Studies on correlated electron materials have largely been confined within the realms of condensed matter materials physics to date. With recent advances in thin film oxide growth and reproducible electronic properties obtainable by robust synthesis, it is natural to ask the question if there are any particular opportunities for such exotic electronic systems in solid state chemical energy conversion. In this regard, we will discuss two interconnected problems, one relating to ultra-thin membrane solid oxide fuel cells that can operate below 500C and the other pertaining to realization of new modalities in micro-power sources such as hybrid energy storage-conversion devices for mobile energy. We will consider perovskite structured nickelates as representative correlated materials with potential for applications as electrodes in solid oxide fuel cells. Some basic science questions gain extraordinary significance when they have to be fabricated in free standing membrane form, ie the connection between structural distortions and electronic phase transitions, synthesis of phase pure oxides on non-lattice matched electrolytes, accommodation of defects under extreme chemical potential gradients and membrane cracking, redox reactions at surfaces and the problem of irreversible phase transitions. Finally, I will give some examples of functional solid oxide fuel cells that incorporate such correlated materials and compare their performance to traditional SOFCs.

During the past decade, the research activities related to the employment of thin-film materials to solid oxide fuel cells (SOFCs) have rapidly surged. The main reason for this is to reduce the operating temperature of the SOFC by introducing thin electrolytes and nano-structure electrodes, in order to compensate the resistance increase due to the temperature decrease. Other important research effort regarding thin films in the SOFC would be that to understand the phenomena happens at the surfaces and interfaces of the SOFC by building simpler model systems using thin film materials.
For the former, the major challenge is to overcome the fatal structural instability of the thin films and nano-structures at elevated operating temperatures, although these temperatures (le; 500 °C) are substantially lower than those (ge; 800 °C) of the conventional SOFCs. Owing to years of the efforts to improve thermomechanical stability of thin-film and nano-structure based SOFCs (TF-SOFC), both high-performance and reliability of the TF-SOFC could be obtained based on the common SOFC materials by means of multi-scale architecture in our group. Now the remaining challenge would be related to the realization of novel SOFC materials as a form of thin films. This is also closely related to the latter, the modeling experiments to understand the phenomena.
In general, novel SOFC materials are of complex composition (e.g. lanthanum gallate-base electrolyte materials, new cathode materials), and contains elements that have high vapor pressure (e.g. Ba in proton conductors, Sr in cathode materials). These characteristics provide extremely difficult conditions for realizing stoichiometric thin films out of these materials. Even for pulsed laser deposition (PLD) which is known to be the most powerful thin-film deposition technique for transferring target composition, it is not trivial to fabricate the thin films of the novel SOFC material with right composition. This is the main reason why it is often observed that the property of the thin film does not reflect that of the bulk materials.
In this presentation, we will report how significant these issues are, especially in novel electrolyte materials like lanthanum gallate-base materials and Ba-containing proton conductors. The problems including poor chemical stabilities and electrical properties induced by off-stoichiometry will be presented. The potential origins and solutions of these problems, as well as their impact on the properties of the thin films and the performance of the TF-SOFC based on those materials will be discussed.

Fuel cells can convert chemical energy directly to electrical energy, and solid-oxide fuel cells (SOFCs) are promising in applications for high efficiency, long-term stability, and low emissions. However, the high operating temperature (typically between 800 and 1000^oC) makes SOFCs limited for further potential in many device applications. Highly corrugated and ultrathin solid electrolytes by other researchers have shown enhanced current density due to the larger surface reaction area and efficient ionic transport. Here, we present the fabrication of 3D scalable, nanostructured membrane-electrolyte-assembly for low-temperature and high-efficiency SOFCs. First of all, yttria-stabilized zirconia (YSZ) nanotube (NT) arrays were fabricated by atomic layer deposition (ALD) into ordered anodic aluminium oxide (AAO) templates. YSZ is used as the solid oxide electrolyte, which enables oxygen ions to conduct. Tetrakis(ethylmethylamino) zirconium, TEMAZr, and Tris(methyl cyclopentadienyl)yttrium, Y(MeCp)₃, were used as reactants, and H₂O was used as oxygen source. As the electrically conductive porous electrode, various candidates such as Ni, Pt, and their alloys were tested and achieved by ALD with high-aspect ratios. The structures were carefully characterized by high-resolution transmission electron microscopy, x-ray diffraction, x-ray photoelectron spectroscopy, and energy-dispersive x-ray spectroscopy. Finally, we show the resulting performance of SOFCs by investigating electrochemical impedance spectroscopy at different temperatures and oxygen partial pressures.

In previous publications we introduced two different processing routes for nanoscaled anode and cathode thin-films and reported first-rate performance values compared to state-of-the-art electrodes of the same nominal composition.
However, their potential for boosting performance of a “real life” SOFC, which is typically an anode supported cell (ASC), was still pending. Therefore, and for the first time, we aimed to combine both processing routes and integrated nanoscaled electrodes into an ASC fabricated by Forschungszentrum Jülich.
Subsequent analysis by electrochemical impedance spectroscopy (EIS) and current/voltage (C/V) curves at different temperatures revealed a remarkable performance. In numbers, the area specific resistance (ASR) was as low as 560 mOmega;cm2 at 600 °C and 40% hydrogen fuel utilization. C/V curves showed a maximum power density of 528 mW/cm2 at 600 °C and 5.5% hydrogen fuel utilization. Unfortunately, as many research groups working on nanoscaled electrodes, do not measure performance values at reasonable fuel utilization, these values cannot be compared with other types of nanoscaled electrodes, i.e. such introduced in micro-SOFCs. However, when compared to the state-of-the-art cell developed at Forschungszentrum Jülich, the nanoscaled electrodes boost the performance by a factor of almost 4. The durability of the nanoscaled electrodes was confirmed by 500 h of operation under load at 250 mA/cm2 at 600 °C.
This study has created one of the most promising approaches for IT-SOFCs with supreme performance and a simple manufacturing procedure also applicable on the larger scale, so that we see great potential in this “nano” ASC design.

Thin film electroceramics are of great interest in gas sensors and solid oxide fuel cells (SOFCs), where the small dimensions lead to more rapid response times in gas detection and low ionic resistance in SOFC electrodes and membranes. Additionally, they provide a geometrically well-defined platform for systematic electrochemical investigations. In this presentation, an in situ optical absorption relaxation technique for studying the oxygen exchange rate of thin films, a key figure of merit of SOFC electrodes, will be introduced. This technique makes use of changes in concentration of optically active defect centers with step changes in environmental variables such as oxygen partial pressure and temperature. The absence of potentially catalytic metal contacts, commonly used in electrical conductivity relaxation and through-plane resistance measurements, is a key advantage of this technique, thereby allowing study of the bare film. Interestingly, oxygen exchange kinetics, which often decay with time, have been found to more rapidly degrade with this technique. The origin of this decay in (Pr,Ce)O2 and Sr(Ti,Fe)O3 thin films will be discussed, and new surface treatment techniques that reverse the degradation will be presented, the study of which are facilitated by this contactless approach. The optical absorption technique is a promising in situ method for investigating the impact of surface chemistry on oxygen exchange.

It has been suggested that tensile strain can facilitate ionic transport. However, there is a lack of systematic studies to explain the coupling between strain and ionic transport. Here, we present tensile strain-driven oxidation of brownmillerite SrCoO_2.5 grown by pulsed laser epitaxy. We performed a low temperature topotactic oxidation of the brownmillerite films in an oxidizing condition (PO₂ ~ 500 Torr) to investigate the strain-driven oxidation and subsequent physical property changes. Based on dc transport and magnetization characterizations, we found that highly oxidized SrCoO_x (x > 2.9) films can only be obtained from the brownmillerite thin films grown on (LaAlO₃)_0.3-(SrAl_0.5Ta_0.5O₃)_0.7 (LSAT) under high tensile strain. However, an insufficient oxidation has been observed from the epitaxial thin films on SrTiO₃ and GdScO₃ substrates due to strain relaxation. Therefore, our results not only clearly provides a proof that tensile strain promotes the oxygen reactivity, which has not been observed in bulk forms, but also acts as a milestone to search for ionically active oxide systems.
The work was supported by the U.S. Department of Energy, Basic Energy Sciences, Materials Sciences and Engineering Division.

The efficient electro-reduction of CO₂ to chemical fuel, and the electro-oxidation of hydrocarbons are critical towards a carbon-neutral energy cycle. The simplest reactions involving carbon species in solid-oxide fuel cells and electrolyzer cells are CO oxidation and CO₂ reduction, respectively. In catalyzing these reactions, doped ceria has been employed as a robust and highly active electrocatalyst. However, many controversies still remain about the reaction intermediates and the rate-determining step, despite numerous theoretical and experimental investigations.
We employed synchrotron-based ambient pressure X-ray photoelectron spectroscopy and electrochemical impedance spectroscopy to investigate CO/CO₂ electrochemical reactions on ceria under technologically-relevant conditions. By exposing ceria thin film in 300 mTorr of CO₂ at various temperatures, we confirmed that carbonates are the primary adsorbate. Furthermore, using a microfabricated electrochemical cell, we applied various anodic and cathodic voltages to drive the reactions away from equilibrium at elevated temperature. We followed the changes in oxygen, carbon and Ce³⁺ species under anodic and cathodic biases. In this talk, we will provide a new perspective on the electrochemical CO/CO₂ conversion pathway on the surface of ceria.

Materials with ordered porosity are gaining interest for applications beyond their use as photonic crystals. Ceria-based inverse opals are currently being investigated to assess the architectural influence on thermochemical hydrogen production. Low tortuosity and continuous interconnected pore network allow for facile gas transport and improved reaction kinetics. Ceria-based ordered materials have recently been shown to increase maximum hydrogen production over non-ordered porous ceria.
One limitation with the use of pure ceria is that it coarsens under the optimal thermal conditions required for hydrogen production, causing a breakdown of ordered porosity. Adding zirconium to the cubic ceria lattice improves both reduction and oxidation capacity, and hinders coarsening as well. However, in some cases, even small zirconia additions result in the formation of undesirable tetragonal phases.
Here, we utilize Fast Fourier Transforms (FFT) to quantify the degree of order lost by the coarsening process. We fabricate zirconium-doped ceria (ZDC) inverse opals with varying zirconia atomic percentages (i.e. ZDC20) by modifying a nanoparticle synthesis method. These nanoparticulate sols are infiltrated within micron-sized polystyrene colloidal crystal templates which are subsequently removed upon calcination. The resulting crystalline ceria-based inverse opals are subjected to heating at temperatures up to 1100C, followed by air quenching, to mimic the conditions of concentrated solar heating cycles. FFT carried out on scanning electron microscopy (SEM) images verify that all samples with at least 10 atom percent zirconium maintain ordered porosity at 1000°C for 12 hours, and samples with higher zirconia content maintain ordered porosity at 1100°C.
In addition to the microstructural changes, we also detected subtle changes in the oxygen lattice using Raman spectroscopy. The prominent cubic phase peak is located at 464cm 1, and is usually accompanied by a broad peak at 625cm-1 due to the presence of oxygen vacancies. Additionally, a peak at 305cm-1 is indicative of a presence of t” tetragonal phase. Although no tetragonal phase can be detected for pure ceria exposed to elevated temperatures, a small tetragonal peak is apparent which increases in intensity as zirconium doping content increases. Interestingly, the oxygen vacancy concentration also increases with increasing zirconia content, a highly desirable effect. There is a three-fold increase in relative vacancy concentration as the composition changes from that of pure ceria to ZDC40 heated to 1100C for 12 hours. In summary, the zirconia addition to cerium oxide leads to a compromise between macroscopic and microscopic thermal stability.

In situ soft X-ray absorption spectroscopic (XAS) technique, which enables us to analyze electronic structures of oxides under polarization in controlled temperature and atmospheric conditions, was developed. The technique was applied to evaluate electronic structures and electrochemical properties of an La2NiO4+δ dense thin film electrode on a Ce0.9Gd0.1CoO1.95-δ electrode
An La2NiO4+δ dense thin film electrode was fabricated on a Ce0.9Gd0.1CoO1.95-δ sintered compact by the pulsed laser deposition technique. In situ soft XAS measurements with the fluorescence mode were performed at the beam line BL27SU at SPring-8, JASRI, JAPAN. Inside the sample chamber, the mixture of He and O2 gases was flowed to control the partial gas pressures and minimize the X-ray absorption by gases. XAS spectra at Ni LII- and LIII-edges and O K-edge were measured from 843 to 877 eV and from 525 to 533 eV, respectively, in p(O2) = 10-3 bar under various applied voltages and in p(O2) = 10-3 and 10-4 bar under OCV condition, at 773 K.
The observed absorption peak at the O K-edge around 528 eV were assigned to the unoccupied orbitals of O 2p orbital hybridized with Ni 3d orbital. The intensity of this peak decreased with lowering oxygen partial pressure, with decreasing cathodic polarization, and with increasing anodic polarization. The peak intensity under p(O2) = 10-3 bar with -46 mV of cathodic overpotential was almost equal to that observed under OCV in p(O2) = 10-4 bar. Such spectrum changes showed that the applied cathodic overpotential induced the decrease of the effective oxygen partial pressure, indicating that the rate-determining step on the La2NiO4+δ dense thin film electrode was a surface reaction.
X-ray absorption spectra of La2NiO4+δ at Ni LII- and LIII-edges were also measured. Compared with the absorption spectra at the Ni L-edges, the change of the spectra due to the polarization was negligible. This might show that the change of the oxygen nonstoichiometry affected not significantly the valence change of Ni ions but the change in the hole concentration.

