Solid oxide cell R&D at Risø National Laboratory—and its transfer to technology

Assessment of co-sintering as a fabrication approach for metal-supported proton-conducting solid oxide cells

Proton conducting oxide electrolyte materials could potentially lower the operating temperature of metal-supported solid oxide cells (MS-SOCs) to the intermediate range 400 to 600 °C. The porous metal substrate provides the advantages of MS-SOCs such as high thermal and redox cycling tolerance, low-cost of structural materials, and mechanical ruggedness. The cell structure for fuel cell and electrolysis operation is shown in Figure 1. To incorporate proton-conductors into metal-supported cell configuration through co-sintering fabrication, a key factor is that the proton conductors must be compatible with reducing atmosphere sintering, as the porous metal substrate can not be fired in oxidizing atmosphere. In this work, viability of co-sintering fabrication of metal-supported proton conducting solid oxide cells is investigated. Candidate proton-conducting ceramics including perovskite-type oxides BaZr0.7Ce0.2Y0.1O3-δ (BZCY), SrZr0.5Ce0.4Y0.1O3-δ (SZCY), and Ba3Ca1.18Nb1.82O9-δ (BCN), pyrochlore-type oxides La1.95Ca0.05Zr2O7-δ (LCZ) and La2Ce2O7 (LCO), and acceptor doped rare-earth ortho-niobates La0.99Ca0.01NbO4 (LCN) were synthesized via solid state reactive synthesis or sol-gel synthesis. These ceramics are sintered at 1450°C in reducing environment alone and supported on Fe-Cr alloy metal support, and their key characteristics such as phase formation, sintering property, and chemical compatibility with metal support are determined. Most electrolyte candidates suffer from one or more challenges identified for this fabrication approach, including: phase decomposition in reducing atmosphere, evaporation of electrolyte constituents, contamination of the electrolyte with Si and Cr from the metal support, and incomplete electrolyte sintering. BZCY is particularly interesting because of its high proton conductivity, but it is found to suffer from reaction with Si and Cr present in the metal support, and loss of Ba via evaporation during sintering. Various approaches to overcome these limitations are proposed, and preliminary assessment indicates that the use of barrier layers, low-Si-content stainless steel, and sintering aids warrant further development. In contrast, La0.99Ca0.01NbO4 is found to be highly compatible with the metal support and co-sintering processing in reducing atmosphere, allowing dense electrolyte and porous electrode made of LCN to be co-sintered on porous stainless steel, Figure 2. A metal-supported cell is fabricated with La0.99Ca0.01NbO4 electrolyte, ferritic stainless steel support, Pt air electrode and nanoparticulate ceria-Ni hydrogen electrocatalyst. The total resistance is 50 Ω∙cm2 at 600 °C. This work clearly demonstrates the challenges, opportunities, and breakthrough of metal-supported proton-conducting solid oxide cells by co-sintering fabrication. Figure 1

The present paper refers to an innovative SOFC concept named EVOLVE after a running FP7 project which integrates advanced materials and an inventive anode current collector made from a NiCrAl preoxidized foam impregnated with a conductive perovskite. Even if the concept is the beneficial combination of Metal supported cell (MSC) and Anode supported cell (ASC) technologies, strong improvements in terms of reliability and durability are expected. In order to enhance anode efficiency, the 3D microstructure needs to be precisely described at the microscopic scale. ISTEC has produced LST-GDC active anode layers using screen printing process. Back-scattered electrons (BSE) SEM images that contain relevant microstructure information have been realized by Centre des Matériaux. All the original images have been enhanced by Centre de Morphologie Mathématique using a series of morphological operators to remove artefacts. Finally, a 3D pluri-Gaussian model has been generated from 2D images, and then computed to predict volume fraction of the 3 phases, 3D tortuosity, gas permeation and ionic conductivities.

