Oxygen electrodes for protonic ceramic cells

.


Solid oxide cells
The environmental issues caused by the combustion of conventional fossil fuels require immediate attention [1][2][3].Many technologies have been developed to tackle the challenges.Among these, the solid oxide cells (SOC) are considered a promising and efficient approach for energy conversion and storage, given their advantages such as high energy efficiency, wide fuel options, low pollutant emissions, and good scalability [4].The SOCs can be operated as either a solid oxide electrolysis cell (SOEC) or a solid oxide fuel cell (SOFC).In SOFC mode, the chemical energy of fuels is converted to electricity and heat, or vice versa in SOEC mode, offering great convenience and high efficiency.Fig. 1 shows a conceptual diagram of the role of SOC technology in a future energy system based on renewable energy sources.Excess electricity generation from intermittent renewable wind and solar energy is converted into hydrogen using SOEC, and hydrogen is stored for later use.When electricity is needed, the stored hydrogen can be reconverted into electricity using SOFC for different applications, ranging from portable devices to power plants.By utilizing the reversible heat produced during SOFC operation in the SOEC operation, the theoretical limit for ideal roundtrip efficiency can increase to 98% [5].In this SOC-based system, no pollutant gasses are generated, and no fossil fuels are consumed [6][7][8].

Protonic ceramic cells
Generally speaking, an SOC cell consists of a dense electrolyte sandwiched between two porous electrodes (one oxygen electrode and one hydrogen electrode).Depending on the type of electrolyte, SOCs can be classified as either oxygen ion-conducting SOCs (O-SOCs; solid oxide fuel cells: SOFCs; solid oxide electrolysis cells: SOECs) or protonic ceramic cells (PCCs; protonic ceramic fuel cells: PCFCs; and protonic ceramic electrolysis cells: PCECs).Fig. 2 illustrates the working principles of PCCs and O-SOCs.PCCs are being developed as alternatives to O-SOCs because they offer several unique advantages [9].First, due to the lower activation energy for proton transport, PCCs can be operated at lower temperatures (400-600 • C) compared to that for O-SOCs Abbreviations: ALD, atomic layer deposition; DFT, density functional theory; EHP, electrochemical hydrogen pumping; EIS, electrochemical impedance spectroscopy; MIEC, mixed ionic and electronic conductor; MPEC, mixed proton and electronic conductor; OER, oxygen evolution reaction; ORR, oxygen reduction reaction; O-SOC, oxygen-ion conducting solid oxide cell; PCC, protonic ceramic cell; PCEC, protonic ceramic electrolysis cell; PCFC, protonic ceramic fuel cell; PLD, pulsed laser deposition; R-PCC, reversible protonic ceramic cell; SOC, solid oxide cell; SOEC, solid oxide electrolysis cell; SOFC, solid oxide fuel cell; TCO, triple conducting oxide; TGA, thermogravimetric analysis; TPBs, triple phase boundaries; WSR, water splitting reaction.
(typically 700-900 • C).This allows for more flexible material selection as well as possibly better durability.In addition, the lower operating temperature enables stack cost reduction and eases challenges related to the sealing issue [10][11][12].Second, dry and pure hydrogen can be directly produced and even pressurized in PCEC without further gas separation [13].The electrolyte materials possess a small fraction of oxygen ion conduction in the operating temperature range [14,15].Third, the degradation of Ni at the hydrogen electrode, which is a challenge in O-SOCs, is less critical in PCCs [16].Therefore, PCCs appear as next-generation renewable energy conversion and storage devices.
Besides fuel cell and electrolysis cell applications, PCCs offer several potential industrial applications such as the conversion and utilization of CO 2 by hydrogenating or electrochemically reducing CO 2 (Co-PCEC), electrochemical hydrogen pumping (EHP) for hydrogen extraction and purification, and ammonia synthesis directly from water, nitrogen, and electricity and so on [17][18][19].More detailed reviews on the applications of proton-conducting oxides can be found in literature [20,21].

