ECS Meeting Abstracts

Mixed ionic and electronic conductor (MIEC) oxides have been proposed as candidates to replace Ni/YSZ composites as anodes for Solid Oxide Fuel Cells (SOFC) due to their good stability under C-based fuels. Some MIECs have also demonstrated a good electro-catalytic activity both for oxygen reduction and hydrogen oxidation, making them suitable for symmetric configurations (S-SOFC). This approach presents remarkable advantages for reducing manufacturing and operational costs, as well as for extending the cell's lifetime by reversing gas flows and thus partially reversing the negative effects of sulphur poisoning and carbon deposition that may happen during operation. Also, the catalytic activity of MIEC electrodes can be improved by functionalizing the oxide surface with active nanoparticles. In this work, we study the formation of Ni-Fe alloy nanoparticles by exsolution from a Sr_0.93(Ti_0.3Fe_0.63Ni_0.07)O_3-δ (STFN) perovskite in reducing atmospheres, and also the process of reoxidation when the exsolved material is exposed to an oxidizing atmosphere. The initial Sr-deficient composition was chosen to alleviate the segregation of Sr [1], which typically can occur in these materials.

Exsolution has previously been reported to improve the electrochemical performance of STFN anodes [2], but the mechanisms underlying the exsolution process and the solid/gas interface are still not well understood. The possibility of using S-SOFC materials that undergo exsolution also raises the question of whether the material is regenerated during reoxidation. While oxidation-induced redissolution of exsolved nanoparticles has been observed for Fe-Co exsolution on La_0.8Sr_1.2Fe_0.9Co_0.1O_4−δ perovskites [3], for the Ni-Fe exsolution in Sr₂(Fe_1.4Ni_0.1Mo_0.5)O_6−δ, nanoparticles remained at the surface even after reoxidation [4]. The first case is a very interesting result to achieve larger cell lifetimes, and the latter case is interesting as it opens an additional route to increase the cathode performance. In fact, in ref. [5] Ni exsolution in SrTi_0.1Fe_0.85Ni_0.05O_3−δ is deliberately employed as design strategy, fully exploiting the non-reversibility of the exsolution of nanoparticles.

However, it is not clear whether Ni-Fe nanoparticles oxidize to form a (Ni,Fe)O_x phase or if Fe is reincorporated into the lattice leaving only NiO particles at the surface. It is also not clear how Sr segregation is affected by the exsolution/reoxidation treatments, or how the reoxidized STFN perovskite is modified compared to the pristine sample.

To address these questions directly, ambient pressure X-ray photoelectron and near-edge X-ray absorption fine structure spectroscopy (AP-XPS and NEXAFS) is used to study the chemical structure of STFN in a complete redox cycle in-situ. Based on the measurements, we can provide insights into the chemical states of Fe and Ni and can differentiate the surface and bulk species for Sr and O in each stage of the cycle. We observe that Ni exsolves readily, but we also note that the amount of surface Fe⁰ increases with increasing H₂ content in the reducing atmosphere; Fe⁰ also increases with the reduction time following an exponential trend until a plateau value is reached within ~1h. Further, we find a significant Sr segregation in reducing atmospheres, which we presume occurs to compensate for the B-site cation exsolution. The amount of Sr segregation remains constant in the nearest surface after reoxidation, but is partially reversed for larger penetration depths; there is also a rapid reversibility in the Fe oxidation state during reoxidation. These observations were complemented with transmission (TEM) and scanning electron microscopy (SEM) studies, with simultaneous energy dispersive spectroscopy (EDS) analysis.

In conclusion, we propose a reoxidation-induced reconstruction which forms a Fe- and Sr-rich STF perovskite in the near-surface region, leaving the Ni segregated from the perovskite. Finally, we link the results to the electrochemical impedance spectroscopy (EIS) response of the STFN electrode, observing that this STFN-reoxidized sample shows a significant improvement in its cathode performance compared to the pristine STFN.

[1] Fagg, D. P. et al, J. Eur. Ceram. Soc. 21, 1831–1835 (2001).

[2] Zhu, T., Troiani, H. E., Mogni, L. V, Han, M. & Barnett, S. A. Joule 2, 478–496 (2018).

[3] Zhou, J. et al. Chem. Mater. 28, 2981–2993 (2016).

[4] Liu, T. et al. J. Mater. Chem. A 8, 582–591 (2020).

[5] Yang, G., Zhou, W., Liu, M. & Shao, Z. ACS Appl. Mater. Interfaces 8, 35308-35314 (2016).

