing of the electrodes included impedance analysis, long term degradation testing at 850^oC for 500 and 1000-hours, high temperature electrical conductivity measurements, scanning electron microscopy and energy dispersive X-ray analysis.

The cost benefits of the electroless co-deposition process compared with other conventional processes are also discussed.

The TU Clausthal and CUTEC institute have teamed up to design, build and test a new stack architecture, based on repeating units with cells connected in parallel. This approach allows the use of electrically conducting seals like brazes and can be realized without any glass sealing.

Unlike conventional (serially connected) stacks, where all cells operate at the same current but individual voltage, in this setup each cell contributes as much current as possible at a common voltage. Thereby accelerated degeneration caused by undervoltage and the associated oxidation of the nickel catalyst can be completely overcome.

The concept uses two cells in electrical parallel connection with the anodes facing each other to form a repeating unit. Two designs have been realized:  An all ceramic housing made from 3YSZ, sealed with reactive air brazing (RAB) and a metallic one with a Crofer 22 APU housing, joined by laser beam welding. The housing materials have been chosen with respect to their thermal expansion coefficient (TEC) matching the TEC of the used ESC cells. Special interconnectors have been produced via hydroforming to create an even flow field while providing sufficient contact area.

The paper will focus on the test results of the repeating units under various operation conditions and the analysis of the degradation of components and joints.

Polarization curves for the twincell repeating unit and the individuell cells under H₂ and air at 880°C are depicted in figure one.

Figure 1

For application of highly active La_0.6Sr_0.4CoO_3-d  (LSCO) and Pr_0.6Sr_0.4CoO_3-d (PSCO) cathodes in the systems with yttria stabilized zirconia (YSZ) membrane the ceria based chemical barrier layers have been used to prevent active reaction between Sr containing cathode and YSZ. However, even in thin chemical barrier layer the Sr has some mobility. The aim of this study was to clarify how the microstructure of chemical barrier layer influences the Sr mobility and how the electrochemical parameters of SOFC are related with these properties.

Gadolinium doped ceria Ce_0.9Gd_0.1O_2-d (GDC) chemical barrier layers with thickness approximately 0.7 mm were deposited to the YSZ electrolyte using pulsed laser deposition (PLD), magnetron sputtering (MS) and spray pyrolysis (SP) method. One set of chemical barrier layers were studied as prepared (PLD GDC prepared at 600 °C; MS GDC prepared at 300 °C; SP GDC pre-sintered at 950 °C) and other set of samples were sintered for 30 h at 1300 °C before printing  and sintering of the cathode paste.  PSCO and LSCO cathodes were used to compare the Sr mobility in chemical barrier layer.  Both cathodes were sintered at two different temperatures, at 950 °C and at 1100 °C, supported onto differently synthesized and thermally treated GDC layers. Concentration profiles of Zr ion in GDC, Ce in YSZ, Sr in GDC and in YSZ were recorded using TOF-SIMS analysis.  XRD, SEM and electrochemical studies were also used to establish formation of new phases, GDC layer thickness and effective electrical parameters of studied systems, respectively.

            Systematic analysis of TOF-SIMS data demonstrated that thermal treatment has a significant influence on the behavior of chemical barrier layers. If non-sintered GDC chemical barrier layer was used, then at both cathode sintering temperatures (at 950 °C and 1100 °C) mobility of Zr, Sr and Ce ions was detected, less at lower and more pronounced at higher sintering temperatures. Very slight accumulation of Sr onto the interface of YSZ and GDC was detected in the case of SP layer, which is, without correct sintering, most nonhomogeneous. When cathode sintering at 1100 °C was carried out, the mobility of ions increased which led to intensive accumulation of Sr onto the interface between YSZ and GDC layers. In addition, a slight increase of thickness of chemical barrier layer was observed when high cathode sintering temperature was used, most likely caused by the formation of SrZrO₂ crystallites in GDC film. This phenomenon was also confirmed with analysis of XRD and HR-SEM data.

            When GDC barrier layer, deposited on YSZ, was sintered for 30h at 1300 °C, the mobility of Zr ions in GDC and Ce ions in YSZ was caused. When Sr containing cathode was sintered at 1100 °C on top of GDC layer contaminated with Zr ions, an intensive increase of GDC layer thickness occurred caused by the formation of SrZrO₂, but no Sr accumulation at interface between GDC and YSZ has been established. When cathode sintering at 950 °C was carried out, a slight Sr mobility in highly sintered GDC layers was observed in the case of LSCO cathode, but not in the case of PSCO if PLD and MS layers were applied.  However, in SP layers some Sr mobility was observed in the case of both cathode materials studied. 

            Electrochemical tests have been carried out using LSCO | CGO | YSZ | Ni-YSZ system using impedance spectroscopy method. High frequency resistance R_ex, dominantly characteristic for electrolyte and interlayers, depends significantly on synthesis method and sintering temperature of layer (microstructure of layer) and cathode. Results of electrochemical tests confirm the result obtained using TOF-SIMS analysis and show that better homogeneity of layer prevents the mobility and accumulation of Sr onto the interlayer between GDC and YSZ. The growth of SrZrO₂ crystallites inside of the GDC layer seems to be less harmful compared with the accumulation of Sr onto the interlayer of GDC and YSZ initiating formation of relatively dense SrZrO₂ film. 

Solid oxide fuel cells (SOFC) have developed to promising energy conversion devices for many applications. However, operation conditions under elevated temperatures, especially high temperature regimes and repeated temperature changes imply extreme demands for the used materials, questioning durability and reliability of such devices.

The scrutinized SOFC isolation layer system as shown in Figure 1 consists of a staple of ceramic layers sprayed on a steel substrate (interconnector). The main focus in this project is investigating application induced failure mechanisms under service conditions.

A newly developed ceramic of Mg-spinel type offers self-healing capabilities (Fig. 2). The self-healing effects provide fracture annihilating potential and promising mechanical capabilities. The manufacturing process of these ceramics causes a complex microstructure with structural defects such as pores or debonded splats. Splats are rapidly solidified molten ceramic particles on a substrate. Residual stresses, caused by quenching stresses during manufacturing as well as thermo-mechanical mismatch stresses due to temperature change may lead to crack initiation in the ceramic and subsequent mechanical weakening effects, respectively.

We present a microstructure-based approach to simulate the influence of thermal cycling on the ceramic coating stress evolution. This stress evolution and, thus, the damage behaviour is affected by the above mentioned influences. The numerical model represents a SOFC system which consists of a ceramic layer on a steel substrate and a thermally grown oxide (TGO) interlayer (Fig. 3). Assuming fracture mode I, the load case normal to the crack plane, the focus is set on the evolution of principal stresses in the ceramic microstructure.

The results of micromechanical simulations show how stress localisations due to structural defects such as pores or microcracks heavily influence this stress evolution and, therefore, the crack initiation in the ceramic layer.

We found the directions of the principle stresses in the microstructure facilitate delamination of the layer. Parameter studies show how residual stresses, TGO and interface roughness affect the stress evolution in the ceramic layer. A comparison of the new self-healing ceramic with a standard reference material point out the promising properties of this newly developed material. Detailed microstructure analyses and insights provide an understanding of the main failure mechanisms of the ceramic layer in depth and deliver clear optimization potentials for sprayed ceramic layer systems of SOFCs.

Figure 1

Metal-supported solid oxide fuel cells (SOFCs) use less expensive materials than traditional anode-supported or electrolyte-supported SOFCs.  The active layers of a metal-supported SOFC can be manufactured by plasma spraying without the need for high-temperature sintering.  Plasma spraying is a rapid-fabrication process that is scalable for large cell areas and for mass production.

Porous ferritic stainless steel (Sanergy HT, Sandvik, Sweden) supports were manufactured from powders by pressing and sintering pellets. The active cell layers consisted of a composite nickel - yttria-stabilized zirconia (YSZ) anode, a YSZ electrolyte, and a La_0.6­Sr_0.4Co_0.2Fe_0.8O_3-δ (LSCF) – Ce_0.8Sm_0.2O_1.9 (SDC) cathode.

Plasma spraying produces unique microstructures, as the cell layers are built up by the rapid solidification of molten (or semi-molten) splats. These microstructures were tailored by adjusting the plasma spray process parameters and the ingredients in the feedstock.  Dense electrolytes were made using a high-power plasma, a very high torch pass speed (to minimize the plasma heat impulse to the substrate), and a feedstock consisting of YSZ suspended in a mixture of water, ethanol, and ethylene glycol. Cathodes with four different microstructures were produced by varying the plasma spray parameters and mixing carbon or starch-based pore forming agents into the ceramic feedstock powders.  Resultant microstructures are shown in Figure 1.

Electrochemical performances of  metal-supported button cells with varying cathode microstructures were measured.  At 750°C, the cells had open circuit potentials of up to 1.057 V and peak power densities as high as 562 mW/cm².  Electrochemical impedance spectra were measured at an operating point of 0.7 V, varying the cathode gas with different air-nitrogen mixtures.  With 100% air as the cathode gas, the lowest cell polarization resistance was 0.29 Ωcm² at 750°C. Equivalent circuit modeling was performed to analyse how the cathode microstructure affected the cell performance.

Figure 1

High material and production costs are key barriers to the widespread commercialization of solid oxide fuel cells (SOFCs). Thermal spray techniques are a low cost alternative for the production of SOFCs. The objective of this work was to fabricate single cells by thermal spraying and to evaluate the impact of electrolytes and anodes, differing in terms of thickness, composition and microstructure, on the cell performance.

In this work, the anode layer was deposited on a planar ferritic steel support with in-plane dimensions 50 mm X 50 mm, thickness of 550 µm and 35-40% porosity by atmospheric plasma spraying (APS). The anode material consisted of nickel, yttria stabilized zirconia (YSZ) and a pore former. The anode layer thickness was around 60 µm. The electrolyte layer was deposited by a low pressure plasma spray technique called plasma spray-thin film (PS-TF) which can produce dense coatings at high deposition rates. The electrolyte material consisted of 3YSZ and 8YSZ consisting of different powder morphologies such as fused and crushed and agglomerated and sintered. In order to ensure sufficient gas tightness, electrolytes thicknesses of 50-90 µm were deposited. The cathode layer was screen-printed and sintered in-situ during the start-up of the electrochemical tests. The cathode material was La_0.58Sr_0.4Co_0.2Fe_0.8O_3-δ (LSCF). It must be noted that no diffusion barrier layer was deposited between the YSZ electrolyte and the LSCF cathode which will affect the performance and stability of the cells. The electrochemical tests were performed in a temperature range of 600-800°C at atmospheric pressure under various gas compositions of H₂, H₂O and N₂ on the anode side and of O₂ and N₂ on the cathode side while keeping the total gas flow constant. Current-voltage characteristics and impedance spectra were measured.

Porous anodes and dense electrolytes were achieved in all cases. The measured open cell voltage indicated acceptable electrolyte gas tightness. Power density of produced cells at an operating voltage of 0.7 V were measured as high as 600 mW/cm². In this work, the impact of spray parameters and applied layer thicknesses on the gas tightness of the electrolyte and the area specific resistance of the cell will be discussed.

Keywords: plasma spraying, solid oxide fuel cells, electrochemical characterization, impedance spectroscopy

Abstract

Solid Oxide Fuel Cells (SOFCs) are promising candidates for future energy conversion devices that transform chemical energy of fuel into electricity [1]. One of the main challenges in improving the performance and cost-effectiveness of SOFC is the control of the degradation processes, such as Sr diffusion and chemical interactions which can contribute to the overall reduction of the cell performance [2-4]. Several authors have evaluated the importance of cation diffusivities for surface segregation of Sr and thus for a major degradation mechanism of SOFC cathodes [5, 6].

In the present work, a 3000 hour degradation test was carried out at 780 ºC in a short stack under real time operation conditions. After the test, a cell from this stack was disassembled and samples from nine different areas were analyzed, looking for evidences of degradation phenomena taking place. A pristine sample was also analysed as reference.

The stack consist on coated ferritic stainless steel, anode supported cells and seal. The cells are supportend on a Ni-YSZ cermet anode with yttria-stabilized zirconia (YSZ) as electrolyte and a lanthanum strontium cobalt ferrite (LSCF) oxide as cathode material. A gadolinia-doped ceria (GDC) is used as barrier layer between cathode and electrolyte to prevent the formation of poorly conducting secondary phases, such as SrZrO₃ or La₂Zr₂O₇[7].

The aim of this work is to study the Sr diffusion, the effectiveness of GDC barrier layer and the evolution of LSCF cathode during operation. The deterioration of the performance measured in the cell is correlated with degradation mechanisms observed in post mortem experiments carried out in pristine and aged cells. The evolution of the Sr and other species in the cathode is examined by X-ray diffraction (XRD), confocal laser Raman spectroscopy, electron probe micro analyzer (EPMA-WDS), scanning electron microscopy equipped with an Energy-dispersive X-ray analyzer.

Besides, the concentrations of pollutants in cells were obtained by inductively coupled plasma optical emission spectrometry (ICP-OES) after dissolution in acid. A local microstructural and phase distribution study is carried out by means of transmission electron microscopy (TEM) and a high resolution scanning electron microscope coupled with electron energy-loss spectroscopy (EELS).

The results throw light on the evolution of the cathode/barrier layer/electrolyte system of SOFC cells working under real conditions after long operating time (Figure 1).

Figure 1. SEM images of the cross section and EPMA elemental distribution maps of the LSCF/GDC/YSZ in pristine and aged cells.

References

[1] K. Hilpert, W. J. Quaddakers, L. Singheiser, in "Handbook of Fuel Cells- Fundamentals, Technology and Applications", ed. W. Vielstich, H. A. Gasteiger and A. Lamm, John Wiley & Sons, New Jersey, USA, vol. 4, (2003) 1037-1051.

[2] R. Knibbe, J. Hjelm, m. Menon, N. Pryds, M. Sogaard, H. J. Wang, K. Neufeld, J. Am. Ceram. Soc., 93 [9] (2010) 2877-2883.

[3] D. E. Vladikova, Z. B. Stoynov, A. Barbucci, M. Viviani, P. Carpanese, J. A. Kilner, S. J. Skinner, R. Rudkin, Electrochimica Acta, 53 (2008) 7491-7499.

[4] F. Wang, M- E. Brito, K. Yamaji, D-H. Cho, M. Nishi, H. Kishimoto, T. Horita, H. Yokokawa, Solid State Ionics 262 (2014) 454-459.

[5] M. Kubicek, G. M. Rupp, S. Huber, A. Penn, A. K. Opitz, J. Bernardi, M. Stöger-Pollach, H. Hutter, J. Fleig, Phys. Chem. Chem. Phys., 16 (2014) 2715-2726.

[6] J. S. Hardy, J. W. Templeton, D. J. Edwards, Z. Lu, J. W. Stevenson, J. Power Sources, 198 (2012) 76-82

[7] A. Arregui, L. M. Rodriguez-Martinez, S. Modena, M. Bertoldi, J. van Herle, V. M. Sglavo, Electrochimia Acta, 58 (2011) 312-321.

Acknowledgements

The research leading to these results has received funding from the European Union's Seventh Framework Programme (FP7/2007-2013) Fuel Cells and Hydrogen Joint Undertaking (FCH-JU-2013-1) under grant agreement No 621207.

Figure 1

Solid oxide fuel cell short stacks for dezentralized stationary power delivery have been operated with special thin-film electrolyte anode-supported cells. After operation times of up to 3,000h the cells were post-test characterized. The cells have besides very thin electrolyte films (≤ 3µm) either LSCF or LSC cathodes (LSCF: La-Sr-Co-Fe oxide, LSC: La-Sr-Co oxide). Typical operation parameters were temperatures between 600 and 750°C, a current density of 500mA/cm², hydrogen with 20% water vapor and a fuel utilization of approx. 40%. Typical chromium evaporation protection layers like wet powder sprayed manganese oxide and cathode contact layers (a perovskite based on La-Mn-Co-Cu oxide) were used.

Microstructural post-test characterization reveals in the case of using a 1µm 8YSZ electrolyte prepared via sol-gel route and on top a sputtered diffusion barrier layer composed of GDC, the existence of SrCrO₄ either on top of the LSCF cathode as expected or, unexpectedly the formation of the spinel phase at the borderline cathode / barrier (electrolyte) layer. In case of using a 3µm thin screen printed 8YSZ electrolyte and an LSC cathode, the chromite phase has not been detected at the cathode / barrier border.

SEM and TEM post-test analysis of the cells show in case of the sol-gel electrolyte and sputtered GDC also the formation of micro-pores within the GDC layer. Theoretical calculations reveal that the chosen current density may lead to either preferred oxygen reduction or preferred chromium reduction depending on the real local current density within the cathode / barrier (electrolyte) interface.

Six solid oxide fuel cell stacks manufactured by Kyocera, MHPS, TOTO, NGK spark plug, NGK Insulator, Murata Manufacturing have been investigated to identify their performance degradation behavior, extract the major sources of degradation factors, clarify the degradation mechanism, and predict the time dependent contribution of such degradation to life time.  Stack performance was analyzed into several contributions, and an interesting correlation of overpotential values and ohmic loss has been obtained in addition to identification of respective degradation rates.  Respective degradation issues have been examined experimentally with SIMS, FIB-SEM, STEM and described as key simulation techniques to predict the long life up to 90,000 h.  Emphases have been placed on degradation of cathodes with Cr or S; to ensure the appropriateness for overcoming such poisoning effects, care is made to measure the impurity level after long term operation in various environments.  For simulation, focus was made to overcome some gaps between knowledge on stacks and on single (or buttons) cells appearing in several aspects.

A clinical correlation of multicomponent ceramic oxide based nanocrystalline cathodes is attempted to establish for conventional Sr-doped lanthanum manganite (La_1-xSr_xMnO_3, LSM) and a composite cathode comprising of Sr-doped lanthanum ferrites and cobaltites (La_1-xSr_xCo_1-yFe_yO_3, LSCF). For both cases nanocrystalline cathodes were synthesized using auto combustion technique. Precursors and powders were characterized thoroughly by thermal, structural and microstructural analyses. Powder characterizations of the perovskite-based cathode compositions exhibit crystallite size for LSM of 23 nm with average particle size ~50 nm and the same for cobaltite and ferritic based compositions revealed crystallite size 15-30 nm with particulate size ranging in between 50 nm and100 nm. Effect of cathode sintering was studied for all the compositions where it was found that doped ferrite and cobaltite based cathodes require lower temperature of sintering in the temperature range of 1000 – 1050 ^oC compared to that for doped lanthanum manganite (> 1100 ^oC). Sintering characteristics of all the cathode compositions were correlated with cathode bulk microstructures. Detailed electrical characterizations of the cathode compositions revealed that composite cathodes of ferrite and cobaltite based systems exhibit higher electrical conductivity of 480 S/cm at operating temperature 800 ^oC when sintered at 1050 ^oC compared to that for doped lanthanum manganite based system 195 S/cm sintered at 1100^oC. While the symmetrical cell performance with doped manganite cathode [LSM/LSM-YSZ/YSZ/LSM-YSZ/LSM] revealed electrode polarization of 0.23 Ω-cm² at 800 ^oC, the same for doped ferrite and cobaltite cathode [LSCF composite/Co Doped CGO/ LSCF composite] exhibits electrode polarizations of 0.02 W-cm² at 800 ^oC. Symmetrical cell study had also been performed for another composite material to be used as a composite interlayer containing 80 % LSCF and 20 % Co doped CGO with the cell configuration [LSCF + CoCGO20-LSCF 80/ Co Doped CGO/ CoCGO20-LSCF80 + LSCF]. It revealed that the electrode polarization is found to be even lower (0.012 Ω-cm²) than that found to be CoCGO-based interlayer. Oxygen reduction reaction (ORR) kinetics of the perovskite based oxides are evaluated from the impedance spectroscopy and calculated based on the low field approximation technique. Detailed comparison on exchange current densities of the perovskites are correlated with electrode compositions and their application using CoCGO-LSCF based composite interlayer. Systematic studies on the electrochemical performance of single cell with configuration [Ni-YSZ/YSZ/LSM-YSZ/LSM], [Ni-YSZ/YSZ/Co doped ceria/LSCF composite] and [Ni-YSZ/YSZ/doped ceria/CoCGO20-LSCF80/LSCF] exhibit the current density of 2.1 A/cm², 2.86 A/cm² and 3.32 A/cm² respectively. Though the current density is significantly higher for the LSCF based composite cathode with respect to LSM, the cell durability revealed the minimal degradation rate of the cell voltage (ΔV=1.7 % /1000 h) at constant current density of 0.5 A/cm² with LSM compared to LSCF based composite cathode (ΔV = 8.1 % /1000 h ). Use of the bi-layer concept viz. one layer of CoCGO and another composite interlayer containing CoCGO20-LSCF80 increases the cell stability with composite LSCF electrodes. The cell degradation is found to be nominal in the first region till 75 h. However, the cell degradation is found to be within 7% (ΔV < 7 % /1000 h) at constant current density 0.5 A/cm² when observed for a span of 500 h. The performance of cells fabricated with composite LSCF-based oxide systems have also been studied in details with doped ceria based composite interlayers and correlated with the microstructural analyses of SrO diffusion with respect to cell stability aspect.

SOFCs are a potential candidate for use as Auxiliary Power Units (APU's) on board heavy duty trucks. Heat from the SOFC exhaust is generally used either for cabin heating or for recuperative heat exchange and even after that there is still a considerable amount of high quality heat available. The unique selling point of an SOFC is the availability of both electricity and high quality heat. Not using the heat from an SOFC stack is tantamount to using only half the available useful energy from the fuel.

The ongoing research work reported here focuses on the design & development of a compact Solid Oxide Fuel Cell –Vapour Absorption Refrigeration System (SOFC-VARS) unit for refrigerated truck applications. This report presents a numerical model and simulation for thermal integration of an SOFC stack with a plate heat exchanger desorber via a specially designed tube in tube heat exchanger with internal fins where thermal oil is heated to the required desorber temperature and then serves as the coupling fluid in the plate heat exchanger desorber.

The sizing of the heat exchanger and desorber has been carried out to fit a small refrigerated van, to cater to a 1 kW cooling load. The modelling focuses on the heat transfer aspects at the SOFC and heat exchanger end and on both heat & mass transfer aspects at the desorber end.

The results show that a plate heat exchanger desorber is able to produce the required quantity of refrigerant needed for a 1 kW cooling load. A thorough sensitivity analysis on the plate heat exchanger desorber has also been carried out to identify the parameters that affect desorption performance the most and the parameters that have least effect on the desorption performance.

The use of plate heat exchangers as desorbers not only gives a high heat transfer surface needed for desorption but also leads to considerable reduction in desorber volume when compared to conventional falling film desorbers. Based on the results obtained from modelling, appropriate design maps have been drawn which showcase the sizing of the plate heat exchanger desorber and the SOFC stack and also the mass flow rate of thermal oil needed.

Hexis is developing and manufacturing SOFC-based micro-CHP systems for single-family or small multi-family houses. The current system Galileo 1000 N has an output of 1 kW electrical power. It furthermore covers the full heat demand of a standard single family house. More than 250 Galileo 1000 N systems have been installed up to now and are operated at customer's sites and in the lab.

The newest achievements in the field test and in the lab on durability and cyclability of SOFC stacks and complete micro-CHP systems are reported. Achievements are total system efficiencies of 95% (LHV) and electrical efficiencies of 35% (AC net, LHV). In tests with steam reforming, electrical efficiencies of more than 55 % (DC) have been shown on short-stack level. The longest system test had been running for more than 40 000 h. More recently, power degradation rates of approx. 0.5% per 1 000 h over more than 20 000 hours on short-stack level and over more than 12 000 h on full-system level have been demonstrated which were also confirmed in field tests. Cycling tests have shown a tolerance against 100 complete redox-cycles on short-stack level and 50 on-off cycles on system level. Summarized, stack lifetimes of 7 to 8 years are considered to be realistic. All-in-all, the technical market readiness has been achieved with the Galileo 1000 N. Consequently, an introduction into pilot-markets in Europe was started in autumn 2013.

SOLIDpower provides efficient energy solutions based on its proprietary planar SOFC technology. The first commercial product of the company is the EnGen™-2500. The company deploys the CE certified 2.5 kW combined heat and power generator within the FCH-JU project ene.field and with selected European partners. The EnGen™-2500 uses natural gas for combined heat and power generation (CHP) at high electrical and total efficiencies. The product targets multi-family houses and commercial applications. More than 60 units are to be installed within 2015 and field test results will be presented.

SOLIDpower operates a first "to scale" production cell in Northern Italy with a production capacity of 2MW/yr. Based on an ordering volume of 50 MW, the stack and balance of stack components match target market requirements, allowing selling and operating systems at grid parity prices.

Besides the micro-CHP generator deployment, SOFCpower also pursues strategic development activities to demonstrate biogas and waste-to-energy (WTE) applications on larger than 2.5 kW scale.

SOLIDpower progressed on the technological level for SO Electrolysis. Improved electrode materials and production technology open the door for lifetime for electrolysers stacks beyond 16'000 hrs. SOLIDpower expects that SOFC stack cost reductions stemming from SOFC volume manufacturing can be transferred to a large extend also to SOE stacks. In this field, SOLIDpower intends to work with integration partners on the systems level.

The paper provides an update of the stack and system development, including operational results of SOFC-based micro-CHP.

SOLIDpower comprises the companies SOLIDpower SpA Italy, SOLIDpower SA Switzerland and SOLIDpower GmbH Germany and unites under its new name the companies formerly known as "SOFCpower SpA" and "HTceramix SA".

In the past years, sunfire has increased its product portfolio and production depth. Four product classes (single stacks and stack modules for SOFC and SOEC) are tackling several markets (off-grid, on-grid, mobile). A ceramic center and a new production facility were inaugurated. The R&D activities are strongly driven by market requirements. With the new SOC stack electrolysis operation is introduced aside fuel cell operation. Furthermore, an increase in power density and a reduction of cost was achieved under the constraint of using standard industrial materials. Latest results of SOC single stack testing like I-V curves and reversible operation will be presented. The engineering and testing of stack modules in the power range >25kW will be also shown.

Fraunhofer Institute for Ceramic Technologies and Systems (IKTS) has long record in SOFC technology, stack design and system development for different applications. In this context IKTS is developing a patented LPG-fuelled power generator, branded eneramic^® system, since several years. This SOFC power generator is designed for use in leisure, industrial and surveillance applications. During the development phase, the demands of potential entry-level markets were identified with industrial partners to develop the technical requirements of the first prototypes. Currently the prototypes have a continuous power output of 100 Watts, short peaks of higher power can be buffered in a separate lead battery. The complete SOFC system has a volume of 55 liters at a total weight of 23 kg and achieves a net efficiency of 23% at nominal load. The standard fuel and efficient and long-lasting SOFC energy conversion make eneramic^®the most cost effective solution for off-grid power supply.

In the above applications frequent start-stop cycles are required, thus the system and the stack are consistently being optimized regarding thermal and redox stability. The compact SOFC-battery hybrid system is powered by a planar SOFC stack. The current stack design consists of 40 cells with 90µm electrolyte stabilized cells of partly stabilized zirconia (3YSZ) simply sandwiched between Crofer^® interconnects and printed layers of sealing glass. The design process for a reliable and cycleable stack involved the minimization of thermal mismatch within the stack assembly, a design of redox stable anodes as well as the optimization of the start-up procedure to reduce thermal stress. Different thermal heating rates up to 20 K/min have been imposed on the planar stack either in a furnace or an insulated hotbox environment. Today eneramic^®stacks sustain more than 110 system cycles without cell fractures at a power degradation below 0.3% per cycle in lab tests. The results of thermo- and redox-cyclization of stacks and complete systems are emphasized and discussed in detail. Besides it was found by quality analysis of different commercial electrolytes that especially the quality of the 3YSZ material itself has a predominant effect on the robustness and cycleability in later operation. Current possibilities for quality tests were insufficient and expensive whereof a noncontact inspection process for planar ceramics is under development at IKTS' material diagnostics branch. The measurement system is based on the optical coherence tomography (OCT). This quality inspection is capable to identify structural defects like fractures in ceramic electrolytes and reject them from MEA production at the earliest.

The prototypes have repeatedly shown a high degree of maturity in the laboratory scale. The long-term robustness of the insulated core module including SOFC stack and gas processing module (GPU) highlights the achieved power degradation of only 0.5% per 1,000 hrs. This ongoing propane fuelled test is currently operated for more than 14,000 hrs at full nominal power in a test bench. The ultra-compact gas processing module of the eneramic^® system is built of brazed Crofer^®sheet metal and ceramic components and combines CPOx-reformer, catalytic afterburner and cathode air preheater. Although operated at elevated temperatures up to 900°C in the burner the multilayer sheet metal assembly shows gas-tight behavior under aggressive redox atmosphere and no breakaway degradation.

To demonstrate the progressive technology level and to push commercialization a field trial phase has begun in 2014. For this purpose, a manual pilot scale production was established at IKTS. Currently the 3^rd generation of eneramic^®prototypes is tested under real-life applications in the field of cathode corrosion protection and mobile LED-boards for traffic road signs. In such long-run applications it was shown that the LPG-powered fuel cell system is superior to conventional battery-only powered solutions as is automatically charges the battery as long as fuel is available and thereby reduces the service costs. The results of these ongoing outdoor tests of several thousand hours are described.

The Fraunhofer patented eneramic^® technology is developed with financial support by the Fraunhofer Future Foundation. A spin-off from IKTS in cooperation with industrial partners is in preparation to bring the technology into market.

Since 2002 AVL is working on fuel cells with the focus on mobile and stationary SOFC systems. In 2014 the first completely standalone mobile SOFC power generator fueled with diesel was demonstrated. The system is equipped with all required additional components for autonomous operation like a battery pack and power electronics. During startup, the internal auxiliaries (e.g. the air blower), are consuming energy from the battery pack to heat the stack up to operating temperature. At a certain point the SOFC system starts to produce electricity which is used to recharge the batteries. An external consumer can draw a power of up to 5kW from the system. Short term power peaks (positive or negative) are buffered by a super cap. The operation is automatically controlled by the control unit implementing an energy management system. This enables a very user-friendly operation where the control system takes care of an always sufficient state-of-charge level of the battery pack. The presentation will also give a general status report and outlook of the AVL APU power generator development program.

Elcogen is a manufacturer of solid oxide fuel cell (SOFC) unit cells and stacks. The competitive advantage of Elcogen products is their performance characteristics at reduced temperatures. The unit cell structure is optimized for operating temperatures 600 – 700°C and is made of a material system from anode to cathode of NiO/YSZ – YSZ – GDC – LSC.  Elcogen tailors the unit cells with different thicknesses and shapes according to the customer requirements. Elcogen stacks are optimized for stationary applications in which both the operation temperature and pressure drop can be reduced due to stack characteristics enabling cost savings because of reduced material costs both in stack and system level, and increased system efficiencies due to reduced air compression losses and enhanced Nernst potential.

The article describes the performance characteristics of Elcogen unit cells and stacks. As an example, the area specific resistance of the unit cells has been determined to be 0.17 Ω.cm² at 700°C, U_fuel = U_air = 20%, and the degradation rate below 5 mΩ.cm².kh^-1 in 10,000 hours test. The average stack voltage is measured to be 0.97V at 30A, T_ave = 700°C, and V_air = 2.2 and V_H2 = 1.0 l_N/min/cell. The pressure drop for the air and fuel at the same conditions is less than 5mbar. With the encoring technical results and customer feedback, Elcogen is constantly increasing its unit cell and stack production capacity.

Based on more than 20 years SOFC scientific development, JÜLICH is still continuously improving its technology. Utilizing the proprietary cell manufacturing technology of warm pressing of the anode substrate cells of the dimension 20x20 cm² have been manufactured and tested in various stack sizes with power outputs of up to 5 kW, of which four were operated in a 20 kW system. It was demonstrated in a 1 kW stack that cells of this size can also be successfully operated using tape casted substrates. In parallel to the cell development the stack design was improved aiming at enhanced thermo-mechanical robustness. Thereby a stack in the kW range could be cycled successfully 100 times between 200 and 700°C. To be able to integrate also cells from other manufacturers the design was shifted to a picture frame layout, incorporating currently four 10x10 cm² cells in one layer, a larger number of cells is still optional. First successfully operated stacks showed a comparable performance to short stacks with one 10x10 cm² cell.

The 20 kW system set into operation in 2012 was operated for about 7000 h in total proofing thereby the suitability of the module and system concept developed in JÜLICH.

As special highlights, the long-term tests with short stacks have now reached under continuous operation at 700 °C and 0.5 Acm^-² times of 63,000 h (degradation 0.7%/kh, 0.3%/kh in the last 25,000h) and 33,500 h (degradation 0.3%/kh), respectively.

Hybrid power plants consisting of a gas turbine coupled with solid oxide fuel cells (SOFC) convert chemically bound energy into electrical energy. They can be operated at higher electrical efficiencies than conventional power plants. Theoretical studies suggest electrical efficiencies of up to 70 %.

The German Aerospace Center (DLR) aims to build and operate a hybrid power plant with an electrical power output of 30 kW. The system concept and design of the power plant have been finalized and the specification of all major system components has been carried out. Currently, different system components are being purchased and tested.