Solid oxide fuel cells operated at temperatures of 600°C are in urgent need of new, high-performing electrodes, as the losses resulting from the thermally activated electrode reactions rise sharply with decreasing temperature. Today, even the best electrodes contribute as much as 75% to the area specific resistance of an anode supported cell at 600°C and 60% hydrogen fuel utilization.
Aiming at faster reaction kinetics at surfaces and charge transfer processes at interfaces, e.g., the lateral extension of the active area at both interfaces gas/electrode/electrolyte has to be drastically enlarged. Processing of tailor-made thin-films with nano-sized pores and grains is demonstrated for the composite anode Ni/YSZ and for the mixed-ionic-electronic conducting cathode LSC. Their potential for performance enhancement is simulated by FEM modeling and proven by experiments using electrochemical impedance spectroscopy. However, also chemical inhomogeneities potentially enhance oxygen surface exchange kinetics, as reported in literature for a hetero-interface made of (La,Sr)2CoO4±d and La0.6Sr0.4CoO3-d (LSC). This beneficial effect is demonstrated for a LSC cathode thin-film cathode derived by metal organic deposition.

Interfaces between dissimilar oxides are attracting significant interest for their potential role in accelerating charge transport and surface reaction kinetics. If well understood and controlled, they can provide a new way to enable high-performance solid-oxide fuel cells, separation membranes as well as fast switching memristors. For example, recent studies have demonstrated that cobaltite hetero-interfaces exhibit orders of magnitude faster oxygen reduction kinetics compared with either single phase. The interfacial strain fields, anisotropy, and electronic interactions between the two phases are the likely mediators behind such an unprecedented enhancement. The underlying mechanisms must be understood quantitatively, so that we can go beyond isolated and empirically found interface structures to rationally designing dissimilar oxide interfaces with superior properties. Towards this goal, we have investigated the local electronic structure at nanometer resolution in model multilayer superlattices and vertical nanostructures that are made of dissimilar cobaltites. To accomplish this, we used a novel combination of in-situ scanning tunneling spectroscopy and focused ion beam milling. We found that the wide band-gap cobaltite is electronically activated at elevated temperatures through an interfacial coupling with a reducible cobaltite. Such electronic activation is expected to facilitate charge transfer to oxygen, and accelerate the reduction kinetics on the surface. There remains still a large amount of open questions on how dissimilar oxide interfaces impact oxygen diffusion and oxygen exchange on the surfaces. However, these recent results are encouraging for an improved understanding of oxide hetero-interfaces at elevated temperatures and could enable new interfaces with fast oxygen transport and oxygen reduction kinetics.

The oxygen evolution (OER) and reduction reactions (ORR) are ubiquitous in many energy conversions. Of particular interest are reactions that take place over the solid/gas interface, for example, in elevated-temperature solid oxide fuel cells and electrolyzers operating at 500-800°C. Despite that OER and ORR generally dominate the overpotential in these electrochemical cells, an in-depth understanding of what controls reaction rate has not been achieved yet. In particular, the formation of secondary phases at the surface of electrocatalysts due to cation segregation and phase separation are believed to influence OER and ORR rates strongly.
Here, we combine in-situ and ex-situ microscopy and spectroscopy techniques to follow the formation and growth of such secondary phases at the surface of perovskite (ABO_3-δ) electrocatalysts. A compositional range of dense, epitaxial thin film electrodes were grown by pulsed-laser deposition. The nucleation and growth of secondary phases were observed by in-situ environmental SEM as a function of temperature and oxygen partial pressure. Auger microscopy was also used to link surface morphology to the nanoscale chemical composition. These results were combined with current-voltage measurements to assess the formation pathway of secondary phases and their impact on OER and ORR rates.

Perovskites (ABO3) with high electronic and ionic conductivity and catalytic characteristics1 have been studied intensively for solid oxide fuel cells (SOFCs)2 and oxygen permeation membranes3 at high temperatures. A major barrier that limits the efficiency of SOFCs and oxygen permeation flux is the slow kinetics of surface oxygen exchange on the oxide surface. Ruddlesden-Popper (RP) oxides such as La2NiO4+δ (LNO) can lead to higher oxygen ionic conductivity relative to ABO3 perovskites4 and are interesting alternative cathode materials for SOFCs. By substituting of a larger cation such as Sr2+ (1.31 Å) for La3+ (1.22 Å) in LNO, the structural stresses are released, which leads to decrease in the amount of additional oxygen to stabilize the structure. Consequently, modifying the rock salts layer where the oxygen diffuse results in the reduction of the oxygen diffusion coefficients.5 However, the influence of Sr substitution on the in-plane and out-of-plane surface oxygen exchange kinetics of La2-xSrxNiO4±δ (LSNO) is poorly understood due to difficulties in the growth of single crystals of LSNO with high Sr substitution.
In this study, we investigate how Sr substitution can affect the surface oxygen exchange kinetics of LSNO thin films with a wide range of Sr content (xSr = 0.0, 0.2, 0.4, 0.6, 0.8, and 1.0) grown on (001)cubic-Y2O3-stabilized ZrO2 (YSZ). The structural orientation of the epitaxial LSNO thin films can be modified from the in-plane to the out-of-plane orientation by increasing the Sr content from 0.0 to 1.0. Such a change in the film orientation can be explained by the reduction in the surface energy of the (001)tetra. surface with increasing Sr as revealed from Density function theory (DFT) calculations. Ex situ Auger electron spectroscopy (AES) indicates no formation of secondary phases across the Sr contents. We show the strong dependence of the surface oxygen exchange rate (kq) of LSNO thin films on the Sr substitution, which can be attributed to the structural reorientation and the adsorption energy change of molecular oxygen on La-La bridge sites.
References
(1) Adler, S. B. Chem. Rev. 2004, 104, 4791.
(2) Adler, S. B.; Chen, X. Y.; Wilson, J. R. J. Catal. 2007, 245, 91.
(3) Hashim, S. M.; Mohamed, A. R.; Bhatia, S. Adv. Colloid Interface Sci. 2010, 160, 88.
(4) Amow, G.; Skinner, S. J. J. Solid State Electrochem. 2006, 10, 538.
(5) Skinner, S. J.; Kilner, J. A. Ionics 1999, 5, 171.

The effect of cation nonstoichiometry on the enhancement of surface reaction kinetics of oxide cathode has been studied with bi-layer (La,Sr)CoO3 and LaCoO3 dense thin film electrodes fabricated by an RF-sputtering method for improving the solid oxide fuel cells&’ performance. The bi-layer electrodes with cation (A-site or B-site) deficiency at surface show higher surface reaction kinetics than the cation-stoichiometric one, which is related to the increase of oxygen vacancy induced by cation vacancy. In addition, the PO2 dependence of the interfacial conductivity is analyzed based on a defect chemical model in consideration of “cation nonstoichiometry”, which supports the effect of cation nonstoichiometry. The electronic structure of these materials will be discussed on the basis of PES and XAS results. Further discussion will be made on possible factors that govern the splitting/reduction reaction of oxygen gas molecules on oxide surface.

Thin film electrodes, having controlled geometries, serve as model platforms for fundamental studies of surface exchange reactions. In addition, because of the surface-controlled performance, thin film electrodes are particularly susceptible to aging via surface chemistry changes. In this work thin film cathodes of (Sr_1-xLa_x)(Ti,Fe)O_3-α were studied to investigate the relationship between cathode electronic structure and oxygen surface exchange rate, with attention to the potentially rate-limiting role of the minority electronic charge carrier, electrons. Cathodes with different concentrations of the donor La (0le;xle;0.5) were fabricated by pulsed laser deposition. Their electronic structure was studied through defect chemical modeling, on the basis of thermogravimetric analysis and electrical conductivity, and by optical absorption measurements. Surface chemistry was studied spatially by TEM-EDS studies of FIB cross-sections and in the top monolayer by low energy ion scattering (LEIS). The oxygen exchange rates of as-deposited cathodes could be explained by the changes in bulk defect concentrations with increasing La concentration - in particular, a slight increase in Fermi level and electron concentration counterbalanced by a large drop in hole and oxygen vacancy concentrations. The performance degradation over time could be correlated with changes in the surface chemistry, which in turn was affected by not only the bulk cathode chemistry but also the electrolyte substrate. Here, both bulk film and surface chemistry are related to the surface exchange rate, thereby contributing to a better understanding of the surface oxygen exchange process.

Sr-doped lanthanum manganite (LSM) is a widely used cathode material in commercially produced solid oxide fuel cells (SOFC). Despite being a poor ion conductor, LSM electrodes may reduce oxygen via different pathways: a path which includes surface diffusion of oxygen species (surface path) and a path based on oxygen bulk diffusion (bulk path). The relevance of each path can be expected to depend on geometry and microstructure, temperature, overpotential and partial pressure. However, separation of effective reaction rates on LSM cathodes into contributions of each path is experimentally nontrivial. Hence, a detailed knowledge of the rate limiting steps and their dependence on experimental parameters is still missing.
In this contribution several different methods are employed to identify, visualize and modify the oxygen reduction paths of (La0.8Sr0.2)MnO3 and (La0.8Sr0.2)0.95MnO3 thin films and thin film microelectrodes deposited by pulsed laser deposition (PLD):
i) LSM films on strontium titanium oxide (STO) and yttria-stabilized zirconia (YSZ) with different
microstructures, from epitaxial to fine columnar textured, were investigated by 18O tracer diffusion. Numerical analysis allowed to separate surface resistance and bulk diffusion properties of both, grains and grain boundaries.
ii) Material parameters calculated from 18O depth profiles were compared to results gained from impedance spectroscopy measurements on microelectrodes.
iii) 18O incorporation upon cathodic bias visualizes different reaction pathways and their changing contributions for microelectrodes under different measurement conditions.
iv) Current-voltage studies on microelectrodes with variation of geometry, bias and oxygen partial pressure allows separating reaction pathways, identifying the rate limiting step and gives information on the current voltage characterization of different kinetic steps.
The combination of these complementary tools under various operation conditions thus lead to a substantially improved understanding of the oxygen reduction kinetics on LSM thin films.

We have developed a robotic instrument that can measure the electrochemical impedance of hundreds of thin film electrodes in automated fashion. By measuring electrodes with systematically varied area, thickness, surface decoration, and composition, this instrument can probe reaction pathways, decouple bulk and surface properties, and rapidly screen hundreds of chemical compositions to discover trends and identify new high performing catalysts. Here we introduce the instrument by using gradients in area and thickness to decouple bulk and surface properties in thin films of two common solid oxide fuel cell cathode materials, (La0.8Sr0.2)yMnO3+δ (LSM) and La0.5Sr0.5CoO3+δ (LSC).

Oxygen electrocatalysis is central to the efficiencies of direct solar and electrolytic water-splitting devices, fuel cells, and metal-air batteries. Late transition metal oxides have shown high activity for the oxygen reduction and oxygen evolution reactions (ORR/OER).1-4 However, the lack of fundamental understanding of the oxide surfaces limits mechanistic understanding at the molecular level. Establishing a fundamental catalyst “design” principle” that links bulk electronic structure, surface structure and chemistry to the catalytic activity can guide the search for highly active catalysts that are cost effective and abundant in nature. Although activity descriptors such as eg occupancy and O p-band center, have been shown recently to correlate with ORR/OER activity,2,3,5,6 such descriptors have not been measured directly from experiments. In this study, we will present synchrotron X-ray methods to correlate the measured occupied and unoccupied electron density of states of valence and conduction bands to ORR/OER activities, discuss how surface sensitive techniques such as ambient pressure X-ray photoelectron spectroscopy measurements and Coherent Bragg Rod Analysis can reveal cation segregation and the changes of oxide surfaces as a function of temperature, and oxygen/water partial pressure, which provide insights into reaction mechanisms and active sites that govern the activity of oxygen electrocatalysis.
References
1. J. Suntivich, H. A. Gasteiger, N. Yabuuchi, Y. Shao-Horn. J. Electrochem. Soc. 157, B1263 (2010).
2. J. Suntivich, H. A. Gasteiger, N. Yabuuchi, H. Nakanishi, J. B. Goodenough, Y. Shao-Horn. Nature Chem. 3, 546 (2011).
3. J. Suntivich, H. A. Gasteiger, J. B. Goodenough, Y. Shao-Horn. Science, 334, 1383 (2011).
4. M. J. Kevin, C. E. Carlton, K. A. Stoerzinger, M. Risch, J. Suntivich, Y.L. Lee, A. Grimaud and Y. Shao-Horn, JPCL, 3, 3264 (2012).
5. Y.-L. Lee, J. Kleis, J. Rossmeisl, Y. Shao-Horn, D. Morgan. Energy & Environ. Sci. 4, 3966 (2011).
6. A. Grimaud, K.J. May, C.E. Carlton, Y.L. Lee, M. Risch, W. Hong, J. Zhou and Y. Shao-Horn, Nature Communications, 4, 2439 (2013).

Oxide anode is now attracting much interest because of high stability against reoxidation and coke deposition. Among the several oxide reported as oxide anode, it was found that LaFeO₃ doped with Sr and Mn for La and Fe site respectively (denoted as LSFM) shows reasonable high activity to anodic reaction. In this study, application of LSFM for anode of the solid oxide fuel cell using LaGaO₃ based oxide ion conductor. Although three phase boundary was essential for anode reaction, we found that two phase boundary, i.e., gas phase/LSFM, shows reasonable high activity to electrochemical oxidation of H₂ and C₃H₈. For improving anodic performance, mixed electric and oxygen ionic conductor La(Sr)Fe(Mn)O₃ with a thin film anode was fabricated via pulsed laser deposition. Although La(Sr)Fe(Mn)O₃ is a dense film, the anodic overpotential could be much decreased by deposition of La(Sr)Fe(Mn)O₃ film resulting in the improved power density (ca.3W/cm² at 973K). Mechanism for the decreased anodic overpotential was studied by impedance measurement and tracer diffusion.