Pressurized operation of SOCs increases performance and enables the coupling with a gas turbine in the fuel cell mode and improves the overall process efficiency in producing pressurized hydrogen or syngas in the electrolysis mode. So far, facilities for pressurized operation of solid oxide cells are commonly based on pressure vessels enclosing the entire single cell housing or stack to be operated. This approach results in reduced gravimetric and volumetric power density, enhanced costs and a complex control strategy to equilibrate fuel-, oxidant- and vessel pressure. Thus, limiting the pressurization solely to the stack or the single cell housing is advantageous.In a previous study [1] a pressurized testing set-up without the need for a pressure vessel was presented. This set-up, which is based on existing solid oxide cell testing technology [2], enables testing of incremental planar solid oxide cells at pressures of up to 11 bara. In order to enable pressurization, the cell is sealed in between two metallic housing components by a glass sealant, which has to withstand the pressure difference between the active part of the cell and the ambient pressure in the furnace.This application of metallic components (figure 1 (a)) in single cell testing might affect a reasonable evaluation of the intrinsic cell losses, thus for ambient pressure testing commonly ceramic cell housings and inert contact grids (Au- or Ni-grid) are applied.To overcome these issues, in a first step, an aluminium oxide scale forming steel grade was chosen for the housing body. The cell is contacted with metallic flow fields similar to a stack as shown in Ref. [3]. As Crofer 22 APU/H are extensively studied steel grades for solid oxide cells, in a second step uncoated Crofer 22 H was used for the cell housing, resulting in severe chromium poisoning of the cell. To overcome this issue, a more inert testing environment was realized by a ceramic inlay combined with Au- and Ni- contact grids as shown in figure 1 (b).In this contribution the different concepts will be evaluated regarding their applicability for pressurized testing. The impact of housing and contacting on cell performance and stability will be discussed. Performance losses and poisoning effects are studied by means of electrochemical impedance spectroscopy and the subsequent analysis by the distribution of relaxation times.Figure 1: CAD-drawing cross section of pressurized test bench with metallic cell housing in (a) [1] and ceramic inlay in (b).

Despite long and successful development history of Solid Oxide Cells (SOC) continuous improvement in performance, longevity, manufacturing and system integration is necessary to bring this highly efficient technology to the market. The activities on material development and optimization at IKTS are focused mainly on enhancement of durability in SOFC, SOEC and rSOC operation and on boosting the power density for electrolysis operation. The recent results show considerable enhancement in cell longevity and power density.The SOC priorities in applied research are moving from materials to industrialization and Fraunhofer as applied R&D service provider is following this trend. During recent years considerable efforts on simplification and automation of cell and stack manufacturing processes has been addressed at IKTS. The manufacturing processes for electrode manufacturing have been adjusted for high yield automated printing on thin electrolytes with integrated quality control measures. The efficient ways for reduction of time and energy consumption for sealing process of SOC stacks have been found and shown in pilot production as well as automated assembling of components to stacks demonstrated. Furthermore the stack integration into modules utilizing stacks of different providers has been addressed during last years and reliable solutions for building of compact stack modules of higher power (2 to 10 kW) has been developed and tested with industrial partners.The increasing interest in “green hydrogen” generated multiple opportunities for SOEC technology to be considered as inherent part of industrial and chemical processes. IKTS pioneering work on coupled operation of SOEC module with Fischer-Tropsch reactor provided first demonstration of feasibility of this approach for waxes production. The process design and analysis of steel, using SOEC technology, has been performed and resulted in multiple demonstration activities in last four years. Recently novel approaches for ammonia production are addressed at IKTS showing opportunity for transportable chemicals/fuel production without need of CO2 source.High power electrolysis applications (>10 MW) will need new approaches for stack design and put higher requirements on durability. High power electrolysis applications will also open a new window of opportunities for industrial SOFC power plants. This vision is considered as outline for our future research.