Figure 1

Operation of solid oxide cells in CO/CO₂ fuel atmospheres is important both for renewable energy storage technologies as well as for ongoing, flagship space missions. Nevertheless, the performance of solid oxide cells operating in CO/CO₂ atmospheres is generally diminished compared to an identical cell operating in the prototypical H₂/H₂O atmosphere. Although Ni-cermet electrodes have demonstrated excellent performance for fuel cell and electrolysis operation using H₂/H₂O gas mixtures, their reduced performance and stability issues when operating in CO/CO₂ due to, e.g., the poor coking resistance of Ni, motivates the development of more-stable, alternative electrode materials. Oxide fuel electrodes have been developed partially in response to these constraints, although they possess drawbacks in the form of low electronic conductivity and low surface catalytic activity due to the lack of a metallic phase. However, oxides within the perovskite-family, when doped on the B-site with thermochemically reducible cations such as Ni, Co, etc., undergo an exsolution process whereby B-site dopants are reduced and nucleate as metallic nanoparticles on the oxide surface. These nanoparticles are highly catalytically active and comparatively resistant to degradation compared to similarly sized composite or infiltrated nanocatalysts. Perovskite fuel electrodes with exsolved catalytic nanoparticles have been shown to possess competitive performance compared to conventional Ni-based electrodes, while nevertheless exhibiting higher stability to reoxidation and resistance to coking.

The effect of exsolved catalyst nanoparticles on the performance of perovskite electrodes operating in CO/CO₂ fuel atmospheres is not well understood. While several studies report exsolved catalysts improve the performance of electrodes in CO/CO₂, the source of improvement is not understood, in particular due to important differences in reaction pathway compared to H₂/H₂O fuel.

Here we study two recently reported perovskite fuel electrode materials, Sr(Ti_0.3Fe_0.7)O_3-δ (STF) and Sr_0.95(Ti_0.3Fe_0.63Ni_0.07)O_3-δ (STFN). The latter exhibits the formation of Ni-Fe nanoparticles socketed on the perovskite surface after periods at elevated temperature in reducing atmosphere, as a result of the exsolution of B-site cations from the perovskite. Electrodes are investigated in both symmetric cell and full cell configurations using Hionic electrolyte supports (Nexceris). Electrode reaction kinetics are investigated as a function of temperature and fuel atmosphere in both fuel cell and electrolysis modes using current-voltage measurements and electrochemical impedance spectroscopy. Electrode and catalyst microstructure is examined using scanning electron microscopy. This presentation will examine, for the first time, the performance of STF an STFN fuel electrodes operating across the entire range of CO/CO₂ fuel atmospheres, will interpret the effect of exsolved catalyst particles on the performance of these electrodes, and will compare and contrast with observations in H₂/H₂O atmospheres in order to better understand electrode behavior.

Exsolution is the process in which cations that make up the lattice of a mixed ionic/electronic conductor (MIEC) are reduced to metals and migrate to the surface of the host ceramic. These materials have become quite attractive for solid oxide fuel cells (SOFCs) because they can facilitate oxygen and electron transport through their bulk and act as electrocatalysts on the surface of SOFC anodes. Also, because of the nature of the exsolution process, the highly dispersed nano-catalysts have strong particle-host interactions which have shown resistance to common SOFC failure mechanisms such as agglomeration and coking. However, the practical application of such materials has not been well studied since most research focuses on the electrolyte-supported cells, which offer a quick method for screening electrochemical activity but the large ohmic resistance makes this cell unrealistic for real energy conversion devices, especially at lower temperatures.

Here, we use a scalable ceramic processing technique to fabricate a ceramic-anode-supported SOFC that exsolves nano-catalysts for low temperature operation (600°C). SrFe_0.2Ni_0.4Mo_0.4O_3-δ (SFNM) is used as the anode material due to its high conductivity and exsolution ability. By shifting the mechanically supporting layer from the electrolyte to this anode, we are able to drastically reduce the overall ohmic resistance of the cell. Additionally, we avoid extra processing steps for conventional catalyst deposition by taking advantage of the in-situ catalyst exsolution. During the exsolution process, we show that our cell performance increases over time, up to a peak power density of 300 mW/cm² at 600°C with H₂ fuel. Further, we discuss two major challenges facing exsolution-anode-supported SOFCs along the way and we show our efforts to identify and address the root causes of these problems.

Solid oxide fuel cells have a proven efficiency for the direct conversion of biogas into electricity, and their full potential from European agricultural and waste-water sites still has to be fully evaluated and harnessed. Biogases are methane-rich mixtures whose composition and trace contaminants, especially of sulfur compounds, can vary greatly.

To avoid carbon deposition from the carbon-rich CH₄-CO₂ mixtures, some amount of steam and/or extra CO₂ needs to be added to process the biogas. Ideally this can be recovered from the SOFC exhaust. In a preparatory work, biogas/SOFC integrated systems were studied and compared using Aspen Plus[1]. Best performances are reached with a catalytic methane pre-reformer when a fraction of the anode off-gas is recirculated for mixing with the biogas feed. This work provided the guidelines for the present catalytic study and predicted thermodynamic equilibria of gas mixtures.

Here, the overall performance of a ruthenium exsolution catalyst[2] for the reforming of methane was evaluated for these different biogas compositions at various gas hourly space velocities. Gas chromatography and material balance calculations enabled the full measurement of inlet and outlet gas compositions which, in turn, permitted to calculate precise conversion rates, reforming ratio, and the deviation from thermodynamic equilibrium. Methane conversion reached >95 % for a dry-dominant reforming mixture, which simplifies the overall system. The longer-term stability of the catalyst as well as its tolerance towards different sulfur traces that occur in biogas was then evaluated and analyzed.