Transition metal cations can exsolve from a host oxide lattice under reducing conditions, and this process can be used to decorate the oxide surface with catalytically active metal nanoparticles. Upon oxidation, the exsolved particles re-dissolve into the lattice for regeneration of nanoparticles in the following reduction step. This reversibility makes them attractive for any catalytic application where catalyst coarsening is of concern. ABO3 perovskites are natural choice for host material since they provide wide-ranged tunability in nonstoichiometry and solubility for many metal cations. In solid oxide fuel cell community, many perovskite oxides are also considered as candidates to replace conventional nickel-YSZ (yttria stabilized zirconia) composite anode owing to their appropriate transport property. Accordingly, perovskite fuel cell anodes with exsolution particles were tested and showed great anti-coarsening property [1]. In addition, it has been shown that one can, by redox cycling, recover the initial fuel cell anode performance after performance degradation [2, 3]. In this presentation, we rely on microstructure images and powder X-ray diffraction patterns to investigate the exsolution particle-host lattice interaction for better design of electrochemical catalyst.
References [1] B.D. Madsen, W. Kobsiriphat, Y. Wang, L.D. Marks, S.A. Barnett, Journal of Power Sources 166 (2007) 64-67 [2] David M. Bierschenk, Elizabeth Potter-Nelson, Cathleen Hoel, Yougui Liao, Laurence Marks, Kenneth R. Poeppelmeier, Scott A. Barnett, Journal of Power Sources 196 (2011) 3089-3094 [3] Tae Ho Shin, Yohei Okamoto, Shintaro Ida, Tatsumi Ishihara, Chem. Eur. J., 18 (2012), 11695-11702

Solid oxide fuel cells (SOFCs) have attracted much attention because of their good fuel flexibility and high efficiency. Many efforts have been devoted to developing functional anodes to obtain a better output in methane fuel. Ni/Samaria-doped ceria (SDC) is one of candidates for achieving methane fuel-SOFC. We have reported that only 20wt% of Ni content was optimal composition for achieving the best anodic performance. In the present study, we wish to report how the particle size of Ni supported on SDC affects anodic properties of SOFC with hydrogen and methane fuels.
SDC was prepared by the ammonia co-precipitation method, followed by pre-calcination at 623-1723 K for 4 h. The impregnated NiO/SDC samples were deposited on scandium-stabilized zirconia electrolyte by EPD technique and then reduced by hydrogen to yield Ni/SDC anode before SOFC measurements.
Surface area of the SDC powders was regulated with pre-calcination temperature. The averaged Ni particle size on SDC was affected with surface area of the SDC supports. This means that we can control the particle size of Ni by changing surface area of SDC. The maximum power density increased with decreasing the averaged Ni particle size of Ni/SDC anode. The relationship between maximum power density and Ni particle size was theoretically analyzed. From these analyses, it was concluded that the maximum power density for hydrogen fuel was controlled by the surface area of Ni and it for methane fuel was controlled by the TPB length.

Sintering of nickel nanoparticles during the long-term operation of solid oxide fuel cells (SOFCs) leads to a degradation of Ni/YSZ and Ni/ScSZ anodes. To improve the performance of the anode, understanding of the sintering mechanisms in Ni/YSZ and Ni/ScSZ is necessary. Experimental methods are difficult to fully understand the sintering mechanism, because the atomic forces play roles in the sintering process. In this study, we investigated the sintering processes and the degradation induced by the sintering in Ni/YSZ and Ni/ScSZ by using molecular dynamics simulation method. Since Ni/YSZ and Ni/ScSZ consisting of nickel and ceramic (YSZ and ScSZ) nanoparticles are porous structure, a multi-nanoparticle model methodology [1] is used to model the Ni/YSZ and Ni/ScSZ anodes. In the simulation of Ni/YSZ, the nickel nanoparticles contact with each other, indicating that the sintering takes place. Then, the contact area increases with the sintering. In the simulation of Ni/ScSZ, the contact area is smaller than that of Ni/YSZ, indicating that the degree of the sintering of nickel nanoparticles in Ni/YSZ is larger than that in Ni/ScSZ. Next, we investigate the degradation by assessing the exposed surface area of nickel nanoparticles (ESA) and the triple phase boundary (TPB) length. We find that the ESA and the TPB decrease when the sintering of nickel nanoparticles takes place. The decrease in the ESA and the TPB in Ni/ScSZ is smaller than that in Ni/YSZ. It implies that the decrease in the anodic activity of Ni/ScSZ is small compared to that of Ni/ScSZ. Thus, it is understood that the degradation of Ni/ScSZ anode is smaller than that of Ni/YSZ anode. We successfully revealed the sintering processes and the degradation induced by the sintering in Ni/YSZ and Ni/ScSZ. [1] J. Xu et al., J. Phys. Chem. C 117 (2013) 9663.

Partial oxidation of methanol (POM) was investigated over Au/CeO2 and Au/CeO2-modified ZrO2 catalysts under different pretreatment conditions. FTIR spectra showed that formate and carbonate intermediate species were formed on the surface of the catalysts during POM reaction, which were finally transformed to CO2 and hydrogen. ZrO2-CeO2 solid solution was formed in the Au/CeO2-ZrO2 catalyst which promotes the reducibility of the catalyst and improves the catalytic activity and hydrogen selectivity. The pre-treatment of the catalysts plays a key role influencing the catalytic activity and product selectivity. Under reductive condition, the catalysts exhibit better catalytic activity and higher hydrogen selectivity than that under oxidative condition. Formaldehyde, formix acid, methyl formate and dimethyl ether were not detected in the reaction products.

Mixed-valent oxides, such as ferrites, cobaltites, and ceria, are employed in the state-of-the-art solid-oxide fuel cells. Over the past few decades, the community has developed an extensive understanding of the bulk nature of these mixed-valent oxides. However, as it is appreciated in surface science, the surface often behaves very differently from the bulk. In this talk, I will show several examples of unusual surface behaviors for metal oxides used for solid-oxide fuel cell electrodes. In particular, I will discuss electron localization, defect formation energy, and reconstruction on the surfaces of transition-metal and rare-earth oxides.

Mixed-conducting SOFC anodes may overcome some problems of the currently most used Nickel-YSZ cermets, such as carbon deposition, sulfur poisoning and stability in oxidizing atmosphere. Mechanistic studies on geometrically well-defined model anodes are, however, scarce but crucial for a better understanding of the reaction mechanism and the origin of the polarization resistance. The electronic p-type conductivity of acceptor-doped mixed conductors decreases dramatically in reducing atmosphere. This leads to a complex interplay of charge transport and electrochemical processes that contribute to the electrode impedance, even for geometrically well-defined thin film electrodes. In this contribution, a novel electrode design with applied metallic thin film current collectors was employed to study all relevant charge transport and electrochemical processes of Ce0.8Gd0.2O1.9-δ (GDC) and SrTi0.7Fe0.3O3-δ (STFO) thin films. These films were deposited on single crystalline YSZ substrates by pulsed laser deposition and microelectrodes were obtained from the films by photolithography. Two separated but interdigitating current collectors were applied to each microelectrode, which allows two different electric connection modes for impedance measurements: In one mode, the impedance between the two current collectors on one microelectrode is measured (in-plane measurement). In the other mode, both current collectors are at equal potential, and the impedance is measured against a macroscopic counter-electrode. This second mode is similar to conventional electrochemical impedance spectroscopy. Both measurement modes can be described by transmission line-type equivalent circuits. The impedance spectra of both measurement modes can be fitted in one run with a single parameter set, and all relevant resistive and capacitive processes can be quantified. Electronic and ionic conductivity as well as the area specific resistance of the surface reaction, resistive contributions from the MIEC/electrolyte interface and the chemical capacitance can be determined on one and the same electrode. The local polarization of the MIEC film and therefore the rate of electrically driven oxygen exchange decreases with the distance from the current collector due to limited electronic or ionic conductivity. This could be visualized by bias-driven tracer experiments with 18O enriched water and SIMS analysis and further validates the interpretation of the impedance measurements.

Solid oxide fuel cells (SOFCs) are promising systems to provide electricity at a large scale stationary power generation setting, typically larger than 100 kW. However, since the SOFC systems are required to operate at an elevated temperature of typically around 850°C or higher with high current density and for at least 40 thousands hours, their long-term degradation issues became pronounced. Specifically in one of the most promising anode materials, Ni-YSZ, the coarsening of Ni has attracted great attention for it attributes to the decrease in the cells&’ triple-phase-boundary (TPB) density and then the efficiency of the cells. In the previous studies, different SOFCs are usually subjected to various amount of iso-thermal heat treatment to observe the degradation process at the various stages under the elevated temperature environment. As a result, the sample variation is likely to attribute to the morphology difference in characterization.
Here we first conducted a serial nano-tomography with transmission x-ray microscopy at National Synchrotron Light Source, to continuously follow the morphological evolution of the same Ni-YSZ sample volume. The 3D electrochemical parameters have been quantified and investigated from the very same sample volume during degradation. In addition to the coarsening of the internal Ni structure as reported previously, a very significant coarsening of Ni at the surface of the Ni-YSZ electrode material was observed. This indicates that the presence of YSZ has a significant impact on the Ni-coarsening behavior. The quantitative 3D morphological analysis results will be discussed.

Solid oxide fuel cells (SOFCs) generate electricity with high electrochemical efficiency and low pollutant emissions, but the required high operating temperatures impose a number of challenges in terms of material performance, compatibility, and durability. In particular, the anode reduction-oxidation (redox) reaction that occurs when the uncontaminated fuel supply is interrupted or the system is shut down leads to rapid oxidation of the metallic nickel phase, inducing large volumetric expansion strains. Components of these heterogeneous strains are transmitted from NiO regions to the ceramic oxide (YSZ) backbone, potentially leading to micro-cracking or catastrophic brittle failure of the entire cell. A better understanding of the complex chemical and structural evolution processes that occur over various length and time scales during anode oxidation and their roles in overall mechanical and electrochemical degradation is fundamental to designing more robust and efficient future generation SOFCs. In this talk, a model for SOFC anode materials undergoing redox reactions will be presented. Various aspects of the heterogeneous build-up and evolution of local strains will be examined, with the aim of identifying optimal material properties and microstructural features for SOFC durability. This work is supported by the Energy Frontier Research Center on Science Based Nano-Structure Design and Synthesis of Heterogeneous Functional Materials for Energy Systems funded by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences (award DE-SC0001061).

The chemical processes that occur at the anode triple phase boundary (TPB) between Ni, YSZ and fuel molecules is essential in determining solid oxide fuel cell (SOFC) anode performance. With a growing interest in vibrational spectroscopy for studying such processes, we aim to investigate the surface and vibrational properties of the materials Nickel (Ni) and Yttria Stabilised Zirconia (YSZ) using first principles atomistic simulations based on Density Functional Theory (DFT) and potential models. Here we report on our findings and the methodology used to study YSZ.
Yttria Stabilised Zirconia (YSZ) is a high temperature oxide ion conductor, yet despite extensive research, a simple set of chemical descriptors for its material properties have yet to be developed. Zirconia (ZrO2) is an oxide material and the major chemical interactions are expected to be governed by electrostatics, however the technologically relevant polymorphs of ZrO2 exhibit dynamic instabilities through soft vibrational (phonon) modes. Through “data mining” of DFT energetics a simple set of chemical descriptors are developed that describes YSZ as the interaction between electrostatics and soft phonon energetics.

The electrode behavior of oxide based SOFC cathodes is acknowledged to be dependent on their degree of mixed ionic electronic conductivity that, in turn, often depends sensitively on their oxygen nonstoichiometry. Increasing attention is being directed towards the study of thin film oxide cathodes as model systems, given their well defined geometries and controlled compositions and their use in micro fuel cell devices. Investigators often report of major differences in the nonstoichiometry of bulk and thin film oxides of the same nominal composition. Techniques for the measurement of nonstoichiometry of thin films remain limited. In this presentation, we review in-situ optical and electrochemical methods, developed or refined in our laboratory, to monitor and control the defect equilbria of oxide thin films, with PrxCe1-xO2-δ, as a model system. Transient studies provide further insights into surface exchange kinetics. These results are discussed in relation to their electrode behavior.

Building from earlier work demonstrating that incorporation of Mn cations into YSZ causes partial de-stabilization of the cubic stabilized structure, the present work focuses on the reactivity of (La,Sr)MnO3-x with YSZ under conditions of elevated temperature, humidity, and CO2 partial pressure. Reactions were monitored both in-situ, using x-ray diffraction (including synchrotron studies), and post-situ on mixed powders and optimized LSM cathodes on YSZ. Parallel continuous electrochemical monitoring was performed for some reaction conditions to track the changes in cell performance.
Phase stability experiments were performed at 850, 1200 and 1350C to evaluate the feasibility of accelerated testing degradation using temperatures well above the target operating temperatures. The effects of water vapor and 10% CO2 atmospheres were also evaluated. At 1350C, it is possible to track the destabilization of cubic YSZ to a mixture of tetragonal and cubic YSZ using in-situ XRD. The reactions proceed in only a few hours. By comparison of the reactions of LSM and MnO2 with YSZ, it is possible to unambiguously assign the degradation to Mn diffusion into YSZ. Parallel studies are in progress to evaluate LSCF and related cathode materials.
Role of cathodic bias on the chemical, structural and morphological changes were investigated using symmetric cell of configuration LSM//YSZ/ /LSM. Enhanced cathode degradation was observed under the cathodic bias. Mechanism responsible for above changes has been developed.