Solid oxide cells (SOC) are exciting both for their capability of fuel-flexible electricity generation and their promise in the production of green H2; however, their degradation and long-term performance remain important questions. Operating individual button cells to characterize these changes is prone to cell-to-cell variations, whereas large-area cell/stack life testing requires considerable resources. In both cases, full cell characterization usually must wait until the completion of the life test, which may delay getting results for months or years. While progress has been made in observing microstructural changes over time (termed here time-resolved characterization) using techniques like non-destructive transmission x-ray microscope tomography, such experiments are difficult to implement, especially the effect of current is unexplored due to geometrical and gas environmental limitations of the system.Presented here is a method to rapidly obtain time-resolved microstructural information of the same cell in tandem with electrochemical measurements. While SOC degradation arises due to a range of processes, this work focuses on the microstructural evolution of Ni-YSZ fuel electrodes that occurs during electrolysis operation, which is known to be a key issue. Two key experimental features allowed time-resolved characterization of multiple cells simultaneously: (1) the use of Ni-YSZ symmetric cells allows multiple cells of different size in a single test by avoiding the need for gas seals, and (2) the use of laser milling to allow controlled removal of a portion of each cell during the test.Here we illustrate the method for the case of low steam content and high current density, designed to exacerbate and accelerate electrolysis degradation due to the highly reducing conditions achieved. Tape-cast planar electrode-supported Ni-YSZ symmetric cells were laser cut into precise geometries with well-defined projected areas and easily removable sections. Four cells with projected areas of 1, 1, 0.75, and 0.5 cm2 were connected in series within the same furnace at 800 C and a gas environment of 97% H2 and 3% H2O. A current of 0.75 A was run through three of the cells, resulting in current densities of 0.75, 1.0, and 1.5 A/cm2, while one was maintained with no current. The life test lasted for 500 h, and the microstructure was observed at 0, 200, and 500 hours.Figure 1 shows the microstructures observed at the highest current density by polished cross-sectional SEM and FIB-polished sections under conditions that provided Ni/YSZ contrast. Microstructural degradation had occurred by 200 h and became immense at 500 h, due to high current density in these reducing conditions. By 200 hours, grain boundaries become enriched with Ni, likely due to the formation of Ni-Zr intermetallics in the ultra-low pO2 induced by local overpotential as reported by Chen[1] and Szasz[2]. Additionally, the electrode-electrolyte interface becomes nanoporous and a Ni enriched fracture is present ~2 μm into the electrolyte. By 500 hours, the Ni rich nanoporosity has progressed more than 10 μm into the electrolyte and created islands of large grains surrounded by Ni-rich deposits.Electrochemical impedance spectroscopy measurements show a clear initial decrease in polarization resistance followed by an overall impedance increase by 200 hours. These correspond to the initial electrochemical boost from the production of a nanoporous structure, followed by the deactivation of areas of the cell due to the large fractures through the electrolyte.[1] M. Chen et al., “Microstructural Degradation of Ni/YSZ Electrodes in Solid Oxide Electrolysis Cells under High Current,” J. Electrochem. Soc., vol. 160, no. 8, pp. F883–F891, 2013, doi: 10.1149/2.098308jes.[2] J. Szász et al., “High-Resolution Studies on Nanoscaled Ni/YSZ Anodes,” Chem. Mater., vol. 29, no. 12, pp. 5113–5123, 2017, doi: 10.1021/acs.chemmater.7b00360.Figure 1: Microstructural and electrochemical changes of a symmetric Ni-YSZ SOC undergoing 1.5 A/cm2 in reducing conditions at 0, 200, and 500 hours. Cathode is on the left in all images; anode is on the right. a, b, c) Backscatter electron imaging reveals destruction of the dense YSZ electrolyte over the lifetime of the experiment. Lines of brighter phases on the anode side of fractures are Ni deposits. d, e, f) Focused Ion Beam cross sections reveal nanoporosity forming in the first 10 μm from the cathode-electrolyte interface at 200 and 500 hours. Charging is evident as a result of curtaining from the nanoporous features in the electrolyte. g) Electrochemical impedance spectroscopy shows the initial decrease in resistance from 0 to 48 hours followed by increasing resistance due to microstructure evolution. Figure 1

The project MF-SOFC, partly funded by the European Commission, is aiming to scale up the Rolls-Royce SOFC technology from kW to tens of kW scale. The consortium brings together the wide range of skills and experience required to achieve this goal from fundamental materials science to end user knowledge. Due to a focus on the requirements of a commercial product, the first generation stacking concept was redesigned to achieve a higher power density. The manufacturing processes were refined and scaled to meet the increase in requirements for components. The cell development activity combines the experience of Rolls-Royce and Risø National Laboratory with the aim of improving the durability of the multi-cell module. Mechanical modelling to evaluate and predict the reliability of the ceramic structures is led by Imperial College London and has been carried out in parallel with the design process. Testing has been carried out in purpose-built test facilities, in France and the UK. Results from both locations are in good agreement and the high power density support module demonstrated performance that would meet the requirements of initial commercial products.

Solid oxide fuel cells (SOFCs) have been developed while pursuing low production cost and high performance with anode-supported planar type single cells at Korea Electric Power Research Institute. Anode-supported single cells have been manufactured by press molding, screen printing, and co-firing. They have been formed in the size 5 x 5 cm 2 and about 1.8 mm thick. The structure of anode-supported planar cells was used to investigate internal resistance and anode polarization resistance for improved performance. In our study, an important factor for determining cell performance was found to be the anode raw material quality assurance. The manufactured cells were evaluated on the basis of their microstructure and electrochemical performance. We have significantly improved cell performance during the past few years.

