• Energy Research

AbstractIn this work, we report a 4,400 h test of a state‐of‐the‐art Ni‐YSZ (yttria stabilized zirconia) electrode supported solid oxide electrolysis cell. The electrolysis test was carried out at 800 °C, –1 A cm−2 with 10% H2 + 90% H2O supplied to Ni‐YSZ electrode compartment. Except for the first 250 h of fast initial degradation, the cell showed rather stable performance with a moderate degradation rate of around 25 mV per 1,000 h. The electrochemical impedance spectra (EIS) acquired during the test show that both serial resistance and electrode polarization resistance increased during the durability test. Further impedance analyzes show that both the LSCF (strontium and cobalt co‐doped lanthanum ferrite)‐CGO (gadolinium doped ceria) electrode and Ni‐YSZ electrode degraded and the degradation was dominated by the Ni‐YSZ electrode. Post mortem analysis on the Ni‐YSZ electrode revealed loss of contact between Ni‐Ni grains, Ni‐YSZ grains and increased porosity inside the active layer. The microstructural changes were most severe at steam gas inlet and became less severe along the gas flow path. The present test results show that this type of cell can be used for early demonstration of solid oxide cell operation at electrolysis current densities around 1 A cm−2.

Abstract Future electricity production will use fossil-free sources with zero CO2 emission or closed carbon cycle technologies based on renewable sources. While hydrogen is considered a key energy source, its production at present time relies heavily on fossil fuels. Furthermore, distribution and storage are not well established and require substantial investments. This is a strong motivation to identify alternative, safe, high power density hydrogen carriers, where existing logistics and infrastructure can be utilized. In this contribution, ammonia and biogas are considered for high-efficient electricity production in solid oxide fuel cells (SOFCs). It is demonstrated that the properties and operating conditions of SOFC allow for direct use of these fuels, with fuel pretreatment inside the SOFC anode. The high efficient electricity production using pure ammonia or real biogas was successfully proven on state-of-the-art SOFCs. Even without optimization of operating parameters, electrical efficiencies of 40–50% and high and stable power output were demonstrated.

AbstractSolid oxide cells (SOCs) can be operated either as fuel cells (SOFC) to convert fuels to electricity or as electrolyzers (SOEC) to convert electricity to fuels such as hydrogen or methane. Pressurized operation of SOCs provide several benefits on both cell and system level. If successfully matured, pressurized SOEC based electrolyzers can become more efficient both energy‐ and cost‐wise than PEM and Alkaline systems. Pressurization of SOFCs can significantly increase the cell power density and reduce the size of auxiliary components. In the present study, a SOC stack was successfully operated at pressures up to 25 bar. The pressure dependency of the measured current‐voltage (I–V) curves and impedance spectra on the SOC stack are analyzed and the relation between various system parameters and pressure is derived. With increasing pressure the open circuit voltage (OCV) and the reaction kinetics (electrode performance) increases for thermodynamic and kinetic reasons, respectively. Further, the summit frequency of the gas concentration impedance arc and the pressure difference across the stack and heat exchangers is seen to decrease with increasing pressure following a power‐law expression. Finally a durability test was conducted at 10 bar.

Operation of a Ni–YSZ electrode supported Solid Oxide Cell (SOC) was studied in both fuel cell mode (FC-mode) and electrolysis cell mode (EC-mode) in mixtures of H2O/H2, CO2/CO, H2O/H2O/CO2/CO at 750 °C, 800 °C and 850 °C. Although the SOCs are reversible, the polarisation characterisation shows that the kinetics for the reduction of H2O and CO2is slower compared to oxidation of H2and CO, and that oxidation/reduction in CO2/CO mixtures is slower than in H2O/H2mixtures. The kinetic differences are partly related to the polarisation heating and the entropy change. Also the diffusion resistance is larger in EC-mode as compared to FC-mode and the low frequency concentration resistance (which is affected by diffusion), is asymmetric around the open circuit voltage (OCV), and is significantly higher in the EC-mode. Both the increased diffusion resistance and the asymmetric low frequency concentration resistance result in a decreased activity in the EC-mode. Changing the porosity of the support structure shows a significant change in both the diffusion resistance and low frequency concentration resistance when applying current, showing that diffusion limitations cannot be neglected for SOCs operated in the EC-mode. Also the Ni–YSZ TPB resistance is affected by changing the support porosity, indicating that kinetic investigations under current and even at OCV, and the chase for a general expression for “all” Ni–YSZ electrodes may be pointless. The diffusion limitations through the support and active electrode structure create an increased reducing atmosphere at the interface which may be related to the degradation of the cells.

Two-dimensional (2D) multi-physics models of solid oxide cells (SOC) reproduce distributions inside the cell during operation, and allow to implement interdependencies of heat transfer, mass transport, charge transfer, and electrochemistry. So far, this approach has mostly been utilised for transient and steady-state problems, preventing widespread application to electrochemical impedance spectroscopy, one of the main methods in experimentally analysing SOCs. In the present work, a computationally efficient alternative is outlined, surpassing these shortcomings by transforming the set of equations from the transient into the frequency form. The coupled multi-physics implemented by partial differential equations are solved numerically over a 2D domain by the finite element method, reproducing the cross section of the SOC. The model is validated with experimentally obtained polarisation curves and impedance spectra. The 2D model reproduces the experimental results and further visualises frequency-dependent oscillations of gas phase and potentials. These insights separate between impedance features from electrochemistry, gas diffusion, and gas conversion. In addition, overlapping gas conversion and diffusion impedance features at low frequencies are deconvoluted and transition frequency regions are discussed.