The presentation gives an overview of the current status of the project and illustrates the general concept of the power plant. All major components of the power plant are described. Important specifications and characteristics of the components are presented. Results of component tests are presented where available and the expected operating range is specified.

A case study analysis of economical and energy values of SOFC based fuel cell micro-CHP for residential homes.

Dr Alem Tesfai, Dr Anastasia Mylona and Professor John T S Irvine

KTP Associate Chartered Institution of Building Services Engineers222 Balham High RoadLondonSW12 9BS Tel: +44 (0) 20 8772 3699 at61@st-andrews.ac.uk, ATesfai@cibse.org

Abstract

This work presents the experimental and CFD model analysis of a 1.5 kW prototype SOFC micro-CHP system. The potential application of Fuel cell micro-CHP in a typical office building is investigated experimentally. With the current fuel cell technology the use of such systems in office buildings has the potential to be more efficient and reliable than in a single family home. This study documents the economic and energy related values of the fuel cell micro-CHP. This study will also look at how to optimise the operating conditions of the system in different climate conditions i.e. during the winter, autumn and the summer. To optimise and compere the experimental data a typical single family apartment heated by radiators will be analysed using a three-dimensional CFD model. Different heat transfer coefficients for the outer wall, the window and for the doors were considered. The seasonal heat requirements were also considered and heat recovery of the fuel cell micro CHP system was optimised. Different options i.e. whether to use the hot water supplied by the fuel cell micro CHP for domestic appliances or for space heating possibly by adding a combination boiler as a booster was also considered.

Over a three year period iPower has developed and now launched the UK's first commercially funded fuel cell micro CHP programme. This utilises the SOFC based BlueGEN product. From current information this is also a world first in providing FC CHP free of charge to the user on a commercial basis.

This is not an academic paper. It is a practical paper sett9ng out experience to date with this live commercialisation process.

The presentation will set out key challenges, success factors and learning points from deployment to date under the programme and prospects for future development. It will include independent third party data and case study material.

Solid oxide fuel cells (SOFCs) can, in principle, directly use hydrocarbon fuels such as methane and liquefied petroleum gas (LPG).  It was previously found that nickel-gadolinia doped ceria (Ni-GDC) anode has high durability under LPG utilization in SOFCs.  However, the performance under direct LPG utilization was low compared to hydrogen fuel, because anode polarization resistance increased by internal reforming.  Carbon was slightly deposited on the Ni-GDC anode after direct LPG utilization for 100 h at 650 ^oC.  In this work, a catalytic partial oxidation reformer was developed for LPG utilization, and high performance was obtained using reformate gas as well as hydrogen without carbon deposition on the Ni-GDC anode for 100 h at 650 ^oC.  Furthermore, microtubular SOFC stacks were developed using mass production technologies, and a proto-type 200 W micro power generator was demonstrated via NEDO project entitled "Technology Development for Promoting SOFC Commercialization".

A microtubular design for a solid oxide fuel cell (MT-SOFC) is studied for use in stacks for high strength and portability. Adapting readily available butane as the fuel source and intermediate temperature operation, MT-SOFC can replace batteries for various portable applications. This work studies the fabrication and preliminary performance results of a 5W MT-SOFC stack using butane as the fuel. The tubes were produced using a standard die extruder into a Ni-ceria-10mol% scandia stabilized zirconia (ScSZ) anode support. A ScSZ electrolyte and a La_0.6Sr_0.4Co_0.2Fe_0.8O₃ (LSCF) cathode were added on top of the tube, respectively, via a dip-coating method. The outer and inner diameters of the unit cell were 3mm and 2mm, respectively. The fabrication process was optimized and the electrochemical properties of a unit cell operated at intermediate (600~700^oC) temperature are described. Unit cells were then stacked for an output of 5 watts.

Fig. 1. (a) Pre-sintered anode support MT-SOFC; (b) Fabricated anode support MT-SOFCs; (b) Schematic of 5W MT-SOFC stack.

Acknowledgement

This work was sponsored by KETEP No. 4.0010459.01.

Figure 1

To investigate possibility to enhance electrical efficiency, performance of methane-fueled 10 kW SOFC systems was numerically simulated with calculated voltage-current-density curves at 750ºC of a single cell with a bipolar plate, which approximated a stack performance. Operation at 90% fuel utilization (U_f) with anode off-gas recycle (AGR) and with steam reforming (no recycle) led to 75-67%-LHV and 72-66%-LHV DC electrical efficiency at 0.05-0.40 A/cm², respectively. System simulations considering heat loss from a cell stack clarified that 750ºC stack temperature can be maintained at 0.12-0.40 A/cm² and 64-58% and 63-59% net AC electrical efficiency will be attainable at U_f = 90% with and without AGR, respectively. Furthermore, enhancement of net AC efficiencies was found to be 11-10 percentage points at those current densities against ones of a Ene-farm-like conventional system using steam reforming at U_f = 73%.

A thermodynamic and electrochemical framework for the analysis of the degrading fuel cell heat engine hybrid is developed and explored for various working reactions.  Fuel cell hybrids are a combination of energy conversion sub-systems – fuel cells and heat engines.  Fuel cell hybrids are important for the future for they are the most efficient devices when converting chemical energy of methane from natural gas and renewable fuels to electricity. While the perfect fuel cell would undergo no degradation, practical fuel cells with irreversibilities will degrade.   The common practice is to linearize degradation. However, experimental evidence shows that ASR(t) is commonly an ohmic parabolic function.   By the principle of superposition one can easily develop the equations for degrading hybrids from those developed for hybrids.  The path of maximum power can be developed for degrading hybrids as a function of ASR and temperature. The maximum power for a degrading fuel cell hybrid is given by the Gaussian hypergeometric function. The range of the operation of the SOFC is likely most conducive to hybrid operation for hydrogen oxidation as the working reaction. Each working reaction has a characteristic transition temperature and a characteristic optimal operating temperature.

Solid oxide fuel cells (SOFCs) are known for having high power generating efficiency among various types of fuel cells and being able to utilize various fuels such as fossil fuels. In the case of applying SOFC to vehicles, improvement of vehicle efficiency can be expected, using fossil fuels.

However, in the case of using fossil fuels for fuel cells, fuels must be reformed on board using H₂O, CO₂, and so on. Therefore, in order to utilize SOFCs to vehicles, methods of supplying these gasses on board need to be considered. To solve this problem, one of the promising methods is the anode off-gas recycle (AGR). The AGR is a technology in which SOFC exhaust gas, which contains residual fuels, is partially mixed with supplied fuels into the system prior to entering a fuel reformer, and recirculates these gasses in the system. By introducing AGR to SOFC system, i.e., by utilizing H₂O and CO₂included in the SOFC exhaust,, fuel utilization of the system can be improved and the steam reforming of fuel become possible without mounting water tanks on vehicles.

In addition, in the case of utilizing SOFC to vehicles, complicated responses to precipitous load variations must be considered, unlike stationary systems operated under moderate load variations. For example, in the case of increasing fuel supply in order to increase output power from the system, it could be possible that the output power is temporarily lowered because of decrease of temperature of the SOFC and reformer due to increase of heat requirement for fuel reforming. This behavior is due to transient variation of heat balance in the reformer. Therefore, for the analysis of transient response of SOFC systems, the response of reformer is very important and needs to be paid attention.

Especially, further complicated behavior of reformers is expected in SOFC systems with AGR (AGR SOFC systems) because of off-gas recycle. For instance, in the case of increasing fuel supply to the system in order to increase the output power from the SOFC, possible delay of fuel reforming in the reformer may cause the decrease in the ratio of increase of recycled gas to that of fuel supply. In such a case, the amount of H₂O and CO₂in the recycled gas decreases compared to the fuel supply, and the carbon deposition occurs, depending on the ratio of increase of fuel supply.

Hence, in this research, the transient behavior of the fuel reformer in AGR SOFC systems under various fuel supply rate is discussed by numerical simulation. Specifically, the effect of the change ratio of fuel supply and the delay of fuel reforming in the fuel reformer on system response is analyzed, by modeling composition of input gas to the fuel reformer. As a result of numerical simulation, the operating conditions in which the carbon deposition occurs are clarified. In addition, because ratio of increase of fuel supply needs to be limited in order to prevent carbon deposition, mitigation methods for this limitation are also discussed.

SOFC exhaust gases, such as methane and carbon monoxide, cause environmental problems. However, extra energy can be realized from oxidizing these out-gases into CO₂ and H₂O. This extra energy can be recycled in an SOFC system and increase the thermal efficiency. Thus the development of new catalysts that lower the oxidation temperature will make the SOFC system more efficient.

           CULLEN et. al.[1] reported high conversion rate of SOFC exhaust gas using a La-Cu-Mn impregnated alumina catalyst and a Pt-Pd catalyst in series. In this work, Cu-Mn catalysts that are used in the oxidation of methane or carbon monoxide were tested for SOFC exhaust gas oxidation.

           For determining the best ratio of Cu-Mn in catalysts, different ratios of La₂O₃-MnO-CuO/γ-Al₂O₃catalysts were prepared. The result shows approximtely 0.5 to 2 Cu/Mn ratio is the best composition for SOFC exhaust gas oxidation[Fig. 1].

Fig. 1. Conversion of CO(160℃), H₂(240℃), CH₄(460℃) with various Cu-Mn ratio.

           Cu-Mn perovskite and spinel supported catalysts were also tested. The perovskite supported catalysts showed much higher activity than the spinel supported. This was especially so for the case of methane conversion; the perovskite supported catalysts showed similar performance to alumina supported. The methane conversion of reference, CP and MP were 53.1%, 55.1% and 56.3% at 500℃, respectively.

References

[1] Greg CULLEN, Jon P. Wagner, Georg Anfang, Chandra Ratnasamy, METHOD FOR REMOVING CO, H₂ AND CH₄ FROM AN ANODE WASTE GAS OF A FUEL CELL AND CATALYST SYSTEM USEFUL FOR REMOVING THESE GASES, US 2012/0128563 Al, 2012

Figure 1

The solid oxide fuel cell (SOFC) is recognized as one of high efficiency plants of electricity generation for a sustainable future system with low environmental effects. In term of fuel flexibility, SOFC is suitable for direct use of syngas from a coal gasification plant.  The application of integrated SOFC in coal gasification based power plant, such as a light integrated gasification fuel cell (L-IGFC) power plant, could be one of the most promising coal energy conversions system providing high electrical efficiency with low CO₂ emitted to the atmosphere. The system consists of a coal gasifier and intermediate temperature SOFC module on the top of a steam turbine. The syngas produced by pressurized coal gasifier is used in SOFC module after heat recovery units and cleaning processes. To minimize exergy loses in the plant, the dry gas desulfurization (DGD) unit which can operate at elevated operating temperature and pressure are adopted. The heat generated by SOFC module can be utilized by the heat recovery steam generator providing high quality steam for a steam turbine. In this work, thermodynamic analysis is carried out to develop process integration of the L-IGFC plant. The overall plant analysis of a baseline system design is performed by identifying the major factors effecting plant performance by using the Aspen Plus simulation. The Aspen plus simulation is then used to perform parametric analysis to identify optimum values for the maximum system efficiency. Modifications to the baseline system design are made in order to determine the most viable system designs based on maximizing total system efficiency.

Solid oxide fuel cells (SOFCs) allow us to directly convert the chemical energy of fuel into electricity. With their emissions consisting mainly of Н₂О and СО₂, SOFC-based power generation systems are highly efficient (with efficiency up to 70%) and environmental friendly, which makes them increasingly attractive for the use in stationary power generation systems. The mixture of CO and H₂(synthesis gas) supplied to the SOFC anode can be produced from the natural gas using the dry or steam reforming or partial oxidation methods. The partial oxidation of methane, that is typically done with Ni catalysts, seems the best option for start-up systems of power plants with the anode gas recirculation or other applications where no water supply can be possibly organized. In this case, carbon deposition on the parts of reformer can be a problem, particularly, where the air-gas mixture supply system works abnormally.

The simple method of calculation of thermodynamically equilibrium composition of partial fuel oxidation products was used to calculate the degree of air consumption  in a reformer which permits to avoid soot formation. The value  ensuring auto thermal process in CPOX reformer was determined.  Air flow rate used in reaction with synthesis gas and for cooling the reformer, was calculated also.

An experiment is carried out to simulate the abnormal operation of the CPOX reformer with further soot formation on the Ni-catalyst and its walls. The results show that even in the event of the abnormal operation and carbon deposition in it, the CPOX reformer can continue working. Specifically, the carbon gasification takes place. The CO-content of the synthesis gas in this case is twice as much, on the average; however, this is not of critical character for SOFC systems.

A solid oxide fuel cell-gas turbine (SOFC-GT) hybrid system run on renewable fuels, such as biogas and ethanol, is an interesting power generation technology due to its high efficiency. To avoid a carbon formation on the anode catalyst by the direct feed of hydrocarbon fuels to SOFC, a pre-reformer is needed to integrate with the SOFC system for hydrogen production. As the steam reformer and air preheater are units that require high external energy input, the efficient heat management of the SOFC system is necessary. Regarding the SOFC system, the anode and cathode recirculations are employed to recover high-quality waste heat for reformate gas and gas turbine cycle, affecting the electrical and thermal system efficiency. In this study, a steam reformer and SOFC-GT hybrid system with anode and cathode recirculations is investigated. The aim of this work is to analyze the effects of the anode and cathode recirculations on the performance of SOFC-GT systems. The mathematical model of SOFC, gas turbine, and auxiliary units based on mass and energy balances under steady-state operation is employed for this study. The parametric analysis of key operating parameters, such as fuel utilization, steam to carbon ratio and operating pressure, on the system performance is performed. The results indicate that the SOFC hybrid system with the anode recirculation can reduce the supplied heat for the external steam reformer; on the other hand, it degrades the gas turbine performance due to a decrease in the turbine inlet temperature. When considering the hybrid system with the cathode recirculation, it is found that the efficiency of the SOFC-GT system is higher. However, the turbine inlet temperature is extremely high, leading to the cooling problem of a micro turbine.

This study examines the performance of brazed plate heat exchanger (BPHE) in an experimental air loop. The high temperature experimental system used for testing the heat transfer and pressure drop for simulation exhaust gas applicable for SOFC system is developed. The inlet temperature and mass flow rate were varied from 300 to 700 °C /0.0015~0.0041kg/s in the hot side to simulate high temperature exhaust gas and 25°C /0.0013~0.005 kg/s in the cold side. A total of  three tests are conducted (Test1: T_h=300~700°C, T_c=25 °C, m_h=0.0041kg/s, m_c=0.005kg/s; Test 2: T_h=700°C, T_c=25 °C, m_h=0.0034kg/s, m_c=0.0013~0.005kg/s; Test 3: T_h=700°C, T_c=25 °C, m_h=0.0015~0.0041kg/s,  m_c=0.005kg/s;). The experimental results showed that the effect of inlet mass flow rate on the temperature performance is quite significant, and the heat transfer coefficient and pressure drop are related to the mass flow rate.

Abstract

In this study, the cathode Ce_0.8Sr_0.2Fe_0.9Ir_0.04Co_0.06perovskite material was prepared by sol-gel technique and then tested for its characteristic application in solid oxide fuel cell. The textural properties were tested using X-ray diffraction, Raman spectroscopy, Fourier transform infrared spectroscopy. The morphology was examined using scanning electron microscopy and high resolution transmission electron microscopy. The electrochemical properties were measured using Kittec Squadro muffle furnace from Fiaxell SOFC technologies connected to a Nuvant ^TM potentiostat and galvanostat. These properties were tested for its potential application in low temperature solid oxide fuel cells within 300 -500 °C. It exhibited a current density of 801.90 mA/cm², a power density of 471.59 mW/cm² and an area specific resistance (ASR) of 0.349 Ω/cm².

A new Y_xBa_2-xCo₂O_5+δ cathodes were successfully prepared by solid-state reaction method. The structure, microstructure, conductivity, coefficient of thermal expansion and electrochemical properties of these Y_xBa_2-xCo₂O_5+δ cathodes were conducted by XRD, SEM, 4 DC four-terminal method, thermomechanical analysis and AC impedance, respectively. The YBaCo₂O_5+δ is a single phase with double perovskite structure. The second phases of YBa₂Co₃O_9-δ and Y₂O₃ were observed as the x>1 and x<1, respectively. The conductivity of all Y_xBa_2-xCo₂O_5+δ amples increased as temperature increased, and show a p-type conductor characterization. The YBaCo₂O_5+δ exhibited a maximum conductivity of 256 S/cm at 300 ^oC. The coefficient of thermal expansion is 13.8*10^-6, which is close to that of Samaria-doped Ceria (SDC) electrolyte. The YBaCo₂O_5+δdid not react with SDC electrolyte below 1050 ^oC. The YBaCo₂O_5+δ is a potential cathode material for intermediate temperature solid oxide fuel cell.

Introduction

Doped lanthanum cobaltite and its derivatives are used as cathode materials in solid oxide fuel cell (SOFC). For a high catalytic activity, the improvement of properties governing the cathode performance is important. It is generally regarded that one of the important factors is electronic structure of cathode material. It is known that the Co ion in doped lanthanum cobaltite shows three spin states, such as low spin (LS), intermediate spin (IS), and high spin (HS). Hong et al. pointed out that the change of electronic structure due to the difference of spin states affect the formation energy of oxygen vacancy in LaCoO₃ bulk system. Although it is expected that the high catalytic materials for oxygen adsorption and dissociation on oxide surface are proposed by the controlling the electronic structure of doped lanthanum cobaltite, the effect of electronic structure due to the difference of spin states of Co ion is unclear for the surface reactions on doped lanthanum cobaltite. In this study, we analyzed the stability of oxygen vacancy and surface reaction of oxygen under the different spin states of Co ion in Sr doped LaCoO₃by using density functional theory (DFT) calculation.

Computational Details

We used LaSrO-terminated (001) surface of cubic LaSrCoO₃ crystal structure. Half of La atoms were substituted by Sr atoms in LaO layers to analyze the effect of segregation of Sr in La₆Sr₆Co₈O₂₈ slab model. GGA-PBE with PAW potentials were applied with cutoff energy of 600 eV and 2x2x1 k-points. The initial spin configuration was constrained, such as LS, IS, and HS states, to obtain the different electronic structures. All DFT calculations were performed by using VASP.

Results and Discussion

We first analyzed the oxygen vacancy formation energy on the surface of La₆Sr₆Co₈O₂₈. The vacancy formation energies of one oxygen on surface under the LS, IS, and HS states were 0.92, 1.03, and 0.95 eV, respectively. The vacancy formation energies of two oxygen on surface under the LS, IS, and HS states were 3.27, 2.36, and 1.90 eV, respectively. The oxygen vacancy formation became unstable when the number of oxygen vacancy increases in all spin states. The oxygen vacancy formation energy for HS was the smallest compared with LS and IS states for two oxygen vacancy systems. We found that the difference of spin state of Co ion affects the oxygen vacancy formation energy on surface of LaSrCoO₃. In addition, we analyzed the stability of oxygen vacancy under the different Sr configurations to understand the effect Sr segregation. Details are discussed in the presentation.

Acknowledgement

The activities of INAMORI Frontier Research Center are supported by KYOCERA Corporation. All calculations are performed on the HA8000 computer systems in Research Institute for Information Technology, Kyushu University.

Reference

[1] W. T. Hong, M. Gadre, Y. -L. Lee, M. D. Biegalski, H. M. Christen, D. Morgan, and Y. Shao-Horn, J. Phys. Chem. Lett., 4, 2493 (2013).

Introduction

REBaCo₂O_5+δ (RE = Rare Earth) layered perovskite oxides are promising mixed ionic-electronic conductor cathodes for Intermediate Temperature SOFCs, thanks to their high electronic conductivity, high oxygen vacancy concentration, good ion transport capability and high oxygen surface exchange coefficients [1]. Their structure is characterized by the ordering of RE and Ba cations in a double perovskite structure (LnCoO_3-δ–BaCoO_3-δ), wherein Ln and Ba occupy alternate layers instead of being randomly distributed on the A site. Consequently, oxygen vacancies are located in the Ln-O layer. In this work, NdBa_1-xCo₂O_5+δcathodes with different levels of Ba deficiency were prepared and characterized by application of different techniques. The kinetics of the Oxygen Reduction Reaction (ORR) were investigated by Electrochemical Impedance Spectroscopy (EIS) tests. The results were numerically modeled to derive the main kinetic dependence.

Materials and Methods

NdBa_1-xCo₂O_5+δ samples (x = 0, 0.05 and 0.10, NBC0, NBC5, NBC10) were prepared via solid state mixing and firing method. X-Ray Diffraction (XRD) was applied to verify the phase purity. Lattice parameters were refined using the Pawley method. XRD analyses were also conducted on mechanical mixtures (50/50 wt%) of the samples with CGO (Ce_0.9Gd_0.1O₂) to evaluate the occurrence of chemical interactions after firing to 1100°C. Thermogravimetric Analyses (TG-DTA, air, 25-850°C, 3°C/min) and Temperature Programmed Oxidation tests (TPO, 2% O₂ in He flow, 20 Ncc/min, 25-850°C, 2°C/min) were performed to investigate the oxygen exchange activity. The oxidation state of Co and the O content in pure and Ba-deficient compounds were estimated by cerimetric redox titration. The electrical conductivity and the polarization resistance of the samples were measured as a function of temperature with a potentiostat/galvanostat equipped with a FRA. The electrical conductivity was measured via a four-electrode DC method on sintered bars between 25 and 850°C. EIS tests were performed using a symmetric cell configuration with CGO as the electrolyte. Pellets were fabricated via die-pressing and calcination (1400°C, 12 h, air). A slurry of the cathodic material was applied on each side of the pellet, dried at 110°C for 12 h in air, and calcined at 1100°C to reach adhesion. The EIS tests were performed in air and O₂/N₂ mixtures (O₂= 100 – 5% v/v) between 550 and 750°C. Scanning Electron Microscopy allowed to access the morphologic characteristics of both powder samples and symmetric cells.

Results and Discussion

The XRD patterns of NBC0, NBC5 and NBC10 show that all the samples have an A-site ordered tetragonal crystal structure (space group P4/mmm). Trace of an impurity phase (NdCoO₃) was detected in the NBC10 sample, indicating that the maximum Ba deficiency accepted is slightly lower than 10%. The Ba deficiency causes a shrinkage of the structure: the deficiency mainly affects the a lattice parameter, while the c parameter is much less affected. The cerimetric titrations show that the oxidation state of Co increases at decreasing Ba content, with a decrease of the total O content of the sample. For all the samples, the TG (Fig. 1A) and the TPO analyses show that the O content decreases at increasing temperature. As well, the Co oxidation state decreases, reaching an average value of 2.9 above 850°C. The transition from a semiconductor- to metal-type conductivity is observed for all samples in the electrical conductivity tests. The estimation of the ASR from EIS tests (Fig. 1B) revealed that the lowest apparent activation energy (1.30 eV) is achieved on the stoichiometric sample NBC0, and that the activation energy progressively increases at decreasing Ba content (1.58 eV for NBC5 and 1.73 eV for NBC10). The kinetic investigation was performed for each sample and the results were analyzed based on equivalent circuits (Fig. 1C). The main reaction steps of the ORR mechanism were identified (O^2- ion transfer at high frequency, O₂ adsorption at medium frequency), and their activation energy and O₂reaction order were quantified. The kinetic analysis reveals that the NBC0 sample is the most active, showing the lowest activation energy for both the steps. The EIS results and that of the conductivity suggest that Ba-site deficiency generates negative defects in the crystalline structure, which are balanced either by the formation of further oxygen vacancies or by increased fraction of electron holes, thus influencing the oxygen and electronic conduction properties. Good stability was observed in the ageing tests (500 h, time on stream, 500-750°C).

Conclusions

The temperature range (500-700°C) required by IT-SOFCs is especially detrimental for the kinetics of the oxygen reduction reaction. Double-layered perovskites based on Nd and Ba cobaltite reveal very interesting cathodic electrocatalysts.

References

[1] J.H. Kim, A. Manthiram, Journal of the Electrochemical Society, 155 (2008), B385.

Figure 1

Metal-supported solid oxide electrolysis cell integrated with metal interconnect by joining process (so-called, with fuel electrode(cathode)-supported electrolysis cell) was fabricated and its electrical performance are investigated under both operation mode of fuel cell(hydrogen) and co-electrolysis(steam and CO₂) mode at the intermediate temperature of 750 ^oC~850 ^oC. The fuel(or steam) electrode layers are prepared using NiO and YSZ(8 mol% Y₂O₃-doped ZrO₂) at the ratio of 1:1. In order to obtain pores in the cathode, graphite powders of 24 volume % with average particle size of 44 mm are added. In the case of a metal-supported solid oxide  electrolysis cell(SOEC) with heat-treated cathode support, 10 mol % Sm₂O₃-doped CeO₂ (SDC) paste is screen-printed on the surface of YSZ and subsequently (La_0.6Sr_0.4) (Co_0.2Fe_0.8)O₃ (LSCF) paste mixed with SDC at 1:1 weight ratio as anode(air electrode) is screen-printed on the SDC layer and heat-treated for desirable adhesion. The single electrolysis cell performances are compared by analyzing AC impedance spectra and product gas composition by GC(gas chromatography)of metal-supported electrolysis cell integrated with metal interconnect and cathode-supported electrolysis cell, and the current-voltage characteristics and cross-sectional microstructure of cathode-supported cell integrated with metal interconnect are tested. The stable long-term performance with co-electrolysis of carbon dioxide (CO₂)and steam(H₂O) is shown for about 800 h at 800 ^oC under the constant current density of 300 mA/cm², 600 mA/cm², 800 mA/cm², and 1000 mA/cm². In the fuel cell mode operation of metal-supported SOEC cell, the maximum power density of 0.8W/cm² is indicated with hydrogen fuel. The microstructures are analyzed using optical microscope for observing a diffusion of metal ions at the cathode, adhesion layer and metal support(ferritic stainless steel, STS430) with EDAX analysis. In the present work, we have investigated the performance and microstructures of metal-supported SOEC for the co-electrolysis of CO₂ and H₂O steam. The exemplary works with AC impedance and the composition of product gas are analyzed in the operation of metal-supported co-electrolysis cell.

The layered samarium and strontium doped perovskite oxide SmBa_0.5Sr_0.5Co₂O_5+δ (SBSC) was synthesized by glycine nitrate combustion process (GNC). The structure, microstructure, conductivity, coefficient of thermal expansion properties were investigated. The crystallinity of SBSC powder showed pure perovskite phase above 1000 °C according to the XRD and TGA results. With the calcination temperature increased, the crystallinity improved and the powders grew larger and agglomeration occurred. The conductivity of SBSC powders increased with elevated temperature while the coefficient of thermal expansion was 22.6 × 10^-6 K^-1. SBSC powders were utilized as cathode materials in Sr- and Mg-doped LaGaO₃ (LSGM)-based electrolyte-supported solid oxide fuel cell for power performance test. The cell with structure of (NiFe + CMF) | LSGM | SBSC was operated from 700 to 800 °C with humidified H₂ as a fuel and ambient air as oxidant. The maximum power densities are 550 and 170 mW cm^-2 at 800 and 700 °C, respectively. The experimental results indicate that SBSC is a promising cathode material for intermediate temperature solid oxide fuel cell (IT-SOFC).

Lowering the working temperature of solid oxide fuel cells (SOFCs) is necessary to avoid undesirable reactions between the cells' elements and to guarantee long-term stability and a short start-up time. At low temperatures (<650°C), the cathode is considered the main resistive factor for global cell performance, because the oxygen surface exchange is a rate-determining step in the oxygen reduction reaction at the cathode. Modifying the cathode surface with catalytic nanoparticles or thin films has been very effective in improving the surface kinetics of the cathode. Among the surface modification techniques, atomic layer deposition (ALD) has many advantages, such as a low deposition temperature (<250°C), precise control of layer thickness, and uniform distribution. In this study, we modified the surface of La_0.6Sr_0.4Co_0.2Fe_0.8O_3-δ (LSCF) with ALD of La_0.6Sr_0.4CoO_3-δ (LSC). The ALD LSC was added to the surface of the LSCF cathode by pulsed laser deposition in an anode-supported SOFC cell. As a result, at 600°C, the maximum power density of the ALD LSC-treated cell increased to 250 mW/cm² compared to the 198 mW/cm² density displayed by the bare cell. We will discuss the performance enhancement of the LSCF surface-treated with ALD LSC layers in terms of electrode kinetics and fuel cell performance.

Lowering operating temperatures is a promising strategy to improve the durability and reduce the cost of solid oxide fuel cell (SOFC) systems. However, the ohmic and polarization resistance would increase significantly at low temperatures (500-700 ^oC), particularly that of cathode electrode due to sluggish oxygen reduction reactions (ORRs). To reduce polarization loss at low temperatures, cathode materials have been studied extensively. The high performance cathode electrodes were also obtained through nano-scale engineering enabled by infiltration technique. Essentially the infiltration is an additive approach to add catalytically active materials into porous cathode to enhance the ORRs. Here we report the fabrication and evaluation of a novel interfacial nanospike-structured cathode. The growth of interfacial nanospikes on the pore walls of the cathode was enabled by a discharge treatment under low-temperature sintering process. The electrochemical performance of the corresponding button cells was well demonstrated at low operating temperatures of 350-550 ^oC. The stability of interfacial nanospikes was also examined under thermal-cycling conditions.

DC conductivity experiments were carried out in order to characterize the area specific resistances (ASR) of various multi-layer-structures that represent the contacting interface between cathode and interconnect of state-of-the-art planar anode-supported SOFCs. The investigation focused on quantifying the influence of various chromium evaporation protection layer materials (MnO_x, MnCo_1.9Fe_0.1O₄ (MCF)), perovskitic cathode contact layers (LCC10, LCC12, LSCF), operational parameters during stack joining and the effect of pre-annealing of multi-layer samples on the overall ASR of the model system. The results demonstrated the influence of different material combinations as well as the duration of heat treatment during the joining process on the cell resistance, whereas we did not observe an obvious effect from pre-annealing. During the microstructural post-test analysis we found a coarsened and inhomogeneous microstructure of the MnO_x barrier layer, which indicates a strong interaction with the adjacent materials. In addition, the adhesion of those MnO_x layers deposited by wet powder spraying is lower when compared to MCF layers deposited by atmospheric plasma spraying, which contributed to long term behavior of the resistances obtained. With these results it is possible to give a recommendation for the choice of contacting and chromium evaporation layer system and the operational parameters during the first heat up of an SOFC stack.

The microstructure of porous electrodes has a strong impact on a power generation performance of solid oxide fuel cells. Sufficient gas diffusion, electron transport and oxide-ion transport to the reaction sites is necessary for active electrochemical reactions. Recent advancement of 3-D microstructure observation by a focused ion beam and scanning electron microscope (FIB-SEM) or micro X-ray computed tomography enabled quantification of electrode microstructure. The three dimensional datasets of porous electrodes are also applied to numerical simulations of the electrode's polarization characteristics. Understanding the correlation between the electrode performance and microstructure is important to obtain guidelines for optimal microstructure. The numerical simulation requires a rate equation for the electrochemical reaction occurring at the reaction sites. It is often expressed by Butler-Volmer like form that includes the density of three-phase or two-phase boundary and the exchange current density per unit reaction site, i.e. unit triple phase boundary length for three phase boundary (TPB) reactions and unit area for two phase boundary reactions. Generally, the exchange current density per unit TPB length was derived from experiments using flat and dense patterned electrodes since there is an advantage that TPB length can be determined from its geometric shapes. However, we often observe a discrepancy between the polarization characteristics obtained from the simulation and the experimental measurements. One of the possible reasons for the discrepancy is the rate equation evaluated from the patterned electrodes which structure is very different from the real porous electrodes. This study focuses on LSM (Lanthanum Strontium Manganite) porous cathode. Although the number is still limited, we can find reports on the exchange current density per unit TPB length obtained from patterned LSM electrodes typically with a thickness of several hundred nanometers. It is also reported in literature that the thickness of patterned LSM electrodes affects activation overpotential. Since LSM has small oxide ion conductivity, oxide ion can be transferred through LSM phase, and the charge transfer at interface of LSM phase and pore phase (two phase boundary) can occur. This phenomenon may bring overestimation of the exchange current density per unit TPB length in the experiment with patterned LSM. In this study, we derive a model of exchange current density per unit TPB length considering its temperature and oxygen partial pressure dependency from experiments using LSM porous cathode with a particle diameter of approximately 3 micrometers. We assume that the charge transfer occurs only at TPB in this large particle. We first conducted power generation experiments using LSM/YSZ/Ni-YSZ button cells. For the reference electrode, a platinum wire was attached so as to surround the side edge of the electrolyte disk. Therefore the potential difference between the cathode and the reference electrode can be obtained. Activation overpotentials were obtained by impedance spectroscopy varying temperature (1073 - 1223 K) and oxygen partial pressure (5 - 21 %). After the power generation experiment, the cathode microstructure was observed by FIB-SEM and three-dimensional microstructure dataset with dimensions of 18 μm / 25 μm / 12 μm was obtained. From the geometry of its microstructure, TPB length was extracted and quantified. The exchange current densities per unit TPB length were calculated by substituting current density, activation overpotential and TPB density into Butler-Volmer like form. Eventually, obtained exchange current densities per unit TPB length are fitted into a power-law equation that has temperature and oxygen partial pressure term. As a result, obtained exchange current densities per unit TPB length in this study is found to be smaller than the ones estimated from patterned electrode at any condition. The discrepancy between them increased with decreasing temperature and/or with increasing oxygen partial pressure. For instance, the exchange current density derived in this study was less than half of that derived by Radhakrishnan et al. from patterned electrode at 1023 K for 21 % of oxygen. For actual LSM or LSM/YSZ composite cathodes using relatively large LSM particles, the equation of exchange current density per unit TPB length derived in this study is appropriate.