Layered perovskite type oxides are promising cathode material for SOFC, because of their high oxide ionic and electronic conductivity. In this study, defects species and crystal structure of layered perovskite type oxides, La2(Ni0.9,M0.1)4+d (M=Fe, Co, Cu), were evaluated. The effect of doping on defect structure and structural change due to defect formation will be discussed in the presentation.
La2(Ni0.9,M0.1)4+d (M=Fe, Co, Cu) were synthesized by solid state reaction method. Interstitial oxygen concentration and crystal structure were evaluated by high temperature gravimetry and in-situ X-ray diffractometry at 873 to 1173 K in 1 x 10^-4 to 1 bar oxygen atmosphere.
La2(Ni0.9Fe0.1)O4+d and La2(Ni0.9Fe0.1)O4+d showed higher interstitial oxygen concentration than non-doped La2NiO4+d, while La2(Ni0.9Cu0.1)O4+d showed lower interstitial oxygen concentration than non-doped La2NiO4+d. The difference of defect structure may significantly affect the electrochemical activity towards oxygen reduction reaction.
According to the in-situ XRD results, lattice parameter along a axis decreases, that along c axis increases and the cell volume slightly increases as interstitial oxygen concentration increases for specimens studied in this work. Temperature and chemical expansion coefficients were apparently constant. Linear expansion model can explain the temperature and chemical expansion behavior of La2(Ni0.9,M0.1)4+d very well.

Acceptor doping of oxides generally results in the formation of charge-compensating oxygen vacancies, leading to enhanced ionic conductivity. Ca-doped LaFeO3 is one such example which has been considered for use in cathode materials of solid oxide fuel cells. In the present work, we determine the structural, electronic, and magnetic properties of individual intrinsic and extrinsic point defects and defect clusters in Ca-doped LaFeO3. We find that the concentration of oxygen vacancies can increase through the formation of dopant-vacancy clusters. Oxygen vacancy concentration will influence the diffusivity of oxygen which in La perovskites is shown to be governed by the oxygen vacancy concentration. We also calculate the interaction between dopant ion and oxygen vacancies and find that oxygen vacancies will not be trapped by the dopant species as found in other oxide materials. These findings suggest an enhanced conductivity of oxygen in LaFeO3 under Ca doping.

La_xSr_1-xCo_yFe_1-yO_3-δ (LSCF) is of interest for application as a cathode material in solid oxide fuel cells (SOFCs) because of its mixed ionic and electronic conduction behavior, and its activity for reducing oxygen under an electrochemical potential. We are performing in-situ grazing incidence x-ray scattering and spectroscopy studies at the Advanced Photon Source (APS) on epitaxial LSCF films of composition x=0.6 and y=0.2. Films are grown on Y₂O₃-stabilized ZrO₂ (YSZ) (001) substrates with an intermediate layer of GdO₂-doped CeO₂ (GDC). Our focus is on identifying and understanding the structural and conduction changes that occur during elevated temperature exposure of films to electrochemical potentials at various oxygen partial pressures. We observe large and rapid changes in the LSCF lattice parameter normal to the film surface when small DC potentials (on the order of +/- 1 V) are applied between the film surface and the back surface of the substrate, and also observe transient behavior of the electrical current flow through the sample. The behavior is consistent with oxygen exchange at the LSCF/atmosphere interface being the rate limiting process in transporting oxygen through the heterostructure under both anodic and cathodic potentials. The effects of controlled carbon dioxide and water vapor exposure of the LSCF surface on oxygen exchange behavior will also be discussed.

In solid oxide fuel cells (SOFC) and electrolyzer cells (SOEC), the oxygen reduction and oxygen evolution reactions, respectively, often limit the device efficiency. The chemical versatility of the perovskite structure provides a vast toolbox to tailor the composition of oxides to desired electrochemical properties, such as ionic and electronic conduction and surface reactivity. It has been shown that the bulk stoichiometry of the compounds impacts strongly their electrochemical activity.[1] Attempts were also made to identify simple activity descriptors across a wide bulk compositional range.[2] However, it is the oxide surface - which can differ significantly from the bulk - where these reactions take place.
To understand directly the evolution electronic structure of perovskites-oxide surfaces under reaction conditions, we employed surface-sensitive and element-specific techniques including X-Ray absorption and photoelectron spectroscopy at close to operating conditions[3]. We systematically investigated model thin-film electrodes of the general formula (La, AE)(TM)O_3-δ, where AE = alkali earth ions and TM = first-row transition metal ions under reaction conditions.
We found that near surface lattice oxygen is electrochemically active during the oxygen reduction and evolution reactions. Our result suggest that transition metal and oxygen redox chemistry simultaneously control the electrochemical activity of perovskites oxides. This new framework gives an improved understanding on why certain complex perovskite compositions yield the highest electrocatalytic activities.

Solid oxide fuel cells (SOFCs) are one of promising high-efficient energy-conversion devices. SOFCs are now in commercialized stage, but their performances are still required to be improved for their full-scale commercialization. It is pointed out that the practical performance of SOFC systems is frequently determined by the performance of the electrodes, especially the cathodes. Therefore, enhancement of oxygen reduction kinetics at the SOFC cathodes is a key challenge to improve the performance of SOFC systems.
To achieve high performance of SOFC cathodes, it is important to understand the electrode reaction mechanisms, and to optimize the materials and the structures. When a mixed ionic-electronic conducting oxides, such as (La,Sr)CoO3 and (La,Sr)(Co,Fe)O3, is used as a cathode material, it is believed that electrochemical oxygen reduction reaction occurs not only at the electrode/electrolyte interface but also further inside the porous electrode. The effective reaction area in a porous mixed-conducting oxide electrode is considered to be determined by the ration between the resistances of the surface reaction at the oxide surface and of the oxide ion diffusion in the oxide. Based on this idea, many numerical studies have been conducted to evaluate the effective reaction area in SOFC electrodes. However, as far as we know, there exist no experimental studies to directly evaluate the effective reaction area in SOFC electrodes.
Our group has succeeded to develop the in situ micro XAS technique, which enables us to measure X-ray absorption spectra with the position resolution of micron meter under SOFC operating conditions, i.e. at elevated temperature, in controlled atmosphere and under polarization. In this work, we applied this novel in situ spectroscopic technique to experimentally determine how the reaction area is distributed in SOFC cathodes. A porous and a patterned thin film electrodes of (La,Sr)CoO3 were chosen as model SOFC cathodes. From the shift of the absorption energy in Co K-edge XAS spectra, we could evaluate the oxidation state of Co ions in (La,Sr)CoO3, that is, the oxygen chemical potential to which the oxide is exposed. The effective reaction area was then clarified from the distribution of the oxygen chemical potential. The results suggested that the effective reaction area in a porous (La,Sr)CoO3 electrode distributed only in the vicinity of the cathode/electrolyte interface, a few micron meters from the interface, and the most part of the porous electrode did not contribute to the electrode reaction.

Mixed ionic and electronic conducting (MIEC) perovskites like La_0.58Sr_0.4Co_0.2Fe_0.8O_3-δ (LSCF) are reported as best performing cathode materials for intermediate temperature solid oxide fuel cells (SOFCs). However, little is known about durability of phase composition and microstructure and its influence on cell performance.
This contribution couples both characteristics combining electrochemical impedance measurements (EIS) and investigations by focussed ion beam (FIB) tomography. Performance of anode supported cells was evaluated by high resolution EIS studies at temperatures of 600, 750 and 900 °C for 1000 h, and ohmic and polarization losses arising from electrolyte, anode and cathode were separated. The cathodic polarization resistance ASR_cat show a remarkable high degradation rate (e.g. at T = 750 °C, ASR_cat increases by 0.286 % / h within a period of t = 1000 h). The coefficients k^δ and D^δ of the LSCF cathode, resolved versus measurement time and temperature, were calculated using (i) the Gerischer impedance and (ii) the FIB tomography data. Obviously both, the exchange of oxygen with the gas phase (described by the surface exchange coefficient k^δ) and the transport of oxygen ions (described by the chemical diffusion coefficient D^δ) are changing over measurement time. Based on these results, durability issues of the LSCF cathode are discussed.

The study of the kinetics of oxygen exchange in transition metal oxides is of increasing relevance for understanding the fundamental mechanisms for oxygen reduction and evolution reactions (ORR/OER) of oxide catalysts as well as for novel cathode materials for intermediate-temperature solid oxide fuel cell (SOFC) technology.
In most transition metal oxide materials the variation in the oxygen stoichiometry produced by oxygen exchange with the atmosphere is often accompanied by subtle cell volume changes. Generally, the incorporation of charged defects (oxygen vacancies or interstitials) is expected to cause a cell expansion from the equilibrium intrinsic cell because of variations in the radii of the transition metal ions as well as Coulomb repulsion between those defects. Any slight variation in the oxygen content would proportionally result in a measurable volume increase. The ability to dynamically follow the cell volume variations allows establishing a direct correlation with oxidation/reduction kinetics.
In this talk we will describe recent development of a new methodology to analyze oxygen surface kinetics in epitaxial thin films by using X-ray diffraction time-resolved measurements in conventional lab diffractometer [1]. With this technique we have been able to measure average cell parameter changes as small as 0.1 ppm in a time scale of a few seconds in a variety of SOFC cathode materials like La2NiO4+d, GdBaCo2O5+d and Ba0.5Sr0.5Co0.8Fe0.3O3-d, as well as in epitaxial bilayers. We will describe the usefulness of this methodology as complementary to other well-established techniques such as Electric Conductivity Relaxation (ECR) and 18O Isotopic Exchange Depth Profiling (IEDP) experiments.
[1] R. Moreno, P. García, J. Zapata, J. Roqueta, J. Chaigneau and J. Santiso, Chem. Mater. 25 (2013) 3640-3647

The performance of Solid Oxide Fuel Cells (SOFCs) depends critically on the ability of the perovskite oxide cathode to catalyze the oxygen reduction reaction, transforming O2 (gas) into solid phase O2-. In this talk we discuss how the rate of this reaction can be predicted from first-principles methods. We first consider a simple descriptor approach, where we find that the bulk oxygen 2p bands strongly correlate with surface exchange, area specific resistance, and a number of molecular scale properties (e.g., oxygen vacancy formation and migration energies). We then describe a detailed prediction of the surface exchange rate for (La,Sr)CoO3, a widely explored cathode material. Through a comprehensive search of different reaction paths and consideration of absorption rates from the gas we predict a specific surface exchange path that shows good agreement with experiments for activation energy, charge transfer steps, rate limiting step, and overall predicted surface exchange rate. This study suggests that our understanding in this class of material is progressing to the point where full molecular prediction of the catalytic rates on general systems may soon be possible.

Among advanced materials for clean energy, non-stoichiometric BaxSr1minus;xCo1minus;yFeyO3minus;δ (BSCF) and LaxSr1minus;xCo1minus;yFeyO3minus;δ (LSCF) are considered as promising materials for cathodes in solid oxide fuel cells (SOFC) and oxygen permeation membranes [1-3]. BSCF exhibits the best oxygen exchange performance among similar materials, mixed ionic and electronic conductivity, high oxygen vacancy concentration, and low diffusion activation barrier, which largely define the oxygen reduction kinetics. However, it tends to decompose at low temperatures into a mixture of cubic and hexagonal perovskite phases, which strongly affects its use.
To understand the mechanism(s) of this unwanted process, the first principles quantum mechanical calculations of BSCF and LSCF crystals with different non-stoichiometry were performed and possible decomposition scenarios were studied. It is shown [3,4] that formation energies of oxygen vacancies in the cubic and hexagonal phases of BSCF differ considerably and also behave in quite different ways depending on non-stoichiometry; in fact, it is the oxygen non-stoichiometry that makes the cubic phase more stable than the hexagonal phase. In comparison, LSCF is shown to be much more stable with respect to both the phase transformation and phase decomposition. The first principles calculations are accompanied with a thermodynamic analysis of the conditions under which a cubic phase is stable, in good agreement with experimental data.
[1] L. Wang et al, J. Mater. Res. 27, 2000 (2012).
[2] M. Kuklja, et al, Phys. Chem. Chem. Phys. 15, 5443 (2013).
[3] Yu. Mastrikov et al, Phys. Chem. Chem. Phys. 15, 911 (2013).
[4] M.M. Kukla et al, J Phys Chem C 116, 18605 (2012).