PCC components 1.3.1. Proton-conducting electrolyte
Protonic defect.Mobile protons can be incorporated into the perovskite structure as hydroxide defects in hydrogen-containing gasses and/or in the presence of steam.The formation of hydroxide defects occurs via the following reaction: Where O × O is the oxygen in the normal lattice site, OH ⋅ o is the proton located on the oxygen ion (protonic defect), e ′ is the electron.When equilibrated in a steam-containing atmosphere, the steam from the gas phase dissociates into a hydroxide ion and a proton; the former fills an oxygen vacancy, while the latter can form a covalent bond with the lattice oxygen [3].This reaction is given by: The above-described mechanism of the appearance of mobile protons in the oxide structure requires the presence of oxygen vacancies (V ⋅⋅ O ), which can be formed by doping the B site of the perovskite with a trivalent cation M 3+ .Oxygen vacancies are generated extrinsically as shown in Eq. (3) [22]: The concentration and mobility of protonic defects depend on several factors, including e.g.host material characteristics (crystal structure, perovskite, double perovskite, etc.) and the nature of dopants.Thermogravimetric analysis (TGA) is often used to measure the protonic defect concentration after water uptake, which leads to a significant weight increase upon switching from dry gasses to humidified gasses.
Proton transport.Due to the small size and highly polarizing nature, protons (H + ) are always covalently bonded to some electronegative atoms/ions (e.g., to oxygen ions in oxides), which is different from oxygen ions and other charge carriers.Currently, the Grotthuss mechanism is the likeliest to describe the transport mechanisms of protons.The proton conduction between fixed oxygen sites occurs mainly by rotational diffusion of a proton around an oxygen ion, followed by proton transfer (migration) toward a neighboring oxide ion.The rotational diffusion stage is faster and easier than the proton transfer stage, thus exhibiting low activation energy (below 0.1 eV) [23].According to the Grotthuss mechanism, the proton transfer reaction is the rate-determining step, and the activation energy is usually about 0.5 eV [24].
Mixed ionic-electronic conductivity.The predominant charge carrier depends on operating conditions, i.e., atmospheres and/or temperature range.In general, protonic ceramics are not pure proton conductors.They typically possess mixed proton, oxygen ion, and electron-hole (Osite polarons) charge carriers.Since the reaction in Eq. ( 2) is exothermic, the proton concentration increases with decreasing the temperature [25].Han et al. showed an increase in the dehydration temperature in Fig. 1.Role of the SOC technology in a future sustainable energy system based on renewable energies.SOFC and SOEC stand for solid oxide fuel cell and solid oxide electrolysis cell, respectively.
Q. Wang et al.BaCe 0.8-x Zr x Y 0.2 O 3-d in 3.1% moist oxygen from 375 to 620 • C using HT-XRD measurements with increasing the cerium content [26].When the samples dehydrate, the concentration of protonic defects decreases, and so does the protonic conductivity.At temperatures above 600 • C, part of the ionic conductivity comes from the oxygen ions.Furthermore, proton-conducting materials might display n-and/or p-type electronic conductivities, which are governed by oxygen partial pressure (PO 2 ) [27,28].Usually, the n-type conductivity is formed due to a partial reduction of transition elements existing in the structure, which can increase at higher temperatures and lower PO 2 [29].P-type electronic conduction appears as a consequence of oxygen incorporation that produces holes (h ⋅ ), according to (4): Recent work shows that the electron holes are located on the oxygen, creating oxygen polarons [29,14].
The evidence of n-type electronic transport in proton-conducting oxides under strongly reducing atmospheres is extremely limited as compared to that of the O-site polaron transport, which occurs in oxidizing atmospheres [14,3].The small-polaron hopping mechanism is activated by temperature [30].Transport numbers of each charge carrier depend on the material composition, operating temperature, and atmosphere.In oxidizing atmospheres, while the total conductivity increases gradually with temperature, the tendency for the apparent activation energy to increase is due to changes in the predominant partial conductivity (from proton to oxygen-ionic and electronic) [31].When using proton-conducting oxides as electrolytes for electrochemical devices, electronic transport becomes a crucial factor.Recent studies have confirmed that electronic leakage through the proton-conducting electrolyte drastically affects the Faradaic efficiency in electrolysis mode and is not so significant in fuel cell mode [32,14].The development of low-temperature electrochemical devices is a rational approach for preventing the negative effect of electron transport on the electrochemical behavior of such devices.