Figure 1

In the present study, La0.6Sr0.4Co0.2Fe0.8O3(LSCF)-Gd0.1Ce0.9O1.95(GDC) composite cathodes with different volume fractions were fabricated by the screen printing method, and their polarization characteristic and microstructure parameters are evaluated. Cathode overpotentials were evaluated at 700 ^oC and 100% O₂. Microstructures were reconstructed by focused ion beam-scanning electron microscope (FIB-SEM). In addition, current and potential distributions inside the reconstructed microstructures were calculated by a Lattice Boltzmann method. The relationship between polarization characteristics and microstructure is quantitatively investigated.

La0.6Sr0.4CoO3 (LSC) powder synthesis by Sol-gel process, together with the appropriate proportion of dispersants, binders, plasticizers, solvents and sintering aids, after thoroughly mixed, tape casting,and laminated to obtain the desired thickness. Then flow channel layer according to the design of the plain mesoderm laser cutting, the cathode current collector can be obtained LSC actually used - flow channel layer, as shown in Fig 1.

Figure 1

Murata Manufacturing Co., Ltd. has been developing planar-type solid oxide fuel cells (SOFCs) based on new design concept.  The cell is fabricated by a single-step co-firing process and consists of an electrolyte, electrodes, current collectors, gas-separators with manifolds, and gas-flow channels. The membrane electrode assembly (MEA) is sandwiched between ceramic gas-separators with via-hole electrodes providing electrical paths between cells. Removing metal interconnects around cathode in the cell enables reducing Cr vapor from metal and achieving high durability. A long-term operation has been carried out at 750°C over 8000h under galvanostatic condition. Single cell voltage degradation rate was less than 0.20%/kh after 2000h operation, and Cr poisoning in the LSCF cathode after long term operation over 8000h was not observed. From these results, the cell structure is effective to reduce Cr vapor in the cathode. But Sulfur distribution in the cathode was observed by analyzing in detail. Larger amount of Sulfur to form SrSO₄ was distributed at the air channel on the upstream than that on the downstream. And furthermore, Co segregation was observed in LSCF near the GDC barrier layer at the air channel on the upstream. To study sulfur poisoning mechanism, SO₂ poisoning test was carried out with various temperature and current density. Distribution of Sulfur and Co depended on operating temperature and current density.

Solid oxide fuel cells (SOFCs) offer great promise for the efficient and cost-effective conversion of chemical energy of fuels to electricity. However, sulfur in the air stream is one of the major contaminants affecting the performance stability of cathodes of SOFCs such as La_0.6Sr_0.4Co_0.2Fe_0.8O_3-δ (LSCF) and La_0.8Sr_0.2MnO₃ (LSM) perovskite oxide. In this report, the effects of operating temperatures, current densities, and SO₂ concentrations on the electrochemical performance of LSCF and LSM cathodes are investigated. Sulfur poisoning of LSCF is more pronounced at lower temperatures, i.e., 700 ^oC. According to SIMS result, sulfur deposition occurs inside the LSCF electrodes and is most pronounced on the surface of LSCF at 700 ^oC, forming primarily SrSO₄, while sulfur deposition occurs mainly near the electrode/electrolyte interface for LSM electrodes. Similar electrode resistance change of porous LSCF electrodes after being poisoned under open circuit condition and cathodic current passage at 700 ^oC indicates that sulfur poisoning effect of LSCF has no direct relationship with polarization current. The concentration of SO₂ has a significant effect on the performance degradation of LSCF and LSM electrodes. Sulfur poisoning is not reversible. The reaction mechanism between sulfur and LSCF and LSM electrodes is discussed.

Metal-supported solid oxide fuel cells (SOFCs) exhibited significant advantages, for example the superior mechanical strength, easy-processing and lower cost. In this paper, we report a new fabrication method for metal-supported SOFC single cell. The single cell with NiO and Fe₂O₃ (Ni:Fe=1:1) anode support, Gd_0.1Ce_0.9O_1.95 (GDC) electrolyte and La_0.4Sr_0.6Co_0.2Fe_0.7Nb_0.1O_3-δ (LSCFN) cathode was fabricated by traditional tape-casting, drop-coating, co-sintering and screen-printing techniques. Then the as-prepared single cell was reduced at 500, 600, 700 ^oC under 5%H₂-95%N₂. Results show that after treated the whole cell at 700 ^oC in the reducing atmosphere, the oxide support was mostly reduced into metal, while the LSCFN cathode maintained its perovskite structure. Then the cell performance was tested using humidified hydrogen, by comparison, the as-prepared cell without reduction was also tested. Electrical impedance spectroscopy tests were carried out with symmetrical LSCFN electrodes reduced at 500-700 ^oC. Results show that the polarization resistance lowered down with the reduction temperature increased, which may possibly due to the more oxygen vacancies created at higher reduction temperature. The fabrication method to reduce the whole cell is a promising way to fabricate the easy-processing metal-supported SOFC single cell.

The electrochemical performance of Ag-Y_0.5Bi_1.5O₃ (YSB) cathodes on Ce_0.78Gd_0.2Sr_0.02O_2-δ (GDCS) electrolyte with an epitaxial Y_0.16Zr_0.84O_1.92 (8YSZ) thin film outer layer was estimated using electrochemical impedance spectroscopy (EIS). A series of bismuth oxides with different silver contents were prepared and measured by an impedance spectroscopy on symmetrical cells in the temperature range from 500 °C to 650 °C. The microstructural analyses of porous composite cathodes and bi-layer electrolyte with fluorite structures were investigated by using scanning electron microscopy (SEM) and transmission electron microscopy (TEM). The results showed that silver in the cathode reduces the surface resistance which improves the efficiency of current collection from the cathode. Besides, the bi-layer electrolytes have lower activation energy than GDCS electrolytes. Without topping Pt or silver metallic working electrode, the 50-70 vol% of Ag powder mixed with a submicron size YSB powder exhibited much lower overpotential and higher exchange current density than Lanthanum Strontium Cobalt Ferrite perovskite cathodes below 650 °C.

A high ionic conductivity is desireable property for cathode materials for

solid oxide fuel cells. LaMnO₃ is a well studied candidate for this applica-

tion. Hence, it is important to understand the oxygen transport mecha-

nism. We find increase in oxygen vacancy formation energy and diffusion

under compressive and tensile strain as well as under Sr-doping using the

density functional theory and nudged-elastic band calculations. The oxy-

gen vacancy sites and diffusion paths are studied. We investigate the effect

of Sr-doping on the anisotropy of the diffusion. Modified strengths of

the Mn and O bonding as well as volume change are identified as fac-

tors that affect the formation/diffusion of vacancies. Furhermore, tensile

strain facilitates enhanced oxygen vacancy diffusion along the [011]/[0 ̄11] direction. The weakening of the Mn- O bonds as well as the distortion of

the MnO6 octahedra hinder diffusion along the [110]/[1 ̄ 10}direction.

The LaCoO₃ class of materials is of interest for cathodes of solid oxide fuel cells. Spin-polarized

density functional theory is applied to cubic La_0.75 (Mg/Ca/Ba)_0.25 CoO₃ . The effect of this cation

doping on the electronic and magnetic properties as well as oxygen vacancy formation energy is

studied. Oxygen vacancies with proximity to the dopant are energetically favourable in most cases.

We discuss the effect of distortions of the CoO6 octahedron on the electronic structure and the

formation energy of oxygen vacancies. The order of formation oxygen is found to be Mg > Ca >

Ba. Cation doping incorporates holes to the Co-O network which enhances the oxygen vacancy and migration.

Solid oxide fuel cells (SOFCs) have received a lot of attention as promising electrochemical power generation devices due to their high electrical efficiency and adaptability to a variety of fuels such as hydrocarbons, ammonia, and hydrogen. The utilization of LaSrTiFeO_3-δperovskite oxide was proposed as an oxygen permeable membrane and electrode materials of SOFCs. Durability in thermal cycling test is an important factor for applying SOFCs to power sources. The reduction of operating temperature and enhancement of thermal-cycling capability are indispensable requirement for industrial-scale commercialization of SOFCs.

   A new cathode material for SOFCs has been developed in R&D Center of Noritake Co., Ltd. In which, it contain La_0.6Sr_0.4Ti_xFe_1-xO_3-δ (0<x<0.4) and La_0.6Sr_0.4Co_yFe_1-yO_3-δ (0<y<1). Here, we report the performance evaluation of our developed cathode material. It was found that the developed cathode material had a high power density and extremely high stability to hydrogen and ammonia fuel.

The polycrystalline oxide (La_0.8Sr_0.2)(Cr_0.2Fe_0.8)O_3-d  or LSCrF is a mixed ionic and electronic conductor (MIEC) of a type called "acceptor-doped transition metal oxide" that has been explored for application as an electrode material in SOFC and other electrochemical devices working at intermediate to high temperatures (500°C to 900°C) [1]. The strontium acceptor doping on the A-site of this perovskite ensures high oxygen ion conductivity whilst the chromium substitution on the B-site ensures the stability of the perovskite structure for a wide range of oxygen partial pressures at the elevated temperatures.  The suitability of this material as an electrode material is determined through the measurement of its ionic conductivity for oxygen ions and the rate of oxygen incorporation into the oxide at elevated temperatures using stable oxygen isotopic methods together with secondary ion mass spectrometry (SIMS).

The isotopic exchange, depth profiling method (IEDP) method allows the bulk oxygen transport parameters, the diffusion coefficient, D(T) and surface reaction coefficient, k(T) both functions of temperature, T, to be determined once an appropriate solution of the diffusion equation [2] has been identified and assuming compliance with the boundary conditions of that solution.  The IEDP results dataset, normalised isotopic fraction measured at N depth intervals, (x_i, C_xi ) for i=1 to N is matched by the 'least squares' process that determines the most probable values of the two parameters, D^Fit and k^Fit, that minimises the 'sum-of-squares' of the residual differences, r_xi  between the measured profile data C_xi and the values calculated from the solution, C_xi^Fit , where r_xi = C_xi-C_xi^Est .  Ideally the conditions,  ∂/∂D(∑(r_xi)²)=0 and ∂/∂k(∑(r_xi)²)=0 are satisfied.  This process assumes that the variance associated with the depth measurement set, V_x is small compared to the variance of the measured isotopic fraction set,  V_Cx^Fit . V_D^Fit , the variance in D^Fit is calculated as a product, ( ∂D/∂r_x)²V_Cx^Fit.  where the partial differential, (∂r_x/∂D) , is the gradient of the sum of the residuals squared along the D axis for a fixed value of k^Fit in D and k space. The variance in k^Fitis calculated from a similar product, V_k^Fit=( ∂k/∂r_x)²V_Cx^Fit .

This statistical quantification of the variance of D and k for each IEDP measurement of the LSCrF oxide is reported in this contribution for three different exchange anneal ambients with an oxygen activity range between 1 and 1x10^-14 [3].  All the calculations are done by numerical evaluation so the effect of the finite element size, (ΔD,Δk) and termination conditions for the 'best-fit' is also reported. .  A comparison is made with the 'goodness-of-fit' tests using the often quoted "R-squared" and "chi" values from their probability distributions.

The variances in D^Fit and k^Fit are known to be sensitive to the surface isotopic fraction, C_x=0 , which for the semi-infinite solution becomes, C_x=0=1-exp(h'²).erfc(h')   where h'=k(t_anneal/D)^1/2 and t_anneal is the time for the oxygen-18 exchange anneal.  Values of h' around 0.2 are preferred for similar variances in D^Fit and k^Fit whilst values of 0.02<h'<2 lead to higher errors with dissimilar upper and lower limits.  This effect is shown from the results set for LSCrF with two cases where h' is 7.3 in one dry oxygen exchange and h'is 0.073 in a steam and forming gas exchange reaction.

The variance of D and k for each IEDP measurement is used to weight the plotted points in the Arrhenius plot and in the 'best straight-line fit' for an estimate of the activation energy for both diffusion and the surface exchange reaction. In this situation with often just a few points it is essential to apply the F-test or t-test significance tests for the null-hypothesis.  The results for LSCrF are plotted with the standard 2-theta confidence limits derived from the 'best-fit' so that a reasonable assessment can be made with data in the literature that has been measured under similar conditions [4].

1.            Mizusaki, J., Solid State Ionics, 1992. 52(1-3): p. 79-91.

2.            Crank, J., The mathematics of diffusion. 2nd ed. 1975, Oxford: Clarendon Press.

3.            Chater, R.J. Ph.D. Thesis, Imperial College London, 2014

4.            Ramos, T., Solid State Ionics, 2004 170(3-4): p. 275-286

Mixed ionic-electronic conductors (MIECs) show promise as next-generation solid oxide fuel cell (SOFC) cathode materials with improved performance at intermediate temperatures (500-800°C) as compared to materials lacking oxide-ion conductivity. The complex interplay of electronic and ionic conductivity in these strongly correlated systems often derives from the mutual influence of transition metal cation mixed valency and oxygen nonstoichiometry. Among MIECs tested as SOFC cathodes, the Ruddlesden-Popper phase La₂NiO_4+δ exhibits rapid oxygen transport at low temperatures, attributed to loosely bound interstitial oxides (0 < δ < 0.3). Key to understanding the high oxide-ion conductivity in La₂NiO_4+δ is experimental confirmation of hypothesized interstitial- and vacancy-mediated mechanisms at the atomic level.

In this study, solid-state ¹⁷O MAS-NMR spectroscopy of La₂NiO_4+δ at temperatures up to 800°C is supported by a theoretical methodology equipped with results from periodic hybrid DFT calculations. Three distinct ¹⁷O resonances are observed and assigned to equatorial, axial, and interstitial oxygen environments in La₂NiO_4+δ. Moreover, with high-resolution MAT-PASS experiments, the axial feature splits into several resonances, consistent with local structural distortions due to nearby interstitials. Loss of the interstitial oxygen feature in the NMR spectra upon heating to ~150°C is attributed to onset of exchange of interstitial and axial oxygen sites. Structural rearrangements due to interstitial motion also manifest as linewidth changes in the axial and equatorial resonances. At operational SOFC temperatures (600-800°C), interstitialcy and vacancy mechanisms of oxide-ion conduction are tentatively confirmed for the first time, showing that interstitial-axial exchange continues to dominate transport at the highest temperatures. Slow vacancy diffusion (equatorial-axial exchange) is surmised to limit the overall three-dimensional ionic conduction. This work is among the first examples of dynamics in a paramagnetic oxide-ion conductor studied by ¹⁷O solid-state NMR.

Background

La₂NiO_4+δ and Pr₂NiO_4+δ are known as mixed conductors and are candidates for SOFC cathodes [1]. SOFC cathodes are frequently used with a ceria buffer layer to avoid any reaction between the cathode and the zirconia electrolyte. We investigated the properties of Pr₂NiO_4+δ and La₂NiO_4+δcathodes and the influence of the ceria buffer layer on the cathode properties. We also investigated the phase compatibility of those cathodes with ceria (GDC) and zirconia (8YSZ).

Experiments

First, La₂NiO_4+δ and Pr₂NiO_4+δ cathode materials were synthesized by employing a solid-state reaction at 1300^oC. The cathode materials were ball-milled. The mean diameter of the grain of the cathodes was about 1 micron. Each material was in a single phase. The cathode slurries were printed on a Ce_0.8Sm_0.2O_1.9 (SDC) spin coated buffer layer on an 8YSZ (0.2 mm thick) electrolyte. The cathode sintering temperature was 1100^oC. Pt was used for the anode and reference electrodes. Three-terminal AC impedance measurements were conducted before and after the DC current loading processes (106 mA/cm² and then 354 mA/cm²) at 800^oC in air. The phase compatibility of the cathode with ceria or zirconia electrolyte was examined with x-ray diffraction analysis. The cathode and Ce_0.9Gd_0.1O_1.95 (GDC) or 0.92ZrO₂-0.08Y₂O₃ (8YSZ) powders were mixed and sintered under the same conditions as the cathode sintering process (1100^oC for two hours). Then their crystal structures were examined with an x-ray diffractometer.

Results and discussion

The impedance plots for those cathodes at 800^oC are shown in Fig.1.

They were measured after a DC current loading of 354 mA/cm² for 55 hours. The cathodes had a similar grain size, but the Pr₂NiO_4+δ cathode had a smaller interface resistance than the La₂NiO_4+δ cathode. To examine the influence of the reaction between and 8YSZ electrolyte, we also fabricated cells without the SDC buffer layer. The impedance plots for the cathodes without the buffer layer are also shown in Fig.1. The interface resistances for Pr₂NiO_4+δ and La₂NiO_4+δ cathodes without the SDC buffer layer are smaller than those with the SDC buffer layer. This shows that the SDC buffer layer degraded rather than improved the cathode performance. Fig.2 shows the X-ray diffraction patterns of the mixture of Pr₂NiO_4+δ powder and electrolyte powders (GDC or 8YSZ) sintered at 1100^oC. The Pr₂NiO_4+δ reacted completely with the GDC and produced Ce(Pr)O_2-δ or Ce(La)O_2-δ and NiO. This shows that Pr₂NiO_4+δ is highly reactive with cerias including GDC. The reaction products of the mixture of cathode and 8YSZ are also shown in Fig.2. After sintering, Pr₂Zr₂O₇ was formed, but Pr₂NiO_4+δ remained. This shows that the Pr₂NiO_4+δ reacts with zirconias in the cathode sintering process. But the interface between Pr₂NiO_4+δ and zirconias are more stable than that of Pr₂NiO_4+δ and cerias.

These results reveal that Pr₂NiO_4+δ and La₂NiO_4+δ cathodes without an SDC layer outperform those with an SDC layer.

  References

[1] J. C. Grenier, et al.ECS Transactions, Vol. 57 (1) pp. 1771-1779 (2013).

Figure 1

Recent studies related to SOFC technology involve materials design on the electrodes and electrolyte layes in search of candidate materials with high O ion conductivity even at lower operating temperatures. This involves understanding the detailed mechanisms of O ion diffusivity at materials which are known to be good O ion conductors at low temperatures. From the many candidate materials, a group of perovskite materials in the K₂NiF₄ structure has been eyed as promising materials for these purpose, specifically Pr₂NiO₄-based materials. Interest in the study of Pr₂NiO₄ and Pr₂NiO₄-based materials rooted from its potential application as a mixed ionic and electronic conductor (MIEC) as a cathode material for SOFC.

In this study, first principles calculation based on density functional theory (DFT) was used to analyze the structural and electronic properties of Pr₂NiO₄ and doped Pr₂NiO₄ systems (Pr_2-yR_yNi_1-x-y-zE_xT_zO₄ (R=La, and E=Cu and T=Ga)). With the aim of determining possible Pr₂NiO₄-based candidate materials for the electrolyte layer, the effects of doping on the migration of O^-2 on an assumed promising system was analyzed.

Two types of cation dopant sites were assumed in this study: (1) a large cation dopant substituting Pr, and (2) a smaller cation dopant substituting Ni. For the first case, the effect of La dopants, with different concentrations, on the change in the structural and electronic properties of the host bulk structure was analyzed. It was observed that La dopants have a dominant effects on the change in the lattice parameter along the the c direction due to the difference in ionic radius. This is assumed to be a factor affecting the O ion conductivity within the material. For the second case, co-doping of Cu and Ga (Pr₂Ni_0.5Cu_0.25Ga_0.25O₄), and doping of Ga (Pr₂Ni_0.5Ga_0.5O₄) with different structures were analyzed. For the co-doped system, it was observed that due to nearly similar atomic radii of Cu and Ga, no cation-O atoms elongation were observed. Cu dopants are considered to increase the electronic conductivity of the system. On the other hand, Ga doping shortens the Ga-Oap distance while elongating the Ni-Oap distance, which affects O ion migration through interstisialcy. Furthermore, electronic property analysis also shows that there is a minimal state for the Ga orbitals along the Fermi level as compared with the host Ni atom. And that higher oxidation state of Ga than Ni will tend to cause increased concentration of interstitial O atoms. These results show favorable characteristics for an electrolyte material.

The development of a cathode that is stable against CO₂ present in the atmosphere still remains a challenge. Over the past ten years layered perovskite La₂NiO_4±δ has been the subject of careful research as far as also concerns their potential use as cathodes in electrochemical devices. So far as La₂NiO_4±δ is a p-type conductor, appropriate acceptor doping on A-site could improve the electronic conductivity by generating extra electron holes for charge compensation. Experimental results reveal that the electronic conductivity does indeed increase with a certain amount of Sr or Ca-doping. However, the substitution of La by alkali-earth elements deteriorates its electrochemical activity due to a decrease in the sub-stoichiometric oxygen amount in the structure and ionic component of conductivity [1]. One way to increase ionic conductivity is development of composite electrodes with ionic conductor as a second component, usually CeO₂-based electrolyte [2]. The composite electrodes with proton-conducting electrolyte, especially in case when such electrodes are proposed to be used in contact with this electrolyte, could provide more extended area of electrochemical reaction.

     The actual work focuses on the electrochemical performance of the La_1.7Ca(Sr)_0.3NiO₄-based composite cathode materials with Ce_0.8Sm_0.2O_1.9 (SDC) and BaCe_0.89Gd_0.1Cu_0.01O₃ (BCGC) ceramic component (in wight ratio 1:1) which were applied as a functional layer in two-layered electrodes with 98 wt.% LаNi_0.6Fe_0.4O₃ + 2 wt.% CuO LаNi_0.6Fe_0.4O₃ and 99.4 wt. % La_0.6Sr_0.4MnO₃ + 0.6 wt.% CuO as a collector layer. Studies by impedance spectroscopy were performed in contact with BCGC and BaCe_0.5Zr_0.3Y_0.2O₃ + 1 wt.% CuO (BCZY) proton-conducting electrolytes in comparison with traditionally used 50 wt.% La_0.6Sr_0.4Fe_0.8Co_0.2O₃ + 50 wt.% SDC and 50 wt.% La_0.75Sr_0.2MnO₃ + 50 wt.% SSZ composite electrodes.

Keywords: La_1.7Ca_0.3NiO₄, cathode, electronic conductivity, proton-conducting electrolyte

Acknowledgments

The authors are grateful for the financial support to: the Ministry of Education and Science of the Russian Federation (Mega-grand contract no. 14.Z50.31.0001), the Russian Foundation for Basic Research (grants no. 13-03-00065, 14-03-00414) and the Russian Foundation for Basic Research (grants no. 13-03 96098 and 14-03-00414).

 References

[1] E.Yu. Pikalova, N.M. Bogdanovich, A.А. Kolchugin, D.A. Osinkin, D.I. Bronin, Electrical and electrochemical properties of La₂NiO_4+δ-based cathodes in contact with Ce_0.8Sm_0.2O_2-δelectrolyte, accepted in Procedia Engineering

[2] B.L. Kuzin, N.M. Bogdanovich, D.I. Bronin et al, Elecrochemical Properties of Cathodes Made of (La,Sr)(Fe,Co)O₃Containing Admixtures of Nanoparticles of Cupric Oxide and Intended for Fuel Cells with A Solid Electrolyte Based on Ceric Oxide, Rus. J. Electrochem. 43 (2007) 920.

In the last decade, thanks to their mixed ionic and electronic conductivity (i.e. MIEC properties), the lanthanide nickelates, Ln₂NiO_4+δ (Ln = La, Pr, Nd) have attracted much attention as cathode for solid oxide fuel cells. Simultaneously, Metal Supported Cells (MSCs) for IT-SOFC application (typically 600 – 700 °C) have been developed in respect of several advantages including reduced cost and better thermal cycling resistivity. Nevertheless, the use of porous metal implies to sinter the constitutive elements of the cells under low pO₂ (e.g. under N₂, pO₂ » 10^-4 atm). In that respect, the chemical stability, thermal and chemical expansion when exposed to temperatures as high as 1400 °C, under pO₂ as low as 10^-4 atm of Ln₂NiO_4+δ phases were studied through TGA (Thermal Gravimetry Analysis), XRD analyses, high temperature X-ray diffraction and Thermal Expansion Coefficient measurements. 

Electrochemical Impedance Spectroscopy measurements on Ln₂NiO_4+δ//GDC//YSZ symmetrical half-cells sintered either in air or under argon at various temperatures have been performed in the range 500 °C to 800 °C, in air, at i_dc = 0 A. Pr₂NiO_4+δ, either sintered in air or under nitrogen at 1150 °C and then re-oxidized, shows the lowest polarization resistance values (R_p = 0.12 Ω.cm² at 600 °C) while La₂NiO_4+δexhibits smaller polarization resistances when sintered under nitrogen, compared to air. Finally, the suitability of lanthanide nickelates in MSCs will be demonstrated and the relationship between cathode performances and structural and microstructural features will be discussed in details. Besides,modelling of the impedance data will be emphasized.

Solid oxide fuel cells (SOFCs) are the most efficient technology to directly convert chemical energy to electrical energy via electrochemical reactions. However, commercialization of SOFCs has been impeded due to the high cost associated with high temperature operation over 800 ^oC with conventional stabilized zirconia electrolytes. Thus, lowering the temperature of solid oxide fuel cells (SOFCs) is essential for the commercialization via availability to use cheap stainless steel interconnector, rapid start-up time, and higher mechanical and chemical stability, leading to significant reduction of system cost. To achieve this goal, at reduced temperatures the cathodic polarization mainly caused by oxygen reduction reaction at triple phase boundaries (TPBs) should be effectively reduced.

  At lower temperatures below 700 ^oC, however, the cathode polarization exponentially increases due to its thermally activated nature, thus dramatically decreasing the performance. Previously, stabilized bismuth oxide-based composite cathodes (ex: LSM-ESB) have been reported as promising cathodes for LT-SOFC applications with their low area specific resistance (ASR) at LT region. This high performance would be explained that LSM has low activation energy for dissociative adsorption of oxygen in ORR on its surface and that ESB has exceptionally high ionic conductivity as well as excellent surface exchange properties. Therefore, one can expect that the microstructural optimization of these cathode could further increase the cathode performance as well as enhance the their durability.

In this study, we employed the infiltration process to tailor the surface morphologies of stabilized bismuth oxide-based composite cathodes for enhancing surface activity and their stability to achieve higher SOFC performance. Through infiltration, for example, LSM was infiltrated on the porous ESB scaffold with different manners. The microstructural evolution and electrochemical performance was characterized and their cross-effect will be discussed.

The single phase double perovskite PrBaCo₂O_6-δ and composites (100-y) PrBaCo₂O_6-δ - y Ce_0.8Sm_0.2O_1.9 (y = 10 - 30 wt.%) were investigated as cathode materials for intermediate temperature solid oxide fuel cells (IT-SOFCs). The chemical compatibility, thermal expansion behavior, DC conductivity and electrochemical performance of these materials were studied. The results on chemical compatibility showed no reaction between PrBaCo₂O_6-δ and Ce_0.8Sm_0.2O_1.9 electrolyte even at temperature as high as 1200 °C. The average thermal expansion coefficient and overall conductivity of (100-y) PrBaCo₂O_6-δ - y Ce_0.8Sm_0.2O_1.9 (y = 10 - 30 wt.%) composite materials were shown to decrease with increasing content of Ce_0.8Sm_0.2O_1.9 in the composite whereas the polarization resistance was found to increase somewhat at the same time. Although the thermal expansion coefficient of the composite cathodes remains too high as compared to that of the electrolyte, the values of their total conductivity and electrochemical activity meet requirements for successful application of such cathode materials in IT-SOFCs.

The La_1-xBi_xMnO_3+δ (x=0.0-0.5) solid solutions were synthesized using three different methods: solid-state synthesis, citrate-nitrate combustion synthesis and mechanochemical activation. All obtained compositions crystallize in a rhombohedral perovskite structure, independent of the synthesis method. Two ranges with linear temperature dependence of the unit cell parameters have been observed during in-situ heating X-ray powder diffraction experiments. The transition between the two ranges occurs at 730°C and is explained by chemical expansion at high temperatures. The chemical compatibility between the obtained solid solutions and bismuth-containing electrolytes (Bi₄V_1.7Fe_0.3O₁₁, Bi₁₃Mo₅O₃₄ and Bi₃Nb_0.8W_0.2O_7.1) was studied. Interaction between these materials has been observed at temperatures higher than 600°С.

X-ray diffraction data collected at room temperature confirm that the La_1-xBi_xMnO_3+δ solid solutions form in the whole range of the investigated bismuth concentrations independent of the synthesis method. All samples crystallize in a rhombohedral perovskite structure with a trigonal unit cell (S.G. R-3c).

For a detailed microstructural characterization by transmission electron microscopy, the La_0.8Bi_0.2MnO_3.15 composition obtained using the solid-state synthesis was chosen. All obtained diffraction patterns can be indexed in a trigonal unit cell with a=5.5300(1) Å, c=13.382(1) Å. An analysis of systematically absent reflections revealed the presence of two types of crystals. Extinction conditions for the fist type coincide with those obtained from XRD (hkl: -h+k+l=3n; h-hl: h+l=3n, l=2n; hhl: l=3n; 00l: l=6n), indicating two possible space groups: R-3c and R3c. A second type of crystals obeys the same extinction rules except for the 00l reflections for which l=3n and h-hl reflections for which l¹2n, pointing towards the following possible space groups: R3, R-3, R32, R3m, R-3m. The formation of two different types of crystallites with trigonal crystal symmetry has been observed in perovskite systems before and explained by the presence or absence of antiphase octahedral tilting about the [111] axis of the cubic perovskite sub-lattice. It is worth to mention that the X-ray powder diffraction pattern of this sample displays asymmetric line broadening of peaks, which can point towards the presence of an additional structurally related compound.

A loss of oxygen was observed in the TG curves of La_1-xBi_xMnO_3+δ (x=0.0-0.5; Δх=0.1) samples at temperatures >700°С, supporting the previously discussed results. TG/DSC curves of La_0.9Bi_0.1MnO_3+δ are shown in Figure 4. The loss of mass was found to be 0.6%.

The excess of oxygen in La_1-xBi_xMnO_3+δ (x=0.0-0.5; Δх=0.1) was determined using redox titration. The value of the oxygen nonstoichiometry (δ) was found to be positive and equal to 0.15 at all concentrations of dopant, which can be explained by the identical oxidation states of La³⁺ and Bi³⁺. To determine the concentration of metal atoms in the obtained solid solutions, surfaces and fractures of pellets have been studied using EDS and AAS. The elemental content of 15–20 crystallites of La_1-xBi_xMnO_3+δ (x=0.1, 0.2) compositions was studied by EDS analysis inside the scanning electron microscope. It was found to be La_0.9(1)Bi_0.07(4)Mn_0.9(7) and La_0.9(8)Bi_0.1(6)Mn_0.9(6) for x = 0.1 and 0.2, respectively. The obtained results showed a slightly inhomogeneous cation distribution and a lower manganese concentration compared to the nominal value for both the bulk and the surface of the pellets. AAS measurements gave similar results for all synthesized solid solutions, showing the manganese content to be 0.9, and thus indicating that all obtained solid solutions lie in the homogeneity region of lanthanum manganite with a small manganese deficiency. Taking into account the obtained values of manganese deficiency and oxygen nonstoichiometry, the chemical formula of the solid solutions can be rewritten as La_1-xBi_xMn_1-yO_{[(3+(1-y))/2]} (where V_Mn is the average oxidation state of manganese). For example, for the La_0.9Bi_0.1MnO_3+δ compound with δ=0.15 and y=0.1, the average oxidation state of manganese (V_Mn) will be equal to 3.67. Thus, the ratio of manganese in different oxidation states (Mn⁴⁺/Mn³⁺) in the studied solid solutions is equal to 2, which means that 67% of the manganese cations in these compounds have oxidation state +4. This fact is probably the reason why we observe the formation of solid solutions with a rhombohedrally distorted perovskite structure already at room temperature.

A study of the chemical compatibility between La_1-xBi_xMnO_3+δ (x=0.0-0.5; Δх=0.1) solid solutions and bismuth-containing electrolytes shows that depending on the chemical nature of the electrolyte material, the formation of additional phases begins at different temperatures: at T³700°С for the studied niobate, at T³600°С for the studied vanadate and at T³500°С for the studied molybdate.

The work was financially supported by Russian Fund of Fundamental Research, grant 14-03-92605.