ABO3-x type perovskite oxides are potential candidates for solid oxide fuel cells. Among them those which crystallize in the orthorhombic brownmillerite-phase (ABO2.5), such as SrCoO2.5, are particularly interesting due to their promising crystal structure which contains ordered channels of oxygen vacancies. Recent experimental work [1,2,3] successfully demonstrated that the two phases, perovskite SrCoO3 and brownmillerite SrCoO2.5, could be reversibly switched without destroying the parent framework. In this work, we complement the experimental studies with first-principles calculations. The strong correlations in the Co d orbitals are treated within the local spin density approximations of Density Functional theory (DFT) with Hubbard U corrections (LSDA+U), U being an empirical parameter. We also compare our results with the Heyd Scuzeria Ernzerhof (HSE) functional. We first perform a theoretical investigation of the electronic structure and magnetic ground state properties of SrCoO2.5, which is interesting due to variety of valence and spin states of Co. Consistent with experimental observation, the G type antiferromagnetic structure is found to be the most stable. By mapping the total energies of different magnetic configurations onto a Heisenberg Hamiltonian, we compute the magnetic exchange interaction parameters J, which are then used to compute the spin-wave frequencies and inelastic neutron scattering intensities. Our study of the structural and electronic properties of SrCoO2.5 allows us to select the parameter U used to perform the calculations of oxygen diffusion. The diffusion energy barriers are found to be highly anisotropic which is a direct consequence of the oxygen vacancy channels in SrCoO2.5 for fast oxygen transport. The effect of tensile and compressive strain on the activation energy barriers have also been investigated.
[1] H. Jeen, W. S Choi, M.D. Biegalski, C. M. Folkman, I.C Tung, D.D. Fong, J.W. Freeland, D. Shin, H.Ohta, M.F. Chisholm & H.N. Lee; Nature Materials (2013) doi:10.1038/nmat3736
[2] H. Jeen, W. S. Choi, J. W. Freeland, H. Ohta, C. U. Jung, H. N. Lee; Adv. Mater., 25: 3651-3656. doi: 10.1002/adma.201300531
[3] W.S. Choi, H.Jeen, J.H Lee, S. S. A. Seo, V.R. Cooper, K.M. Rabe, and H.N Lee; Phys. Rev. Lett. 111, 097401 (2013)

Solid Oxide Fuel Cells (SOFCs) have demonstrated high efficiency in full scale trials and are a possible partial solution to maximizing dwindling fossil fuel resources. Functionally graded electrodes have previously been investigated to improve SOFCs performance with optimized microstructure. However, little investigative focus has been put on optimal power output for given electrode microstructures. In this work, a multiscale and multiphysics full cell electrode polarization model of SOFCs has been expanded and developed to incorporate both electrodes microstructure. The macromodel describes the overall cell behavior through activation, Ohmic, and concentration losses based on chemical and concentration potentials. The micromodel outputs effective resistivity of the porous electrode based on microstructural parameters such as pore diameter, particle size, and reaction area. The integration of macro- and micromodels is achieved by passing the micromodel parameters to the macromodel during the solution procedure. The cell-level SOFC model has been utilized to reveal the complex relationship between the transport phenomena, which includes the transports of electron, ion and gas molecules through the electrode and the electrochemical reaction at the triple phase boundaries. The work contributes to our understanding of the cell performance in relation to graded microstructures. The performance of functionally graded electrodes has been analyzed to understand the implications of varying the electrode microstructure. The design guidelines provides in the study can aid selection and development of fabrication process that can further improve SOFC performance.

Electrical Conductivity Relaxation (ECR) is a common technique used in the initial characterization of new mixed-conducting electrode materials and structures for solid oxide fuel cells. The advantages of the technique are its relative simplicity and its ability to isolate the behavior of the electrode of interest from that of other components of the fuel cell. The normal procedure is to fit an analytical solution of the parabolic diffusion equation with a linearized chemical reaction-controlled flux at the gas-exposed interface to a time series of electrical conductivity measurements; the result is an estimate for the effective surface reaction rate k* and the effective diffusivity D*. Typically, a straightforward least-squares fit is used to estimate parameters. However, such fit procedures with a nonlinear function in a parameter space that is not guaranteed to be convex can be prone to identifiability problems and the existence of local minima. This work presents a Bayesian approach to parameter estimation in ECR that quantifies uncertainty in parameter estimates and efficiently explores the parameter space with approximately the same computational effort as is deployed in most least-squares fitting routines. Propagation of uncertainty to the scale of electrode behavior is natural, as the method returns not just one, but a distribution of plausible parameter estimates that can be directly incorporated into electrode models. The method can also be used to rationalize -- and quantify uncertainty arising from -- disparate results for the same materials that appear in the literature.

Proton conducting oxide electrolytes offer several advantages for their use in intermediate-temperature fuel cells such as high ion conductivity at moderate temperatures, and water formation at the cathode avoiding dilution of the fuel. Few is known so far about the mechanism of the oxygen reduction reaction mechanism at the cathode, e.g. if typical cathode materials exhibit a sufficient proton conductivity to allow for oxygen reduction to extend beyond the triple phase boundary and to occur on the whole cathode surface ("bulk path"). In contrast to cathodes on oxide ion conducting electrolytes, the oxygen reduction on a proton conductor does not necessarily involve incorporation of the oxygen into the lattice of the cathode material.
For Ba0.5Sr0.5Fe0.8Zn0.2O3-d (BSFZ) ceramic samples the proton concentration and mobility is extracted from thermogravimetry (analyzing also the transient behavior) in different atmospheres (various pO2, pH2O) [1]. The proton uptake is discussed which can either occur by hydration of oxygen vacancies (acid-base reaction), or including a valence change of iron (redox reaction). A comparison of the obtained proton conductivity with the ionic conductivity of (La,Sr)MnO3 films on YSZ indicates that indeed the bulk path for oxygen reduction is possible on BSFZ cathodes. This question is further investigated with the help of impedance spectroscopy on dense thin-film microelectrodes on a Ba(Zr,Y)O3-d electrolyte.
When measurements with Ba(Zr,Ce,Y)O3-d proton conductors as the electrolytes are made with high pO2 on both sides (in contrast to a real fuel cell with the anode exposed to H2), the nonnegligible hole conductivity in Ba(Zr,Ce,Y)O3-d in oxidizing atmosphere has to be taken into account. This can offer a short-circuit in parallel to the "Faradaic path" comprising the oxygen reduction resistance, and thus pretend a too high oxygen reduction activity [2]. While this effect is most severe for pore-free cathode films, it can still have some effect for porous cathodes, and it will affect also the apparent pO2, pH2O dependence of the oxygen reduction resistance.
[1] D. Poetzsch, R. Merkle, J. Maier, submitted
[2] D. Poetzsch, R. Merkle, J. Maier, J. Power Sources 242 (2013) 784

Perovskite-type structures based on transition metals have been the object of intense scientific research, after the first report 40 years ago, mainly for their electronic and magnetic behaviour. Materials with this structure are commonly used as interconnection materials, as gas sensors, as catalysts and extensively as low resistive electrodes in solid oxide fuel cells (SOFC). It was proposed for the first time as a potential SOFC cathode material by Hrovat et al. The perovskite-type LaNiO3, exhibits a high electronic conductivity at room temperature but is instable above 850 C. Its performances were improved by doping the structure with cobalt. Later, Chiba et al., in a set of experiments with LaNiMO3 [M = Al, Fe, Ga, Cr, Mn, Co], found that cobalt had the highest conductivity at 812 C, just followed by iron, while the rest of the elements produced conductivity values of up to 2 orders of magnitude lower.
Several methods exist to prepare perovskite materials, some by the solid state route, such as, mechanical milling, powders calcined in air, or by wet chemistry such as; spray-freeze-drying, spray-drying, surfactants, Pechini, sol-gel, reverse micelle, co-precipitation, flash microwave and polymerization. The perovskite lattice can tolerate multiple cation substitution with small changes; that means many properties can be exploited to obtain the desired material without adding too much distortion to generate an instable structure. The modified Sol-Gel route has several advantages, these being of the main importance: homogeneity can easily be achieved, control over the morphology, submicrometrical particles and a specific composition can be obtained.
Perovskite-type structures with the composition LaNixCo1-xO3 (x= 0.3, 0.5, 0.7) have been synthetized by modified Sol-Gel method. To produce different perovskite materials via a modified Sol-Gel, metallic salts (MS) and complexing agents (CA) with a CA/MS rate of 3 are used. The powders obtained are calcined at 1000 °C over a period of 5 hours to form the crystalline structure. HT-XRDX evidence 350 °C as the beginning point of the perovskite phase formation. Scanning Electron Microscope images show the microstructure changes (grain size) due to the amount of dopant (Cobalt). The influence of composition on the structure was investigated; finding out that LNC37 has an open structure with rounded grains (307 nm), while LNC73 (432 nm) has a more sintered structure with less open pores and showing grains with planar faces. LNC55 has an intermediate structure (416 nm) showing the possibility of controlling the microstructure through the composition of the material, opening the possibility to design a material with the desired properties.

In recent years, the infiltration method has become a popular approach for preparation of composite electrodes for solid oxide fuel cells (SOFCs) and good performances obtained by incorporating nanoparticulate catalysts via infiltration have been reported. These cathodes have very low polarization resistance and are promising candidates for use in low and intermediate temperature (500 -700 °C) SOFCs. In terms of overall mechanical stability under thermal cycling, infiltrated cathodes appear to be more robust than conventional composite cathodes; the structure of the infiltrated cathodes where the electrocatalyst form a thin nano-particulate layer covering the surface of a porous and rigid backbone that adheres well to the electrolyte provides a better balance between achievable strengths and stresses arising from material TEC mismatches.
It is one of the aims of the EUDP project “SOFC accelerated” funded by the Danish Energy Agency to further develop such electrodes. Here, the development of a porous backbone cathode structure (co-fired with the half-cell) with a “post firing” infiltration of the electro-catalytically active cathode material is pursued.
In this presentation the i) stability of the porous cathode backbone structure and ii) the electrochemical performance of the infiltrated cells will be discussed.
Thermo-mechanical properties of porous YSZ (yttria stabilized zirconia) and ScYSZ (scandia and yttria co-doped zirconia) layers were characterized by means of the combined use of optical dilatometry, (thermo-) mechanical analysis and scanning electron microscopy. The results from the different techniques were found complementary and can be used to understand the different properties of individual layers/interfaces.
Several electro-catalytic active materials (La_0.6Sr_0.4CoO₃ (LSC), LaCo_0.6Ni_0.4O₃ (LCN), La_0.6Sr_0.4FeO₃ (LSF)) were infiltrated and screened as cathodes in symmetrical cells. The most promising materials were tested in anode supported SOFCs under realistic operation conditions. Next to a detailed electro-chemical performance, the stability and reactions occurring at the interface of the cathode backbone and the infiltrate will be presented.

Solid oxide fuel cells (SOFCs) are a promising technology for the efficient conversion of hydrocarbon fuels to electricity. However, the long term operation of SOFC systems is affected by a number of stability issues. Among these is the Cr poisoning of the state-of-the-art (La,Sr)MnO₃ cathode by volatile Cr species originating from Cr-containing interconnects. Thermodynamic calculations have shown that one of the mechanisms of Cr poisoning is the substitution of Cr for Mn on the B-site of (La,Sr)MnO₃ at high temperatures.¹ To elucidate the possible effects of such a Cr poisoning mechanism within (La,Sr)MnO₃ SOFC cathodes, we measured the electronic conductivity and electrochemical oxygen reduction activity of the model series La_0.8Sr_0.2Cr_xMn_1-xO₃ (x = 0, 0.02, 0.05 or 0.10) in air over the temperature range 650-850 °C.
The results of X-ray diffractometry (XRD) of powders prepared by solid state synthesis were consistent with the formation of a rhombohedral perovskite for all compositions. Electronic conductivity (σ) was measured on dense pellets using the van der Pauw method and followed a small polaron hopping mechanism.² ln(σT) vs. 1000/T plots revealed that Cr substitution resulted in lowered σ (for example from 174 S cm^-1 (x = 0) to 89 S cm^-1 (x = 0.10) at 850 °C) and raised activation energy (E_a) (from 0.105 eV (x = 0) to 0.139 eV (x = 0.10)). These effects were attributed to decreases in the number of available Mn B-sites for thermally activated hopping of charge carriers.
Electrochemical oxygen reduction activity was probed by carrying out electrochemical impedance spectroscopy (EIS) on symmetrical cells composed of a dense yttria-stabilised zirconia (YSZ) electrolyte and composite (La,Sr)(Cr,Mn)O₃/YSZ electrodes. Compared to the x = 0 composition, significant rises in ohmic (R_s) and polarisation (R_p) resistances were observed upon Cr substitution, although no clear trends in R_s or R_p were observed with extent of Cr substitution. The rise in R_s was attributed to lowered electronic conductivity, while the rise in R_p was likely due to retardation of oxygen reduction reaction (ORR) processes following the substitution of catalytically active Mn B-site cations with less catalytically active Cr cations. Overall E_a for the ORR was independent of x and ranged between 1.41-1.49 eV.
The La_0.8Sr_0.2Cr_xMn_1-xO₃ model series was also characterised using laboratory-scale X-ray photoemission spectroscopy (XPS), and near-edge X-ray absorption fine structure (NEXAFS) spectroscopy and resonant photoelectron spectroscopy (RESPES), which were carried out at a synchrotron light source. Electronic structure information obtained using these X-ray spectroscopic methods was related back to electronic conductivity³ and electrochemical oxygen reduction activity results.
1. Yokokawa, H., et al, Solid State Ionics, 177 (2006) 3193.
2. van Roosmalen, J.A.M., et al, Solid State Ionics, 66 (1993) 279.
3. Braun, A., et al, Appl. Phys. Lett., 94 (2009) 202102.

Predicting and improving a solid oxide fuel cell&’s performance over a lifetime of thousands of hours requires an understanding of a multitude of materials degradation processes that are intrinsic and extrinsic to the normal operation of the fuel cell system. The microstructure and composition of electrode and electrolyte materials can change due to the static or dynamic operating conditions or from the presence of contaminants in the fuel or materials supply stream. Many of the processes, such as cation segregation to the cathode surface or across the electrolyte/cathode interface, have been studied in more ideal samples (e.g., thin films) that are easier to analyze structurally and chemically; however, the nature of those samples makes it difficult to directly link the observed changes in the sample to changes in the electrochemical performance in real fuel cell samples. In this study, a more direct connection is established between changes in the at the composition and microstructure of the cathode/air interface of La0.8Sr0.2MnO3 (LSM) electrodes and changes in the electrode&’s performance while the electrode is thermally cycled at open circuit between 700°C and 850°C. Electrochemical impedance spectroscopy measurements of porous LSM electrodes demonstrated an aging effect, on the scale of tens of hours, on the electrode&’s polarization resistance as the sample was thermally cycled within this temperature range. The surface of dense pellets of the same electrode material exposed to the same thermal cycling conditions were then analyzed by SEM, TEM, and XPS to identify the corresponding reversible changes at the LSM surface. The results indicate that within the normal operating conditions of an SOFC, there are thermodynamic and physical driving forces on the electrode surface that must be considered when analyzing electrode degradation.