Review: 22 refs.

The degradation of metal interconnects (ICs) is one of the main lifetime limiting issues in Solid Oxide Cells (SOCs). Chromium oxide thermally grows on the substrate surface that causes ohmic loss and its vapors poison the oxygen electrode reaction (OER). Mn-based spinels are presently among the materials of choice to protect ICs to prevent Cr diffusion.As the target lifetime is around 40000h, it is of interest to design accelerated ageing tests, under harsher than nominal operating conditions, to simulate the long-time ageing behavior at nominal conditions.In this study, the performance of IC coatings was investigated with accelerated ex-situ corrosion ageing tests, in which the operating condition of the air-side interconnect was simulated. Temperature and humidity were selected as test variable parameters. Area Specific Resistance (ASR) and Cr depletion/migration were studied for the stressed conditions. Tests were carried out on stainless steel substrate with Mn-Cu-Oxide and Mn-Co-Oxide as protective coatings.Ageing tests over 1000h at different temperatures (750-950°C) and humidity (above 15%RH) were run while monitoring the ASR of the samples. Fig. 1 shows the ASR results for the samples run at different temperatures and in dry condition. Based on Fig. 1a-e, it is observed that the ASR trends remain constant during the test time. Although 916°C is well above nominal operating temperature, the ASR showed constant value within a reasonable range. At a test temperature 950°C (Fig. 1f), the ASR behavior of samples containing Co-Mn Oxide protective coating is the same found for lower temperatures; however, a sharp ASR increase in Mn-Cu Oxide coated samples can be observed. Applying humidity to high-temperature conditions also may help to accelerate the test, and its results will be presented. SEM/EDX post-test characterization provides information on the scale morphology and helps to understand the different behaviors at different stressed conditions.Keywords: SOC; interconnect; protective coating; ageing test; accelerating test; Cr diffusion; ASR; Figure 1

A corrugated mesoscale structure on electrode–electrolyte interface for enhancing cell performance in anode-supported SOFC

Stress development in metallic interconnects during operational cycles of solid oxide cells (SOCs) could create permanent plastic deformations, which can cause contact losses in the interface between the interconnect and cells. The magnitude of these stresses and their distribution within the SOC stack are dependent on the overall design of the stack and its operational parameters/conditions. In this study, analysis of stress generation in metallic interconnects due to temperature and stack assembly loadings for different types of generic SOC stack designs operated under the same operational cycles is performed. The analysis includes stack design concepts with and without contact layers having cross-shaped and/or corrugated metallic interconnect. To consider the effect of thermal gradient over the surface of the stacks (due to electro-chemical reaction), stacks with two types of temperature profiles are studied. The analysis is performed numerically using finite element modeling of a standard repeating unit of SOC stack. Special focus is given to stresses that can cause possible delamination of the interconnect from the cell that subsequently leads to loss of electrical contact. Comparison of stress for different types of stack design concepts are presented. An effective way to minimize stresses in the interconnect is found to be through designing the micro-architectures of the interconnect and us eof contact layers materials. Key words: Solid oxide fuel cells, Metalic interconnect, Creep, finite element modeling