Introduction

Single cell performance of solid oxide fuel cell (SOFC) is largely limited by oxygen reduction reaction (ORR) at cathode, 1/2O₂ + V_O¨ + 2e' → O_O^×.  Therefore, understanding the mechanism of ORR is fundamental issue in improving SOFC performance. The above reaction equation tells us that higher chemical potential of electron is advantageous to the ORR kinetics, which implies that catalytic activity is significantly influenced by electronic band structure of cathode material. The authors also point out that the electronic structure is related to the basicity (a(O^2-)) determined by the reaction, 1/2O₂ + 2e' → O^2-, which also has an impact on the cathode activity. Though the discussion based on electronic structure has been initiated, the comprehensive understanding on the relationship between ORR activity, electronic structure and basicity has not been achieved.

To better understand the ORR mechanism, PrO_x-based cathode is employed as a model system, which is a simple material and sensitive to doping and surface modification.  In order to examine the influence of density of surface-adsorbed oxygen, the authors also investigate the PrOx-based cathode surface-modified with BaPrO_x (BPO), which includes a strong basic element, Ba. The cathode reaction properties of those cathodes are evaluated and the ORR mechanism will be discussed.

Experimental

PrO_x thin film electrodes have been prepared by an RF-sputtering technique in an oxygen-including atmosphere on both sides of the substrates (electrolyte) of single-crystal yttria-stabilized zirconia (YSZ). For surface modification with BaPrO_x (BPO), a thin layer is deposited on the PrO_x thin film by RF-sputtering under the same condition. A reference electrode is fabricated by painting Pt paste and firing at 1273K before the above thin film fabrication. The performance of these thin film electrodes has been evaluated in terms of interfacial conductivity by three-terminal AC impedance measurement under controlled temperature and oxygen partial pressure (Po₂). The samples are analyzed by the X-ray diffractometry (XRD), Raman spectroscopy and X-ray photoemission spectroscopy (XPS).

Results and Discussion

The Nyquist plot typically shows, in addition to a contribution of the electrolyte resistance at highest frequency, a small semi-circle at intermediate frequency and a large one at the lowest frequency.  The latter can be ascribed to the surface reaction, and the resistance is normalized as area-specific conductivity (σ_LF) to serve for a measure of cathode reaction activity.In the Po₂ dependence of σ_LF at 700˚C, all of the examined electrodes, with two different thickness (90 and 270nm) and a BPO modification, show a 1/2 power dependence.  In addition, this behavior is perfectly the same as that of (La,Sr)CoO₃-based cathodes [1].  Although the bulk properties such as defect structure and electrical transport are totally different among those oxides, the universally observed Po₂ dependence suggests that the rate-determining step is the following reaction, O_2(ad)^2- + V_O^●●→O_O^x + O_(ad)^- [2].

Figure 1 shows the temperature dependence of　σ_LF at Po₂=1 atm.  It is noted that the activation energy indicated is averaged value for the temperature above 550˚C.  The PrO_x electrodes with 90nm- and 270nm-thickness show different magnitude and activation energy in σ_LF, suggesting that especially at lower temperatures, these may have different rate determining steps. For the 270nm-thick and 270nm+BPO electrodes, a discontinuous drop of σ_LF are observed between 500-550˚C, which can be related to the phase transition of PrO_x from β-phase (Pr₆O₁₁) to α-phase (PrO_x).  The 270nm+BPO electrode shows higher σ_LF than the single-layer PrO_x electrodes. This implies that the ORR activity of the BPO-modified electrode is improved by enhancement of surface basicity and resultant increase in the density of surface-adsorbed peroxide ion.

[1] A. Takeshita et al., ECS Trans. 57 (2014) 1733.

[2] A. Takeshita et al., to be submitted.

Figure 1

It has been clarified that chromium deposition occurs significantly onthe electrolyte surface near the cathode reaction sites in consequence of cathode polarization. The deposited chromium probably remains on the surface, not diffusing into the electrolyte. Therefore, focusing on the behavior of the deposited chromium, such as crystal growth or dissipation by cathode polarization change is suitable for clarification of the deposition mechanism in detail.

   In this study, we investigated influence of cathode polarization change on the deposited chromium on the electrolyte surface by using NiO/YSZ or NiO/GDC for the cathode material. We used NiO for the cathode material because of the following two reasons. 1) Effect of cathode polarization can easily be studied because the predetermined cathode polarization is generated at the cathode reaction sites with smaller current flow. 2) Lower reactivity of NiO with chromium vapor than conventional cathode materials (e.g. LSCF, LSM) can make the analysis simpler.

   Although ohmic resistances of these cathodes using NiO are higher than usual, they showed enough performance to control cathode polarization. Chromium poisoning tests were conducted at 700 ^oC for 100 hours keeping cathode polarization constant at 200 mV. At the end of the chromium poisoning tests, several different procedures to stop the cell tests were executed concerning the timing to remove the cathode polarization

   After the chromium poisoning tests at 200 mV, chromium deposition was observed on the surface of the electrolyte near the cathode/electrolyte interface as observed in LSM cathode. The deposited chromium segregated at the interface of NiO and YSZ for the NiO/YSZ cathode by decreasing cathode polarization, which suggests nucleation under cathode polarization at 200 mV and growth of chromium compounds after decreasing the cathode polarization. On the other hand, the deposited chromium decreased for the NiO/GDC cathode after decreasing cathode polarization.

(La,Sr)(Co,Fe)O_3-δ (LSCF) perovskite oxides have been considered to be a promising cathode material for SOFCs operating in the intermediate temperature range due to their significantly higher electrochemical activity for the O2 reduction reaction and higher oxygen ion conductivity than the conventional (La,Sr)MnO₃ (LSM) cathode [1,2]. In practice, metallic materials have become a preferential choice for the interconnect due to their excellent physical and chemical properties. However the presence of chromium in all commonly used metallic alloys has been found to cause poisoning of the cathode under operating conditions leading to rapid electrochemical performance degradation of the cathodes including LSCF [3–5].

Development of a Cr tolerant cathode material for increased long-term durability of SOFCs relies on a fundamental understanding of the mechanism of the chromium deposition and poisoning of the cathode materials. Extensive research on the chromium deposition and poisoning processes on SOFC cathodes has been carried out but careful microstructural studies especially on the nanometer to atomic scale are very limited. This is however a subject that can give valuable information on the detailed process of Cr incorporation and evolution in the cathode materials, providing an important insight for the mechanistic understanding of the Cr poisoning.  

LSCF/CGO/LSCF symmetrical cells were prepared by screen printing the LSCF cathode on Gd doped cerium oxide (CGO) electrolyte pallet, and the effect of Cr poisoning at 900 ˚C in air on the electrochemical behavior of the cell was assessed by impedance spectroscopy. The change in nano/microstructure and chemistry between cells with and without Cr poisoning were studied in parallel by TEM/STEM/EDX/EELS. To facilitate the TEM sample preparation using focus ion beam thinning (FIB), bulk samples containing the porous LSCF layer were firstly embedded in an epoxy resin.

Due to the peak overlap between the interested La and Cr and inherently poorer spatial resolution in EDX, EELS was employed with the aim for an improved accuracy in the analysis.  Figure 1a shows an example of EDX artefacts where the spectrum from an epoxy area in between LSCF grains exhibit the presence of the LSCF elements. On the other hand, EELS could suffer a higher detection limit compared to EDX due the generally poorer signal to background (S/B) ratio, which is typically worse when the sample thickness increases. Figure 1b shows an example of EELS spectrum, which fails to detect the presence of Co in the thicker area of a LSCF grain (t/λ ~ 0.86). It was found that all the expected elements (apart from Sr) in LSCF can be successfully detected by EELS when t/λ < ~ 0.7, and this criteria was kept during all the later EELS measurements.

Our results show that Cr is incorporated in LSCF in more complicated ways than a simple formation of SrCrO_x as suggested by previous studies. Cr was found to segregate in Sr rich phases that also contain other elements such as Fe and Co (Figure 2). In addition, Cr was incorporated in the LSCF perovskite structure and repelled sometimes the other B site elements from the lattice. As a result, Cr rich phases take the form of various compositions such as (La,Sr)(Fe,Co,Cr)O_x, (La,Sr)(Fe,Cr)O_x and (La,Sr)CrO_x (Figure 3). It is interesting to note that Cr appears to repel more readily Co than Fe. The Cr poisoning also promotes the formation of Co and Co-Fe rich phases that may contribute to the deficiency of Co in the LSCF grains (Figure 4). It can be seen that the area adjacent to the Fe-Co rich particles contains Cr and are deficient in Co with a composition close to that of (La,Sr)(Cr,Fe)O_x. Careful examination of Cr incorporation was also extended to LSCF grain boundaries with both structural and chemical analysis. The high-resolution microscopy results and their implications are discussed in relation to the degraded electrochemical properties of the LSCF cathode and the Cr poisoning mechanisms.

[1]      S. Jiang, Solid State Ionics 146 (2002) 1.

[2]      A. Esquirol, N.P. Brandon, J.A. Kilner, M. Mogensen, J. Electrochem. Soc. 151 (2004) A1847.

[3]      M.C. Tucker, H. Kurokawa, C.P. Jacobson, L.C. De Jonghe, S.J. Visco, J. Power Sources 160 (2006) 130.

[4]      L. Zhao, J. Drennan, C. Kong, S. Amarasinghe, S.P. Jiang, ECS Trans. 57 (2013) 599.

[5]      S.P. Jiang, X. Chen, Int. J. Hydrogen Energy 39 (2014) 505.

Figure 1

One of the most critical issues causing degradation of SOFCs is poisoning of cathodes by a trace amount of impurities transferred with air such as sulfur (S) and chromium (Cr), and therefore, we are continuously investigating the test method on accelerated degradation by these impurities.  In our previous studies on Cr poisoning, chromium metal covered with Cr₂O₃ was directly placed on the current collector of the cathode as a source of Cr vapor.  However, the setting accelerated the performance degradation in surplus.  In order to perform accerelated poisoning test at an appropriate rate, we improved the cell operating equipment to control the Cr vapor pressure during the cell tests.  The Cr vapor pressure was controlled by adjusting the place and temperature of the Cr source. The cell tests were operated at 1073 K under controlled feeding rate of Cr vapor after the cell performance became stable. The degradation behavior was analyzed with AC impedance analysis, field emission scanning electron microscopy (FE-SEM) with energy dispersive X-ray analysis (EDX) / wavelength-dispersive X-ray spectrometer (WDS).  The vapor pressure of Cr species was estimated from the amount of Cr trapped by (Sm,Sr)CoO₃cathode in the same equipment of cell tests.  The degradation rate decreased with decreasing the Cr vapor pressure, and the amount of Cr species deposited in LSCF cathode decreased with decreasing the Cr vapor pressure.

Rare earth elements like La and Ce are known to improve the corrosion resistance of coated ferritic steels [1, 2]. The solid state reaction between La-rich oxide and Cr₂O₃ has been studied by x-ray diffraction (XRD) and in situ heating TEM with electron diffraction. The La-rich oxide powder was found to be porous and polycrystalline and would start by healing the pores and forming a transition phase of orthorhombic CrLaO₃ [3] before migration of La is found at the interfaces (Figure 1). After full reaction pure La₂O₃ has been observed by HRTEM in a region initially starting as porous La-rich oxide (Figure 2).

1.            S. Canovic, et al., Oxidation of Co- and Ce-nanocoated FeCr steels: A microstructural investigation. Surface & Coatings Technology, 2012.

2.            M.W. Lundberg, et al., Precoated AISI441 for SOFC interconnectors studied by TEM methods. ECS Transactions, 2013. 57(1): p. 2321-2329.

3.            C.P. Khattak and D.E. Cox, Structural Studies of the (La,Sr)CrO₃ System. Materials Research Bulletine, 1977. 12: p. 463-472.

Figure 1

SO₂ is one of the most major impurities in air and it induces a performance degradation of SOFC cathodes.  Our group has investigated the enforced degradation behaviors of LSCF cathodes by SO₂ containing air.  It has been clearly observed that SO₂ induced degradation is affected by the SO₂ concentration in air, the operating temperature, the Sr content in LSCF cathode, the flowrate of air, the humidity of air as well as cell operation.  We also find out that sulfur is a tendency to preferentially exist in the vicinity of the cathode/electrolyte interface.  In this presentation, we try to provide a comprehensive description of SO₂ induced degradation of LSCF cathode from a thermodynamical perspective.  In progress of degradation, sulfur reacts with Sr component in LSCF and forms SrSO₄.  As a result, both conductivity and oxygen reduction reaction activity of LSCF decrease.  There seems to be a relationship between increments of ohmic resistance and reaction resistance in each degradation step

We investigated property of a FeCrAl type stainless steel as a porous alloy substrate of metal supported SOFCs especially on the cathode side. We confirmed not only good heat resistance but also low electrical resistance at the interface between the porous substrate and La_0.6Sr_0.4Co_0.2Fe_0.8O₃ (LSCF) coating at 700^oC in air. Morphology and crystal structure of the surface oxide layer of the FeCrAl alloy was analyzed by STEM-EDS and TEM in detail to clarify the cause of such a low electrical resistance. Long-term stability of the oxidation resistance of the porous FeCrAl alloy substrate was investigated by increasing operating temperature up to 900^oC. Oxidation rate of the alloy at 700^oC in air was estimated by the increase in weight, thickness of the surface oxide and electrical resistance.

Mn_1.5Co_1.5O₄ (MCO) protective coating was deposited by Electrostatic Spray Deposition on SS446 alloy for application as interconnects in Solid Oxide Fuel Cells. The main purpose of this work is to verify the feasibility of a thin coating layer prepared by this technique to inhibit the Cr volatilization and the subsequent cathode poisoning. Phase crystallization was obtained after deposition by thermal annealing at 800 °C in air. The corrosion resistance behavior of uncoated and coated steels was investigated by electric and thermogravimetric measurements. The coating films, 400 nm thick, are found efficient for reducing the oxidation rate by limiting the outward Cr³⁺ diffusion. The area specific resistance of the coated steel is found to be considerably low (6.7 mW.cm²) and stable compared to uncoated SS446 alloy (80 mW.cm²) after oxidation for 200 h at 800 °C in ambient air. These results evidence the excellent performance of the deposition technique for fabricating thin, crack free and dense conductive Mn_1.5Co_1.5O₄-coated ferritic alloys.

Sandvik Materials Technology develops nanocoatings for SOFC stainless steel interconnects. In this study a multilayered PVD-coating has been investigated. Cyclic oxidation for more than 20,000 hours at 800°C, chromium volatilization and area specific resistance studies has been conducted on three different coatings and compared with the uncoated steel, in this case AISI441 (EN 1.4509). In this work we have shown that pre-coated AISI 441 show very promising results for SOFC interconnectors at 800°C. We have studied and compared state of the art CeCo coating with a new four layered coating that shows slightly higher cumulative chromium volatilization, similar ASR values and reduced mass gain in the cyclic oxidation experiments.

Figure 1

One of disadvantages of SOFCs is their mechanical strength. Ceramic supported SOFCs are weak and they can be failed by mechanical load. It can be supplemented by applying metal support. The sealing issue is another disadvantage of ceramic support SOFCs, but it can be solved easily using metal support by welding its boundary. Therefore, it minimizes sealants usage in SOFC stack fabrication, and this allows SOFCs to be applied to various applications including transportation facilities such as vessels and automobiles.

The joining process has been used to fabricate metal supported SOFCs. In the conventional process a metal supported SOFC is fabricated under reduction atmosphere to avoid the oxidation of metal support. When the metal support is exposed to SOFC operating temperatures, 600-800°C, the metal substrate is oxidized, and then Cr oxide layer is formed on its surface. Although Cr oxide layer can protect substrate's surface from high temperature, it decreases SOFC stack performance due to low electrical conductivity. Moreover, Cr oxide layer reacts with oxygen at high temperature and forms Cr gas species which can lead cathode poisoning. As a result, SOFC stack performance is decreased significantly. Therefore, the sintering process should be performed at high temperature under reduced atmosphere to prevent Cr poisoning. However, lots of problems can be caused by the conventional joining process as the following: 

• Thick thickness of a single cell

• Delamination between anode and metal substrate due to redox process

• Difficulty in sealing between anode and metal substrate

The performance of metal supported SOFCs is worse than that of ceramic supported SOFCs. The joining process has a critical limitation fundamentally caused by oxidation problem. Therefore, an alternative fabrication process is required in order to overcome facing challenges.

The structure of metal supported cell (Fig. 1(a)) using the conventional joining process has some limitations due to the oxidation problem. Many researchers have manufactured metal supported cells using the conventional method, and none of them can avoid serious degradation of stack performance because of problems mentioned earlier such as low activity of cathode, delamination of anode and sealing issue.

This work suggests an alternative method to fabricate metal supported SOFCs which is the interconnect-coating-based cells. This technology enables sintering at high temperature under oxidation atmosphere. Fig. 1(b) shows the interconnect-coating-based metal supported SOFC. It is manufactured by a series of coating processes. Firstly, the interconnect-coating material is coated on both sides of a metal substrate, and then cathode, electrolyte and anode are sequentially coated on one side of the coated substrate. Although the structure of interconnect-coating-based metal supported SOFC is multilayer, it can be well manufactured by some sintering steps. Recently, an interconnect-coating-material which can be stabilized at around 1000°C was developed. It was validated that it can prevent Cr evaporation and Cr poisoning of cathode.

In order to accomplish successful development of this technology, forming dense or porous films is indispensable. Generally, high temperature sintering process has to be performed for densification of SOFC electrolyte. Therefore, the research on developing a sintering process which can make electrolyte dense at relatively low temperature such as 1000°C should be performed at the beginning. However, this work is too difficult for accomplishment, because the substrate's surface for coating electrolyte has high porosity and is not smooth. Furthermore, conventional methods to dense layer at low temperature are very expensive and have to take coating for long period.  Therefore, it needs new approach to solve the problem mentioned above. Firstly, solution which has high viscosity is coated on substrate surface to block the porous surface. Then, another solution which has low viscosity is coated within electrolyte layer to densify. Fig. 2 shows the results of new method which can make dense layer at relatively low temperature.

As the next step, a metal supported SOFC with a newly designed structure is manufactured by interconnect-coating-based method using a new coating technology for electrolyte. The performance and characteristics of the metal supported SOFC were measured by solartron 1260 which is an electrochemical measurement device.

Figure 1

Ferritic stainless steel with high chromium content (>16 wt%) are regarded as promising interconnects materials for solid oxide fuel cell applications. However, particularly in oxidizing atmospheres, the formation of chromia scales leads to high contact resistances harmful to SOFC performance. At the same time, up to now one of main challenges in SOFC technology developments is to suppress the degradation processes, often associated with the stainless-steel interconnect materials, and to provide low contact resistivity in oxidizing atmospheres. In previous works [1,2] we proposed a new approach for forming barrier layers impeding Cr diffusion to the metallic current-collector surface. The deposition of Ni-based layers with subsequent thermal treatments was shown to result in unique interfacial microstructures, where the formation of an internal oxidized Cr layer prevents the segregation of insulating Cr₂O₃ phase on the stainless steel surface and reduces electrical resistivity [2]. The present work, focused on the analysis of near-surface interdiffusion phenomena in Crofer 22 APU ferritic steel interconnects with Ni-based protective layers, summarizes our developments in this field. Particular emphasis is centered on the studies of time dependencies of the area-specific resistance (ASR) between the current collector and standard La_0.8Sr_0.2MnO₃(LSM) cathodes, depending on the protective inter-layer composition. A theoretical model describing the contact resistivity behavior was proposed and validated.

The deposition of protective layers onto the Crofer 22 APU interconnect plates was described elsewhere [2]. The microstructural and compositional analysis of the junctions "current collector – LSM cathode" was performed using scanning electron microscopy and energy dispersive spectroscopy (SEM/EDS, Supra 50/VP instrument). The studies of element distribution profiles across the interfaces were carried out after prolonged isothermal treatments at 850°C in air, coupled with electrical measurements as a function of time. In order to detect trace separation of new phases such as Cr₂O₃, micro Raman scattering spectroscopy was also employed. The area-specific resistance of the "current collector – LSM cathode" junctions was studied over long periods of time under the SOFC cathode operation conditions (atmospheric oxygen pressure, 850ºC, current density of 0.5 A/ cm²).

The results showed that the electrical resistance variations of the assemblies "Crofer 22 APU | LSM" with and without surface modification of the metallic plates can be quantitatively described in framework of the Schottky barrier model [3,4] for metal– semiconductor interfaces. The microscopic mechanism governing these changes involves metal interdiffusion between the cell components leading, in particular, to the formation of essentially immobile Cr₂O₃grains at the boundary between Crofer 22 APU and deposited Ni-based layer. These interfacial alterations make it possible to preserve low contact resistances during, at least, 20000 h. The junction between the current collector and SOFC cathode should be considered as forward-biased. In the case of relatively thin blocking layer, the current-voltage relationship for such a junction is

J=AT²(Ve/kT)exp( -F/kT),

where the Richardson constant A=4πemk²/h³ comprises the Planck (h) and Boltzmann (k) constants as well as electron charge (e) and mass (m); V is the voltage drop across the forward-biased junction; F = Φ_metal – Χ_LSMis the difference between the metal work function and electron affinity of the LSM cathodes. Time dependencies of the contact ASR between Crofer 22 APU and LSM can be adequately described by the Schottky barrier changes, originating from changing the current-collector work function due to metal interdiffusion between the Ni-based coatings and stainless steel.

This work was supported by the Ministry of Education and Science of the Russian Federation (project 14.610.21.0007)

References

[1]. N. Ledukhovskaya, E. Frolova, G. Strukov, D. Matveev, S. Bredikhin "New Type of Current Collectors with Modified Near-Surface Layer" ECS Transactions, 25 (2) 1523-1528 (2009).

[2]. N. Demeneva, S.Bredikhin "Improvement of Oxidation Resistance of Crofer 22 APU with Modified Surface for Solid Oxide Fuel Cell Interconnects" ECS Transactions, 57 (1) 2195-2201 (2013)

[3]. S. Bredikhin, T. Hattori, M. Ishigame "Schottky barriers and their properties in superionic crystals" Rhys. Rev. B, 50, p. 2444 – 2449 (1994)

[4]. S.I. Bredikhin, V.N. Bondarev, A.V. Boris, P.V. Pikhisa, W. Weppner "Electronic conductivity and current instability in superionic crystals" Solid State Ionics 81. 19 -28 (1995).

Abstract

A Fe-22Cr mesh/LaNi_0.6Fe_0.4O_3-δ composite, obtained via dip coating technique, was machined using a femtosecond laser as a possible gas permeable contact layer for solid oxide fuel cells (SOFCs). Both electrical performance and processing reproducibility of {Crofer22APU channelled interconnect/contact layer/La_0.6Sr_0.4FeO₃ cathode} structure were studied through ASR testing on four replicas. In addition, the long-term behaviour of this system at 800 ºC was determined using energy dispersive X-ray spectroscopy (EDX). The metallic and ceramic parts were quantitatively analysed using EDX, focusing on different positions of the sample. A chromium content reduction in both the interconnect and mesh was observed, accompanied by a Cr enrichment in LaNi_0.6Fe_0.4O_3-d (LNF). Considering the obtained results, the developed setup showed an adequate reproducibility and only a short-range damage caused by the laser irradiation.

Keywords: SOFC; contact layer, metallic interconnect, contact resistance; chromium poisoning; femtosecond laser.

Acknowledgements

This research has been funded by Ministerio de Economía y Competitividad (MAT2013-42092-R) and the Dpto. Educación, Política Lingüística y Cultura of the Basque Goverment (Research Group of the Basque University System IT-630-13). The authors wish to thank SGIker-UPV/EHU technical and human support. Dr. Sergio Fernandez Armas is acknowledged for useful scientific help in discussion of EDX measurements. The authors would also like to express their gratitude to Dr. Raúl Montero Santos for helping with femtosecond laser measurements. A. Morán-Ruiz thanks UPV/EHU for funding her PhD work.

Reduction of operational temperatures from 800°C towards an intermediate temperature (IT) range of 600-750°C is one major trend in current Solid Oxide Fuel Cell (SOFC) research as a cost-effective approach to higher reliability and durability of fuel cell components and systems [1]. Intermediate temperatures expand, in fact, enormously materials selection in SOFC design allowing the possibility of using conventional ferritic stainless steels for the realization of important components such as cell housings, gas manifolds, support structures and plate interconnects (IC). In particular, 18Cr ferritic stainless steels are currently the most evaluated materials to realize cost-effective metallic IC plates due to their excellent mechanical stability, ease of fabrication and high formability characteristics. However, excessive contact resistance due to oxide scale growth and evaporation of the protective chromia scales are two primary sources of cell performance degradation that are related to an insufficient long-term corrosion resistance of 18Cr ferritic stainless steels. In order to alleviate these problems, protective ceramic oxide coatings with spinel or perovskite layers are usually applied to provide better electrical contact and reduced Cr volatility [2]. For further cost reduction, 13Cr stainless steels have received some attention in the recent literature as a potential substitution alternative at the low end of IT-SOFC operation temperatures [3,4]. At temperatures below 700°C, the oxidation of these steels gives raise to protective oxide films mainly composed of mixed (Fe,Mn,Cr) spinel oxide phases [4]. As compared to chromia-forming alloys, stainless steels that form protective (Fe,Mn,Cr) spinel oxides have an expected advantage of lower Cr volatility and contact resistance issues with only a small trade-off in corrosion resistance properties, under appropriate temperature conditions. Recently, we have proposed a novel approach for high temperature corrosion protection of stainless steels, which is based on producing LaFeO₃-based perovskite layers by chemical conversion reactions taking place in a molten carbonate salt bath. Our previous results indicated that LaFeO₃ perovskite coatings were able to protect stainless steels with various Cr content in typical Molten Carbonate Fuel Cell (MCFC) environments [5,6]. In order to extend the application areas of this novel coating method, the aim of this work was to evaluate the suitability of LaFeO₃ perovskite coatings also for the improvement of corrosion resistance of a 13Cr ferritic stainless steel under typical IT-SOFC conditions. Techniques such as XRD and SEM/EDX analysis were used to characterize morphology and structure stability of the perovskite coating after prolonged exposure at 700°C in ambient air atmosphere. Detailed results from these studies will be presented at the conference venue.

Bibliography

[1] M. C. Tucker, Progress in metal-supported solid oxide fuel cells: A review, . J Power Sources 195, 2010, 4570-4582;

[2] N. Shaigan, W. Qu, D.G. Ivey, W. Chen, A review of  recent progress in coatings, surface modifications and alloy developments for solid oxide fuel cell ferritic stainless steel interconnects. J Power Sources 195, 2010, 1529-1542.

[3] J. Fergus, Y. Zhao,  Low-Chromium Alloys for Solid Oxide Fuel Cell Interconnects. ECS Trans 25, 2011, 2447-2453;

[4] S. Frangini, A. Masci, S.J. McPhail, T. Soccio, F. Zaza, Degradation behavior of a commercial 13Cr ferritic stainless steel (SS405) exposed to an ambient air atmosphere for IT-SOFC interconnect applications. 144, Mater Chem Phys 2014, 491-497;

[5] S. Frangini, A. Masci, F. Zaza, Molten salt synthesis of perovskite conversion coatings: A novel approach for corrosion protection of stainless steels in molten carbonate fuel cells. Corros Sci, 53, 2011, 2539-2548;

[6] S. Frangini, F. Zaza, A. Masci, Molten carbonate corrosion of a 13-Cr ferritic stainless steel protected by a perovskite conversion treatment: Relationship with the coating microstructure and formation mechanism. Corros Sci, 62,  2012, 136-146.

The monoclinic Bi-Sr-Co-O compound was successfully synthesized by ethylene glycol-citrate acid polymerization method. The prepared sample was evaluated by various experimental tests. Powder X-ray diffraction (XRD) data shows the Bi-Sr-Co-O powder with good crystallinity. Its good thermal stability and chemical compatibility with Ce_0.9Gd_0.1O_2−γ electrolyte was also proved by XRD data. The chemical composition of Bi-Sr-Co-O powder was confirmed by ICP-AES and XPS analysis. The compatibility with the electrolyte, along with the encouraging electrical conductivity (around 10 S·cm⁻¹ in the temperature range of 600-700 °C), allows us to conduct further study on Bi-Sr-Co-O as a potential novel cathode material for intermediate temperature-solid oxide fuel cells.

The metal interconnect oxidation and interaction with the cell are issues limiting the performances and the durability of the SOFC stacks. Despite the application of additional surface treatments or protective coatings such problematic are still affecting the system reliability because of hardly avoidable interactions since the operating conditions to which such material are exposed. Beside the high temperatures and oxygen partial pressures one should consider also the presence of a current load which acts locally changing the working conditions and the kinetic aspect of the  oxidation processes. In this work the current density effect has been investigated on Crofer 22 APU in air at 750°C. The same material has been pre-oxidized and coated with a state-of-the-art Co_1.5Mn_1.5O₄ spinel coating to test the effect of such treatments on the interaction with a La_0.6Sr_0.4Co_0.2Fe_0.8O₃ contacting layer simulating the interconnect interface at the cathode compartment. The behavior of such samples has been compared with the results on a working reproduction of a stack cathode compartment.

In this study, plasma sprayed (Mn,Co)₃O₄ (MCO) coatings with spinel structure are used to form protective layer for metallic interconnects of solid oxide fuel cells.

The temperature during thermal spraying is high enough to change the crystal structure and conductivity of protective oxides. The structure change is caused by the removal of oxygen ions at higher temperature. This behavior is so-called thermal reduction. Furthermore, the transformation of MCO Spinel to a NaCl-type structure causes the degradation in its conductivity by several orders of magnitude. With better understanding and control of thermal reduction and/or phase transformation, it is believed that the proper heat treatment is an adequate method to obtain desired properties for protective oxide.

Thus, the objective of this study is (1) to investigate the influence of thermal reduction on the conductivity and crystal structure; (2) to improve the performance of interconnect through proper annealing at desired temperatures. XRD, SEM, and impedance spectroscopy are used to characterize the protective oxides before and after annealing and thermal reduction.

Tubular type of solid oxide fuel cells (SOFCs) is durable under high temperature operation compared to planar type SOFCs where the use of metallic interconnect is inevitable, which deteriorate electrochemical performance by Cr evaporation. On the contrary, development of ceramic interconnection has been a hurdle to fabricate tubular type SOFCs. The ceramic interconnection material must have high electronic conductivity both in oxidizing and reducing atmosphere. A gas-tight microstructure should also be obtained mostly by co-firing with anode or cathode substrates. We have been developing flat-tubular SOFCs from extruded NiO-YSZ tubular substrates. In this study we investigated (La,Sr)FeO₃ (LSF)-based composite materials as the interconnection for tubular SOFCs. La_0.8Sr_0.2FeO_3-δ, a mixed ionic and electronic conductor, was mixed with Gd- or La-doped ceria (GDC or LDC). LSF is a mixed ionic and electronic (hole) conducting material and thus shows low electrical conductivity in reducing (H2) atmosphere. However, composite materials with LSF and ceria showed moderate electrical conductivity by electronic leakage current through doped ceria in reducing atmosphere. LSF-ceria composite and LSF layers were also successfully deposited on NiO-YSZ anode substrate by conventional co-firing process. The area specific resistance of LSF/LSF-GDC bilayer on Ni-YSZ measured in air//H₂ condition was initially ~50 mΩ cm².

Ni-Y₂O₃ stabilized ZrO₂ (Ni-YSZ) cermet is a widely used hydrogen electrode for solid oxide fuel cells (SOECs), but detailed studies regarding the water splitting processes are scarce. Here we report the preliminary results on the electrocatalytic activity of Ni-YSZ hydrogen electrode in the SOEC mode as a function of operating temperature and water concentration. The electrode activity for the water splitting reaction is strongly influenced by the water concentration and dc bias. The electrochemical performance for the water splitting reaction increases with the increase of water concentration. On the other hand, the electrode polarization resistance increases gradually with the increase of dc bias regardless of the H₂O content, which is particularly pronounced in the low frequency impedance process related to the H₂O dissociative adsorption and diffusion processes. The reaction mechanism and kinetics of the H₂O dissociation reaction on Ni-YSZ hydrogen electrodes are discussed.

Solid oxide electrolysis cell (SOEC) is regarded as one of the most promising hydrogen generation devices. However, electrolysis operated at high temperature has some problems; component inter-diffusion and difficulty in the selection of sealant etc.. SOEC (IT-SOEC) at intermediate temperature (550~650^oC) may reduce problems, however, most of the overpotential occurs in air electrode. In this study, we have tested La₂NiO_4+δ (LNO) as an air electrode and compared the performance with that of La_0.6Sr_0.4Co_0.2Fe_0.8O_3-δ (LSCF) which shows degradation due to Sr segregation. LNO has advantages due to its good thermal-expansion coefficient (TEC) match with electrolyte and little cation segregation. Electrochemical performance of SOEC at 650^oC under electrolysis current shows that LNO is a promising as an air electrode.

The present paper investigates the variation in performances in co-electrolysis of steam/CO₂ on SOEC single cells caused by the progressive increase of CO₂ inlet concentration and current density.