La₂NiO₄ and related materials with the K₂NiF₄ structure are known to support high electronic conductivity (about 100 S*cm^-1) and oxygen ionic conductivity (comparable to or exceeding that of yttria-stabilized zirconia). Unlike other well known oxygen ion conductors, these materials conduct via an interstitial rather than a vacancy mechanism. While the K₂NiF₄ crystal structure is robust across a fairly wide solid solution compositional space, nearly all known compositions exhibit mixed ionic electronic conduction due to one or more transition-metals on the B-site. One exception, LaSrAlO₄, was previously found to be unstable to the introduction of ionic defects. For these reasons, despite high oxygen ion conductivity, oxides with the K₂NiF₄ structure have been limited in application as electrode and not electrolyte materials in solid oxide fuel cells (SOFC). In this work, we aimed to develop a new type of SOFC electrolyte material from transition-metal-free oxides with the K₂NiF₄ structure. The first successful composition, La_1.6Sr_0.4Al_0.4Mg_0.6O₄, was created by solid-state reaction. Oxygen ion defects were able to be created in the lattice by adjusting the La/Sr ratio. We hypothesize that defects are stabilized in this composition but not LaSrAlO₄ due to higher charge separation between the rocksalt and perovskite layers in the crystal lattice. Similar compositions substituting Al, Mg or Sr with Ga, Zn or Ca, respectively, were also synthesized and were confirmed to maintain this crystal structure. X-ray diffraction revealed that the crystal structure was consistently more stable to higher concentrations of vacancies relative to interstitial defects. Four-electrode conductivity measurements indicated that the B-site composition had a larger effect on total electrical conductivity than the A-site composition. Specifically, replacing Al or Mg with Ga or Zn, respectively, improved the conductivity by 1 to 2.5 orders of magnitude. Unfortunately, further electrochemical studies suggested significant hole conduction in these samples. Substituting Sr with Ca in the baseline composition had no effect on the total conductivity value, and La_1.6+xCa_0.4-xAl_0.4Mg_0.6O₄ series samples were all found to be pure ionic conductors. Nevertheless, the oxygen ion conductivity values achieved in the compositions to date remain less than technologically desirable. Results from experiments on newer compositions will show that Li can be placed on the B-site. Such compositions may have utility in both Li-ion and Li-air batteries.

As found in recent investigations, a high oxygen ion conductivity is not only necessary to allow for the oxygen incorporation to proceed on the whole surface of a mixed-conducting perovskite SOFC cathode material, but is also highly beneficial for a fast oxygen exchange surface reaction. Experimentally, it is found that within the group of (La,Sr)(Mn,Fe,Co)O3-δ perovskites the oxygen vacancy mobility does not depend much on cation composition. Interestingly, a decreased migration barrier and increased absolute vacancy diffusivities are observed when Ba is introduced.
To understand the atomistic origins of these changes, DFT calculations were performed for (Ba,Sr)(Co,Fe)O3-δ (BSCF) and (La,Sr)(Co,Fe)O3-δ (LSCF) perovskites with different cation com-position. It was shown that for BSCF the barrier does not only depend on geometric criteria (ion distances in the "critical triangle"), but also on the oxygen vacancy formation energy which is a measure of the reducibility of the transition metal ion.
In contrast, for LSCF perovskites the barriers are almost independent of the vacancy formation energy. This can be understood when looking at the charge transfer from the migrating oxygen O* to the neighboring transition metal ion in the transition state. This electron transfer corresponds to a partial reduction of the transition metal ion and is thus more favorable for perovskites with lower vacancy formation energy (more reducible materials). Since this partial electron transfer decreases the effective size of the migrating oxygen, it facilitates its passage through the "critical triangle" and thus lowers the barrier.
Thus, the low oxygen migration barrier in BSCF perovskites can be traced back to the low va-cancy formation energy, which is related to the expansion of the lattice by the large Ba2+ that stabi-lizes Co ions in lower oxidation states. Unfortunately, this size mismatch decreases also the stability of the cubic perovskite lattice, leading to the detrimental hexagonal perovskite phase formation.
Mastrikov et al, Phys. Chem. Chem. Phys., 2013, 15, 911-918
Merkle et al, J. Electrochem. Soc. 2011, 159, B219-B226
Kuklja et al. Phys. Chem. Chem. Phys. 2013, 15, 5443-5471

LaCrO3 is chemically quite stable in both oxidizing as well as reducing atmosphere. The doped LaCrO3 shows high electronic conductivity as interconnect material in solid oxide fuel cells. But, because of the different densification characteristics of a La0.8Ca0.2Cr0.9Co0.1O3-δ(LCCC) interconnector and a NiO added yttrium-stabilized zirconia(YSZ) anode, complete adhesion between the two layers is hardly obtainable.
In our previous report, LCCC-YSZ composite layer was screen-printed on the NiO-YSZ substrates and the layer was firmly attached to the substrates. It was proposed that tensile stress was decreased on the composite layer due to the relatively delayed densification of the LCCC-YSZ composite, compared to the densification of the NiO-YSZ substrate.
In this study, we have investigated effect of LCCC sol infiltration on LCCC-YSZ composite layer to improve densification and electrical conductivity. Precursor solutions for La0.8Ca0.2Cr0.9Co0.1O3-δ (LCCC) infiltrations are prepared by adding nitric acid and ethylene glycol into an aqueous solution of lanthanum, calcium, chromium and Cobalt nitrates. By controlling the synthesis time of sol, different solute concentration were obtained. X-ray diffraction and TG-DTA were employed to monitor the crystal structure and phase transition of heated precursor solution at 1400°C. The improved densification was observed by FE-SEM as LCCC sol infiltration on LCCC-YSZ composite layer. The 4-probe DC electrical characteristics of the layers were measured using a Van der pauw technique. Not only that formed dense layer observed stable in reduction atmosphere by FE-SEM.

Strontium and magnesium doped lanthanum gallate (LSGM) perovskite-type compounds are considered as very promising solid electrolytes for intermediate temperature fuel cell (IT-SOFC) applications. The cermet-type anode and nickelate-type cathode materials are used as electrodes while their chemical compatibility with LSGM is not well known. For the practical use of these materials not only the thermodynamic stability of the electrolyte and electrodes themselves, but also the reactivity between the electrolyte and electrodes are needed. In this study, thermodynamic modeling of La-Sr-Ga-Ni-O system, which plays a crucial role for the understanding of interface reactions between LSGM electrolytes and nickalate cathodes has been done. Using calculated phase diagrams the compatibility of nickelates as cathode material and also of Ni-based cermets as anode material with LSGM electrolytes were investigated. Chemical potential diagrams representing fabrication and operation conditions are presented.

For improving fuel cell performance and efficiency, understanding of oxygen reduction reaction (ORR) on perovskite oxides surfaces is essential. LaMnO₃ (LMO) has long been studied as an archetypical oxide catalyst and many theoretical analysis have been carried out to explain ORR on LMO(001) surface. Due to alternating positively charged LaO and negatively charged MnO layers along the [001] direction, the Coulomb potential diverges unless half of an electron is compensated near the surface region.[1] However, there has been no study on the charge compensation mechanism on the polar LMO(001) surface. In this study, therefore, we investigate the charge distribution and electronic structure of LMO(001) surface using the first-principles methods. We carefully examined the magnetic moment, Bader charge, and layer-decomposed density of states, and confirmed that charges near the oxygen atoms in the subsurface layer are depleted as a form of electronic reconstruction. This is a rather unexpected result because the valence top is mainly contributed by Mn eg orbital in bulk LMO. This also means that the Mn atoms maintain its valence state of +3 as in the bulk. The possible limitation of the current theoretical framework will be explored by employing other methods such as GW and hybrid functional. We will also discuss the influence of charge compensation on the catalytic effect during ORR reaction.
[1] W.-j. Son, E. Cho, B. Lee, J. Lee, and S. Han, Phys. Rev. B 79, 245411 (2009)

A three-dimensional (3D) finite element method (FEM) model for the performance simulation of Ni/YSZ anodes for solid oxide fuel cells is presented. The 3D FEM anode model enables the spatially resolved simulation of the electrochemical processes at the triple phase boundary (TPB) and the intrinsic coupling with the ionic, electronic and gaseous transport in the complex three-phase microstructure. Experimental data obtained from measurements with patterned model anodes are used for a detailed description of the electro-oxidation kinetics at the TPB.
Moreover, the 3D FEM anode model is able to use reconstructed anodes obtained via focused ion beam (FIB) tomography as model geometry. Hence, all critical microstructure parameters are intrinsically considered in the 3D FEM anode model. The simulated performance showed a good agreement with experimental data both of patterned model anodes as well as technical Ni/YSZ cermet anodes.

Development of a reliable sealant or sealing system remains one of the top priorities in planar SOFC technology. Recently, a compliant glass seal such as an alkali silicate glass was proposed for SOFCs. In contrast to the conventional glass-ceramic sealant which develops a rigid or semi-rigid microstructure after sealing, the compliant glass shows so called “self-healing” behavior and will remain vitreous without substantial crystallization after sealing and during operation at elevated temperatures. The objective of this study was to investigate the effect of sealant composition on the high temperature viscosities, electrical conductivities and thermal cycle stability of glass composite sealants for solid oxide fuel cells (SOFCs). Alkali/alkaline-earth borosilicate glass-based sealants, both with and without fillers, were heat treated at various temperatures for periods of up to 240 h. The effects of filler and alkali oxide content in sealants on the viscosities, electrical conductivities and cyclic sealing performance of the sealants were investigated. The results showed that glass composite sealants containing 4 mol% alkali oxide showed good thermal cyclic seaingl behavior in air for 8 deep thermal cycles during the 1,800 hour of operation. Upon 48 hour heat treatment electrical conductivity of filler-containing glass were higher than those of a base glass. This was attributed to the partial crystallization and densification of sealing glass, and the increasing mobililty of the alkali and alkali-earth ions in glass.

Acceptor-doped barium zirconate (BaZrO₃) is one of the most promising high-temperature proton conductors for protonic ceramic fuel cells due to its high proton conductivity at the bulk and sound chemical stability. Since fuels, such as hydrogen or methane molecules, under operation of fuel cells should reduce at the anode surface, the generated protons migrate through the BaZrO₃ electrolyte, and react with oxygen molecules at the cathode surface with electrons coming from the anode through an external electrical wire, detailed information on BaZrO₃ surfaces are quite important to understand the fuel cell operation at an atomic scale. Surface stability diagram and surface Gibbs free energies were studied as functions of oxygen and barium chemical potentials to determine the most stable surface structure among the three low-indexed (100), (110), and (111) surfaces. The (100) surface with ZrO₂-termination was the most stable among them.

Nickel (Ni) catalyst that shows high melting temperature has been used in solid oxide fuel cells and solid oxide electrolyser cells, due to its reasonable cost and excellent catalytic activity. However, its sulfur (S) poisoning has been a serious problem because the fuels used in the cells always contained a small amount of S even after the desulfurization process. On the other hand, copper (Cu) catalyst that is known for high tolerance to S has been used in steam reforming cells due to its cheap cost and excellent catalytic activity. However, because Cu shows relatively low melting temperature, it can only be used at intermediate and low temperatures. In order to utilize both advantages of the catalysts, we investigated the effect of Cu alloying on S poisoning of Ni based catalyst via ab initio thermodynamics calculations. The low indexed surface of (100), (110) and (111) for Ni, Cu, Cu 100% and 75% over-layer surfaces were considered to understand the effect of Cu segregation on the S poisoning. The adsorption energies of S were investigated as function of d-band center with consideration of the weighted d-band scheme. The increase of Cu concentration at the surface shifted the b-band center downward from the Fermi-level, and decreased the S adsorption energies, indicating that Cu segregation at the surfaces played a key role for S tolerance. The phase diagrams for Ni, Cu, Cu 100% over-layer and Cu 75% over-layer were constructed as functions of temperature and the partial pressure ratio of H2/H₂S. The 75% and 100% Cu over-layer surfaces showed S tolerance of 1 part per million H₂S at 1000K. We also investigated the morphologies of Ni, Cu and NiCu alloys as functions of surface energy and S chemical potential by using Gibbs-Wulff theorem. The nano-particle morphologies of Ni, Cu and NiCu alloys were varied as a function of the S chemical potential.

Cation segregation to the surface of materials for SOFC applications is known to impact on overall device performance and hence is of great interest both fundamentally and practically. In the past many segregation studies have focused on the surface of bulk polycrystalline materials [1, 2], yet developments in thin film fabrication techniques allow greater control over the surface properties such as crystallographic orientation. Experimental studies focused on observing changes in the transport properties of ion conductors when grown in low dimensional systems such as thin films have recently attracted considerable interest [3, 4], yet cation segregation is rarely considered. However, changes in the chemical composition cannot be neglected especially when the dimensions of the films approach those distances (tens of nanometers) over which cation segregation is expected to occur.
Low energy ion scattering (LEIS) is a powerful technique capable of determining the elemental composition of the very outer monolayer of a surface. It also allows a fully quantitative profile of the ratio of cations in a material as a function of depth from the surface to be measured. In this work we have employed LEIS to study the (100) surface of gadolinia-doped ceria (CGO) films on STO, YSZ and LAO substrates and the (111) surface grown on sapphire fabricated by pulsed laser deposition (PLD). We show that as-grown films show minimal segregation despite high temperatures used during fabrication (600 - 800°C), yet the surface composition varies drastically from the nominal bulk when heated post growth to temperatures associated with processing or conductivity measurements (1000°C). The Gd/Ce ratio has been found to be 4 times higher than the bulk composition for the (100) surface and over 7 times for the (111) surface, and in both cases segregation effects extends up to 5 nm into the films. The effects of lattice strain in the films due to lattice mismatch with the substrates will also be discussed along with the composition profile as the film thicknesses are decreased to the length scales associated with the segregation effects.
1. de Ridder, M., et al., Journal of Applied Physics, 2002. 92(6): p. 3056-3064.
2. Scanlon, P.J., et al., Solid State Ionics, 1998. 112(1-2): p. 123-130.
3. Rupp, J.L.M., Solid State Ionics, 2012. 207(0): p. 1-13.
4. Kant, K.M., et al., Applied Physics Letters, 2012. 100(3): p. 033105-3.