Proton-conducting oxides.Many candidate materials have been investigated.Among them, the most well-studied proton-conducting materials are perovskite-type ABO 3 oxides, namely barium cerate (BaCeO 3 ) and barium zirconate (BaZrO 3 ).In 1999, Kreuer [33] reported a proton-conducting BaZr x Y 1-x O 3-δ (BZY), which is now widely used as an electrolyte material for PCCs due to good chemical stability and high bulk proton conductivity.However, BZY materials are difficult to densify and require very high sintering temperatures (above 1600 • C).
An efficient way to obtain dense samples and reduce grain boundary resistance is to add sintering aids, such as NiO, CuO, and ZnO.Sintering aids help reduce the BZY sintering temperature and enhance grain growth [25,26].Bonanos [27] reported that the addition of 1 wt.%ZnO can significantly improve the sinterability of BZY.Previous studies have shown that improvement in sinterability is achieved at the cost of partially sacrificing in electrochemical performance by introducing electronic conductivity [34,35].However, some recent studies suggest that the overall electrochemical performance is enhanced by using sintering aids.Moreover, the sintering aids can be exsolved from the lattice of the proton conductor to recover the conductivity [36].Overall, sintering aids offer a compromise solution to advance the development of proton-conducting oxides, although an electrolyte without sintering aids is still the preferred choice [37].Another breakthrough in the fabrication of densely doped barium zirconates is solid-state reactive sintering, which consists in a single high-temperature step that combines synthesis and sintering [38,39].
To further improve the sinterability and proton conductivity of BZY, Katahira et al. [40] developed proton-conducting BaCe 0.9− x Zr x Y 0.1 O 3− δ (BZCY, with 0.0 ≤ x ≤ 0.9) oxides in 2000.Substituting Zr in BZY with Ce (BZCY) improves the sinterability and conductivity but decreases the stability.Systematic studies have been conducted to tailor the chemical stability and electrical performance of BZCY.For example, Fabbri et al. [41] studied the effects of Zr content in BaCe 0.8− x Zr x Y 0.2 O 3− δ (BZCY, with 0.0 ≤ x ≤ 0.8) on cell performance (CO 2 tolerance, electrical conductivity, etc.).Ricote et al. [42] investigated the water vapor solubility and conductivity of BaCe 0.9− x Zr x Y 0.1 O 3− δ (BZCY, 0.0 ≤ x ≤ 0.9) compounds.They demonstrated that BZCY exhibits higher ionic conductivity with increasing Ce content but improved chemical stability with increasing Zr content.
Additionally, some higher conductivity proton conduction materials based on the BZY and BZCY series have also been developed by substituting a variety of cations at the B site.Yang et al. [43] synthesized Fig. 2. Schematic of the working mechanism of SOCs based on either proton-conducting or oxygen ion-conducting electrolytes.

Hydrogen electrodes
The hydrogen electrode should possess adequate electronic, ionic conductivity as well as sufficient catalytic activity for hydrogen oxidation and evolution reactions.Moreover, the hydrogen electrode materials are required to be chemically stable in reducing atmospheres.So far, Ni-based cermets (ceramic-metal composites) are widely used as hydrogen electrodes in PCCs due to their low price and high activity.It consists of Ni and the proton-conducting phase (typically the same material as the electrolyte) to extend the reaction area and enhance the chemical and mechanical compatibility between the hydrogen electrode and the electrolyte.Ni has two main functions in PCCs: electrocatalyst and electron-conducting current collector.Ni-BaCe 0.55 Zr 0.3 Y 0.15 O 3-δ [56], Ni-BaCe 0.2 Zr 0.7 Y 0.1 O 3-δ [57], Ni-BaCe 0.7 Zr 0.1 Y 0.1 Yb 0.1 O 3-δ [44], Ni-BaZr 0.8 Y 0.2 O 3-δ [58], and Ni-BaZr 0.85 Y 0.15 O 3-δ [59] hydrogen electrodes were successfully fabricated, and showed good performance.The proportion of electron-and proton-conducting phases, microstructure, and particle size of the materials must be well chosen because they have a significant influence on the electrode and cell performance.