Steam electrolysis and co-electrolysis were investigated on circular Ni-YSZ supported cells produced by SOLIDpower ("ASC-700" type cells). The thin film electrolyte is constituted by YSZ and the oxygen electrode by LSCF-GDC. The active area corresponds to 12.56 cm² (2 cm of radius).

The cells were fed with different H₂O/CO₂ mixtures, while keeping 10% H₂ of the inlet molar flow in order to avoid re-oxidation of the Ni-YSZ support.

A sensitivity analysis was performed looking in particular at the effects of inlet gas composition, gas flows and temperature on the performances. I-V characterization and EIS measurements at different current densities were carried out for each sequence.

In steam electrolysis mode, the reference inlet molar flow was equal to 150 Nml/min (90% H₂0, 10% H₂), at 750°C.  The I-V curve showed a voltage of 0.861V at OCV, 1.049V at 0.5 A/cm² and 1.253V at 1 A/cm². A sensitivity analysis was performed varying the temperature (700°C, 750°C and 800°C), the inlet air flow (100 Nml/min, 300 Nml/min and 500 Nml/min) and the total inlet fuel flow (from 8 to 16 Nml/min/cm²). In this case, the steam conversion factor was kept constant equal to 50% (adjusting the current imposed).

EIS analysis exhibits that ohmic and polarization resistances decrease with current densities lower than 0.4 A/cm² and then rise with higher values. As expected, the increase of temperature has a positive impact on the performances, while the change of air and fuel flow slightly affects them.

In co-electrolysis mode the CO₂ and H₂O content was varied, keeping constant the total molar flow and the hydrogen fraction as explained before. As shown in the enclosed image, the area specific resistance (ASR) increases with the CO₂ concentration in the mixture and gas transport limitation occurs at lower current densities.

Also in this case a sensitivity analysis was conducted changing the temperature (from 700°C to 800°C) and the resulting EIS plots demonstrate different ongoing processes favored at high current densities. In particular, the reverse water-gas-shift reaction plays an important role on the overall balance of the gas species, and it directly depends on temperature and inlet fuel composition.

Figure 1

Reversible solid oxide fuel cells (RSOFCs) have been considered and developed for power generation from a variety of fuels and hydrogen/syngas production from steam/mixtures of steam and carbon dioxide. When the RSOFC is operated in electrolysis mode (referred to as solid oxide electrolysis cell or SOEC mode), it is critical that electrode performance is reversible and system design is configured for efficient operation in both SOFC and SOEC modes. This paper summarizes recent results on electrode reversibility in SOEC mode for planar hydrogen electrode-supported RSOFCs and discusses a preliminary system design for electrolysis operation.

Four solid oxide electrolysis cells (SOEC) with different cathode microstructure and chemical composition (Ni-YSZ and Ni-GDC) and electrolyte thickness (ScSZ) have been prepared using tape casting and screen printing methods. Electrochemical performance of the single cells at various working temperatures and gas mixtures in the cathode compartment has been analyzed using cyclic voltammetry, impedance spectroscopy and other electrochemical methods. It was established that the ratio of the rate-determining steps (mass transfer, charge transfer and adsorption of intermediates and fuel products) changes from mixed kinetics process at 1 V (solid oxide fuel cell regime) towards mainly the diffusion (δ = -45°) limited process at 1.4 V (electrolysis regime).  It was shown that the total polarization resistance decreases with the replacement of Ni-YSZ to Ni-GDC. The lowest total polarization resistance (and highest current density values) has been measured for single cell consisting on the supportive ScSZ electrolyte and thin Ni-GDC cathode layer.

A tubular Ni-8YSZ/SSZ/GDC/SSC (Strontium-doped Samarium Cobaltite) cell was prepared for high temperature co-electrolysis of H₂O and CO₂. This cell enhanced a current density at750ºC and 1.28 V by two-fold compared to a previous type of our cell. Measurements of voltage-current density (V-J) curves revealed that 0.55 A/cm² was attained at 750ºC and a thermoneutral voltage (1.33 V). The V-J curves were not influenced by change of H₂/C (= 2-4) and flow-rate ratio (= 1-4) of positive electrode to negative one. In addition, V-J curve measurement at constant 80% utilization of H₂O and CO₂ clarified that the cell was stable at the high utilization and 0.41 A/cm² was achieved at the thermoneutral voltage.

Syngas production through co-electrolysis of steam and carbon dioxide has been shown as an effective method of CO₂ utilisation, however little is known about the complex surface reaction mechanisms involved.  Further understanding is needed in order to optimise SOC operation parameters and inform materials development.  Currently many investigative techniques are ex-situ, some of which are even destructive: e.g. SEM imagining of the microstructure.  To fully understand reaction intermediates, characterise electrochemical performance, and develop novel materials and microstructures for co-electrolysis SOCs; in-situ analysis methods are required.

The EPSRC funded research programme, 4CU aims to enhance the fundamental understanding of CO₂ and its role in co-electrolysis by developing a suite of in-situ investigative methods for high temperature SOC operation.  Using Diffuse Reflectance Infrared Fourier Transform Spectroscopy (DRIFTS) we can monitor both adsorbed and gas phase species in-situ.  Variation of potential and temperature conditions allow for visualization of surface mechanisms associated with single atmosphere fuel cell operation.  This, along with dual atmosphere measurements, will be discussed.  This work illustrates the applicability of the new technique for SOCs operated in a range of atmospheres using Au, Pt and Ag electrodes.  The comparison of potential dependent DRIFTS combined with electrochemical measurements provides an important insight into the roles of oxygen ion and proton transport within the systems studied.

Figure 1

The increased interest in stable and low cost electrodes for solid oxide cells (SOC) has driven the research of electrode preparation to infiltration of catalyst material into porous backbone material. The infiltration method enables a reduction of amount of catalyst material and increases its activity, due to high surface area of  catalyst nano particles. Advantage of infiltration is also separate production of electrolyte backbone structure with good ionic connectivity and mechanical properties.

With this study we present the results of a solid oxide cell with infiltrated porous yttria stabilised zirconia (YSZ) backbone air electrode and Ni/YSZ cermet fuel electrode. The SOC was tested at electrolysis conditions under high current (up to -1 A/cm²). The porous YSZ electrodes was infiltrated with gadolinium-doped ceria oxide (CGO), to act as a barrier layer between the catalyst and the backbone, and perovskite catalyst material. Cobalt doped lanthanum nickelate was used as the perovskite catalyst due to its excellent performance.  The cell was tested in steam electrolysis for at least 2000h. This initial test indicate that a stable air electrode was formed, and that the cell performance and stability matches that of a state-of-the-art SOC.

Solid oxide electrolysis cells (SOECs) are expedient to solve the site-specific and intermittent problems for current renewable energies, such as solar, wind and tidal power, as SOECs can convert these renewable energies into chemical energy in the form of H₂ that is clean and easily to be transported. Most of the studies on SOECs are focused on oxygen-ion conducting SOECs, using mainly yttria-stabilized zirconia (YSZ) as electrolyte. However, problems such as the dilution of produced H₂, high temperature operation, and oxidation of Ni-based fuel electrode occur, which the use of proton-conducting SOECs can circumvent. In addition, the high conductivity of proton-conducting oxides with respect to YSZ enables proton-conducting SOECs operating at intermediate temperatures [1].

The few works reporting on proton-conducting SOECs show the use of BaCeO₃-based electrolytes, which has been demonstrated to be chemically unstable in H₂O. Although proton-conducting BaZrO₃-based materials are the most promising electrolyte candidates for proton-conducting SOECs due to the excellent chemical stability and high bulk conductivity [2], their processing difficulties due to poor sinterability and high-resistive grain boundaries prevented the deployment in SOEC devices. In this report, we pioneeringly used BaZrO₃-based electrolytes for SOECs, demonstrating that tailoring the electrolyte and electrode materials results in further improving the electrolysis cell performance. 

Anode supported BaZr_0.9Y_0.1O_3-δ (BZY) electrolyte films were fabricated using an ionic diffusion method [3]. Both thermodynamic calculation and experimental results suggest that BZY has an excellent chemical stability in H₂O, which is critical for practical applications. A current density of 119 mA cm^-2 was obtained at 600°C, with an applied voltage of 1.65 V for the electrolysis cell, which is comparable or even higher than that for proton-conducting SOECs with BaCeO₃-based electrolyte at similar conditions, while BZY electrolyte offers much better chemical stability. This cell also operated at 600 ^oC for more than 80 h without any obvious degradation, while the BaCeO₃-cells are reported to only last tenths of minutes or a few hours [4].

To improve the cell performance, BaZr_0.7Y_0.2Pr_0.1O_3-δ (BZPY10), which is a more conductive proton-conducting electrolyte material, wass used for the SOEC. With La_0.8Sr_0.2MnO_3-δ (LSM)-BZPY10 composite air electrode, the SOEC with BZPY10 electrolyte reached a current density of 576 mA cm^-2 with an applied voltage of 1.6 V at 600 ^oC. This cell performance is much improved compared with the above discussed BZY cell. Further cell performance improvement was achieved by tailoring the air electrode material with BaZr_0.5Y_0.2Pr_0.3O_3-δ (BZPY30), which is a mixed protonic-electronic conductor rather than a pure protonic conductor [5]. By coupling BZPY30 with LSM, the triple phase boundary (TPB) is much improved due to the mixed conducting behavior of BZPY30, which is beneficial to the reaction. As a result, the cell with BZPY30-LSM air electrode produced a current density of 1007 mA cm^-2 with an applied voltage of 1.6 V at 600 ^oC. An obvious cell performance improvement is obtained with the tailored electrode.

References

• L. Bi, S. Boulfrad and E. Traversa, Chem. Soc. Rev., 2014, 43, 8195-8300.

• D. Pergolesi, E. Fabbri, A. D'Epifanio, E. Di Bartolomeo, A. Tebano, S. Sanna, S. Licoccia, G. Balestrino and E. Traversa, Nature Mater., 2010, 9, 846-852.

• L. Bi, E. Fabbri, Z. Q. Sun and E. Traversa, Energy Environ. Sci., 2011, 4, 409-412.

• L. Bi, S.P. Shafi and E. Traversa, J. Mater. Chem. A, 2015, DOI: 10.1039/c4ta07202b.

• E. Fabbri, L. Bi, D. Pergolesi and E. Traversa, Energy Environ. Sci., 2011, 4, 4984-4993.

To overcome global warming issue and energy shortage problem, economical hydrogen production technologies from renewable energy or unused fossil fuel with carbon capture technology are being developed all over the world. However it becomes the large problem that it is necessary to reconstruct energy infrastructures for production, transportation and usage of hydrogen and to entail a vast cost.

Against this problem, for effectively utilizing existing energy infrastructures, we are developing a methane producing technique from water and carbon dioxide by using renewable electricity instead of hydrogen production. Specifically, solid oxide electrolysis cell (SOEC) technology for producing syn-gas with high efficiency has been developed and we have succeeded in trial manufacture tubular SOECs whose sin-gas production rate is more than 3.5 sccm/cm² at the operating temperature of 750 ^oC and electrolytic voltage of 1.35V.

Therefore, in this paper, the numerical models of an SOEC methane production system and a conventional methane production system were made in reference to performance of the tubular SOECs and dependence of performance of the each system on operating conditions, such as operating temperature, feed gas composition, is analyzed and compared with each other.

The methane producing system which combined a water electrolysis system with a high temperature water-gas shift reactor was considered as a conventional system.

From the result of the analysis, advantage of SOEC methane production system is clarified.

The pollutant NO mainly from diesel/gasoline engine exhausts without purification into environment can give rise to the acid rain and photochemical smog^[1] which causes great potential safety hazard to human health and global environment. Accompanied by NO, there exists excess O₂ in exhausts. Electrochemical reduction of NO and O₂ can save the huge reducing agents storage system with the low operation temperatures simultaneously^[2-3]. The exhaust temperature is low that high O^2- conductivity is needed to guarantee high conversions of NO and O₂ in the low temperature range.    

In this study, we fucused on the electrochemical reduction of NO and O₂ in solid oxide electrolyte cell(SOEC) on symmetric Pt electrode with the temperature varying from 250℃ to 500℃ using three different electrolytes including YSZ, Ce_0.9Gd_0.1O_1.95(GDC) and La_0.9Sr_0.1Ga_0.8Mg_0.2O_2.85(LSGM). The GDC and LSGM electrolyte were prepared by tape-casting method and characterized by SEM after electrochemical tests. The symmetric cells were characterized by electrochemical measurements including linear sweep voltammetry(IV) and electrochemical impedance spectroscopy(EIS) at open circuit voltage(OCV). The polarization range was set from 2V to 0V containing 800ppm NO and 8% O₂. The experiment testing temperatures and the concentrations of the NO and O₂ are choosen according to the exhausts of the diesel or gasoline engine. The GDC and LSGM electrolyte showed higher electrochemical performance than YSZ by polarization measurements especially at low temperature of 250℃ to 350℃. The EIS experiment results were fitted well with equivalent circuit model R(C(RW)) containing the serial resistance R_s, electron transfer resistance R_t, double layer capacitance C and Warburg component W. It can be observed that the activation energy of the R_s and R_t were close to each other and the activation energy of the R_s were 0.581eV, 0.702eV and 0.719eV in GDC, LSGM and YSZ electrolytes. The EIS results yielded typical semi-infinite diffusive character in three electrolytes. The activation energy of the electron transfer process for the YSZ, GDC, and LSGM electrolytes were 1.057eV, 0.933eV and 1.031eV in 800ppm NO with 8% O₂. And the YSZ electrolyte displayed the maxmium activation energy of electron transfer resistance R_t indicating the larger resistance of the charge transfer process. The gases in the Pt/YSZ/Pt cell showed the lowest diffusion coefficients obtained from the Warburg component in low frequency in the electrode resulted from largest charge transfer resistance that blocked the diffusion of the gases. This studies demonstrated that GDC and LSGM can be promising electrolytes for low temperature electrochemical removal of NO and provided theoretical guidance for the design of practical NOx elecrochemical application device at low temperatures.  

References

[1] M.T. Lerdau, J.W. Munger, J.D. Jacob, Science 289, 2291 (2000).

[2] T.J. Huang, C.Y. Wu, S.H. Hsu, C.C. Wu, Energy Environ. Sci., 4, 4061 (2011).

[3] R.M.L. Werchmeister, K.K. Hansen, Electrochim. Acta, 114, 474 (2013).

This study investigated the electrochemical reduction of NO and O₂ on Pt symmetric electrode by adding NOx adsorbents for the removal of NO from the exhausts of diesel engine or gasoline engine. O₂ is excess when compared with NO. Thus, a large mount of external electric field should be consumed by electrochemical decomposition of O₂ which gives rise to a low selectivity towards NO. Electrochemical reduction of NO has confronted with the difficult of improving NO selectivity in NO/O₂ mixtures.

Two cells were prepared and tested: single Pt electrode, Pt electrode with Ba(NO₃)₂ impregnation. Linear sweep voltammetry(0V~2V), cyclic voltammetry(CV:0.5V~-0.5V) and electrochemical impedance spectroscope(EIS) measurements were carried out between 350℃ and 550℃. The Pt electrode was covered with fine particles on the impregnated electrode surface by scanning electron microscope(SEM). It could be observed from the polarization curves that the Ba impregnated electrode showed higher electrochemical performance than the single Pt electrode especially at low temperatures and high voltages (1.25V to 2V). It can be obtained from the CV tests that the Ba impregnated electrode was superior to the Pt electrode in 800ppm NO. When scanned from 0.5V to -0.5V, the electrochemical performance of NO with O₂ was far better than NO especially at high voltages and was close to the performance of O₂. While changed the sweep direction backward, NO showed the best electrochemical performance than NO with O₂ and O₂ near open circuit voltage.  

EIS results at open circuit voltage revealed that the Ba impregnated electrode exhibited lower resistance than the single Pt electrode due to the decreased polarization resistance in the low-frequency region that dominated the impedance spectra in all atmospheres. The polarization resistances in 800ppm NO showed the smallest while in 8% O₂ showed the largest in both electrodes possibly resulted from the generated NO₂ formed by NO oxidation on Pt sites, the storage of NOx in the form of Ba(NO₃)₂ in active sites or the direct electrolysis of Ba(NO₃)₂ in external electric field that enhanced the reactivity and selectivity of NO reduction.

From the impedance analysis using by equivalent circuit model R(CR)(CR)(CR) consisted of serial resistance R_s(related with the conductivity of the electrolyte) and 3 (CR) elements. Results showed that the activation energy of R_s for the two electrodes were nearly the same. The activation energy of the total polarization resistance R_p­ for the Ba impregnated electrode and Pt electrode were 0.874eV and 1.396eV respectively in 800ppm NO and 8% O₂. The results were very close to the results obtained in the pure 8% O₂ atmosphere. And the activation energy were 0.92eV and 1.50eV for the Ba impregnated electrode and Pt electrode. This was very close to the results of O₂ reduction on Pt electrode conducted by Bauerle^[1] with the activation energy between 1.43eV and 1.78eV demonstrated that the dissociative adsorption of O₂ can be the controlling reaction step. The activation energy in the low frequency(0.01Hz-2Hz) for the three electrodes were close to the results showed by Bauerle resulted from the adsorption, surface diffusion, transfer of O₂ and NOx intermediates near/at the triple phase boundary(TPB) and the dissociative adsorption of O₂^[1]. The activation energy for the Ba impregnated electrode and Pt electrode in the high frequency were 0.608eV and 0.646eV respectively. The high frequency arc indicated the diffusion of oxide ions to the electrode/electrolyte interface and charge transfer of oxide ions from electrode/electrolyte interface to the electrolyte^[2-3]. The activation energy in the middle frequency may be attributed to the charge transfer reactions in the electrode and the dissociative adsorption of O₂. And the activation energy were 0.657eV and 1.06eV for the Ba impregnated electrode and Pt electrode. Results showed that the Ba impregnated electrode exhibited lower activation energy in middle and low frequency region indicating the improvement mechanism.

Reference

[1]J.E. Bauerle, J. Phys. Chem. Solids, 30, 2657 (1969).

[2]M. J. Jørgensen, M. Mogensen, J. Electrochem. Soc., 148, A438 (2001).

[3]M. L. Traulsenz, K. K. Hansen, J. Electrochem. Soc., 158, P156 (2011).

Figure 1

Recent investigations indicated that oxygen reduction is accelerated at hetero-contacts between perovskite (La,Sr)CoO_3-d (LSC) and Ruddlesden-Popper phase (La,Sr)₂CoO_4+d(LSC-RP) [1-3]. In the present study, we want to quantify the activity of triple phase boundaries (TPB) and distinguish it from the activity of the pure LSC and LSC-RP phases, and evaluate under which conditions LSC/LSC-RP composites could be more active cathode materials compared to pure LSC.

Dense thin films of LSC/LSC-RP composites on single crystalline YSZ were prepared by a single PLD run from a two-phase PLD target with overall Sr content z=n_Sr/(n_La+n_Sr) of 0.3 and phase ratio w=n_LSC-RP/(n_LSC+n_LSC-RP) of 0.25-0.75. On the heated substrate, the incoming material from the plasma plume crystallizes into a two-phase mixture (miscibility gap between LSC and LSC-RP). The resulting dense films are a nanocomposite of LSC and LSC-RP phases. For comparison, films of pure LSC and LSC-RP only were deposited under identical conditions. Furthermore, composite samples were deposited as amorphous films on unheated substrates, and crystallized by subsequent annealing.

XRD shows crystalline LSC and/or LSC-RP phases in the single-phase and composite films deposited at 770 °C. The amorphous films developed crystal-like morphology (SEM) at annealing temperatures >500 °C. The SEM images in Fig. 1 show a crystallite size of 50-100 nm for films deposited at 770 °C, and 5-10 nm for amorphously deposited, post-annealed films. The area specific resistance of oxygen reduction (ASR, measured by impedance spectroscopy on microelectrodes) of single phase samples and composites is shown as a function of the volumetric LSC-RP content. The lines show the expected behavior of the ASR under the assumption that TPB effects are negligible. The grey line assumes a Sr content of 30% in both phases, the red curve takes into account that LSC and LSC-RP phases in thermodynamic equilibrium exhibit different Sr content (Sr accumulation in LSC-RP, for overall Sr content up to 40%). This Sr distribution can happen during the PLD films growth, and was evidenced by TEM. Its quantification by XRD on powder samples (glycine-nitrate process, annealed at 1200 °C) is shown in Fig. 2. Since the oxygen exchange activity of LSC decreases more steeply with decreasing Sr content than the activity of LSC-RP increases [4,5], the expected overall ASR of composite films is higher considering this distribution.

The post-annealed sample with high TPB length (about 10⁶ cm/cm²) showed the lowest ASR of all composites, but also deactivated most quickly (under morphological changes such as coarsening). The composites deposited at high T with shorter TPB lengths (10⁵ cm/cm² or less) exhibit a higher ASR, and after initial deactivation most of them are situated at or above the red curve. XRD/TEM indicated formation of a higher-order Ruddlesden-Popper phase at expense of LSC in some samples, which could explain an activity below the red curve. Taking the red curve as reference ASR (absence of TPB effects) for the two most active composites, their length specific resistance was in the range of 1-2 MΩcm. At 500-600 °C their ASR is lower than the values obtained in [3] for a slightly larger overall Sr content.

To summarize: dense composite films can be deposited in a single PLD run, with crystallite sizes down to 5-10 nm. The effects of Sr distribution increase the overall ASR compared to single phases with same overall Sr content, and unless very high TPB lengths are achieved (or high overall Sr/La ratios are used which diminish this effect) this effect prevails over the small activity enhancement which is specifically caused by the LSC/LSC-RP hetero-contacts.

[1] M. Sase, K. Yashiro, K. Sato, J. Mizusaki, T. Kawada, N. Sakai, K. Yamaji, T. Horita, H. Yokokawa, Solid State Ionics 178 (2008) 1843.

[2] E.J. Crumlin, E. Mutoro, S.-J. Ahn, G. Jose la O', D.N. Leonard, A. Borisevich, M.D. Biegalski, H.M. Christen, Y. Shao-Horn, J. Phys. Chem. Lett. 1 (2010) 3149.

[3] W. Ma, J.J. Kim, N. Tsvetkov, T. Daio, Y. Kuru, Z. Cai, Y. Chen, K. Sasaki, H.L. Tuller, B. Yildiz, J. Mater. Chem. A 3 (2015) 207

[4] A.V. Berenov, A. Atkinson, J.A. Kilner, E. Bucher, W. Sitte, Solid State Ionics 181 (2010) 819

[5] F. Zhao, X. Wang, Z. Wang, R. Peng, C. Xia, Solid State Ionics 179 (2008) 1450

Figure 1

A perovskite-type oxide (La,Sr)CoO₃ (LSC) has been extensively studied as a cathode material of SOFC due to its high performance.  In comparison to the other cathode materials, the bulk property of LSC can be characterized by a large oxygen nonstoichiometry, high oxide ion diffusivity, and high electronic (hole) conductivity, while it is not clear which is the factor governing the high catalytic and cathode performance.  On the other hand, discussions have recently been made on the correlation between electronic structure and cathode properties.  After evaluation of surface exchange coefficient and defect chemical analyses on a series of solid solution SrTi_1-xFe_xO₃, it was suggested that the overall cathode reaction kinetics is mainly determined by the density of minority electronic carrier in bulk, in other words, the position of the Fermi energy relative to the electronic band structure [1].  It is also proposed that the oxygen reduction reaction (ORR) activity of a series of perovskite oxides, although the reaction in aqueous electrolyte is concerned, is primarily correlated with the electronic state of the B-site transition element, that is, the occupation of e_g orbital and extent of covalency against oxygen [2].  The fundamental understanding on the electronic structure of cathode materials is thus required.

While there have been many studies on the electronic structure of LSC by means of x-ray spectroscopies and DFT calculations, most of them are concerned with the electronic transport and magnetic property, and sometimes lack determination of chemical situation of the specimens.  In this study, the authors report the comprehensive analyses of the electronic structure for LSC by using soft x-ray absorption and photoemission spectroscopy in consideration of chemistry in bulk and the cathode reaction process.

Dense specimens of LSC have been prepared via a citrate process, and the surface has been polished finally with 0.1um diamond abrasive.  The specimens have been heat-treated under some conditions of temperature and oxygen partial pressure (Po₂), and subjected to x-ray spectroscopic measurements.  O K-edge and Co L-edge XAS spectra have been collected at BL-11A and -19B of Photon Factory, KEK, Japan.  XPS spectra of the constituent core level and valence band have been collected with an x-ray photoemission spectrometer (PHI5000 / ULVAC-PHI).

Fig. 1 shows the O K-edge XAS spectra of La_0.6Sr_0.4CoO_3-d (LSC40) collected in electron yield mode.  The main absorption feature (C) observed around 536eV is rather insensitive to the heat-treatment condition, and can be assigned mainly to La5d and Sr4d hybridization orbital.  On the other hand, the low energy feature (A-B) ranging 526-531eV show a significant change depending on the heat-treatment condition.  It is noted that among the T-Po₂ conditions examined, the oxygen nonstoichiometry (3-d) changes in the following order: 2.95 under (T, Po₂)=(873K, 1atm), 2.89 (1073K, 10^-1atm), 2.83 (1073K, 10^-3atm(Ar atmosphere)) [3].  The absorption intensity at the lowest energy part (A) drastically decreases upon reduction.  From the XPS measurements, it is observed that the binding energies of the constituent core level, except for Co2p, increases upon reduction.  The above observations suggest the following picture.  With decreasing oxygen content, the electron hole introduced by Sr-doping reduces, leading to the decrease in the unoccupied density of state corresponding to the absorption feature (A).  At the same time, the occupied density of states in the valence band increases and the Fermi energy upward shifts.

In contrast to the O K-edge XAS spectra, the Co L-edge absorption spectra are overall less sensitive to the heat-treatment condition, apart from slight energy shift and appearance of shoulder structure.  It is widely known that L-edge absorption structure of Co and also the other transition elements significantly changes according to the valence and spin states.  It is therefore suggested that the charge compensation for extrinsic defects is practically taken by oxide ion rather than by Co; that is, the electron hole may be expressed as O^-.  This behavior can be ascribed to the large extent of covalency, and has been observed also for Fe-doped BaZrO₃[4].

[1] W.C. Jung and H.L. Tuller, Adv. Energy Mater.1 (2011) 1184.

[2] J. Suntivich et al., Nature Chem.3 (2011) 546.

[3] M. Kuhn et al., J. Solid State Chem.197 (2013) 38.

[4] D. Kim et al., Chem. Mater.26 (2014) 927.

Figure 1

Non-stoichiometric oxides used for the electrodes and electrolytes of solid oxide fuel cells (SOFCs) typically exhibit chemical expansion behavior due to the large defect concentrations required for high ionic conductivity or gas reactivity. Robust SOFC design requires knowledge of how chemical expansion contributes to mechanical strains at interfaces, as such deformation promotes dislocation mobility, delamination, and fracture. Direct measurement of chemical expansion via X-ray diffraction or dilatometry typically requires long time scales (minutes to hours) to allow equilibration of gas atmospheres or bulk samples, or to allow adequate signal detection in the absence of synchrotron access. Here, we describe a new approach to directly and rapidly quantify dynamic chemical expansion of non-stoichiometric oxides in situ at elevated temperatures up to 650^ºC, and demonstrate this method for the mixed-ionic-electronic conducting Pr_0.10Ce_0.90O_2-δ (PCO) as a model SOFC electrode.  The activity of oxygen is modulated via sinusoidal electrical bias signal, while amplitude and phase lag of film expansion are detected with second-scale temporal resolution and sub-nanometer displacement resolution. These dynamic chemical expansion measurements are coupled with defect models of PCO to advance understanding of defect concentration and mobility under oscillating electrical potential. Extensions of this approach to estimate activation energies, lateral expansion differences, and breathing modes of multilayered electrodes are discussed. This approach provides facile access to dynamic chemical expansion under in operando conditions of SOFC electrodes, facilitating improved design of materials that withstand large chemomechanical changes during SOFC start-up, shut-down, and redox cycling.

Mixed Ionic Electronic Conductors (MIEC) are increasingly used as oxygen electrode of Solid Oxide Cells (SOCs). The main advantage of these materials lies in their electrochemical reactivity which is not restricted to the Triple Phase Boundary lines (TPBls) but can extend to the whole MIEC surface. This results in a significant decrease in activation overpotentials. Several studies have been dedicated to cathodic polarization (fuel cell mode) to understand the MIEC reactive mechanisms. In anodic polarization (electrolysis mode), much fewer studies are available and the electrochemical process occurring remains unclear. Moreover, the role of the electrode microstructure in the reactive mechanisms has to be clarified. Thanks to a coupled modelling and experimental approach, the present study intends to analyze the predominant reaction mechanisms in both polarizations and to investigate the impact of electrode microstructure on the cell performance.

The electrode material is of La_0.6Sr_0.4Co_0.2Fe_0.8O_3-δ compound. In order to obtain relevant morphological properties, a 3D reconstruction was performed by X-ray nano-holotomography at the European Synchrotron Radiation Facility (ESRF) on the Nano-Imaging beamline ID16A. This new beamline has been designed to investigate large volumes of materials with a high spatial resolution. The obtained 3D reconstruction of a sample prepared with a plasma Focused Ion Beam (PFIB) presents a large field of view of 50 µm and a voxel size of 25 nm (Fig. 1). These characteristics allow being representative of the porous medium and thus computing accurately all the electrode morphological properties such as the specific surface area, TPBls, tortuosity factors, etc...

These morphological properties are introduced as input data in a specific electrochemical model developed in both anodic and cathodic polarizations. The model takes into account the oxygen reactivity by surface reactions at the gas/MIEC interface as well as the direct oxidation of oxygen adsorbates at the TPBls. This model has been validated on the basis of experimental results performed on symmetrical cells (with a three electrode set-up) (Fig. 2). Simulations have allowed determining the rate-limiting steps in each polarization. As expected in fuel cell mode, the reactive pathway is found to be controlled by the oxygen adsorption and incorporation followed by solid state diffusion. In electrolysis mode, it appears that the mechanism is governed by the direct oxidation at the TPBls followed by the adsorbates diffusion on the surface of MIEC particles. In addition, the specific role of microstructure on the electrode performance has been investigated in both polarizations.

The research leading to these results has received funding from the European Union's Seventh Framework Programme (FP7/2007-2013) Fuel Cells and Hydrogen Joint Undertaking (FCH-JU-2013-1) under grant agreement No 621207 and 621173.

Figure 1

In a conventional SOFC porous electrode consisting of mixed ionic and electronic conductor, the electrochemical reaction does not homogenously occur, and the reaction faradaic current decreases with the distance from the electrode/electrolyte interface. It is generally believed that the reaction distribution in the electrode is basically determined by the material properties, e.g. chemical diffusion coefficient D and surface reaction rate k, as well as the electrode microstructure, e.g. porosity, tortuosity, and particle size. However, so far, the contribution of D and k to the formation of the electrochemically active area was not quantitatively confirmed in an experimental manner. In this work, we experimentally evaluated the electrochemically active area in an La_0.6Sr_0.4CoO_3-δ (LSC64) electrode by using ¹⁸O/¹⁶O isotopic exchange experiments and the following second ion mass spectroscopy (SIMS) analysis, and the contribution of D and k was discussed.

Conventional SOFC porous electrodes have very complicated microstructures. For quantitative evaluation of the electrochemically effective area while eliminating the influence of structural complexities of the electrode, a columnar-type electrode (Fig. 1 (a)) is useful. However, this columnar electrode is difficult to be fabricated in practice. Therefore, we proposed the patterned film electrode, which is schematically shown by Fig. 1 (b). In this electrode, an Al₂O₃ insulating layer is partially inserted between electrode and electrolyte, and oxide ions can pass only through the electrode/electrolyte interface where the insulating layer does not exist.

¹⁶O/¹⁸O isotopic exchange experiments were performed with the patterned thin film electrode. The electrode was heat-treated at 973 K under P(¹⁶O₂) = 1 bar for 1 hour and thereafter under the same pressure of ¹⁸O₂ for 5 minutes. The dashed curve in Fig. 2 shows the observed profile of the oxygen isotopic ratio as a function of the distance from the electrode/electrolyte interface. The remarkable decrease of ¹⁸O concentration was found near the interface. This is because ¹⁶O can diffuse from electrolyte to electrode while surface-exchanged ¹⁸O can diffuse from electrode to electrolyte through the interface.

Similar isotope exchange experiments were performed with the patterned thin film electrode under polarization. Cathodic bias of 220 mV were applied under P(O₂) = 1 bar at 973K. The observed profile of the oxygen isotopic ratio was given by the solid curve in Fig. 2. The ¹⁸O concentration was gradually increased with approaching the electrode/electrolyte interface and drastically decreased in the vicinity of the interface. The increase of ¹⁸O was observed within about 20 μm from the interface. Such an increase of ¹⁸O near the interface is considered to be caused by electrochemical oxygen incorporation. This suggests that the electrochemically active area was approximately 20 μm from the electrode/electrolyte interface in the case of the LSC64 patterned thin film electrode under P(O₂) = 1 bar at 973 K. In the presentation, the results obtained under various P(O₂) and temperatures will be also presented, and the contribution of D and k to the formation of electrochemically active area will be discussed.