Small-scale energy-converting devices promise increased run-time and functionality of portable electronic devices such as laptops, mobile phones or PDA`s. Micro-Solid Oxide Fuel Cells (microSOFCs) integrated on micromachinable substrates such as silicon can operate on pressurized or liquid fuels and independent of the electricity grid. These show high potential to replace today`s batteries due to their increased power density. Despite their promise, microSOFCs reveal alternating open circuit voltages and power outputs on a cell level with respect to processing and cell microstructures (1. The active fuel cell components - anode, electrolyte, cathode - are deposited as thin films of several 100 nm in thickness. Here, the nanoscopic interfaces dominate the charge and mass transport: interfacial space charge potentials and lattice strains in the structures require attention. In the first part, we summarize recent work directed towards nanostructuring of thin film microstructures and cells for the field of microSOFCs. In the second part we focus on strain-oxygen ionic conduction interaction for electrolyte thin films (2,3 and multilayer structures. Recent findings on lattice strain and its impact on oxygen ionic migration/defect association and near order changes are presented for doped ceria single thin films relative to processing (4. As alternative approach of strain-oxygen ionic transport tuning the model multilayer electrolyte system erbia/gadolinia-doped ceria is presented (5. Here, new routes on how to use microfabriation to probe ionic transport for small multilayer oxide dots are discussed relative to strain state and near order Raman characteristics. Also, the role of substrate dependent impurities such as Si and processing-dependent porosity and amorphous phases in electrolyte thin films are introduced (6,7,8. In the third part we focus on cathode-electrolyte thin film interfaces based on (La,Sr)(Co,Fe)O3 cathode/ceria electrolytes. For this, conventional area specific resistance (ASR) tests based on single crystal electrolyte or pellet set-ups are discussed towards new microSOFC half cell tests.
References
1) A. Evans et al., J. of Power Sources, 2009. 194: 119.
2) J.L.M. Rupp, Solid State Ionics, 2012. 207: 1.
3) A. Chroneos et al., Energy&Environmental Science, 2011.4:2774.
4) J.L.M. Rupp et al. Scalable oxygen-ion transport kinetics in metal-oxide films: Impact of thermally induced lattice compaction in acceptor doped ceria films, Adv. Funct. Mat., 2013 in press
5) S.Schweiger et al. Tuning the strain-conductivity interaction in gadolinia-doped ceria/Er2O3 multilayers: new insights on the Raman near order probing in micro-devices, J. Materials Chemistry C, 2013 in review
6) B. Scherrer et al., Adv. Funct. Mat., 2011, 21:3967.
7) B. Scherrer et al., J. of Power Sources, 2011. 196: 7372.
8) J.L.M. Rupp et al., Adv. Funct. Mat., 2010, 20:2807.

In recent years, there have been a number of reports of enhancements in the ionic conductivity of yttria-stabilised zirconia (YSZ) as the dimensions of the material are reduced, such as when grown as thin films or heterostructures. Improvements in the ionic conductivity are often attributed to an expansive lattice strain caused by a coherent interface between the conducting and insulating layers. Alternatively, regular networks of dislocations are often claimed to be the cause due to the locally strained lattice regularly surrounding a dislocation core. To date, experimental investigations in these systems have led to conflicting reports with conductivity measurements ranging from an enhancement of orders of magnitude [1] to a modest reduction compared to bulk behaviour [2]. One of the reasons for the disparity in the literature may be due to the large number of differences between nominally similar systems due to variations in fabrication, processing and measurement techniques employed. Another cause of concern when assessing the transport properties of nanoscale ionic conductors is the increased danger of electrical current leakage through either the substrate or experimental setup effecting the measurement, leading to an erroneously interpreted result [3].
In order to mitigate the issues above we have fabricated highly textured YSZ thin films oriented in the (111) direction onto MgO and sapphire, and the (100) direction onto MgO, LAO and NGO using pulsed laser deposition (PLD). These correspond to a range of lattice mismatches from 4.5% tensile to 18% compressive, and have been grown at a number of thicknesses in order to isolate interfacial effects. Using x-ray diffraction (XRD), secondary ion mass spectroscopy (SIMS), and high-resolution transmission electron microscopy (HR-TEM) the structural and chemical properties of the thin films have been characterised. Electrochemical impedance spectroscopy (EIS) combined with isotope tracer diffusion will be shown to directly and unambiguously measure the oxygen ion transport properties in these films as a function of thickness. This allows compositional, micro- and nano-structural variations observed in the film interfaces to be associated with changes in the conduction properties. We will present evidence to show that a regular dislocation network at YSZ/substrate interfaces does not in fact drastically alter the conduction properties despite being linked to enhanced conduction properties previously [1, 4].
1. Sillassen, M., et al., Advanced Functional Materials, 2010. 20(13): p. 2071-2076.
2. Gerstl, M., et al., Physical Chemistry Chemical Physics, 2013. 15: p. 1097 - 1107.
3. Kim, H.-R., et al., Physical Chemistry Chemical Physics, 2011. 13(13): p. 6133-6137.

A Solid Oxide Fuel Cell is a device in which a solid electrolyte is sandwiched between a cathode and an anode. As such, the interface between these different components is very important, as it can affect the materials properties and, therefore, the device&’s performance. Furthermore, the lattice mismatch at the interface between different materials has also been recently suggested as a means to improve materials properties, especially the ionic conductivity and oxygen reduction reaction kinetics.
Dislocations are ubiquitous in these materials, and their concentration is high close to an interface. For instance, Chen et al have observed an array of equally spaced dislocations (~ 4 nm) at the interface between yttria-stabilised zirconia and ceria [1]. The role of dislocations on the materials properties is, however, still unclear and an object of controversy. For instance, dislocations have been suggested to either increase [2] or decrease [3] the ionic conductivity of YSZ, without a clear conclusion.
In this paper, we hypothesize that the distribution and mobility of charged defects can change because of the elastic and electrostatic field around dislocations in oxides. We take SrTiO3 as a model perovskite, which is considered to be a promising cathode material when doped with Fe [4] and also used in red-ox based resistive switching memories [5], and assess the impact of dislocation on redistribution and migration of defects by atomistic simulations. By using pair potentials, parameterized with respect to first-principles calculations, we could perform simulations on a fairly extended system (~100,000 atoms). This allowed us to capture one dislocation and to study its effect on the defect chemistry and mobility in this material. The results are rationalized in terms of the strain field around a dislocation.
[1] Chen et al., Applied Physics A 76, 969 (2003)
[2] Sillassen et al., Advanced Functional Materials 20, 2071 (2010)
[3] Li et al., PCCP 15, 1296 (2013)
[4] Jung et al., Advanced Energy Materials 1, 1184 (2011)
[5] Waser et al., Advanced Functional Materials 21, 2633 (2011)

Highly strained yttria-stabilized zirconia (YSZ) sandwiched between strontium titanate (STO) in epitaxial heterostructures has been shown to exhibit up to a remarkable eight orders of magnitude increase in ionic conductivity near room temperature [1]. Scanning transmission electron microscope images show the YSZ layers to be coherent and highly strained, while electron energy loss spectroscopy indicates the presence of disorder in the YSZ O sublattice near the interface [2]. Density functional theory (DFT) simulations reveal that a combination of strain and interface effects cause the O sublattice in the YSZ to effectively melt into a highly disordered and mobile form, with a greatly reduced migration energy barrier consistent with experiment [3]. Recent DFT simulations also show that vacancies prefer to be in the YSZ over the STO to the extent that vacancies placed at or near the interface in the STO immediately move into the YSZ upon relaxation, even at zero temperature. The self healing nature of the interface is important as it rules out vacancy migration from the YSZ to the STO, which could p-type dope the STO. Even if vacancies are accidentally grown in to the STO near the interface, they would spontaneously move into the YSZ, explaining the inability to measure any significant electronic component to the conductivity.
Research sponsored in part by the UK Engineering and Physical Sciences Research Council through the UK National Facility for Aberration-Corrected STEM (SuperSTEM). Research at Vanderbilt was supported in part by the U.S. Department of Energy Grant DE-FG02-09ER46554 (TJP, STP) and the McMinn Endowment (STP). Research at Oak Ridge National Laboratory was sponsored by the U.S. Department of Energy, Office of Science, Materials Sciences and Engineering Division (SJP). Computations were performed at the National Energy Research Scientific Computing Center at Lawrence Berkeley National Laboratory.

[1] J. Garcia-Barriocanal et al., Science 321, 676 (2008).
[2] T. J. Pennycook et al., Eur. Phys. J. Appl. Phys. 54, 33507 (2011).
[3] T. J. Pennycook et al., Phys. Rev. Lett., 104, 115901 (2010).

Ceria (CeO2) is one of the most active oxide electro-catalysts for hydrogen oxidation and water-splitting reactions at elevated temperatures. The surface concentration of oxygen vacancies and Ce3+ species were found to be significantly greater than the bulk, and can plausibly explain ceria&’s high electro-catalytic activity. These defect concentrations and activities of ceria are expected to depend strongly on the surface termination, as already observed in nanostructured ceria with different facets. To better understand electrochemical reaction mechanisms and the relationship between surface structure and activity , we investigated on the dependence of ceria&’s electrochemical activity on the surface termination, morphology, and roughness. We fabricated doped ceria thin films on yttria-stabilized zirconia (YSZ) substrates by pulsed-laser deposition. By varying the growth conditions, we can modify the film surface from atomically flat to roughened. Next, we fabricated an electrochemical device by depositing current collectors and measured the current-voltage characteristics as a function of temperature and gas atmosphere. We will present the activities of atomically-flat and roughened ceria grown on (100)- and (111)-terminated YSZ. These results provide insights to design more efficient ceria-based electro-catalysts with the desired surface morphologies.

In recent years a growing interest in tailoring the properties of nanostructured oxygen ionic conductors has been increased. The development of oxide electrolyte materials with enhanced ionic conductivity, particularly with possibility to operate at lower and intermediate operating temperatures is of prime importance for devices based on electrochemical reactions such as solid oxide fuel cells (SOFC).
In this talk I will present our recent results on the strained epitaxial doped ceria thin films which are deposited on diverse substrates to clarify the effect of strain on chemical stability and electrical properties. Particularly, diverse level of strain in doped ceria epitaxial thin films were controlled by varying the thickness of designed buffer layers. The induced strain is found to enhance the in-plane ionic conductivity in doped ceria epitaxial thin films. Finally, I will show that by controlling the growth of the heteroepitaxial thin films is possible to obtain fluorite electrolyte films with superior properties.

Enhancement of ionic conductivity in thin films or multilayers of oxide materials, i.e. doped zirconia and ceria, has sparked great interest in the search for fast ion conducting structures for fuel cells. The enhancement in ionic conductivity in such structures could be attributed to elastic strain arising from the lattice mismatch at the interface. However, this assumes that the interface between two materials is perfectly coherent, while in most cases dislocations are observed, which relax the interfacial elastic strain. The strain field and the electrostatic field that arises from the dislocations can also impact the defect stability, distribution and mobility in these materials; and yet, the role of dislocation on the ionic conductivity is not consistently reported in the literature nor is it clearly understood. The aim of our work is to quantitatively assess the dislocation's influence on the ionic conductivity, especially in fluorite oxides, such ceria and yittria stabilized zirconia.
In this work, we combined Monte Carlo and Molecular Dynamics simulations in a hybrid algorithm to find the distribution of dopant cations and oxygen vacancies, around a 1/2<110>{100} edge dislocation in doped and reduced ceria. Different dopant type (Gd, La) and concentrations are studied at varying temperatures. The dopant cations and oxygen vacancies segregate to the tensile region around the dislocation core. This is because the oxygen vacancy formation energy is lower in the tensile strained region of the dislocation core. The open space and low coordination in dislocation core can attract the dopant cations by a steric effect. At the same time, a depletion zone of dopants is observed on the compressive region around the dislocation, since the dopant cation (usually bigger than the host cation) is not easy to be accommodated at the compressive region. The outcome is an asymmetric space charge zone around the dislocation. A higher concentration or a higher temperature was found to lessen the association of dopants and vacancies with the dislocation, and subsequently change the degree of segregation and depletion. Our work demonstrates that dislocations can significantly change the distribution and mobility of dopant cations and oxygen vacancies, and their effects must be considered in quantitatively interpreting the experimental results.

The role of materials&’ interfaces/grain boundaries on enhancing anion conductivity is a hotly debated issue that has exposed limited understanding on point-defect energetics at interfaces. Using density functional theory (DFT) and interatomic simulations on zirconia, ceria and other related materials, we unravel key interface issues, i.e., oxygen vacancy migration barriers at interfaces in the absence and presence of dopants, and oxygen vacancy-dopant binding energies at interfaces. We show that pure, strained interfaces indeed possess very low oxygen migration barriers; however, the segregated dopants counteract and raise the barriers significantly. In addition, the dopants bind oxygen vacancies much more strongly at the interfaces than in the bulk, thereby further lowering oxygen diffusivity at interfaces. From DFT calculations, we also show that that strained interfaces could lead to formation of metastable phases that can ultimately dictate unintuitive oxygen vacancy stability. We conclude that the concept of strained interfaces to enhance anion conductivity prevails primarily in the absence of segregated dopants, and strategies that prevent dopant segregation need to be considered in the design of anion-conducting interfacial and nanocrystalline materials.