Abstract Low energy conversion efficiency and high storage costs still hamper a successful implementation of sustainable energy systems. Recent theoretical studies show that reversible electrochemical conversion of H2O and CO2 to CH4 inside pressurized solid oxide cells combined with subsurface storage of the produced gases can facilitate seasonal electricity storage with a round-trip efficiency reaching 80% and a storage cost below 3 ¢/kWh. Here we present test results with a 30-cell SOFCMAN 301 stack operated with carbonaceous gases at 18.7 bar and 700 °C in both electrolysis and fuel cell mode. In electrolysis mode the CH4 content in the stack outlet gas increased from 0.22% at open circuit voltage to 18% at a current density of −0.17 A cm−2. The degradation observed by scanning electron microscopy studies correlate well with the observed electrochemical stack degradation. The degradation rates in both fuel cell and electrolysis mode were comparable to previously reported SOFCMAN stack degradation rates measured at ambient pressure operation with H2/H2O gas mixtures.

Many research facilities and industrial companies worldwide are engaged in the development and the improvement of solid oxide fuel cells/stacks (SOFC) and also of solid oxide electrolysis cells/stacks (SOEC). However, the different stacks cannot be easily compared due to non-standardized test programs. Therefore the EU-funded project SOCTESQA which started in May 2014 has the aim to develop uniform and industry wide test modules and programs for SOC cells/stacks. New application fields which are based on the operation of the SOC cell/stack assembly in the fuel cell (SOFC), in the electrolysis (SOEC) and in the combined SOFC/SOEC mode are addressed. This covers the wide range of power generation systems, e.g. stationary SOFC μ-CHP, mobile SOFC auxiliary power unit (APU) and SOFC/SOEC power-to-gas systems. In order to optimize the test programs, which consist of different test modules, several testing campaigns have been performed. The project partners apply the developed test procedures on identical SOC stacks. In this project 5-cell shortstacks with anode supported cells (ASC) were used, which were provided by an established stack supplier. Altogether 10 pre-normative test modules were developed: Start-up, current-voltage curve, electrochemical impedance spectroscopy, reactant utilization, reactant gas composition, temperature sensitivity, operation at constant current, operation at varying current, thermal cycling and shut-down. The presentation compares the results of the test modules of the different project partners. Important aspects are the evaluation of the results in terms of repeatability of the different testing campaigns and the reproducibility of the results among the partners. Electrochemical properties, e.g. open circuit voltage (OCV), area specific resistance (ASR), power density, fuel utilization (FU) and impedance values of both stack and repeat units (RU) are presented. Moreover, the results are discussed in context to the test input parameters. Another important issue is the reproducibility of the different test methods, e.g. jV-characteristics and EIS-spectra. Finally, important aspects for the determination of reproducible degradation rates of SOC stacks will be presented and discussed.

Concentration impedance in the context of solid oxide cells (SOC) has been discussed in the literature [1-6]. We divide the concentration impedance into diffusion impedance and conversion impedance. Diffusion impedance is defined as impedance due to a reactant concentration gradient caused by a slow (relative to the electrode reaction rate) diffusion through a porous boy of a stagnant gas layer. Conversion impedance is defined as impedance caused by a decrease in an electrode reactant concentration due the conversion of the reactant by electrode reaction. Mainly the diffusion impedance has been treated in spite of the fact that the conversion impedance often is of greater importance. Actually not all workers within the SOC field recognize the concept of conversion impedance. Other workers claim that the two kinds of concentration impedance cannot be distinguished. It is correct that the two phenomena will interact, and thus often it may be difficult to separate them, but we have tools that make it possible to distinguish between the two types of impedance. The diffusion impedance due to diffusion through a porous body, e.g. an electrode support layer, will not be sensitive to reactant gas flow rate, whereas the conversion impedance will be very sensitive to the reactant flow rate. Diffusion impedance results in the simple, clean case in a Warburg element, i.e. a 45 ° slope followed by a relative sharp decrease by decreasing low frequencies down to the x-axis in a Nyquist plot. In contrast to this, clean conversion impedance can always be modeled by a resistor in parallel with an ideal capacitor, which most often has a rather huge value in the order of 0.1 – 1 F cm-2. The paper and presentation will explain how the two types of concentration impedance may be extracted from impedance spectra in a manner that they are separated. Both simple examples and more difficult cases with very high fuel utilization at high current density, where the interaction between diffusion and conversion impedance is in particular strong, will be given. Special emphasis will be put on the case of plug flow condition, which is the relevant case for practical cells and stacks. References: S. Primdahl, S., M. Mogensen, “Gas conversion impedance: A test geometry effect in characterization of solid oxide fuel cell anodes”, J. Electrochem. Soc., 145, (1998) 2431-2438. S. Primdahl, M. Mogensen, “Gas Diffusion Impedance in Characterization of Solid Oxide Fuel Cell Anodes”, J. Electrochem. Soc., 146, (1999) 2827-2833. T. Jacobsen, P. V. Hendriksen, and S. Koch, Electrochim. Acta, 53, (2008) 7500. Søren Højgaard Jensen, Anne Hauch, Peter Vang Hendriksen, Mogens Mogensen, “Advanced Test Method of Solid Oxide Cells in a Plug-Flow Setup”, J. Electrochem. Soc., 156, (2009) B757-B764. B. Liu, H. Muroyama, T. Matsui, K. Tomida, T. Kabata, K. Eguchi, “Gas Transport Impedance in Segmented-in-Series Tubular Solid Oxide Fuel Cell, J. Electrochem. Soc.”, 158, (2011) B215-B224. Y. Tanaka, M.P. Hoerlein, G. Schiller, “Numerical simulation of steam electrolysis with a solid oxide cell for proper evaluation of cell performances”, Internat. J. Hydrogen Energy, 41 (2016) 752 -763.