Figure 1

Solid Oxide Fuel Cell (SOFC) cathode materials rely on the catalytic processes at their surfaces to reduce and incorporate oxygen. For many common cathode materials surface passivation occurs through 'A site' segregation at common operating temperatures [1]. These processes have a negative effect upon the oxygen reduction rates and therefore overall cell performance. Quantifying the true extent these segregation effects have on the exchange properties has not been fully explored but it is clear that the properties of the surface are essential.

Many techniques have been employed to study cell properties in-situ, however, surface studies are typically restricted to post mortem or ex-situ analysis due to the necessity of a high vacuum. In this study High Temperature Environmental Scanning Electron Microscopy (HT-ESEM) has been employed to analyse the morphology changes of common SOFC cathode materials in-situ. Samples of La_0.6Sr_0.4Co_0.2Fe_0.8O_3-x (LSCF) and La_0.6Sr_0.4CoO_3-x (LSC) were annealed from room temperature to 1000^oC at pressure ranges between 3 – 6mbar in atmospheres of Oxygen and Water. Using secondary electron imaging during the thermal annealing cycle morphological and compositional changes were able to be documented at temperature for the first time (Figure 1).

This technique has enabled detailed analysis of the surface degradation that occurs during SOFC fuel cell operation. Secondary phase precipitation rate and formation behaviour has been quantified over a range of temperatures and time scales furthering the understanding of SOFC cathode surfaces significantly.

[1] M. Kubicek, A. Limbeck, T. Fromling, H. Hutter, and J. Fleig, "Relationship between cation segregation and the electrochemical oxygen reduction kinetics of La_0.6Sr_0.4Co_0.2Fe0.8O₃ thin film electrodes" Journal of The Electrochemical Society, vol. 158, no. 6, pp. B727–B734, 2011.

Figure 1 - Showing in-situ growth of secondary phase particulates on a polished surface of LSCF

Figure 1

The present study reports on the surface segregation phenomena in La0.6Sr0.4Co0.2Fe0.8O3-δ (LSCF) subjected to mechanical stress. LSCF samples with mirror-polished surfaces were annealed under uniaxial compression and also four-point bending to investigate the effects of the compressive and tensile stresses on the surface segregation. The annealing was performed at up to 1173 K under up to 100 MPa of mechanical stress. The surface segregaion was examined by using SEM, EDX, and micro-Raman spectroscopy. The Sr-rich phase segregates on the surface, especially where specific surface area is large. The size of the segregated particles increases with increasing annealing time, whereas their number reveals no clear dependence. The number of the segregated particles is fewer, whilst the particle size is larger for the sample annealed under the uniaxial compression, compared to the one annealed without compression. There is a significant difference in the segregation phenomena between the tensile and compressive surfaces annealed under the bending. The effects of the stress distribution as well as the porous structure on the segregation are also demonstrated. In addition to the segregaion, high-temperature creep deformation due to Sr diffusion and grain boundary sliding is observed. Based on the above results, the mechanism for the Sr surface segregation in LSCF under mechanical stress is discussed, considering Sr diffusion under microscopic and macroscopic stress fields. The surface reactivity of LSCF under mechanical stress is briefly discussed.

Pressurized operation is of interest for solid oxide fuel cells used in combined cycle applications, and for solid oxide electrolysis cells to influence the chemical constitution of the fuel produced.  Here we present results on the effect of oxygen pressure on the electrochemical performance of a number of solid oxide cell (SOC) oxygen electrodes, studied in symmetrical cells by impedance spectroscopy. Electrodes investigated include single-phase (La_0.8Sr_0.2)_0.98MnO_3-δ (LSM), LSM - Zr_0.84Y_0.16O_2-γ (YSZ) composite, LSM-infiltrated YSZ, single-phase La_0.6Sr_0.4Fe_0.8Co_0.2O_3-d (LSCF), and LSCF-Ce_0.8Gd_0.2O_1.95 (GDC) composite. All electrodes exhibited reduced electrode resistance with increasing oxygen pressure, with all but the single-phase LSM showing a resistance decrease of around pO₂^-0.18. The single-phase LSM electrode displayed a pO₂^-0.28 dependence. Two primary features were found in the impedance spectra equivalent model fits. A higher frequency peak around 10³ Hz had around a pO₂^-0.25 dependence and was attributed to charge transfer reaction limitations. Of all the electrodes, only the single-phase LSM sample did not exhibit the higher frequency peak. A lower frequency peak around 10² Hz had around a pO­₂^-0.15 dependence, which was attributed to either ionization of an adsorbed oxygen atom or oxygen ion transport from the electrode material to the electrolyte material. Only the LSM-infiltrated electrode did not display the lower frequency peak. Overall, these results indicated that performance losses due to oxygen electrode resistance can be reduced by operating a SOC at elevated pressure.

The Pr_0.6Sr_0.4CoO_3-δ (PSC) and La_0.6Sr_0.4CoO_3-δ (LSC) SOFC cathodes were prepared and deposited onto Ce_0.9Gd_0.1O_2-δ | Zr_0.85Y_0.15O_2-δ | Ni-Zr_0.85Y_0.15O_2-δ anode supported half-cells for cyclic voltammetry, electrochemical impedance and chronoammetry analysis. The electrochemical analysis has been conducted under various working temperatures from 550 °C to 800°C and under various fuel and compositions. The performances of PSC and LSC based single cells at 800 °C were comparatively high with maximum power output 1.48 W/cm² and 1.28 W/cm², respectively. It was established that the power output under the same conditions for PSC based SOFC single cell showed slightly higher values comparing to LSC based cells. Short term stability tests show that the both materials have good stability at 700 °C and at cell potential 0.8 V and power output 540 mW/cm².

Figure 1

Most current cathodes for SOFCs such as LSM, LSC and LSCF contain lanthanum oxide. The tendency for the lanthanum oxide to absorb moisture and degrade the materials make processing a critical issue. In this study, we have explored the fabrication of cathode materials free from lanthanum oxide. Compositions with high electrical conductivity in BaO-Y₂O₃-Fe₂O₃ and BaO-Y₂O₃-Co₂O₃-Fe₂O₃ systems have been obtained. Selected compositions with high electrical conductivities up to 600Scm^-1 have been composited with GDC10 solid oxide electrolyte to form composite cathodes with matching thermal expansion to GDC10 for solid oxide fuel cells(SOFCs). Symmetrical cells with GDC10 between the experimental composite cathodes, and small SOFCs with GDC10 as electrolyte, GDC10.NiO as anode and experimental composites as cathodes have been constructed and evaluated up to 800^oC. The electrical conductivity, thermal expansion, impedance and power densities have been studied. The power densities between 400^oC and 800^oC have been measured. The power density obtained at 800^oC is found to be about 600mW/cm². The results are presented and discussed in relation to the processing of the materials, properties, microstructures and performance of the solid oxide fuel cell. Ways to enhance the performance of the SOFCs are highlighted.

In this study, continuous manufacturing processes including spray drying, compression molding and sintering are developed and constructed to produce molybdenum (Mo)-containing nickel (Ni)-based porous alloy as a supporting component for metal-supported solid oxide fuel cell (SOFC). The porous interconnected networks of Ni-Mo alloy are made of introducing pyrolyzable filler during fabrication processes. The particle size distribution analysis results showed d₅₀ of starting material (E.g., Ni, Mo, pyrolyzable filler and binder) after spray drying is 40.7 um. A compression load of 35 ton is applied to form a specimen with size of 60×60×1.2 mm and then sintered at 1200°C to obtain porous alloy. The anode (Ce_0.55La_0.45O_2-δ -Ni, LDC-Ni), electrolyte (La_0.8Sr_0.2Ga_0.8Mg_0.2O_3-δ, LSGM) and cathode (Sm_0.5Sr_0.5CoO_3-δ, SSC) is coated by using an atmospheric plasma spraying (APS) technique. The active electrode area of the cell is 16cm² and the open circuit voltage (OCV) is higher than 1.0 V under cell performance testing from 600 to 750 °C, indicating that a fully dense layer of LSGM electrolyte is successfully fabricated via APS coating process. The measured maximum output power densities (@0.6V) of this cell have reached 1196, 1012, 716 and 415 mW/cm² at 750, 700, 650 and 600 °C respectively, by employing H₂ as fuel and air as oxidant.

Extrusion techniques can be used to produce microtubular SOFCs with a wide range of hierarchical microstructures. It is believed that by controlling the pore structure of the thicker supporting electrode, the transport properties can be optimised.

X-ray micro-tomography has been used to image the finger-like voids in a microtubular anode. These structures are on the order of 10 microns in diameter; however, smaller characteristic lengthscales can be imaged using FIB/SEM, which reveals the packing of the sintered particles, as well as the pore structure within the particles. Quantifying the tortuosity factor at each of these lengthscales offers insight into the effect of tailoring the pores.

The tortuosity factor of porous SOFC electrodes is used to account for the apparent reduction in species diffusion caused by the geometry of the network. It affects both the electrical transport in the solid phases as well as gas diffusion in the pores. The tortuosity factor is ideally calculated by solving the steady-state diffusion equation; however in systems spanning several orders of magnitude it is very computationally expensive. The figure shows the distribution of a scalar field in an interdigitated void structure, calculated using a finite element method.

This work combines the calculated tortuosity factors of each lengthscale to try and understand the significance of the hierarchical pore structure. The straight, radial pores of the microtubes imaged were found to augment the system significantly compared to an isotropic network of equivalent porosity. The importance of recasting the tortuosity factor as a function of distance from the electrolyte for hierarchical pore systems is also discussed.

Figure 1

Keywords: SOFC stack, flat tube, metallic support, Ag-Cu air braze,

  Introduction

Recently, great interests are focused on Solid Oxide Fuel Cells for future energy economy. In order to reduce the cost and increase the lifetime of SOFCs, lowering the working temperature is necessary. To meet the demands, new stack structures and materials are developed. Among them, flat tubular stacks are shown to be promising for their performance and fabrication techniques, and also metallic materials are applied as interconnect and support for their characteristics [1].    

In this work, we design and fabricate a 600W anode supported flat tubular SOFC stack. The stack is composed of 36 flat tubular cassettes made by FeCrAl alloy. Cells composed of NiO-YSZ (anode)/ YSZ (electrolyte)/LSM-YSZ (cathode) are sealed on cassettes by Ag-Cu air braze. Experimental

The schematic diagram of stack structure is shown in Fig 1. A short stack is composed of several metallic cassettes made by FeCrAl (a commercial product: 1Cr21Al4). Five frames are located on the surface of the cassette, on which cells are sealed on the anode side by Ag-Cu air brazing. Ag-Cu brazes also serve as the electrical connect to the anode. FeCrAl interconnects with Ag mesh serve as the electrical connect to the cathode. Ag wires connect th e Ag-Cu braze with interconnects. Fuel inlet pipe is located at one end of the cassette and outlet is on the other end. Pipes are sealed by Ag rings with ceramic tubes. Fuel flows along the pipes and the ceramic tubes.      

Cells of 5x5cm² are produced as described in previous results [2]. Anode substrates of NiO-YSZ are produced by conventional tape casting and sintering methods. YSZ electrolytes of 15 um are produced by spray and cathodes of 4x4cm² LSM/YSZ are produced by screen printing.

Results and Discussion

   Stack of 36 cassettes are produced and tested. Humidified H₂(2-3% H2O) is applied as the fuel and air as the oxidant. A mean open circuit voltage of 1.03V per cassette is obtained under working temperature 750 centigrade. The stack has been tested for 1000 hours. Ag-Cu braze and FeCrAl show good stability and reliability.

  Fig.2 shows the performance of the 36-cassette stack under different fuel flow. When 15slm H2 is supplied to the stack, a maximum output of 614W and an average power density of 213mw/cm²are obtained. The fuel utilization efficiency is 51%.

Conclusions

  An anode supported flat tubular stack is designed and produced. Stack of 36 cassettes are produced and tested with humidified H₂ as fuel. An average power density of 213mw/cm²is obtained, higher than those reported results of the stacks [1].

References

[1] Ludger Blum etc, Int. J. Appl. Ceram. Technol., 2 [6] 482–492 (2005).

[2] Lei Zhang etc, Journal of Alloys and Compounds 586 (2014) 10–15

Figure 1

Murata Manufacturing Co., Ltd. has been developing planar-type solid oxide fuel cells (SOFCs) based on new design concept.  The cell is fabricated by a single-step co-firing process and consists of an electrolyte, electrodes, current collectors, gas-separators with manifolds, and gas-flow channels. The membrane electrode assembly (MEA) is sandwiched between ceramic gas-separators with via-hole electrodes providing electrical paths between cells. Murata's new concept stack achieved volume power density higher than 0.5kW/L. By arranging gas manifolds in the central part of the cell, the temperature distribution in plane of the cell was moderated while keeping high volume power density. As a result, the stack operated stably at high efficiency. A long-term operation has been carried out at 750°C over 8000h under galvanostatic condition. The cell power degradation rate was less than 0.20%/kh after 2000h operation. Cr poisoning in the LSCF cathode after durability test was not observed. That is one reason of high durability of the cell.

The ability to predict the lifetime performance of an SOFC can give guidelines for where improvements and alterations can be made in terms of production, operating parameters and cell design.  Multiple attempts have been made to implement models which can respond to all the varying parameters to give their characteristic performances over longer operational lifetimes than is feasible to physically test.

This study focuses on continuous degradation mechanisms and the impact they have on the components of a single repeating unit level, i.e. the anode, cathode, electrolyte, metal interconnect and sealants. Currently the model being developed deals specifically with anode and electrolyte degradation. Depending on operating conditions, SOFC cells suffer from Ni particle agglomeration, coarsening, and volatilisation which lead to increased overpotential in the anode. A loss in the ionic conductivity in the electrolyte is also observed.

Using models derived for these degradation mechanisms combined with percolation models, Matlab ® coding has been developed to outline the loss of performance in a single repeating unit for a given set of operating parameters. These equations will be tested against real life data to establish their validity and, where necessary, adjusted for accuracy.  Further progress will allow for accelerated testing of components and cells to determine the long term performance of a given material alteration or design change.

Manufacturing of thin and gas tight electrolyte is still a challenging issue for metal supported solid oxide fuel cells (SOFCs). In this work, we present a porosity graded multi-layer structure supporting thin film electrolyte. The substrate is made of NiCrAl metal foam impregnated with La-substituted SrTiO₃ (LST) ceramic. Furthermore, a 20 µm thick composite anode functional layer with LST and gadolinium doped ceria (GDC) was deposited onto the substrate. The surface of the anode functional layer is later engineered with a suspension of nano-particles with 40 nm average particle size of yttria-stabilized zirconia (YSZ). This results in a mesoporous layer with smooth surface able to support a thin film GDC layers. Finally, a gas-tight GDC electrolyte was deposited by EB-PVD method. The thickness of the thin-film YSZ layer and the gas-tight GDC layer was approximately 1 µm and 2 µm, respectively. Button half cells with size up to 25cm² were successfully produced showing an air leakage rate down to 7.0*10^-4 (hPa*dm³)/(s*cm²), satisfying the gas-tightness quality control threshold of the state of the art metal supported SOFCs at DLR. Full cells with LSCF cathodes were then produced and electrochemically tested presenting stable Open Circuit Voltage near to 1V. Results will be presented and discussed in this paper by confronting the performance with the cell microstructure.

Key Words: metal foam, thin-film, electrolyte

The research leading to these results has received funding from the European Union's Seventh Framework Programme (FP7/2007-2013) for the Fuel Cells and Hydrogen Joint Technology Initiative under grant agreement n° 303429

Toho Gas has developed a planar type of scandia stabilized zirconia (ScSZ) electrolyte supported cell, and has been researching the improvement for the performance of cell stack.

We focused on the seal materials in order to improve the performance of Toho's ScSZ electrolyte supported cell stack in this study. The existing seal material, powder type, is mixed with the organic solvent, and printed on the cell stack assembly. However, the seal quality varies in some degree in each samples, and the yield rate of printing process is not high enough. Furthermore, we have to design the cell stack with consideration for the seal contraction, because the seal materials crystallize and contract at the high temperature.

In this study, in order to improve the seal performance, we evaluate the seal performance of two types of seal, and the contraction thickness in some conditions.

We have reported the ultimate performance level that the anode-supported solid oxide fuel cell (SOFC) can reach by employing thin-film electrolytes and nano-structure electrodes, without drastically changing the component materials. Based on the platform named as multi-scale-architectured thin-film SOFC (TF-SOFC), thin-film electrolytes of thickness less than 1 micron and electrodes with particle size in nm scale are successfully realized on NiO-YSZ anode supports, and a peak power density exceeding 500 mWcm^-2 at 500 ^oC is obtained.

Although this approach proves the potential of the thin-film technology to significantly improve the low-temperature performance of the SOFC, it is highly questionable that this platform can be transferred to the commercial sector because pulsed laser deposition (PLD) is employed to fabricate the cells. PLD is the best thin-film deposition research tool for the oxide with complex composition and has a high deposition rate, which is tremedous advantages for the thin film components in the TF-SOFC. Nevertheless, PLD has a small deposition area which cannot be larger than several cm by several cm in general, which is the most critical disadvantage of PLD to be employed as a commerciallization processing technique.

Therefore, to be employed in the commercial sector, it is the prerequisite that the multi-scale-architectured TF-SOFC should be realized by commercially viable techniques other than PLD. In this regard, the present study deals with the realization of the same platform by using other technique, specifically sputtering. Several key issues, such as nano-composite deposition, stoichiometry transfer, crystallinity of the thin films, etc., will be identified and the possible solutions to resolve these issues in sputtering will be suggested. A larger area (~ 5 cm by 5 cm) TF-SOFC using a 2-inch sputtering system will be fabricated to check the fesibility and its chracteristics will be compared with that of the cell fabricated using PLD.

Thermo-mechanical issues in planar solid oxide fuel cells (SOFC) must be understood and overcome to meet the reliability standards for market implementation. This study presents a modelling framework constituted by a thermo-mechanical and electrochemical model of a SOFC stack. Rate-independent plasticity and creep of the component materials have been considered. Modified periodic boundary conditions have been implemented to simulate the behaviour of a single repeating unit (SRU) in a stack.

The analysis starts with the simulation of stack conditioning procedures, which aims at capturing the stress build-up/relaxation throughout the several steps in the stack components production and assembly.

The simulated stresses after the initialisation sequence provide the initial stress state for the simulation of stack operation, which consists in importing into Abaqus software the temperature profiles from thermo-electrochemical simulations. The study includes long-term steady-state operation in both co-flow and counter-flow configuration, interrupted by thermal cycling. The temperature profiles implemented in the long-term operation cases correspond to operation under internal steam-methane reforming.

The simulation results show that the contact pressure is significantly lower at room temperature, after a first thermal cycle after stack qualification. The contact pressure is lost in large regions, which is expected to alter in the long-term the electrical conductivity of the interface. It also progressively decreases during long-time operation. Thermal cycling after long-term operation worsens the extent of loss of contact pressure at room temperature. Most of the active area is close to no-contact pressure.

No critical stresses were found in the sealant joining the cell to the SRU. In contrast, the stresses in the manifold sealants were found to be relevantly high. Similarly to the evolution of contact pressure, long-term operation increases the risks of sealant failure during thermal cycling.

Figure 1: On the top side, temperature field in co-flow configuration plotted on the deformed shape of a short stack (deformation along the z-axis magnified by 300:1), state: "Start Operation". On the bottom side, contact pressure at the interface anode/GDL fuel along the symmetry line of the SRU. Lines 3b, 4 and 5 are referred to the case of long-term operation in co-flow configuration.

Figure 1

The 3YSZ substrates with thickness between 90-200 μm are excellent from mechanical and economical points of view, but its main disadvantage is rather low ionic conductivity when compared with scandia-doped zirconia or fully stabilized 8YSZ. With decreasing electrolyte thickness between 40-60 μm, it is possible to significantly improve the electrochemical performance of the 3YSZ based electrolyte-supported cell (ESC) thereby fully utilizing the available mechanical stability. Development and progress in manufacturing of high power density electrolyte supported cell based on a thin (50 µm) 3YSZ substrate is presented. The cells with improved cathode based on LSMM'/ScSZ and the multilayer anode based on Ni/GDC cermet show very good electrochemical performance. The maximum power density increases up to 750 mW/cm² for developed cell without additional contact layer at 0.7V@860 °C and is greater than one of the commercial cells based on partially scandia stabilized zirconia electrolytes as well as 95 µm 3YSZ. The changes in polarization resistance of tested cells under different operating conditions as well as during redox-cycling and durability tests are discussed on basis of analysis of impedance spectra. The developed cells show a good long-term stability (proved for >1300 h) under high current density (500 mA/cm²@850 °C, N₂:H₂:H₂O=55:40:5, air). The estimated power degradation rate is lower than 0.5%/1300 h. By using Ni/GDC anode the redox cycle ability of cell under real operating conditions is considerably improved. To reduce the contacting losses in stack different contacting layers have been tested and optimized in test bench. The influence of different parameters on electrochemical performance of the cell as well as first results for cells integration in stack are presented and discussed.

One of the most sought after applications for small-scale SOFCs has been the Auxiliary Power Unit (APU) for trucks. A mSOFC could provide heat for the cab whilst powering the electrical loads required when the engine is shut down. Significant development efforts have been made by several large companies including Delphi, AVL and Volvo, all using planar configurations with diesel as the fuel. However the planar construction is prone to damage via cracking and use of diesel fuel is known to require problematic reforming. Despite large research funding from SECA and other sources, directed especially toward the 5kWe unit since 2000, there is still no commercial product in the market. The purpose of this paper is to discuss three novel ideas applied to this attractive market area:-

• Replacing the planar stacks with mSOFCs to give greater robustness

• Downsizing to 100We to allow more rapid evolution

• Focusing on dual fuel trucks using LNG as clean fuel for the SOFC

An EU project was initiated in 2014 to develop these ideas under the heading SAFARI (SOFC APUs For Auxiliary Road-truck Installations) funded by FCH-JU with 6 partners in the UK, Spain, Poland and Switzerland.  This paper describes results from the first 18 months of the project and shows that mSOFCs operating on LNG can give significant impact in the marketplace.

There is increasing interest in small SOFC generators for the conversion of commercial butane/propane fuel into DC electricity. However, with minor modification of the pre-reformer such units can also be used for the conversion of methane, natural gas, biogas or LNG.  A consortium of six groups form the UK, Spain, Poland and Switzerland is presently engaged in the development of a small SOFC power source for road trucks powered by LNG. The project "SAFARI" is funded by the FCH-JU Programme of the European Union.

Two different types of SOFC are simultaneously prepared for comparative evaluation. While the project leader "Adelan Ltd." is optimizing its tubular SOFC design (see separate conference presentation), ALMUS AG is perfecting its planar approach for road truck applications. Stacks of 60 mm x 60 mm foot print are used. They are composed of metal bipolar plates and anode-supported cells. To match the generator output to the power demand, up to four such stacks can be arranged in line, each delivering up to 80 Watts at 24 VDC and 700°C operating temperature. For low heat losses and maximum conversion efficiency, the stack arrangement is surrounded by a 100 mm thick thermal jacket made of the best available insulation material (silica gel). The arrangement is contained in a metal box for safe handling and mounting on the truck platform. A partial oxidation methane reformer is part of the system. Start-up heat is generated by burning small amounts of fuel.    

Technical details and first test results will be presented.

Lightweight SOFC stacks are currently being developed for stationary applications such as residential CHP units, for automotive applications such as APU and for portable devices. Within the EU funded project the so-called Jülich CS-design has been improved for lightweight SOFC stacks using glass-ceramic sealings that decouple the thermal stresses within the stack and at the same time allows optimal sealing and contacting. The design is highly suitable for industrial low-cost manufacturing and automated assembly. To determine the effect on the thermomechanical behaviour of the fuel cell stack, modelling and coupled thermomechanical analysis have been performed. Based on the analysis results in combination with the manufacturing experience, test results, and post-test analysis from previous stack design, substantial changes have been made to the CS-design in order to improve mechanical robustness of the stack.

The manufacturing of single parts, particularly due to the improved design of sheet metal interconnects, as well as the assembling processes are suitable for low-cost mass manufacturing. The novel decal concept of glass-ceramic sealant screen printed on foil in order to produce green tapes is used for joining the stack layers offering an enormous potential for cost savings in industrial assembly process. The new design CS^V(D2.0 in the MMLCR=SOFC project) furthermore has extended the process window of the laser welding joining process.

First CS^V–design stack tests showed a comparable electrochemical performance to the previous CS^IV design. The good adhesion of the glass sealant applied on steel sheet by screen printed tape and good contacting behaviour between cells and interconnects were furthermore confirmed by post-test analysis.

Hybrid power plants consisting of a gas turbine and solid oxide fuel cells (SOFC) promise high electrical efficiencies if both components are directly coupled and the  SOFC is operated at elevated pressure. This contribution discusses various aspects of the pressure influences on electrochemistry at the electrodes to operating strategies of a hybrid power plant.

The influence of pressure on SOFC performance has been investigated theoretically and experimentally. Experiments are carried out using a test rig that allows for characterization of SOFC stacks at pressures up to 0.8 MPa. Performance curves and electrochemical impedance spectra are used for evaluations. In addition to experimental investigations an SOFC stack model is developed based on an existing electrochemistry modeling framework. The stack model is experimentally validated and used for a theoretical analysis of pressure. As expected, Nernst potential increases with increasing pressure causing a higher open circuit voltage. Furthermore, gas diffusion is enhanced with increasing pressure and the charge transfer reaction is facilitated due to higher adsorption rates of reactants at the electrode surfaces. At constant operating conditions and efficiency an increase in SOFC power density of up to 83% is measured. If power density is kept constant, electrochemical efficiency is improved by up to 14 %. Results generally show that pressure influence is stronger at low pressures up to 0.5 - 1 MPa and weakens towards higher pressures.

 The influence of pressure on formation of nickel oxide and solid carbon is investigated. An analytical evaluation of the nickel oxidation propensity shows thatnickel oxidation is more likely to occur at higher pressures because the equilibrium partial pressure of oxygen in the anode gas increases. Carbon deposition is another degradation mechanism that can decrease the performance of an SOFC system. It was investigated via thermodynamic simulations using the software package Cantera. Thermodynamic equilibrium of gas mixtures with different oxygen to carbon ratios is calculated showing that the aptitude for carbon deposition is highly pressure dependent. Carbon deposition should be avoidable if oxygen to carbon ratio is kept above 2 within conditions that are relevant for hybrid power plants.

The developed stack model is integrated into an existing validated gas turbine model that is extended to include further SOFC system components. A system operating strategy is presented that is based on a gas turbine control. Operating conditions of the SOFC are not directly controlled. A sensitivity analysis is carried out showing that the power ratio between gas turbine and SOFC is the most important parameter in order to achieve a high electrical efficiency. Other parameters like the number of SOFC stacks as well as gas and heat recirculation rates are of less importance. Thermal losses can significantly reduce electrical efficiency if they occur downstream of the recuperator.

Finally, the operating range of a hybrid power plant based on the proposed system control is investigated. It is found that high electrical efficiencies above 60% (based on the HHV) are achievable within an electrical power range from 310 to 670 kW if gas turbine speed and SOFC electrical power are adjusted.

Based on various efforts, commercialization of fuel cell technologies has been just started in Japan. Energy Basic Plan in Japan has been revised where hydrogen is clearly stated as one of the important energy carriers. After this revision, roadmap for fuel cell and hydrogen technologies is established up to 2040. While more than 100,000 residential fuel cell units have been sold and the commercialization of fuel cell vehicles (FCV) has just been started, industrial fuel cells as distributed power systems will be commercialized probably from 2017. For such larger-scale stationary applications, SOFC is the promising technology.

In Kyushu University, "Smart Fuel Cell Demonstration Project" is going on where our University Campus will become a campus where fuel cell and hydrogen technologies are widely applied. While FCV, hydrogen fuelling station, and various residential fuel cells are in operation, installation of a few types of SOFC systems is going on, including a 250kW SOFC-MGT power generation system, 5kW SOFC cogeneration systems, and 1kW-class SOFC cogeneration units. These SOFC systems can cover ca. 5 % of the whole electricity demand of this university campus with a maximum electricity contract of 6000 kW. As shown in the Figure, the university campus will be a future energy society where SOFC is playing an important role as the efficient power generation system. The electric power generated from solar, wind, and SOFC will be displayed on the web site of our campus to improve the social acceptance of SOFC technologies. SOFC will act as a base power generator, while hydrogen refueling station will act as an energy storage station for the renewable electricity and FCV will be a zero-emission vehicle using CO₂-free hydrogen. Concept for future energy society will be discussed where SOFC will take a major role, based on our demonstration experiences.

Figure 1

Affordable sanitation around the world with the recovery of valuable energy and clean water from faecal biomass in an off-grid system is a challenge. An energy neutral sanitation  system  concept  based on plasma gasification and solid oxide fuel cell (SOFC) was presented by TU Delft {Mings et al.}. The system is further optimised and  novel small scale CHP toilet system concept  has been developed which can generate power very efficiently and produce clean water. Pre-dryed faecal biomass, is fed to the integrated system. The system combines a dryer, a plasma gasifier, and a gas cleaning unit with an ambient pressure stack and a micro steam turbine (MST). The gasifier uses the heat supplied from the combustion of SOFC depleted fuel. With this concept syngas with high heating value can be produced, which can be efficiently converted in a SOFC system. This is reflected by the net electrical efficiency of the integrated system configuration  of the order of 50%. Additional power is produced combining it with an MST and an efficiency increase of the order of 5% is achieved. Variation of the different operating conditions in the system reveals an optimum for the temperature, steam to biomass ratio and oxidant to fuel ratio in the gasifier. Furthermore, the SOFC operating temperature influences the system performance and self-sustainability. As a result, from a thermodynamic point of view it is demonstrated that a high efficiency power generation through faecal biomass can be realized.

Ming Liu, T. Woudstra, E.J.O. Promes, S.Y.G. Restrepo, P.V. Aravind. System development and self-sustainability analysis for upgrading human waste to power.

A new bi-functional fuel-cell/battery hybrid system has recently been developed in the presenter's group for power generation and energy storage. The new system is comprised of a reversible SOFC as the electrical charger/discharger and a redox-active metal/metal-oxide chemical bed as the energy storage. During operation, the cathode of the system is constantly open to air, whereas the anode configuration defines the true functionality of the system: power generation when the anode chamber is open to a flowing fuel, and energy storage when the anode chamber is closed next to a neighboring multivalent metal/metal-oxide chemical bed.

The presentation highlights recent progress in computational analysis of physics and chemistry governing the performance of the fuel-cell/battery hybrid. These highlights include the identification of performance limiting factors, analysis of heat flow balance and proposal of strategies to achieve energy efficient hybrid system for simultaneous power generation and energy storage.

The daily growing global energy consumption has to be provided in the future by technologies which emit increasingly less greenhouse gases. This requires highly efficient energy systems. Methane fueled SOFC systems are already able to produce electricity with an efficiency up to 60 %. Through the combined use of produced electricity and thermal energy the overall efficiency can be above 90 %.

One possibility for a further improvement of the electrical efficiency is the implementation of an anode off-gas recirculation loop. Challenging is the high anode off-gas temperature of at least 700 °C, which prohibits the use of commercial components. Therefore, the use of ejectors and blowers with or without anode off-gas cooling are discussed in literature. At Forschungszentrum Jülich an anode off-gas recirculation loop operating at temperatures up to 200 °C is favored. To analyze the effect of anode off-gas recirculation, a dynamical model of the system including the recirculation loop was implemented in Matlab/Simulink®. The results show, that the recirculation rate has a huge effect on the electrical efficiency (both positive and negative). In principle, at constant current density high recirculation rates decrease the cell voltage and increase the power consumption of the recirculation blower. Therefore, the highest electrical efficiency can be reached with high systems fuel utilizations, low recirculation rates and in consequence high stack fuel utilizations. On the other hand less amount of steam is available for the reforming reaction at low recirculation rates, therefore the minimum recirculation rate is determined by the formation of carbon. An optimal operation range to avoid carbon formation and to ensure a high electrical efficiency requires recirculation rates between 65 and 70 % and system fuel utilizations above 90 %. This leads to acceptable stack fuel utilizations in the range of 70 to 75 % and electrical efficiencies above 60 %.

The performance of a cathode air preheater with internal off-gas oxidation is presented. This component is a plate heat exchanger, handling both injection and oxidation of depleted anode gas in the hot cathode flow. The gas mixture oxidizes in the heat exchanger, and the reaction heat is used to preheat cathode air.

An initial integration effort using two separate components is reported. The system consists of a close-coupled injector and a downstream plate heat exchanger, which was catalytically coated. The performance was constrained by reaction kinetics: measurements showed that the off-gas and cathode mixture typically ignited within 5 – 10 ms at mixture temperatures above 750°C. 

CFD on this system shows that part of the mixture takes more than 15 ms before reaching the inlet port of the heat exchanger, leading to early ignition. Both CFD and SEM analysis showed that the walls of the heat exchanger overheat to 1000 – 1100 °C. The SEM analysis reveal the effect of this overheating on scale growth and evaporation of Cr VI in the heat exchanger, and its subsequent condensation on the cool, catalytically coated walls.