A number of opportunities exist to use thin film deposition methods to nanoengineer materials so that we can understand and improve their ion transport behavior and other electrochemical properties. Here, we describe experiments on sputter-deposited ceramic films with controlled, heterogeneous chemical composition. Films are created using a technique developed in our lab where pure, single-element targets are reactively co-sputtered to create ceria-zirconia films where both the host and dopant atom concentrations can be controlled at the nanometer scale through the film thickness. The ionic conductivity of all samples was determined using impedance spectroscopy in the temperature range 573 K to 1073 K and over a range of oxygen partial pressures.
Two sets of results using this system will be presented. First, we will discuss films made to have heterogeneous dopant content. By placing the dopants in 2-D atomic planes, defects are trapped in engineered space charge zones surrounding the interface, and vacancy-dopant associations can be studied. Specifically, by comparing samples using heterogeneous La, Gd, or Y dopants, the role of strain in the mobility and concentration of defects in the space charge region is examined, with larger dopant ion correlated to increased space charge conductivity. A second system to be presented will involve modulation of the Ce/Zr ratio through the thickness. Since the lattice parameter is a linear function of Ce/Zr ratio, these samples can be used to smoothly and systematically control lattice strain. Strain was determined by x-ray diffraction and microscopy, including selected area electron diffraction. Clear differences in interfacial structure are found as the lattice parameter differential between the layers is adjusted. The strained multilayers indicate both a strain-based change in vacancy mobility and a general reduction in conductivity as the interfacial density increases.

The proton conducting lanthanide transition metal oxides are of interest for their use as electrolytes in electrochemical devices like fuel cells, electrolysers and gas sensors. We are searching for strategies to control the proton conductivity activation energy and thus make better fuel cell electrolytes. Recent data show that the activation energy increases steadily with the mechanical pressure, suggesting the proton conductivity depends on the crystallographic structure. To comprehend the crystal structure and dynamics, XRD and Raman spectroscopy are studied on the compressively strained BaCe0.8Y0.2O3-δ (BCY20) proton conductors. Higher activation energy could be due to the octahedral distortions, or the increasing phonon frequency in the Ce-O stretching mode. The phonon frequency increase may be anisotropic, and so is the proton mobility. An expanded lattice - for example thin films with epitaxial strain - can promote proton conductivity. This is desirable for the development of epitaxial strained proton conducting thin film systems with high proton mobility and low activation energy using tensile strain. We have employed high pressure XPS together with impedance spectroscopy and found how oxygen vacancies are filled by oxygen with a particular departure of a state in the valence band. With increasing temperature, the OH signature in the oxygen core level spectra disappears while the proton dynamics changes from a rotational mode to an effective translational mode. This project was funded under Swiss National Science Foundation Grant # 124812 and Marie Curie MIRG # 042095.
References
6. Q. Chen, F. El Gabaly, F. Aksoy, Z. Liu, B.S. Mun, S. Yamaguchi, A. Braun, In-situ observation of oxygen vacancy filling under water vapor in ceramic proton conductors with ambient pressure XPS and impedance spectroscopy; submitted.
5. Q. Chen, J. Banyte, X. Zhang, J. P. Embs, A. Braun; Proton Diffusivity in Spark Plasma Sintered BaCe0.8Y0.2O3-δ: in-situ combination of quasi-elastic neutron scattering and impedance spectroscopy; Solid State Ionics; in press, 10.1016/j.ssi.2013.05.009
4. Q. Chen, S. Holdsworth, J. Embs, V. Pomjakushin, B. Frick, A. Braun, High Temperature High Pressure Cell for Neutron Scattering Studies, High Pressure Research 2012, 32(4) 471-481.
3. Q. Chen, T.-W. Huang, M. Baldini, A. Hushur, A. Pomjakushin, S. Clark, W. Mao, M. Manghnani, A. Braun, T. Graule, The Effect of Compressive Strain on the Raman Modes of the Dry and Hydrated BaCe0.8Y0.2O3 Proton Conductor, J. Phys. Chem. C, 2011, 115 (48), 24021-24027.

2. Q. Chen, A. Braun, S. Yoon, N. Bagdassarov, T. Graule, Effect of lattice volume and compressive strain on the conductivity of BaCeY-oxide ceramic proton conductors, J. Eur. Ceram. Soc. (2011) 31 (14) 2657-2661.
1. Q. Chen, A. Braun, A. Ovalle, C.-D. Savaniu, T. Graule, N. Bagdassarov, Hydrostatic pressure decreases the proton mobility in the hydrated BaZr0.9Y0.1O3 proton conductor, Appl. Phys. Lett. 97, 041902 (2010)

There has been growing interest in developing new materials which possess high oxygen ionic conduction at low temperatures for solid oxide fuel cell electrolytes. In our group, we have been developing trivalent element doped ceria/zirconia based thin film structures for this purpose. In this study, we have grown multi-layers of high quality samaria doped ceria (SDC) and scandia stabilized zirconia (ScSZ) epitaxial thin films using oxygen plasma-assisted molecular beam epitaxy (OPA-MBE). These multi-layer films exhibit significantly higher oxygen ionic conduction at intermediate temperatures in comparison to single-layer doped ceria and zirconia thin films. In addition, it appears that the oxygen ionic conductivity of these multi-layer structures depends on the number of interfaces. Although we have demonstrated that these structures possess high oxygen ionic conduction at low and intermediate temperatures, the mechanisms associated with the enhancement in oxygen ionic conduction through these engineered heterogeneous interfaces is not well understood. It is possible that the strain associated with the multi-layered structures in combination with the dopant distribution can play a role in this enhanced ionic conductivity. We employed a state-of-the-art three-dimensional (3-D) chemical imaging analysis capability; atom probe tomography (APT) to characterize these films. APT can provide quantitative 3-D chemical analysis of materials with lateral and depth resolutions in the order of 0.2-0.3 nm and chemical sensitivity up to parts-per-million levels with field-of-view on the order of 100 × 100 × 100 nm3. In this study, we have utilized this capability along with other surface and bulk sensitive capabilities in an attempt to understand the influence of chemical and physical nature of the interfaces on oxygen ionic conductivity through ceria/zirconia multi-layer films.

Grain boundaries in doped ceria electrolytes have a deleterious effect on the total ionic conductivity especially at intermediate temperatures (300 - 550 °C) [1,2]. The high resistivity of grain boundaries has been attributed to a space charge layer which creates a vacancy depletion region emanating from grain interfaces. Other factors may also contribute to high grain boundary resistance. For example, recent high resolution elemental analysis in the transmission electron microscope (TEM) of 20% Gd-doped ceria (GDC) by our group and others shows significant Gd segregation to grain boundaries yielding enrichment zones of approximately 40 at% Gd, far exceeding the optimal Gd concentration (10-20 at%) for maximum ionic conductivity. On the other hand, enhancement of grain boundary conductivity has been demonstrated by selectively doping grain boundaries with transition metal ions such as Fe²⁺, Cr²⁺ and Co²⁺ [2,3]. However, the structure, chemistry and defect distribution responsible for the observed conductivity change of grain boundaries is not well understood. Here we investigate the relationship between electrical conductivity and the nanoscale structure and dopant distribution in high purity Ce_0.9Gd_0.1O_1.95 grain boundaries containing nominally 0 - 2 at% Fe²⁺ and/or Co²⁺. Starting materials are synthesized using a spray drying method, and electrolyte pellets fabricated via traditional ceramic processing techniques. Sintering conditions are determined which yield dense, polycrystalline electrolytes of similar grain size with grain boundary structure suitable for high spatial resolution TEM. For select samples, average grain boundary electrical conductivity is measured via impedance spectroscopy in the temperature regime 150 - 700 °C. The structure and chemistry of grain boundaries in these samples is probed using high spatial resolution electron energy loss spectroscopy (EELS) in a scanning TEM (STEM). We present these data and discuss correlations between structure, chemistry and electrical properties of transition metal-doped GDC electrolytes.
1. Guo, X., Maier, J. Grain Boundary Blocking Effect in Zirconia: A Schottky Barrier Analysis. Journal of The Electrochemical Society, 148 (3) E121-E126 (2001)
2. Avila-Paredes, H.J., Kim, S. The effect of segregated transition metal ions on the grain boundary resistivity of gadolinium doped ceria: Alteration of the space charge potential. Solid State Ionics 177 (2006) 3075-3080.
3. Zajac, W., Suescun, L., Swierchzek, K., Molenda, J. Structural and electrical properties of grain boundaries in Ce0.85Gd0.15O1.925 solid electrolyte modified by addition of transition metal ions. Journal of Power Sources 194 (2009) 2-9.

Electrochemical impedance spectroscopy (EIS) is frequently used in conjunction with simple equivalent circuit models to quantify grain boundary properties in solid ion conductors. In the case of depressed impedance arcs, which are often encountered in experimental studies, an equivalent circuit including one or more constant phase elements (CPEs) is usually employed to obtain a better fit. However, as the CPE is a mathematical abstraction with no clearly defined physical meaning, the ability to accurately estimate grain boundary properties from such a model remains suspect. Depressed grain boundary impedance arcs were previously shown to occur in samples with heterogeneous grain boundary conductivity. Here, we use numerical modeling to investigate the utility of CPE-based equivalent circuit fitting in quantifying heterogeneous grain boundary conductivity and/or permittivity. Specifically, the relationship between the equivalent circuit parameters and the mean value and distribution of the grain boundary properties is studied.
A 2D electrical model of a geometrically ideal polycrystal was developed, allowing impedance spectra to be generated for arbitrary distributions of grain boundary conductivity and/or permittivity. For the present study, the relevant properties of each grain boundary were assigned randomly according to a specified distribution. In each sample, the conductivity, the permittivity, or both, were heterogeneous. Heterogeneous conductivity values followed a log-normal distribution, while heterogeneous permittivity values followed an exponential distribution with a maximum equal to the bulk grain permittivity. The simulated impedance spectra were then fitted to an equivalent circuit consisting of two parallel R-CPE elements in series, and conductivity and permittivity values were calculated from the circuit parameters.
An increasingly depressed impedance arc and decreased CPE exponent were correlated, as expected, with increasingly heterogeneous grain boundary properties. When only one grain boundary parameter was heterogeneous, the range or spread of parameter values could be estimated from the CPE exponent. However, this was no longer possible when both parameters were heterogeneous. In all cases, the calculated conductivity differed from the actual mean grain boundary conductivity by no more than 30% provided a geometric correction factor was included to compensate for edge effects. Simply using the Q parameter of the CPE as a capacitance results in severe error in calculating the mean grain boundary permittivity. Using an equivalent capacitance expression commonly found in the literature, the calculated permittivity could still deviate by up to 60% from the actual mean value. However, a new empirical equation allowed a more accurate estimate which differed from the mean by no more than 35%. Recent extensions of this work to quantification of the impedance of the electrode-solid electrolyte interface will also be presented.

In recent years the interest in yttria-stabilised zirconia (YSZ) as a potential proton conductor has increased significantly. Studies have shown proton transport occurs at low temperatures (<150 °C) and that this transport is driven by the grain boundary structures in the material [1-3]. In work by Kim et al. [2] and Avila-Paredes et al. [3] proton conduction is observed in nanocrystalline YSZ with grain boundary size between ~ 13 and ~ 100 nm at room temperature in humid atmospheres. These works illustrate how samples with large grains exhibit very little protonic conduction leading to the conclusion that protonic conduction in YSZ is interfacially driven [1]. There is, however, ongoing debate over whether protonic conduction does occur primarily in the grain boundaries with Scherrer et al. [4] observing no proton conductivity through grain boundaries in YSZ nanocrystalline, dense thin films. Proton conduction is, however, observed at low temperatures at the inner surface of the porous films. An additional study by Raz et al. [5] also supports the conclusion that proton conductivity occurs at the YSZ surface and not its grain boundaries.
Both classical and quantum mechanical simulation techniques have been applied to investigate protonic defects in yttria-stabilised zirconia. The calculated redox reaction energies are found to be high, although the reduction energies are lower than those of bulk cubic ZrO₂ and are shown to decrease further with increasing Y content. The hydration energies for YSZ are also lower than the values calculated for bulk ZrO₂ and are found to be lowest when the oxygen ion is in close proximity to at least one Y ion. Strong binding (proton trapping) energies are observed between the protons and additional acceptor dopants. These energies are found to vary significantly depending on local configuration. Density functional theory (DFT) calculations are used to determine proton migration energy barriers of 0.32 - 0.42 eV for pathways with the closest O-O interatomic distances and are found to be very comparable to well-established proton conducting materials.
[1] Kim, S.; Avila-Paredes, H. J.; Wang, S.; Chen, C. -T.; De Souza, R. A.; Martin, M.; Munir, Z. A. Phys. Chem. Chem. Phys. 2009, 11, 3035.
[2] Kim, S.; Anselmi-Tamburini, U.; Park, H. J.; Martin, M.; Munir, Z. A. Adv. Mater., 2008, 20, 556.
[3] Avila-Paredes, H. J.; Zhao, J.; Wang, S.; Pietrowski, M.; De Souza, R. A.; Reinholdt, A.; Munir, Z. A.; Martin, M.; Kim, S. J. Mater. Chem. 2010, 20, 990.
[4] Scherrer, B.; Schlupp, M. V. F.; Stender, D.; Martynczuk, J.; Grolig, J. G.; Ma, H.; Kocher, P.; Lippert, T.; Prestat, M.; Gauckler, L. J. Adv. Funct. Mater. 2013, 23, 1957.
[5] Raz, S.; Sasaki, K.; Maier, J.; Riess, I. Solid State Ionics 2001, 143, 181.