[1] Shuai Ma, Gabriele Loreti, Ligang Wang, Andrea L. Facci, Stefano Ubertini, Changqing Dong, François Maréchal, and Jan Van herle, E3S Web of Conferences 2021 238:04002

DOI: 10.1051/e3sconf/202123804002

[2] Muhammad A. Naeem, Paula M. Abdala, Andac Armutlulu, Sung Min Kim, Alexey Fedorov, and Christoph R. Müller, ACS Catalysis 2020 10 (3), 1923-1937

DOI: 10.1021/acscatal.9b04555

The use of syngas to feed solid oxide fuel cells (SOFCs) appears as an attractive alternative due to the possibility of being obtained from renewable sources such as biomass or from other solids of carbonaceous composition such as solid wastes through reforming reactions. The use of syngas in SOFCs has been studied in several previous works, directing the efforts to obtain an anode material with the appropriate electrocatalytic activity for the electrooxidation of syngas, able to tolerate carbon deposition and avoid sulfur poisoning [1]. The Ni-YSZ anode has been studied in the electrooxidation of syngas, being one of its advantages the catalysis of the water gas shift (WGS) reaction, generating more H₂ with respect to CO and therefore improving the reaction kinetics [2]. One of the disadvantages presented by Ni-YSZ is the decrease of its performance in long term tests, due to the catalytic activity of Ni towards the reverse Boudouard reaction, so that carbon deposits block the triple phase boundaries (TPBs) of the anode [3]. CeO₂-based materials are suitable for use as anodes in fuel cells fed with carbonaceous fuels due to their resistance to carbon deposition and sulfur poisoning [4, 5]. Recently it has been reported that the addition of molybdenum to CeO₂ (Mo-doped CeO₂ or CMO) confers catalytic activity to the carbon gasification reaction and an increase in the electrical conductivity of CeO₂ [6]. Furthermore, it has been reported that the addition of copper nanoparticles to Mo-doped CeO₂ increases about 10 times its electrical conductivity [7].

Two nanomaterials for SOFC anodes were synthesized: Mo-doped CeO₂ (CMO) and Cu,Mo-doped CeO₂ (CMCuO). Both nanomaterials were synthesized by the combustion method with 10 wt.% total metal loading following the experimental procedures detailed in [6] and [7] correspondingly. The SOFCs were fabricated using a CMO-YSZ or CMCuO-YSZ composite as anode, a dense YSZ electrolyte and a LSM-YSZ cathode. Both SOFCs were fed with air at the cathode and humidified syngas at the anode (syngas + 3 vol.% H₂O). The morphology of the SOFCs was studied by SEM, while the crystal structure and chemical stability of the materials were studied by XRD and EDS. The cells were characterized by EIS at OCV condition at different temperatures and syngas concentrations, while the cell performance was studied by linear sweep voltammetry.

An ADIS analysis obtained from the EIS data for both fuel cells in Figure 1 clearly shows the presence of two phenomena at high frequencies related to the electrooxidation reactions of H₂ and CO, these phenomena are sensitive to temperature changes, the differences in their charge transfer resistance values suggest that both reactions are carried out with different kinetics. The processes shown at low frequencies are related to diffusional limitations of each gas towards triple phase boundaries (TPBs), processes sensitive to changes in syngas concentration. Cell tests, operating at 800°C, showed a maximum electrical power density of 114 mW cm^-2 at a current density of 208 mA cm^-2 and a cell voltage of 549 mV for SOFC with CMO-YSZ anode and a maximum electrical power density of 120 mW cm^-2 at a current density of 200 mA cm^-2 and a cell voltage of 600 mV for SOFC with CMCuO-YSZ. The addition of copper to the CMO produces a slight improvement in the electrocatalytic activity of the material and reduced ohmic resistance by almost 50% (see Figure 2), however the ASR of the cell with CMCuO anodes does not show a substantial improvement with respect to the cell with CMO anodes. In addition, the ohmic resistance is the most important component in the ASR of both cells. SEM images of the cells after tests show that the CMCuO anode surface exhibits fractures and delamination of the material, which is not observed on the CMO anode surface. The X-ray diffraction spectra do not show the presence of new phases in any of the two anode materials so the delamination of the CMCuO anode is related to the differences in thermal expansion coefficients (TECs) of the materials whose value must be measured to prove this hypothesis.

[1] T.M. Gür (2013) Chem. Rev., 113(8), 6179-6206.

[2] O. Costa Nunez et al. (2005) J. Power Sources, 141(2), 241-249.

[3] T.M. Gür et al. (2010) J. Power Sources, 195(4), 1085-1090.

[4] I. Sreedhar et al. (2019), J. Electroanal. Chem., 848, 113315.

[5] N. Mahato et al. (2015) Prog. Mater. Sci., 72, 141-337.

[6] I. Díaz‐Aburto et al. (2019) Fuel Cells, 19(2), 147-159.

[7] I. Díaz-Aburto et al. (2021), J. Electrochem. Sci. Technol., [epub ahead of print: https://doi.org/10.33961/jecst.2020.01571].

Figure 1