The integrated HEX and oxidizer consists of an internal injector, which delivers the anode gas to each cathode flow path between pairs of heat exchanging plates. The injection occurs adjacent to a catalytically coated zone in the heat exchanger. The CFD effort covers the combined effects of injection, mixing, generation of reaction heat and heat exchange. The results show that the injection and subsequent mixing is uniform. The mixture reaches the cooled sections of the heat exchanger within 3 ms, and thus below the auto ignition time. These calculations are in line with experimental performance data on the integrated component.

Figure 1

In the presented work a medium-sized integrated system based on a sodium heat pipes powered allothermal fluidized bed gasification and SOFC is introduced, simulated and validated in Aspen Plus. Suitable feedstocks for such fluidized bed gasification systems are for example waste and biomass, especially wood. The concept development and evaluation have been conducted as part of the EU FCH-JU project SOFCOM.

During the allothermal heat pipes gasification process heat for the endothermic gasification reactions is transported into the gasification chamber via the heat pipes. The heat pipes allow low temperature differences using mainly latent heat transport. Inside the heat pipes sodium is evaporated on the hot side by a heat source (i.e. combustion chamber) and condensed on the cold side (i.e. gasification chamber). Heat pipes based gasification has proven to be a reliable technology at different installations by the company Agnion energy GmbH and others. The gasifier model has been validated against data from this type of gasifier. The SOFC model applied is a thermodyamic model, which has been developed in the frame of SOFCOM and already described in detail in other works.

In the presented system configuration the product gas from the gasifier is cleaned during several process steps at an intermediate temperature of 350°C. After the cleaning procedure the gas is introduced into the SOFC anode where it is converted into electricity and high temperature heat at a fuel utilization rate of 80%. At the same time preheated air is fed to the cathode of the SOFC. The SOFC exhaust is fed into a fluidized bed post combustion chamber where residual fuel is burned with residual oxygen. Heat is extracted via the heat pipes from the combustion chamber, as well as the SOFC itself. Combustion exhaust gases are used to preheat the cathode air. Residual exhaust heat is used to generate high pressure superheated steam, which in turn is expanded in a steam turbine. At this point, the steam necessary for the fluidization of the gasifier and gasification agent is not produced in a separate steam generator but instead extracted from the steam turbine. Especially by means of utilization of excess heat from the SOFC and its exhaust gases via the heat pipes and full integration of the steam cycle the electrical efficiency of the combined system reaches up to 62%_LHV.

Figure 1

Exposure of solid oxide fuel cell (SOFC) cathodes to atmospheric air contaminants, such as humidity, impacts the oxygen reduction reaction (ORR) mechanism and can result in long-term performance degradation issues. Therefore, a fundamental understanding of the interaction between water molecules and the cathode is essential to develop improved performance cathodes with enhanced durability. To study the effects of humidity on ORR, we used in-situ ¹⁸O isotope exchange techniques to probe the exchange of water with two of the most common SOFC cathode materials, (La_0.8Sr_0.2)_0.95MnO_3-x (LSM) and La_0.6Sr_0.4Co_0.2Fe_0.8O_3-x (LSCF). In these experiments, heavy water, D₂O (with a mass/charge ratio of m/z=20), was used to avoid the overlapping of the H₂O and the ¹⁸O₂ cracking fraction, which both have a peak at m/z=18. A series of temperature programmed isotope exchange measurements were performed to comprehensively study the interaction of water with the cathode surface as a function of temperature, oxygen partial pressure, and water vapor concentration. The results suggest that H₂O and O₂ share the same surface exchange sites. Therefore, the presence of H₂O competes with O₂ for the available surface sites. Our findings show that H₂O prefers to exchange with LSM and LSCF at a lower temperature range, around 300-450°C. For LSM, O₂ is more favorable than H₂O to be adsorbed on LSM surface and the presence of O₂ limits H₂O exchange with the LSM surface. For LSCF, H₂O has two different exchange mechanisms, each dominating in a different temperature region. The experimental data are summarized in a Temperature-P_O2 diagram to visualize the dominant reactions at each temperature and P_O2 for the two cathode materials.

A detailed computational model of the oxygen reduction reaction in composite strontium-doped lanthanum manganite/yttria-stabilized zirconia (LSM/YSZ) SOFC cathodes is presented.  The coupled interactions of elementary heterogeneous chemistry, elementary electrochemistry (at two separate electrochemical double layers) and transport through the porous electrode are studied. Species diffusion and convection through the porous cathode are evaluated using the Dusty Gas Model while a distributed charge transfer model is used to study the ionic and electronic transport through the cathode. Charge-transfer reactions are modeled using a spillover mechanism with no a priori assumption of rate limiting steps unlike the frequently published Butler-Volmer approach. The model is validated against overall area-specific resistance (ASR) values previously reported by Barbucci et al. [Electrochemica Acta, 47 (2002)] over a temperature range of 950-1200 K. A sensitivity analysis reveals the impact of various kinetic parameters on the ASR and subsequently leads to the identification of rate-limiting steps. Preliminary results indicate that the ASR is sensitive to the rate of dissociative adsorption of O₂ on LSM and oxygen ion transfer from the YSZ surface to the YSZ bulk.

Introduction

La_1-xSr_xCo_1-yFe_yO_3-δ (LSCF) cathodes are best suited for high power densities of anode-supported cells even at low operating temperatures (1). This requires the development of a rather sophisticated processing route, because a poorly conducting SrZrO₃(SZO) phase may develop at the cathode/electrolyte interface (2). The Gd-doped Ceria (GDC) interlayer, which blocks the SZO reaction, densifies only at rather high sintering temperatures. This leads to GDC/YSZ interdiffusion with a lower ionic conductivity (3).

This study presents a high-resolution analysis of the temperature-dependent reaction processes between LSCF-cathodes, GDC-interlayers and Zr-based electrolytes.

Experimental

Symmetrical LSCF-GDC-YSZ cells (YSZ - Y₂O₃ stabilized ZrO₂) were prepared, for which the sintering temperature of the screen printed GDC interlayer (T_sinter,GDC) was systematically varied between 1100...1400 °C. Porous LSCF was applied by screen printing subsequently and sintered at 1080 °C for 3h. A comprehensive scanning transmission electron microscope (STEM) analysis was performed, including energy dispersive x-ray spectroscopy mappings (EDXS) for a high spatial resolution of the elemental distribution along the interfaces. The corresponding cell performance was tested by means of current-voltage curves (Fig. 3) and electrochemical impedance spectroscopy measurements (EIS) with symmetrical cells and anode-supported cells (ASC) provided by Forschungszentrum Jülich (JÜLICH).

Results and Discussion

The symmetrical cell tests reveal a complex interaction between LSCF and GDC, depending on sintering temperature and time. Besides Sr, also Co and Fe evaporate and play a yet underestimated role as sintering aid during densification. In the presence of gaseous Co-species cation diffusivity is significantly enhanced (4) and leads to Gd demixing in the Gd_0.2Ce_0.8O_2-δ (GDC) interlayer (Fig. 1). A side effect is the formation of GdFeO₃ and CoO grains inside the GDC layer. Furthermore, Gd³⁺ is likely to diffuse into the adjacent layers YSZ and LSCF (Fig. 1). This seems to be crucial, as already small amounts of Gd in LSCF lead to a drastic performance decrease. We assume that the smaller Gd³⁺ cation tilts the rhombohedral LSCF to the orthorhombic structure (5). At the GDC/YSZ-interface, a gradually increase in SZO phase is observed from high (Fig. 2) to low GDC sintering temperatures (Fig. 1). Understanding the structural and electrical interplay of these phases is essential. We will correlate our findings from STEM, EDXS, C/V (see Fig. 3) and EIS measurements and propose guidelines for processing high-performance anode-supported cells at a high reliability level.

Conclusions

The complex interplay between LSCF cathode and GDC interlayer during fabrication is discussed. LSCF constituents Co and Fe play a yet underestimated role as sintering aid for screen printed GDC layers during LSCF sintering. The consequence is pronounced cation diffusion along the cathode/electrolyte interface which affects the electrochemical performance drastically. In summary, optimum conditions for GDC fabrication will be presented that have practical relevance in every large-scale SOFC fabrication routine.

References

• L. Blum, P. Batfalsky, L.G.J. de Haart, J. Malzbender, N.H. Menzler, R. Peters, W.J. Quadakkers, J. Remmel, F. Tietz, D. Stolten, ECS Trans.,57, 23 (2013).

• H. Yokokawa, N. Sakai, T. Horita, K. Yamaji, M.E. Brito, H. Kishimoto, J. Alloys Compd.,452, 41 (2008).

• A. Tsoga, A. Naoumidis, A. Gupta, D. Stöver, p. 308-311, Mater. Sci. Forum, 794 (1999).

• G.S. Lewis, A. Atkinson, B.C.H. Steele, J. Drennan, p. 567–573, Solid State Ionics, 152–153 (2002).

• A. Mai, V. Haanappel, F. Tietz, and D. Stöver, p. 2103–2107, Solid State Ionics, 177, 19–25 (2006).

Figure 1

Mixed ionic-electronic conductor (MIEC) cathode materials such as La_(1-x)Sr_xCo_0.2Fe_0.8O_3-δ (LSCF,where x=0.2, 0.3, 0.4), Ba_0.5Sr_0.5Co_0.8Fe_0.2O_3-δ (BSCF 5582), Sr₂Fe_1.5Mo_0.5O_6-δ (SFMO), La₂NiO_4+δ (LNO), GdBaCo₂O_6-δ (GBCO), and PrBaCo₂O_6-δ (PBCO) exhibit excellent oxygen transport properties as well as high electronic conductivity at operational temperatures of SOFCs. Their promising electrochemical performance make these materials excellent candidates for potential application as SOFC cathodes. However, their long-term thermodynamic stability in the operating environment of SOFCs, including unexpected and uncommon situations such as CO₂-leakage from the anodic side is of concern and should be thoroughly investigated prior to deploying them in cells and stacks.

This study investigates the thermodynamic stability of promising SOFC cathodes under operating conditions, in molten salt media, and through galvanic cell measurements. The cathode materials of interest are La_(1-x)Sr_xCo_0.2Fe_0.8O_3-δ (LSCF, where x=0.2, 0.3, 0.4), Ba_0.5Sr_0.5Co_0.8Fe_0.2O_3-δ (BSCF 5582), Sr₂Fe_1.5Mo_0.5O_6-δ (SFMO), La₂NiO_4+δ (LNO), GdBaCo₂O_6-δ (GBCO), and PrBaCo₂O_6-δ (PBCO). The stability of these materials is studied in air and at various CO₂ partial pressures.

High polarization losses associated with the oxygen reduction reaction (ORR) that occurs at the cathode and degradation of cathode materials under operating conditions remain serious issues for widespread implementation of solid oxide fuel cells (SOFC). Performance of the cathode is determined by the microstructural/interfacial relationship between the electrode and electrolyte, and the inherent materials properties. Rates of degradation depend significantly on the operating temperature, applied-potential, and environmental conditions, such as the presence of unwanted oxygen-containing compounds, namely H₂O and CO₂. Common degradation mechanisms can be categorized into three main groups: sintering, poisoning, and secondary phase formation. In this study we explore the effects of operating conditions on the above degradation mechanisms for two common cathode materials, (La_0.8Sr_0.2)_0.95MnO_3±δ (LSM) and (La_0.6Sr_0.4) (Co_0.2Fe_0.8)O_3-δ (LSCF) using a multifaceted approach. Three-electrode cells have been aged under various temperatures, applied-potentials, oxygen partial pressures and contaminant conditions in order to observe changes through electrochemical impedance spectroscopy (EIS). In-situ EIS is a powerful tool that allows us to identify changes in conductivity for the reaction steps comprising the overall ORR. We have developed a powerful strategy for identifying the steps in the ORR that contribute to the overall impedance spectra, and how that contribution changes as a function of operating conditions and aging time. Our EIS results indicate a strong correlation between blocking effects caused by CO₂ and H₂O and the operating temperature for LSCF, while similar experiments performed on LSM show significantly less effect caused by the presence of these contaminants. Using EIS to de-convolute the overall cathode polarization helps us identify the mechanisms by which degradation occurs. In addition, focused ion beam-scanning electron microscopy (FIB-SEM) ex-situ analysis of the three-electrode cells has been performed to identify changes in the microstructure and composition of the electrodes, which are directly correlated to changes in the EIS response, and reveal specific regions where degradation has occured. Further, to investigate the fundamental kinetics of these materials, we have used a variety of gas-phase isotope exchange experiments on fresh and aged powder samples of the same cathode compositions to study the change in materials properties. Using impedance spectroscopy, supported by FIB-SEM and isotope exchange, we elucidate how changes in the conductivity, microstructure and kinetics of SOFC cathodes relate to their degradation.

Roddlesden-Popper Phases, such as La2NiO4, have been proposed as promising intermediate temperature SOFC cathodes, due to their high surface oxygen exchange and bulk diffusivities. However, relatively low electrical conductivity of LNO phase is a concern for its application. It has been shown that Sr doping is an effective way to improve LNO's electrical conductivity, but it is important to investigate the effect of such doping on other properties such as oxygen surface exchange coefficient. In this study, Sr doped lanthanum nickelates i.e., La_2-xSr_xNiO_4+δ (0≤x≤0.4) thin and dense films on dense thick GDC (~1mm) substrate, were prepared by using spray-modified pressing method. The surface oxygen transport kinetics was investigated by electrical conductivity relaxation (ECR) technique. Since the thickness of thin films is ~15μm, much less than the characteristic length of LNO, the oxygen transport kinetics are controlled by the surface exchange steps and the oxygen ion concentration is uniform in the film at any time during the ECR process. Such situation enables a more accurate fitting results compared to the traditional fitting procedure based on thick pellets, because only k value is the unknown parameter. The fitted k for LNO is 4.02×10^-5cm/s at 0.2atm 700^oC, which decreases with lowered oxygen partial pressure. Sr doping harmed the surface exchange kinetics, k of Sr40 is about one order of magnitude smaller than undoped one. Interstitial oxygen and Ni oxidation state are suggested to be predominant roles in determining the surface kinetics. Both Sr doping and lowered oxygen partial pressure lead to the loss of interstitial oxygen and the oxidation of Ni²⁺, resulting the lowered surface exchange rate.

Reversible changes in the lattice parameters of SOFC cathode were observed depending on the temperature (T), electrode potential (E) and oxygen partial pressure (p_O2) applied. At fixed T and p_O2, the cathode potential noticeably influences the unit cell volume, thus, the oxygen stoichiometry and concentration of vacancies.

The stabilization time of the XRD peak was in minute scale similarly with the plateau time on chronoculonometric curves. Relaxation time of the changes in crystal dimensions are correlated with electrochemical measurement data. The electrode structure (porosity) has strong influence to the shift time of XRD peak.

There was registered only slight influence of oxygen partial pressure and potential to the crystallographic parameters of LSM (La_0.6Sr_0.4MnO_3−δ). At the same time for LSC (La_0.6Sr_0.4CoO_3−δ) and GSC (Gd_0.6Sr_0.4CoO_3−δ) the crystallographic unit cell volume was a function of electrode polarization and oxygen partial pressure. There is no remarkable vacancy formation detectable with HT-XRD in LSM cathode lattice caused by electrode polarizations applied.

Thermochemical crystallographic expansivity study of LSM, LSC and GSC cathodes with in situ HT-XRD method gives valuable experimental data about the crystallographic changes caused by changes in operating parameters like cell potential, temperature, and oxygen partial pressure in cathode compartment and therefore helps to understand thermomechanical and degradation behavior as a function of different conditions applied for  SOFC cathode.

Figure 1

Composite solid oxide fuel cell (SOFC) cathodes consisting of ionic conducting scaffolds infiltrated with mixed ionic and electronic conducting (MIEC) nano-particles have exhibited some of the lowest reported polarization resistances in the literature.  In this study, symmetric cell cathodes were prepared via wet infiltration of Sr_0.5Sm_0.5CoO₃ (SSC) nano-particles via a nitrate process into porous Ce_0.9Gd_0.1O_1.95 (GDC) scaffolds to be used as a model system to investigate performance under varying infiltration loadings.  Detailed analysis of the infiltrated cathodes was carried out using electrochemical impedance spectroscopy (EIS).  The results presented show that the infiltrated cathode microstructure has a direct impact on the initial performance of the cell.  Small initial particle sizes and high infiltration loadings (up to 30 vol% SSC) improved initial polarization resistance (R_P).  A simple microstructure-based electrochemical model successfully explained these trends in R_P. Further understanding of electrode performance was gleaned from fitting EIS data gathered under varying temperatures and oxygen partial pressures to equivalent circuit models.  Both RQ and Gerischer impedance elements provided good fits to the main response in the EIS data, which was associated with a combination of oxygen surface exchange and oxygen diffusion in the electrode. A gas diffusion response was also observed at relatively low pO₂.

Oxygen electrodes of Solid Oxide Electrolysis Cells (SOECs) have been manufactured using a spray deposition method with strictly controlled parameters to understand electrode transport, improve performance and extend electrolyser life time. The controlled parameters were the; particle size distribution, solid fraction of the ink, and spray deposition rate. The effect on physical properties; porosity, pore size, and the mass fraction of the individual ceramics in the solid phase as a function of distance from electrolyte, caused by controlling the above mentioned parameters was studied. The changes in performance of the cells with differing parameters were observed as; polarization responses, long term current-voltage behaviour, impedance spectroscopy and effective conductivities, measured with the van der Pauw method. The polarization responses gave valuable information to compare cells to those commercially available. The impedance spectroscopy of the cells allowed for the study of the changing physical properties effect on the electrochemical behaviour. Specifically the individual electrochemical processes; cell conductivity, oxygen gas diffusion, and oxygen evolution.

The degradation of solid oxide electrolysis cells (SOECs) is an issue of both scientific and technical importance. In this study, chromium deposition and poisoning at La_0.8Sr_0.2MnO_3+d (LSM) and La_0.6Sr_0.4Co_0.2Fe_0.8O₃ (LSCF) oxygen electrodes are studied under SOEC operation conditions. Significant performance degradation was observed for the reaction in the presence of Fe-Cr alloy metallic interconnect, indicating that presence of gaseous Cr species significnatly poisons and degrades the electrocatalytic activity of LSM and LSCF oxygen electrodes. In the case of LSM electrodes, XRD, XPS and SIMS analysis clearly identified the deposition of SrCrO₄, CrO_2.5 and Cr₂O₃ phases on the YSZ electrolyte surface and LSM electrode inner surface for the reaction under anodic polarization conditions. In the case of LSCF, Cr deposition occurs in the electrode bulk and on the electrode surface, forming SrCrO₄. The results clearly indicate that the Cr deposition is closely related to the increased segregation of SrO species under anodic polarization conditions. The mechanism of the Cr deposition at LSM and LSCF electrodes under SOEC operation conditions is discussed.

The long-term stability of GDC interlayer as a reaction barrier between La_0.6Sr_0.4Co_0.2Fe_0.8O_3-δ (LSCF) cathode and polycrystalline yttria-stabilized zirconia YSZ substrate is essential in improving the performance of the LSCF cathode at the required operating temperature. However, the formation of resistive phases, especially SrZrO₃ at the interfaces that affects the performance of LSCF cathode is still difficult to mitigate even with the use of GDC interlayer. For this purpose, we systematically evaluated the long-term stability of GDC interlayer by performing cation diffusion experiments using porous and dense LSCF layers. The dense LSCF/GDC/poly-YSZ has homogeneous interfaces, enabling simpler analysis of interface stability and cation diffusion. In both heterostructures, a ~1.0 µm–thick, dense GDC interlayer was utilized which was prepared by PLD with subsequent annealing at 1300°C for 5h. The LSCF/GDC/YSZ heterostructures were then annealed in air at temperatures ranging from 800°C to 1000°C starting from 168 hours to 731 hours. We find that in porous LSCF, the GDC interlayer characterized by significant formation of pores while severe microcraking occurred in dense LSCF after prolonged annealing. We also observed shrinkage of GDC interlayer at elevated temperature in porous samples compared to dense. At high sintering temperature, the Sr and Zr diffusivity along the grain boundaries of GDC interlayer is the dominant diffusion pathway in porous LSCF/GDC/YSZ leading to the formation of SrZrO₃, while for dense LSCF/GDC/YSZ, the cracks or crack–walls in the GDC interlayer are fast diffusion paths for cation diffusion. We also observed that the lateral growth of SrZrO₃ at the GDC/YSZ interface is influenced by the microstructure of GDC interlayer and sintering temperature. Our results show that the interfacial microstructure of LSCF/GDC plays an important role in understanding the formation of SrZrO₃ in LSCF/GDC/YSZ triplets as well as the effectiveness of GDC interlayer as a reaction barrier.

Considering the expected lifetimes for Solid Oxide Fuel Cells of 5 to 10 years, durability is still a major issue. Even though quite a number of research institutions and companies have proven an acceptable degradation rate of cells, stacks and systems for periods ranging from a few hundred hours to several years, neither reliable degradation models nor methodologies to evaluate the durability in a short timeframe are available. Thus any kind of modification on the cell or stack level, which might affect the durability, requires an expensive and time consuming repetition of the durability test. Accelerated lifetime tests, which enable a rapid degradation analysis, and degradation models, that enable an extrapolation of the results, would be highly desirable.  

In most durability investigations a cumulative degradation rate is evaluated from the cell voltage decrease in a galvanostatic operating mode. In addition, extensive post test analysis is performed to detect the failure causes. This approach does not reveal any quantitative information on the different underlying degradation mechanisms. If aggravated stress is applied by increasing operating temperature, current density or gas utilization, the degradation of cathode, electrolyte, anode, contact layers and interconnects is accelerated in different ways. Therefore the cell voltage alone is insufficient to understand the performance degradation and its acceleration due to aggravated stress.

To deconvolute the degradation of cathode, electrolyte and anode (i) electrochemical impedance spectroscopy, (ii) impedance data analysis by the distribution of relaxation times and, (iii) a subsequent CNLS-Fit to a physically meaningful equivalent circuit model is suggested. Cell tests were performed at different stress levels by varying temperature and current density as well as fuel and oxidant composition. Based on this extensive data set, the interplay between stress level and impact of the different stresses on degradation mechanisms in cathode, electrolyte and anode will be presented and guidelines for the design of accelerated tests will be discussed.

The solid-state ceramic construction of SOFCs enables high fuel to electricity conversion efficiencies as high as 50 to 60 percent LHV with high temperature operation, allows more flexibility in fuel choice.  Preceding studies have been  reported on multi-stage electrochemical oxidation with SOFCs for further improvement of the electric efficiencies. However there are many parameters for the multi-stage oxidation, and effects of the parameters on the efficiency remains to be identified.

In this study, we have conducted a parametric study of SOFC performances with multi-stage electrochemical oxidation by using a symbolic analysis method developed for the parametric analysis to investigate the possibility of enhancement of the electric efficiencies. In the case of two-stage oxidation, the parameters having important impacts on the efficiency are found to be ratio of active electrode area of a first stack to a second stack, and net fuel utilization which should be decided by the ratio and the fuel utilization limit of the each stack.

This research is supported by The Japan Science and Technology Agency (JST) through its COI Program.

A new concept of a high temperature fuel cell based on a dual H⁺ and O^2- conducting membrane was successfully developed recently ("IDEAL-Cell", FET-Energy/FP7, 2008-2011). It operates in the range of 600-700 °C, and was shown to be superior of standard SOFCs and PCFCs at equivalent overall thickness. It is based on the junction between the anodic part of a PCFC and the cathodic part of a SOFC through a mixed H⁺ and O^2- conducting porous composite ceramic membrane, avoiding all disadvantages associated to the presence of water at the electrodes. In the initial configuration, the porous mixed conductivity central membrane was made of a composite BCY15/YDC15, in which the active sites lay along the triple contact lines (TPB) between the protonic conducting phase (BCY15), the anionic conducting phase (YDC15) and the gas phase in the pores. During the course of the European project cited above, it was discovered and modeled that, in addition to be a protonic conductor, BCY15 was also an excellent oxygen conductor at 600-700 °C provided that it is fed with oxygen. Therefore, a second generation of dual membrane cell was developed in which YDC15 is replaced by BCY15, leading to a strongly simplified cell, much easier to shape and sinter, with potentially higher performances, i.e. the TPB become full active surfaces, magnifying the number of active sites; i.e. the dual membrane being solely made of BCY15, the tortuosity of the conductive phase is strongly diminished and its volume fraction increased, hence the electric resistance becomes much lower. As a whole, this so-called "monolithic concept" (for the cell is almost essentially made of a single BCY15 phase) improves the chemical and mechanical compatibility, increases the global cell conductivity and represents an important step towards simplifying the technology for industrialization. Moreover this new concept shows a good reversibility between the SOFC and SOEC modes; since this cell is based on 3 separate compartments (oxygen at the cathode, hydrogen at the anode, water at the dual central membrane), the dynamic of the device when shifting from one mode to the other is very high (no need to adjust the gas mixture at electrodes).

The present work proposes to further increase the electrochemical properties of this monolithic dual membrane high temperature fuel cell by studying the catalytic properties on the mixed H⁺ and O^2- conducting membrane with or without addition of Pt nanoparticles or Ni foam, which have been evaluated by impedance spectroscopy and polarization measurements. These additions led to slight adjustments of the shaping parameters to obtain flat and comparable samples. Finally, the electrochemical performances of the different configurations of the monolithic cell have been evaluated in real operating conditions via a dedicated 3-chamber set-up named Real Life Tester (RLT℗). Results show satisfactorily OCV and good performances for all samples, but a rapid degradation for the configuration with Ni foam probably due to aqueous electrochemical corrosion.

While the development of new solid oxide fuel cell (SOFC) electrodes must fundamentally occur within a lab setting, computational models which aid in the initial characterization and selection process can help reduce the overall financial cost and development time. A computational performance model, which both evaluates the structural properties and predicts the performance (current production) within electrode microstructures, generated from either experimental or numerical techniques, is presented. The application of the model to investigate the effects of different initial powder manufacturing parameters on the electrode performance is also demonstrated. 

The first part of this work presents the aforementioned performance model, which was developed in a modified version of OpenFOAM, MicroFOAM (1). The model begins by evaluating the total TPB length and the normalized effective species transport of all three phases within the electrode microstructure. The model then predicts the current production within the electrode by coupling the three percolating transport regions (electron, ion and pore) at the electrochemically active triple phase boundary (TPB), with a Butler-Volmer type expression. The identification of the microstructure properties and performance will enable relationships between these properties to be established. 

A particle-based numerical reconstruction model (2) is employed in this study to generate the synthetic electrode microstructures. The generated microstructures consist of a random distribution of overlapping spherical particles placed using a drop-and-roll algorithm. The packing algorithm allows for user specification of the initial starting parameters and is ideal for studying microstructure with a wide range of properties. Among the modifiable initial parameters include; the solid volume faction, porosity, and particle size distribution. 

In the second part of the study the performance model, developed in the first section, and the packing algorithm, mentioned above, are used to study the effects of different manufacturing parameters on the structural properties and performance of a cermet anode active layer. Initially the structural properties and performance of anodes generated with different initial electron-ion phase volume fractions (solid volume fraction) and porosities will individually be examined to identify the microstructure settings which maximizes the current production. Once identified, the two optimal individual settings will be combined and varied again to explore what optimal balance maximizes current production. Once identified, conclusions about the microstructure properties and manufacturing settings will be reviewed and discussed. Further manufacturing parameters including the particle size, particle size distribution, and thickness may also be explore in this work. 

References

• Choi, H.-W., Berson, A., Pharoah, J.G., Beale, S.B. Proc. IMechE. J. Power & Energy, 225(2): 183-197 (2011).

• Kenney, B., Vadmanis, M., Baker, C., Pharoah, J.G., Karen, K. J. Power Sources, 189: 1051-1059 (2009).

Solid-oxide fuel cells (SOFCs) have received significant attention due to their high efficiency, flexible fuel selection, low emissions in exhaust gases and relatively low cost. The SOFCs are also an alternative energy sources which operate at high temperature fuel cell technology [1-2]. Unlike lower temperature fuel cells, any carbon monoxide (CO) formed is transformed to carbon dioxide (CO₂) at the high operating temperature, and so hydrocarbon fuels can be used directly through internal reforming or even direct oxidation. Conventional SOFCs are also excellent devices for efficient power generation. However, they are facing various challenges to overcome high cost, durability problems related to materials degradation and particularly the interconnecting and sealing materials. Single chamber SOFCs and direct flame solid oxide fuel cells (DFFCs) [3-4] are also alternative SOFC concepts that do not face the sealing problem. But the potential explosion in a single chamber SOFCs could be dangerous as a fuel oxidant mixture is fed to the high temperature fuel cell, especially if operating conditions are not well-defined. This problem can be avoided if DFFC is used where the fuel and oxidant are mixed at the point of use in a flame removing the dangerous dead of explosive mix. Besides, the DFFC provides simple cell configuration, rapid start-up and no external heater required, and it is suitable for portable applications. But the performance of the DFFC is still relatively poor, which hinders its practical applications. Therefore, the studies of DFFCs are important for designing the DFFCs for practical applications and investigating the appropriate operating conditions as well as improving the cell performance.

In this study, a direct flame SOFC setup is designed and implemented based on fuel rich ethylene/air premixed flames to investigate the Ni-GDC (gadolinium-doped ceria) electrolyte supported fuel cell's performance. The diameter of Ni-doped ceria anode and LSCF (lanthanum strontium cobalt ferrite)-GDC cathode layer is 10 mm and GDC-electrolyte is 20 mm. A flat flame burner (64 mm outer diameter) along with stainless steel stabilisation plate was used for the investigations. The stabilisation plate was placed above the burner surface and is used to stabilise the flame. The fuel cell was attached to the stabilisation plate in such a way that the anode facing to flame front and the cathode exposed to ambient air. Silver paste and wires were applied as current collectors to both sides of the fuel cells. Fig. 1 shows the schematic of the DFFC experimental setup. I-V (current-voltage), open circuit voltage (OCV), cell temperature and electrochemical impedance of the fuel cell were investigated under different operating conditions.

The cell temperature was measured using a fine-wire R-type thermocouple against different ERs and distances between the burner surface and the fuel cell. The thermocouple was located 2 mm below the anode layer. The temperature was found are in the range of 628 -700 °C for the distance separations at 15 to 30 mm. It was observed that the temperature was decreased with increasing the distances. It can be explained by the fact that flame shape is largely distorted by the distance and more soot was formed in the inner flame. Relationship between OCV and cell temperature at various distances is shown in Fig. 2. Highest cell temperature and flame stability were observed for the ER of 1.52 at the distance separation 15mm. The fuel cell also exhibits an OCV of 0.84 V with the same condition. It was also found that the position of the fuel cell with respect to the flame have a significant effect on the cell temperature and the performance. The fuel cell achieved a peak power density of 41 mW/cm² and a peak current density of 121 mA/cm²with ER at 1.52 and temperature at 700 °C. In near future different hydrocarbon fuels (e.g., methane and propane) and fuel cells can be used for the further investigations over a range of operating conditions.

References

[1]   P. I. Cowin, C. T. G. Petit, R. Lan, J. T. S. Irvine, and S. Tao,  Adv. Energy Mater., 1,314–332 (2011) .

[2]   S. Sengodan, S. Choi, A. Jun, T. H. Shin, Y.-W. Ju, H. Y. Jeong, J. Shin, J. T. S. Irvine, and G. Kim, Nat. Mater., 14, 205-209 (2015).

[3]   Y. Q. Wang, Y. X. Shi, X. K. Yu, N. S. Cai, and S. Q. Li,  J. Electrochem. Soc.,160, F1241-F1244 (2013).

[4]   M. Horiuchi, S. Suganuma, and M. Watanabe, J. Electrochem. Soc.,151, A1402 (2004).

Acknowledgement: The authors thank EPSRC SuperGen Hydrogen Fuel Cells Challenges Flame SOFC Project (Grant No EP/K021036/1) for funding.

Figure 1

An anode-supported Solid Oxide Fuel Cell (ASC) consists of a mechanically supporting, thick, and highly porous anode substrate, coated on top with thin films of a fine-structured porous anode, a dense electrolyte, and a fine-structured cathode. Both, the anode substrate and the anode are made of a cermet of nickel and yttria-stabilized zirconia (YSZ). An increase of the oxygen partial pressure at the anode side will lead to a reoxidation of the nickel and to microstructural changes. This may result in a macroscopic expansion of the anode substrate and in a complete cell failure, due to cracking of the YSZ electrolyte.