Introduction Lifetime expectations and uncertainties regarding end-of-life performance hinder widespread deployment of fuel cell technology. Conventional cell performance diagnostic methodologies rely upon either ex-situ measurements, which lack an immediate connection with the physical degradation phenomena, or upon single cell in-operando electrochemical impedance spectroscopy (EIS), which often is troubled by challenges of feature identification of the spectra and uncomplete representation of the full stack operating conditions in large scale systems [1]. To address these problems, we propose to use electrochemical impedance spectroscopy techniques as diagnostic and prognostic tools to predict the cause-effect relationship between degradation phenomena and stack operating conditions. This experimental methodology can be potentially integrated into commercial systems for stack state of health monitoring during continuous operation. Experimental methods Through experimental study on a solid oxide fuel cell (SOFC) short stack, we explore the potential use of electrochemical impedance spectroscopy (EIS) for in-operando diagnosis. We identify the mechanisms contributing to this system EIS, by carrying out pristine cell/stack tests at different operating conditions (gas partial pressure, current, temperature).We collect experimental data on a 6-cell short-stack with active area of 80 cm2 per cell. Each anode-supported cell features a Ni-YSZ fuel electrode (240 μm), a Sr-doped LaMnO3 (LSM) oxygen electrode (40 μm) and an 8 mol% Y2O3 stabilized Zirconia (YSZ) electrolyte (8 μm). In our experimental setup, the first (reference electrode, i.e., cathode of cell no.1) and last (i.e., anode of cell no.6) electrodes are connected to the counter and working electrode leads of the potentiostat, respectively. This allows connecting the diagnostic system in parallel with the load and the stack, while performing floating ground galvanodynamic EIS measurements. This configuration enables the load to impose the operating current of the stack, while the potentiostat modulates the signal without any current in its main circuit. Reference and working sensing connections to individual cells in the stack are independent on the potentiostat leads.We operate the stack with mixtures of hydrogen, nitrogen and steam at the fuel electrode and mixtures of oxygen and nitrogen/helium at the oxygen electrode. We control the operating temperature via an electric furnace, while monitoring the stack temperatures with a set of six thermocouples, as reported in [2]. Discussion Figure 1 compares the measured EIS spectra for each cell in the short stack at beginning of life and after a major degrading event. With continuous stack monitoring, we are able to identify how degradation phenomena differently impact the state of health of the cells during the stack normal operation. This allows identifying the most degrading operating conditions of the system in real-time, hence informing the operational planning and lifetime maximization strategies of the system. Conclusions Preliminary results suggest that our methodology can provide meaningful insights regarding the cause and nature of the physical processes at the origin of degradation phenomena in solid oxide cells. This process is particularly interesting as a non-disruptive diagnostic and prognostic tool for commercial scale systems that need to optimize their operating conditions to either maximize their lifetime or meet a performance target at end-of-life.Figure 1 – (left) Beginning of life and (right) degraded cell plot comparison of experimental EIS for a 6-cell short stack at 1 A. Tstack=750°C, Anode flow=1.2 Nl min-1 (N2/H2=50/50%), Cathode flow=18.9 Nl min-1 (Air)[1] A. Baldinelli, L. Barelli, G. Bidini, A. Di Cicco, R. Gunnella, M. Minicucci, A. Trapananti, 2019, p. 020012.[2] L. Mastropasqua, S. Campanari, J. Brouwer, J. Power Sources 2017, 371, 225–237. Figure 1

Present results on intermediate temperature-operating solid oxide fuel cells (IT-SOFC) are mainly focused to obtain higher performance of single cell at lower operating temperature, especially with planar type. In order to develop a 1 kW-class SOFC system for Residential Power Generation (RPG) application that can be operated at intermediate temperatures, we have developed an anode-supported and planar type SOFC stack using cost-effective interconnects such as ferritic stainless steels. To improve electrical performance of the anode-supported SOFC, cells with alternative electrode or thin electrolyte layer of YSZ and ScSZ, respectively, were fabricated by slurry coating and their performances were investigated. Mixtures of NiO and 8YSZ powders were pressed into rectangular plates, which were pre-sintered at 1400°C for 1 h. Then the 8YSZ (8mol%Y2O3+ZrO2) or 10ScSZ (10mol%Sc2O3+1 mol%CeO2+ZrO2) electrolytes were coated by slurry coating method on the upper side of anode substrate and cofired at 1550 or 1500°C for an hour. Thereafter, composite cathode pastes of LSM (La0.7Sr0.2MnO3) or LSCF ((La0.6Sr0.4)(Co0.2Fe0.8)O3 ) were screen-printed on thin electrolytes and heat-treated at 1100°C for 2 h. The final size of anode-supported single cell sintered was about 5 × 5 cm2, and the thickness of the electrolyte and the cathode layer was about 20 μm and 30 μm, respectively. The I-V and AC impedance characteristics of these single cells were evaluated at intermediate temperature (650 ∼ 750°C) using hydrogen gas as a fuel. The maximum power density of anode-supported cell was 0.4 ∼ 1.7 W/cm2 at 750°C and 0.12 ∼ 0.55 W/cm2 at 650°C depending on the composition and microstructure of electrolyte/electrodes.

We developed induction furnace brazed anode supported tubular and flat tube SOFCs to improve power density and studied basic technology of key components for making stacks from them. Performance of the flat tube cell showed 210 mW/cm 2 (0.7y 300 mA/cm 2 ) at 750°C. Induction furnace brazing was applied to connect ceramic and metal parts and their gas permeability was tested. The results indicated that the induction furnace brazing method could be used effectively for making anode supported tubular and flat tube cell stack. Through these experiments, we obtained basic technology of such cells and established a proprietary concept of the anode-supported tubular and flat tube cell stack.