Zunic et al. [60] studied the influence of Ni content on microstructure and electrical properties of Ni-BaCe 0.9 Y 0.1 O 3− δ (Ni-BCY10) cermets, with 40, 50, 60 wt.%Ni (corresponding to 36, 45, and 55 vol.%Ni, respectively).They reported good Ni percolation was achieved in all three compositions.It was found that decreasing the Ni: BCY10 ratio has a significant effect on the microstructure of conducting pathways, but the electrochemical performance was only slightly changed.They concluded that the Ni-BCY10 cermet with 40 wt.%Ni can be used for fabricating hydrogen electrode-supported PCCs without significant performance degradation.
Coors and Manerbino [61] suggested that high porosity may not be essential in PCC hydrogen electrodes, as hydrogen was the only diffusing gas species in this case.Rainwater et al. [62] studied Ni-BZCYYb cermet with different porosities (37 vol.%, 42 vol.%,and 50 vol.%)and reported a monotonous decrease in performance with increasing porosity.The cell (Ni-BZCYYb/BZCYYb/La 0.6 Sr 0.4 Co 0.2 Fe 0.8 O 3-d (LSCF)-BZCYYb) with the lowest porosity in the Ni-BZCYYb electrode (37 vol.%) shows the highest performance in PCFC mode (1200 mW cm − 2 at 750 • C).Nasani et al. [63] also investigated the effect of porosity on the polarization resistance of Ni-BZY cermet for PCFC application.Pore fractions within the range of 33% ∽ 56% were realized by adjusting the content of starch porogen as pore former.They observed that the high-frequency resistance in total R P shows the largest increase with increasing the porosity, which could be ascribed to a decrease in the length of available three-phase boundaries essential for electrochemical reactions in the Ni-BZY electrode.These conclusions differ from the observations on O-SOCs, where higher levels of porosity in hydrogen electrodes may be required for optimal polarization behavior (thus requiring the use of porogens).Therefore, these studies suggest that the porosity from NiO reduction may be sufficient for PCC applications (i.e., no pore-forming agents are required).
On the other hand, degradation of the hydrogen electrode due to coking, contaminant poisoning, or catalyst agglomeration, which are challenges in O-SOCs, is less critical in PCCs.Duan et al. [36] examined the stability of PCFCs for thousands of hours, under exposure to 12 different hydrocarbon fuels (hydrogen, methane, domestic natural gas (with or without hydrogen sulfide), propane, n-butane, i-butane, isooctane, methanol, ethanol, and ammonia).They found that PCFCs possess high coking resistance and sulfur tolerance due to the existence of the proton-conducting ceramic phase.Nowadays, only a few alternatives (metals: Pd, Pd-Ag, and Ruddlesden-popper oxides: (LaSr) 0.9- Fe 0.9 Cu 0.1 O 4 ) have been proposed as novel hydrogen electrode materials for PCCs [64,65].Recently, research efforts have been devoted to further improving the Ni-based cermet performance, especially at 500 • C, by using electrode modification techniques (infiltration, exsolution, addition of functional layers, etc.) [66][67][68].It is generally agreed that the Ni-based composite electrodes meet the necessary requirements for PCC hydrogen electrodes.Since this review is intended to provide a comprehensive and in-depth overview of oxygen electrode development in PCCs, the reader is referred to other reviews for a more detailed literature review on hydrogen electrodes and electrolytes [55,21,20,69,70,4,71].