In this work, a redox-stable SOFC with a thin-film sol-gel electrolyte is presented. Using tailored suspensions and polymeric sols of varying target particle size, a 1 µm thick, three layer sol-gel electrolyte is coated on top of a tape cast anode substrate and a screen printed anode by spin-coating and dip-coating techniques. With this sol-gel electrolyte a power output larger than 1.25 W/cm² at 0.7 V and an operating temperature of 600 °C could be demonstrated. In this study, half cells (anode substrate, anode, and electrolyte) are reoxidized in excess air, in order to test the redox stability of these SOFCs. Afterwards the electrolyte is investigated ex situ using optical microscopy and SEM. No cracks are found in the sol-gel electrolyte after reoxidation for 4 hours at 600 °C and 15 minutes at 800 °C, respectively, whereas a lot of cracks are visible in a standard ASC with a 10 µm thick, screen printed electrolyte for both reoxidation conditions. This means the sol-gel electrolyte can tolerate higher degrees of oxidation (DoO) of the anode substrate than the screen printed electrolyte. The residual compressive stresses of both electrolytes are measured to about 600 MPa using XRD and are therefore not responsible for the higher stability against reoxidation of the sol-gel electrolyte. In contrast, the energy release rate has to be larger than a critical value for the creation of cracks in a thin-film on a thick substrate. Due to the fact, that the energy release rate is proportional to the thickness of the thin-film, a thinner film is more stable against cracking than a thicker film at constant tensile stresses. After reoxidation of 8 hours at 600 °C and 60 minutes at 800 °C, respectively, cracks are also found in the sol-gel electrolyte. However, the SOFC with the thin-film sol-gel electrolyte can be considered as stable against reoxidation, because the long reoxidation time of 4 hours at an operating temperature of 600 °C is unlikely to happen under real conditions.

Full ceramic anodes have been developed over the last decade as an alternative to the state-of-the-art Ni/YSZ cermet. The objective is to overcome the technical limitations of the cermet, especially in high fuel utilization operational mode, when hydrocarbon fuels are used, or in systems where rapid heat up is required. Numerous oxide compositions were proposed and investigated. (La,Sr)Cr_0.5Mn_0.5O₃(LSCM) perovskite has been reported as a very promising anode material because of its high tolerance to carbon and sulfur, and its redox and thermal cycling stability. However, similarly to other oxide anode materials, its overall performance still needed further improvement to match that of Ni/YSZ. The relatively low performance of LSCM compared to Ni/YSZ, used to be attributed to its relatively low electronic conductivity. Recently, we have reported that the pre-coating LSCM with Ni nitrate results in a considerable decrease in the anodic polarization resistance and activation energy. Ni dissolves in the perovskite phase at high temperature under oxidizing atmosphere, and then exsolves in the form of nanoparticles under reducing atmospheres.

This type of anode composition with split functionalities (ionic conduction, electronic conduction, and catalytic activity) gives a unique opportunity to investigate the specific influence of each of the above functions on the anodic mechanisms and performance. In this work, electrochemical impedance spectroscopy was used to characterize anode compositions containing LSCM and different levels (15, 40, and 60 wt%) of gadolinia doped ceria (CGO), with and without additional submicron Ni, as well as Ni nanoparticles exsoluted from pre-coated larger LSCM particles. Impedance measurements were performed in a three electrode – half cell configuration from 700 to 900°C under two flow rates of 3% wet H₂(50 and 150 ml/min).

At 900°C and under 150 mL/min of 3% wet H₂, lower values of the polarization resistance (R_p) were measured for the samples containing CGO compared to pure LSCM ones. The R_p showed a slight decrease when the content of CGO was increased (15 to 40 and 60 wt%). This was characterized by a decrease of the high frequency part of the impedance diagrams, likely due to the higher ionic conductivity brought by the addition of CGO. The addition of 5 wt% of submicron Ni to LSCM-CGO led to a remarkable decrease in R_p. This effect was even more spectacular for the samples containing 5 wt% of exsoluted Ni nanoparticles, characterized by a decrease of the whole impedance diagram, and especially the high and the middle frequency parts, which suggests an enhancement of the electrochemical processes related to anode charge transfer. Further analysis of the impedance diagrams, measured for the above anodes at different temperatures and gas flow rates, is in progress in order to define the effect of different anode functionalities on the anodic processes involved, and thus its overall performance.

Electrode microstructure plays an important role in determining the performance and durability of solid oxide fuel cells (SOFCs). It needs to be tailored towards a variety of requirement for the electrodes, such as transport properties and electrochemical activity. Long-term stability of the microstructure is also important to guarantee the durability of SOFC systems. In order to satisfy the growing demand for their performance and durability, an improved understanding of the microstructure-performance relationships is desired, along with optimization of the microstructure, through a design-led approach.

The focus of this study is on electrodes fabricated with the nano-particle infiltration techniques. Porous framework structures (scaffolds) were first fabricated with gadolinium-doped ceria (GDC), and then nickel metal particles were introduced into the structures in the form of nano particles. 3D microstructural analysis using focused ion beam and scanning electron microscope (FIB-SEM) was carried out and revealed that the infiltration technique has a greater potential, compared with the conventional approach based on powder mixing and sintering, to control the electrode microstructure, helping satisfy multiple requirements for the electrodes. Most notably, the triple-phase boundary (TPB) density in the infiltrated electrodes was found to be one order of magnitude larger than that typically found in the conventional electrodes.

A 1D numerical model for Ni-GDC electrodes was also developed assuming that the GDC phase has mixed ionic and electronic conductivity and hence that the electrochemical reaction takes place on the GDC-pore double-phase boundary (DPB), and successfully reproduced the overpotential characteristics obtained from the experiment. Sensitivity analysis was also conducted with the developed model at 700 °C to investigate the effect of electrode microstructure on electrode performance. This revealed that the electrochemical reaction on the DPB is the rate-determining process within the electrodes; therefore increasing the DPB density is recommended as the most effective route to improving performance of ceria-based electrodes, rather than improving species transport rate.

Using the insights from the experiment, microstructural analysis and numerical simulation, guidelines for further optimizing the electrode microstructure are proposed

Figure 1

Biogas is a potentially widely available fuel derived from the anaerobic digestion of biological materials such as animal waste. The nature of this fuel source predicates its availability to areas of a more rural nature and as such can be a valuable resource in remote areas where supply of other energy vectors (such as an electricity grid) can create significant logistical issues. Where Biogas is currently utilised it is often burnt in a reciprocating engine to drive a generator for the supply of electricity. This is a low efficiency method of energy conversion and much of the energy content of the fuel is lost in the process. A better method would be direct electrochemical conversion, taking the higher conversion efficiencies offered by solid oxide fuel cells to improve the utilisation of biogas and provide an efficient source of electricity for remote or off grid locations.

Biogas is mainly a mixture of carbon dioxide and methane, however the ratio of the two gases in many cases is not enough to prevent carbon formation if used directly in a nickel based cermet anode. This is compounded by the presence of hydrogen sulphide as a significant impurity in biogas, which is of course well known to poison SOFC Ni cermets. To look to tackle these issues University of St Andrews has been part of a collaborative UK-India research programme jointly funded by EPSRC (UK ) and DST (India) on improving bigas utilisation in SOFC, other partners are Imperial College London and University of Strathclyde from the UK and CGCRI (Kolkata) and IMMT (Bhubaneswar) from India. In this paper some of the recent activities at St Andrews to improve anode robustness through catalyst impregnation and exsloution are discussed.

Two approaches are described here, in the case of the impregnation, a proton conducting oxide, BaZr_0.1Ce_0.7Y_0.1Yb_0.1O_3-δ (BCZYYb) was impregnated into the Ni-YSZ anode of a thin electrolyte, anode supported cell at various levels up to 1.6wt%. For the exsloution based structures, both Sr and Ca doped lanthanum titanate perovskites were doped with various catalyst materials such as Ni, Ce, Mg, and Rh. These perovskites were A-site deficient so that on reduction the catalyst materials were exsolved onto the surface as fine metallic particles with a typical size in the 10'snm. All anodes were tested on a standard reformed biogas mixture developed as part of the project. This represented the product  of a 63:37 methane:CO₂  input biogas exposed to 25% recirculation of an 80% utilised fuel to result in a gas mixture of 36% CH₄, 26% CO₂, 20% H₂O, 4% H₂ and 4% CO. For impurity testing H₂S was added to this mixture at levels of 4-10ppm. Catalyst function was assessed by both cell testing and reforming activity.

For the impregnated specimens, microstructures revealed coarser, more concentrated distribution of impregnated particles on the surface of the anode as the level of impregnation increased. All of the BCZYYb impregnations improved cell performance, however the optimum performance was at the lowest level of impregnation (0.6wt%)  (<1.5Acm^-2 at 0.8V at 800°C) with performance dropping towards the non impregnated value (around 0.6 Acm^-2 at 0.8V at 800°C) as impregnation levels increased to 1.6wt%.  Short term steady state durability was good with initial performances proving stable over the first 50 hours of operation at 1.25 Acm^-2 at 0.8V at 750°C. However poisoning effects were observed on the introduction of H₂S with a performance drop of the order of 60-70% over 20 hours of exposure. Recovery was observed on return to a non-poisoned biogas mixture, however perfomance was still significantly lower than initial levels. Changing to steam/hydrogen mixtures helped bring performances back to initial levels.

The doped perovskites showed various levels of metallic particle exsolution depending on the dopant type and level.  These have been tested for reforming activity, with all showing activity for reformation of the standard biogas mixture, however they also all exhibited rapid degradation on introduction of H₂S. Only the Rh catalyst retained some catalytic activity in the presence of hydrogen suphide, although all specimens showed recovery when the this was removed from the gas stream.  A comparison was carried out between exsolved and impregnated nickel, with the former showing greater resistance to initial particle coarsening and carbon deposition. This demonstrates that it is not just the catalyst material itself which is important but also the nature and morphology of the catalyst particle.  These results show that both approaches continue to be of great interest in the continued development of robust anodes for challenging fuel environments.

To make solid-oxide fuel cell (SOFC) systems more manufacturable and reduce system costs, SOFC developers, wherever possible, have substituted lower cost stainless steel into the stack design, including replacing ceramic interconnects within the stack and high-cost nickel-based superalloys in balance-of-plant components. However, for successful implementation of these steels, protective coatings are necessary to protect the air-facing metal surfaces from high temperature oxidation and to minimize chromium volatilization from the metal, because chromium volatiles poison the cathode and degrade cell performance.

NexTech Materials has developed a cost-effective coating technology to improve the high temperature performance of stainless steel components. Coatings include an electrically conductive, manganese cobaltite (MCO) coating tailored for the cathode active-area of metallic interconnects and a complementary diffusion-based aluminide coating for the non-active, seal-area of interconnects and also readily amenable to balance of plant components.

In this paper, the suitability of this coating technology for SOFC applications will be demonstrated. Microstructural analysis in combination with long-term electrical area-specific resistance (ASR), oxidation and chromium volatility testing will illustrate the stability and functionality of the coatings under application specific conditions. Stack-level testing will demonstrate the successful scale-up of the technology.

Bipolar plates in solid oxide fuel cells operate under a significant oxygen potential gradient (0.2 to 10^-18 bar pO₂). For air side conditions (pO₂ = 0.2 bar), a PVD technique has successfully been used to apply combination reactive element/Co coatings to inhibit the rapid oxidation of and Cr volatilization from Fe-Cr alloys such as Crofer22 APU and Sanergy HT. However, relatively little work has been done on developing similar coating strategies for fuel side conditions.

In this work, the ferritic steel Sanergy HT was coated with single element (Ce,La, Co and Cu) and duplex (Ce/Co, La/Co) coatings of varying thicknesses and exposed at 850°C in Ar-5%H₂ charged with 3% H₂O in a tubular furnace up to 500 h. Additionally, the effects of a 'sealing' step on corrosion mechanisms were also investigated by pre-oxidising the coated steel in air, prior to exposure in the fuel side environment. Chemical analysis on the samples was subsequently performed with SEM/EDX and XRD.

 It was established that:

(i)                 reactive element (Ce and La) coatings brought about a 2-4x reduction in oxidation rate

(ii)                Co and Cu coatings negatively affected oxidation performance due to their instability in fuel side conditions.

Figure 1

Manganese cobalt spinel coatings have attracted much interest as a protective layer for stainless steel interconnects of solid oxide fuel cells (SOFCs) owing to their effectiveness of suppressing high temperature oxidation and Cr evaporation. In this study, a simple wet powder spraying (WPS) method was applied to prepare spinel layers on commercial stainless steels (Sanergy HT and K41/441) which were evaluated as protective coating of SOFC interconnects. A condition of heat treatment on spray coatings was optimised to achieve dense protective layers with microstructure analysis. Area specific resistance (ASR) and stability were measured in a humid atmosphere at 700 °C for 1000 hours. Post-test analysis was conducted on microstructure of measured samples using SEM-EDS. In addition, effect of a sublayer with spinel coatings was investigated on ASR and Cr retention. The implication for microstructure, ASR and stability of the evaluated coatings is discussed.

Osaka Gas Co., Ltd. has been developing electrodeposition painting method for SOFC metal interconnector coating since 2010. Although the general electrodeposition painting was mainly consisted of resin with small amount of ceramics as a pigment, we have successfully increased the amount of ceramics in the coating and this method was adopted to SOFC metal interconnector coating. Generally, SOFC interconnector had the complex structure with the gas flow path. This method had the advantage of coating thickness uniformity regardless of the interconnector structure and mass productivity.

We have investigated to enhance the durability of the SOFC interconnector with this method.In addition, we aimed to reduce the cost of the metal interconnector by using the normal stainless steel, not stainless steel specially designed for the SOFC component.

The composition of Co-Mn spinel-type oxide widely used in this field was selected as the coating ceramics.

As is often the case with interconnector, the protective coating needed to be partially coated. The wet coating method like electrodeposition painting needed to be fired after coating to remove the resin and sinter the spinel-type oxide.

From the standpoint of processing cost, it was desirable to be fired in air, but non coated area was severely oxidized and Cr2O3 layer showing high ASR was formed on the surface of the stainless steel.

In order to prevent the oxidation of non-coated area and sinter the spinel powder enough, two steps of firing have been normally taken: the resin was removed in air first, then the Co-Mn spinel powder was sintered in the reduction atmosphere. This process cost tends to be high due to changing atmosphere.

However, our electrodeposition painting method did not need to be fired in two steps because its resin was acrylate resin which was removed completely even in the reduction atmosphere.

Cross section images of the post anneal coating showed the dense structure and the uniform layer thickness.

ASR test over 10,000hr showed excellent durability at the both of the continuous condition and thermal cycle condition.

The Cr diffusion from the stainless steel was also evaluated, it showed the high prevention of Cr diffusion.

The variability of coating weight were quite small, it was concluded that this electrodeposition painting method is suitable for mass production.

Chromium poisoning is a widely recognized degradation mechanism in solid oxide fuel cells (SOFC). Stainless steel interconnects (IC), in a direct contact with the cathode, have been identified as the main chromium source. Chromium evaporation from the bulk steel may also cause break-away oxidation of the IC plate, which is equally deleterious for the stack. Thus, protective coatings are needed to prevent chromium from migrating from the steel. A low electrical resistivity and high chemical and physical stability are required of the coatings. This work compares manganese-cobalt and cobalt-cerium protective coatings. The evaluated coatings are fabricated on thin steel samples by commercial companies and research facilities with differing manufacturing methods. The chosen steels are Sandvik Sanergy HT and K41/441 as they are common in SOFC applications. Area specific resistance (ASR) and overall stability were determined with an innovative measurement setup. The steel samples were stacked in a sandwich structure adjacent to thin palladium foils with a screen-printed lanthanum strontium cobalt (LSC) layer. This setup offers two advantages. First, a realistic electrical contact with the cathode material is obtained since the LSC layer is manufactured with the same methods as real SOFC cathodes. Second, the adoption of palladium spacers instead of steel enables electron microscopy analysis on chromium migration into the LSC layer as well as on oxide scale growth. ASR measurements were conducted in a humid atmosphere at 700 °C for a period of 1000 hours. The paper compares the coating solutions in terms of ASR and stability and discusses their usability in SOFC applications.

Sandvik Materials Technology develops nanocoatings for SOFC stainless steel interconnects. The self healing properties of cerium cobalt coated AISI441 (EN 1.4509) have been studied. A selectively coated sample was oxidized in air at 800°C for a total of 504 hours. A FIB trench, milled on the uncoated surface, was used as a reference point for the coated border. A second FIB trench, made after oxidation, revealed that the oxide formed on the uncoated part adjacent to the coated area contained both cobalt and cerium. Cerium was detected with EDS in the surface oxide of the uncoated area as far as 10 mm away from the coated edge, cobalt had diffused even further. These findings are encouraging since they suggests that cracks introduce in the coating during stamping and/or forming, which are usually in the 1-3 mm range, are likely to self heal (i.e. become enclosed by an upper Co, Mn, Cr-spinel)  upon oxidation.

Figure 1

Cr vaporization and oxide scale growth are probably the two most detrimental degradation mechanisms associated to Cr₂O₃-forming alloys as interconnect material in SOFC. Different type of perovskite or spinel coatings are commonly used to mitigate Cr vaporization. Furthermore a number of studies have suggested that reactive element coatings (Ce, La, Hf) can reduce the oxidation rate of Cr₂O₃-forming alloys. However, the use of coatings increases cost due to addition processing and handling costs.

The present study investigates the effectiveness of thin Ce and Co Physical Vapor Deposited (PVD) coatings with respect to Cr vaporization, corrosion resistance and Area Specific Resistance (ASR). The coatings were applied in a continuous roll-to-roll process which allows for high volumes.

The use of pre-coated interconnects is thus more economic than a batch process. However, a consequence of the pre-coated concept is that the coating is subject to deformation when the pre-coated steel is pressed into the interconnect shape.

Therefore this study also investigates the effect of deformation on the coating. Coated steel strips were bi- and uniaxially deformed. Furthermore a commercially used  interconnect shape is investigated and the effects on Cr evaporation and corrosion are studied: Cr vaporization was measured in an air-3% H₂O environment at 850 °C for 336 h. The results revealed that when the pre-coated steel was deformed, large cracks were formed. However, upon exposure those cracks did heal forming a continuous surface oxide rich in Co and Mn. As an effect of the rapid healing, no increase in Cr vaporization was measured for the pre-coated material.

Precoated AISI441 for SOFC interconnectors were investigated by STEM-EELS looking at the diffusion processes in the initial stage of the oxidation. The PVD coating contained Ce and Co/Mn-alloy in two layers with at total thickness of 800nm. The samples were exposed up to 1000 hours at 800°C then characterized by TEM. Multiple linear least squares (MLLS) fitting of reference spectra aids the interpretation of the elemental maps due to overlapping edges (Figure 1). Within 4 min heat treatment the oxidation front has reached the substrate steel and Cr₂O₃ is forming at the stainless steel interface. CeO₂ acts as the barrier line for Cr and is situated between the Cr-containing spinels and the (Co,Mn)-spinel after 72 h heat treatment.

Figure 1

Precoated AISI441 for SOFC interconnectors were investigated by STEM-EELS looking at the diffusion processes in the initial stage of the oxidation. The PVD coating contained Ce and Co/Mn-alloy in two layers with at total thickness of 800nm. The samples were exposed up to 1000 hours at 800°C then characterized by TEM. Trivariate analysis of multiple linear least squares (MLLS) maps show that Cr₂O₃, (Co, Cr)-spinel, (Co, Cr, Mn).spinel and (Co,Mn)-spinel phases can be identified (Figure 1). Oxidation states are characterized from white line ratios of the present transition metals in these phases and can be correlated to changes in the O K-edge fine structure.

Figure 1

Interconnect plates, made of ferritic type stainless steel, are widely used in planar solid oxide fuel cell (SOFC) or electrolysis cell (SOEC) stacks. These interconnect plates serve as current collector and separator for the neighboring fuel and oxygen electrode compartments of two adjacent cells. An intimate contact between the cell component and the interconnect (IC) plate is therefore essential to ensure optimum cell and stack performance. During stack production and operation, inter-diffusion across the cell – IC interface takes place, which under certain circumstances introduces adverse effects on the electrical, mechanical, and corrosion properties of the IC plates. One representative example is diffusion of nickel from the Ni/YSZ fuel electrode or from the Ni contact component into the IC plate, while iron and chromium from the steel diffuse in the reverse direction. Diffusion of Ni into the steel causes transformation of the ferritic BCC phase into the austenitic FCC phase in the interface region, accompanied with changes in volume and in mechanical and corrosion properties of the IC plates. Very few studies have been devoted to investigate this process experimentally. In this work, kinetic modeling of the inter-diffusion between Ni and FeCr based ferritic stainless steel was conducted, using the CALPHAD (CALculation of PHAse Diagrams) approach with the DICTRA software. The kinetics of inter-diffusion and austenite formation was explored in full detail, as functions of steel composition, thickness of Ni contact component or IC plate, temperature and time. The simulation was further validated by comparing with previously obtained experimental results. Growth of the austenite phase in commercial interconnect materials such as Crofer 22 APU and Crofer 22H was predicted under practical SOFC and SOEC operation conditions. The present work provides a proper account of the thermodynamics and kinetics of Ni-IC inter-diffusion and in addition provides input to further analysis of associated changes in the mechanical and corrosion properties of the IC plates.

In recent years, solid oxide fuel cells (SOFCs) have been extensively studied because of its potential as clean and efficient electric generators. Recently, SOFCs coupled with an external reforming have been suggested for the utilization in marine on-board applications. Diesel marine is the conventional fuel actually used; however, the stability of a conventional SOFC fed with a diesel reformate is not yet proven. Furthermore, the testing of a large area SOFC has a specific importance with the aim of the subsequent scaling up to a stack.

This study deals with the investigation of the degradation effects occurring onto the anode of a large area cell in presence of n-dodecane reformate. A commercial anode supported SOFC (10 × 10 cm²) with planar configuration having the composition Ni-YSZ/YSZ/YDC/LSFC, has been investigated in-situ electrochemically. The test bench used for this experiment is the conventional ECN test station showed in Figure 1. The electrochemical experiments concerned I-V curves (Figure 2) and impedance spectroscopy (Figure 3) carried out at OCV and at 0.9 V. Ex situ tests were also carried out in order to evaluate morphological and and physico-chemical modification on the discharged cell.

In the case of anodes (Ni-YSZ cermets), the degradation phenomena are recognized due to Ni agglomeration and carbon deposition. These microstructural changes in anode is expected to be highly dependent on operating conditions, such as operating temperature, operating voltage and water content in the feed.

Figure 1: ECN test bench (a) and commercial large area cell (b)

Figure 1

Solid oxide fuel cells (SOFCs) are one of the promising power generation devices due to their high energy conversion efficiency and low emissions. Hydrogen fuel can be used as an environmentally clean energy carrier for SOFCs. However, the storage and transportation are major obstacles for the large-scale application because of the low volumetric density and boiling point. Thus, high fuel flexibility of SOFCs has attracted much attention for a simplified power generation system.

Recently, ammonia was considered as a promising fuel for SOFCs because of following reasons; high energy density, ease in liquefaction (–33.4ºC at atmospheric pressure or 8.46 atm at 20ºC), no greenhouse gas emission, narrower flammable range than hydrogen. Much effort has been devoted to enhance the performance of direct ammonia fuelled SOFCs employing oxide-ion [1, 2] or proton [3, 4] conducting electrolytes. Meng et al. [2] fabricated a 10-μm thick SDC electrolyte with 50wt% Ni–SDC as an anode and BSCF as a cathode. A peak power density of 1190 mW/cm² was achieved at 700 °C, which was the highest performance reported in literatures for the direct NH₃-SOFC using oxygen ion conducting electrolyte.

In this report, the mechanism of electro-oxidation of ammonia over Ni-based cermet anode was studied. Especially, we focused on the correlation between the catalytic activity of anode for ammonia decomposition and cell performance. This systematic investigation has revealed that ammonia is catalytically decomposed to H₂ and N₂, and then H₂ produced is electrochemically oxidized over anode. Furthermore, the operating temperature and hydrogen concentration had a great impact on ammonia decomposition as well as cell performance. These results provided the strategy for the development of anode materials for direct ammonia-fueled SOFCs.

[1] Qianli Ma, Jianjun Ma, Sa Zhou, Ruiqiang Yan, Jianfeng Gao, Guangyao Meng, Journal of Power

Sources 164 (2007) 86–89.

[2] Guangyao Meng, Cairong Jiang, Jianjun Ma, Qianli Ma, Xingqin Liu, Journal of Power Sources

173(2007) 189–193.

[3] N. Maffeia, L. Pelletierb, J.P. Charlandb, A. McFarlanb, Journal of Power Sources 140 (2005)

264–267.

[4] N. Maffei, L. Pelletier, J.P. Charland , A. McFarlan, Journal of Power Sources 162(2006) 165–167.

We have so far developed Ni-Fe and Ni-Mo anodes with high catalytic activity for ammonia(NH₃)-fueled SOFCs [1], and have proposed that the performance of NH₃ -fueled SOFCs was determined by a balance between NH₃ adsorption (Fe, Mo) and nitrogen(N₂) desorption (Ni) capability [2]. Furthermore, we also found that both Ni-Fe and Ni-Mo anodes react with NH₃ to form metal alloy nitrides (FeN_x, γ-Ni₂Fe₂N and Ni_0.2Mo_0.8N) and metal Ni under the operation condition, while pure Ni anode does not form nitrides. These results suggested that the nitrides enhance NH₃ adsorption at the anode and led to the high anode activity for NH₃ oxidation.

In the present study, we prepared three kinds of anodes, Ni-W, Ni-Ta and Ni-Nb and evaluated their performance as anode in NH₃-fueled SOFCs to investigate the correlation between the nitride formation and the anode activity. These anodes were synthesized by impregnating W, Ta and Nb precursor solutions into a pre-sintered Ni-SDC anode cermet. The addition of W and Ta enhanced the anode activity for NH₃ oxidation, but Nb did not.  XRD measurements of metal oxides (WO₃, Ta₂O₅ and Nb₂O₅) after annealing in NH₃ indicated that W and Ta in the anode reacted with NH₃ to form nitrides, but Nb did not under the operation condition (i.e. at 973-1173 K in NH₃). These results suggest that the easiness of the nitride formation is closely related to the high activity anode for NH₃ oxidation.

Acknowledge

This work was supported by the Kyoto Environmental Nanotechnology Cluster from MEXT, Japan.

Reference

[1] W. Akimoto et al., Solid State Ionics, 256, 1–4 (2014).

[2] W. Akimoto et al., ECS Trans., 57 (1), 1639-1645 (2013).

Long term durability was demonstrated on high-performance direct ethanol solid oxide fuel cell operating in gradual internal reforming. Anode supported single cells, fabricated at Forschungszentrum Jülich, composed of (La,Sr)MnO₃ cathode, 8 mol% yttria-stabilized zirconia (YSZ) electrolyte, and YSZ-Ni anode were used for long term testing on anhydrous (direct) ethanol. Gadolinia-doped ceria (CGO), with 0.1 wt.% of Ir, catalyst was produced by the impregnation of Ir salt in commercial CGO (Praxair) and mixed with organic additives. The CGO-Ir suspension was deposited onto the surface of the YSZ-Ni anode support with an air brush. Such a catalytic layer acts both as a physical barrier to avoid direct contact of the fuel with Ni-based anode and to promote the gradual internal reforming of ethanol. The CGO-Ir catalyst was designed for the reforming of the main compounds resulting from the thermal decomposition of ethanol at high temperature and to inhibit the formation of ethylene, which is a known source of carbon deposits. After deposition the sample was heated at 850ºC in inert atmosphere to ensure good adhesion between the catalytic layer and the anode. Gold grid and paint were used as current collectors in both cathode and anode sides. The single fuel cells were mounted in the test apparatus, sealed with gold rings and alumina cement and heated to the measuring temperature (850°C) under inert atmosphere. After gas tightness was checked by flowing air on the cathode side and argon on the anode side, the anode was reduced under hydrogen. The fuel cell was kept at 0.6 V while monitoring the current drained from the cell at 850 °C. Samples were initially operated in hydrogen for a few hours and fuel was switched to dry ethanol, carried by nitrogen. It is important to mention that no oxidizing agent was added to the fuel and that water produced by the electrochemical reaction of hydrogen at the anode promoted the reforming of the ethanol in the CGO-based layer, in the so-called gradual internal reforming. In such conditions the current output was recorded and observed to not be significantly different form that under hydrogen. Moreover, the direct ethanol fuel cell was continuously operated for 650 hours without major losses due to carbon formation. The experimental results demonstrate that the water released by the electrochemical oxidation of reformate hydrogen at the electrolyte/anode active interfaces is sufficient for the reforming of the primary fuel in the catalytic layer, provided that a minimum fuel utilization is respected. After the durability tests, energy dispersive X-ray spectroscopy and scanning electron microscopy analyses of the samples revealed no carbon deposit formation in the anode. The experimental results provided compelling evidence for the viability of gradual internal reforming of ethanol in high performance solid oxide fuel cells.

This paper presents an experimental study of the impact of toluene on the performance of anode-supported (AS) Ni-YSZ SOFC operating at ~800 °C, 0.25 A/cm² fed with H₂ and reformed biosyngas. Toluene was added to the fuel stream and its concentration increased stepwise from tens to thousands of ppm. Each poisoning test was followed by a recovery step (no toluene in the fuel stream) for the duration of 25 hours. The main goal of this work is to define a concentration threshold of toluene as a model tar compound in the fuel stream and later to identify the degradation mechanism caused by this tar compound. Polarization behavior, current-voltage and electrochemical impedance spectroscopy (EIS) were analyzed to evaluate the cell performance. A linear degradation is observed at concentrations above 1000 ppmv for the cell which was fed with pure H₂ (cf Figure) The degradation was due to local C(s) deposition. In contrast, toluene even up to 3500 ppm did not cause a severe degradation when the cell was fed with reformed biosyngas.

Figure 1

Butane steam reforming with two different types of nickel magnesia catalysts is investigated. The catalytic and electrocatalytic properties of nickel magnesia with YSZ and GDC supports made using different calcinations temperatures; 673K, 873K, 973K, and 1073K are determined. A palladium doped catalyst sample is used to examine the effects of metals on the performance. Moreover, S/C ratio is also controlled from 4 to 6 to elucidate the mechanism of butane steam reforming. The effect of microstructural change and properties are observed using X-ray diffraction (XRD) and Scanning Electron Microscopy (SEM) to characterize the differences in reforming performance.

A small and high-performance solid oxide fuel cell (SOFC) system is promising as a power source for many potential mobile robot applications. To realize the small SOFC system, the direct use of liquid hydrocarbon fuel, which has high energy density and is easy to storage and carry, is desirable. The optimization of type of fuel, operating conditions, and a fuel electrode (anode electrode) is essential. In the view point of the use at high temperatures in summer outdoor, octane having high boiling temperature is a good candidate as the fuel. Steam reforming is the first choice for the internal reforming method because it has been frequently-studied for the direct use of methane fuel. A porous cermet of Ni and yttria-stabilized zirconia (YSZ), which has widely been studied as an anode electrode for the internal reforming operation SOFC, is one of promising anode electrode materials, however, performance degradation due to sulfur poisoning and carbon deposition on the anode electrodes surface is concerned. A porous cermet of Cu and gadolinium oxide-doped ceria (Cu-GDC), which, respectively, have excellent sulfur tolerance and oxygen ion supply property, has high potential for the anode electrode of SOFC operated by internal reforming of methane. Therefore, in this study, the Cu-GDC cermet was focused as an alternative anode electrode for the octane-direct use SOFC. The aim of this study is to clarify the effects of operating conditions on the reforming reaction and electrochemical reaction on the Cu-GDC cermet anode electrode.

In this study, the fuel-cracking reaction, which occurs at elevated temperature in the presence of excess steam, was estimated. In order to understand the effect of difference in branching of carbon chain, n-octane (linear alkane) and iso-octane (branched chain alkane) were used. Octane and steam were supplied to the SOFC cell by bubbling carrier gas (He or Ar) at a constant rate. Temperature of the octane bubbler and the carrier gas flow rate was 30 ^oC and 10 ml/min, respectively. The carrier gas flow rate of the water bubbler was 83 ml/min. Steam/carbon ratio (S/C) was controlled over a S/C ratio from 0.5 to 5.5 by changing the temperature of the water bubbler from 30 to 80 ^oC. The cell temperature range was 700-850 ^oC. I-V and I-P characteristics, and stable power generation property by constant current were investigated to demonstrate the octane direct internal reforming operation of SOFC. The effects of S/C ratio and cell temperature on the open circuit voltage (OCV) and I-V and I-P characteristics were investigated. In order to identify the chemicals generated by the fuel cracking reaction, outlet gas compositions from the cell of open circuit condition and power generation condition were analyzed by gas chromatography.

The following are the main findings that we have revealed in the present study. By raising the S/C ratio and/or the cell temperature, the reforming reaction is facilitated, but fluctuation of the OCV is increased. Chemical species and their ratio generated by the cracking reaction are changed by the difference in the carbon-chain of octane. Mainly H₂, CO, CO₂, CH₄, and C₂H₆ are generated in the reforming reaction of n-octane, and C₃H₈, CH₃OH, and C₄H₁₀ are also generated in addition to those in the reforming reaction of iso-octane. However, the main electrode reaction to generate electrical power is only the electrochemical oxidation of H₂ and/or CO. By using the n-octane fuel, the carbon deposition is reduced. The rate-determining process is dissociation to H and CO.