Oxygen electrodes
The main role of the PCCs oxygen electrodes is to provide the sites for the electrochemical reactionsthe oxygen reduction reactions (ORR)/ oxygen evolution reactions (OER), and the water-splitting reactions (WSR).Also, it must provide sufficient electronic conductivity for the current collection.Due to the distinct cell working mechanisms in PCEC and PCFC, there exist some differences on various aspects of the oxygen electrode as summarized in Table 1, such as the electrode reaction process, operating atmosphere, material requirement, etc.The elementary oxygen electrode reaction processes proposed in literature are listed for both PCEC and PCFC [72,4], with slight changes in these steps depending on the oxygen electrode materials [73].In general, the oxygen electrode reactions in PCEC and PCFC are in a reversed order.In PCEC mode, steam is dissociated into oxygen and protons at the oxygen electrode under a current supply, and pure and dry hydrogen evolves at the hydrogen electrode.As already mentioned in the previous section, dry hydrogen is produced only if the electrolyte is a pure-proton conductor, which is not the case for BZCY materials at high temperatures [14,32,15].In PCFC mode, the protons transport through the electrolyte and react with oxygen forming steam on the oxygen electrode side.In contrast to a typical O-SOFC where steam is generated at the hydrogen electrode, producing steam at the PCFC oxygen electrode improves fuel utilization and efficiency.Furthermore, the absence of steam on the fuel electrode slows down Ni agglomeration, which is otherwise promoted by the presence of steam and high temperature.
As mentioned in the oxygen electrode reaction steps in PCCs (PCEC and PCFC), in addition to the electronic and oxygen ion conductions, the protons play an important role during the oxygen electrode reaction processes.In the absence of proton conduction, the ORR and WSR reactions occur only at the interface between the oxygen electrode and the electrolyte, leading to poor performance [74,75].There are various terms used to describe electrode performance: area-specific resistance (ASR) or polarization resistance (R p ).In this work, the latter term will be used for convenience.The electrode performance, R p , depends on different factors, which can be categorized into two main groups: electrode structure/morphology and material composition.To ensure stable and robust operation, on the one hand, the PCCs oxygen electrode structure requires sufficient pathways for both electrons/ions and gasses, as well as high surface area for better catalytic activity.On the other hand, the oxygen electrode material shall possess sufficient electronic/ionic conductivity (e − /O 2− /H + ), high catalytic activity, good chemical compatibility, and a thermal expansion coefficient (TEC) that matches with other cell components.In PCEC, the oxygen electrode is exposed to atmospheres with high steam content (≥10%), imposing challenges for the hydrothermal stability of the oxygen electrode.When selecting oxygen electrode materials, it is thus very important to consider their steam tolerance and chemical/phase stability under high steam partial pressures.These issues become less critical in PCFC, as the atmosphere there is either dry or typically with 3% H 2 O.In addition, the different elementary steps require specific air-electrode catalytic activities: The PCEC oxygen electrodes should exhibit sufficient OER catalytic activity, while the PCFC ones require high ORR catalytic activity.Thus, it is clear from Table 1 that the best oxygen electrode for PCEC is not necessary the best for PCFC.Rational design of electrode materials for PCCs should thus be made according to the specific electrode working temperature and atmosphere.
It is important to highlight that measuring the R p in a symmetrical cell configuration in an oxidizing atmosphere leads to an underestimation of R p due to the mixed conductivity nature of the proton-conducting electrolyte in different atmospheres.This issue is more severe at high temperatures, as mentioned in Section 1.3.1 (mixed ionic-electronic conductivity).The reader can refer to [29] for a more detailed explanation.The take-home message is that one has to be careful when comparing results on symmetrical cells: only the studies using similar electrolyte materials and tested in similar gas atmospheres and temperatures should be compared directly.

The objective of this review
The major bottleneck in the PCCs development has been the sluggish reaction kinetics at the oxygen electrodes and insufficient stability for commercial application.The requirements for appropriate oxygen electrodes with good performance have become more and more stringent with the development of PCCs technology which aims to operate at 400-600 • C. Table 1 also summarizes the main degradation mechanisms and challenges in PCEC and PCFC.Although differences in the oxygen electrode material choice between PCEC and PCFC are highlighted, one shall keep in mind that the development of PCEC and PCFC (and PCC in general) is synergistic.Reviewing the literature over the last decade evidences this statement: the main achievements in PCEC are attributed to the development of feasible, cost-effective fabrication methods and promising electrode materials for both PCFC and PCEC.That is why the oxygen electrode in PCEC and PCFC was discussed together in this review.
This review aims to provide a comprehensive overview of the research progress and scientific achievements concerning the PCCs oxygen electrodes, revealing important insights into the mechanisms of oxygen electrodes, and establishing a scientific basis for the rational design of high-performance and durable oxygen electrodes.Finally, problems limiting the exploitation of PCCs are analyzed to provide guidelines for further development.

Electrode configuration and processing
The electrochemical reactions can occur only where electron conductive, ion conductive, and gas phases coexist.The simultaneous presence of the three phases allows the conduction of electrons, the migration of ions, and the transport of gas molecules to/from the reaction sites.These regions are called triple-or three-phase boundaries (TPBs).The surface structure and morphology of the oxygen electrode significantly affect the electrochemical reaction active sites and mechanical stability.According to the employed electrode fabrication techniques, the oxygen electrode structure can be classified into different categories: (a) Single-phase/composite structure realized by screen printing or spray painting (Fig. 3a) [76,77]; (b) Nanoparticle-coated skeleton structure formed by infiltration or in-situ exsolution (Fig. 3b) [78][79][80]; (c) Multilayered film using deposition techniques, such as pulsed laser deposition (PLD) and atomic layer deposition (ALD) (Fig. 3c) [49,81,82]; and (d) Three-dimensional (3D) structure formed by electrospinning or self-assembly (Fig. 3d) [83,84].Owing to their significantly different structures, these oxygen electrodes display large variations in local composition, lattice strain, and electrochemical properties.In this section, some advanced electrode fabrication techniques are also discussed.

Composite structure
For composite electrodes, brush painting is a simple and costeffective solution process that has been extensively reported.Its advantages include low material loss, ultra-low-cost, vacuum-free processing, and processability on various flat and textured substrates.It is widely used for the preparation of small PCCs button cells.For example, the SrEu 2 Fe 1.8 Co 0.2 O 7− δ (SEFC) and BZY powder (in a weight ratio of 95: 5) were mixed with terpineol as the oxygen electrode ink, and brush painted on the BZY electrolyte surface with an effective area of 0.385 cm 2 .Here, 5% BZY is added to improve the bonding between the electrolyte and oxygen electrode.The as-prepared samples were heated at 1000 • C for 2 h to complete cell fabrication [17].However, there are still some issues to be solved regarding brush painting such as difficulty in thickness control, limitation of functional ink, and difficulty of automation.
Screen printing is simple, flexible, and economical, and it is the most widely used technique to fabricate oxygen electrodes with a thickness in the range of 10-100 µm.It involves typically the following steps: (1) preparation of electrode material powders with fine particle sizes, (2) preparation of an electrode slurry (with the appropriate set of binders, plasticizers, and solvent), (3) passage of the slurry through the aperture of the screen mesh, and (4) drying and firing at high temperature.
Fabrication of optimized screen-printing inks is critical for the production of high-quality films with improved performance.An et al. [85] screen-printed a Ba 0.5 Sr 0.5 Co 0.8 Fe 0.2 O 3-δ (BSCF) oxygen electrode onto the BZCY electrolyte with an effective area of 4 × 4 cm 2 , which achieves a total cell R P of 0.09 Ω cm 2 and delivers a power density of 1302 mW cm − 2 at 600 • C. Screen printing plays an important role in the up-scaling and commercialization development of PCCs.

Nanoparticle-coated skeleton structure
Infiltration is a promising approach for preparing high-performance nanostructured electrodes for SOCs [86,87].An infiltration process consists of two steps: preparation of the electrode backbone and infiltration of the electroactive catalyst.After being infiltrated into the porous backbone (typically the same composition as the electrolyte) and further calcination, the catalyst solution can form two different morphologies: discrete particles or continuous thin-film coating.The advantages of using the infiltration technique to fabricate the composite electrode include: (1) the particle size of the infiltrated catalyst can be kept in the sub-100 nm range, which creates a large surface area; (2) the extra surface area can be introduced without sacrificing the mechanical compatibility between the electrode and the electrolyte; and, finally, (3) the cation inter-diffusion between the two phases, that would occur during electrode firing, which takes place at much higher temperatures, is avoided.For example, enhanced electrocatalytic activities were reported by Wu et al. [88] for a Sm 0.5 Sr 0.5 CoO 3-δ (SSC)-infiltrated BaCe 0.8 Sm 0.2 O 2.9 (BSC) oxygen electrode in PCFC.The SSC-infiltrated BSC electrode shows a polarization resistance of about 0.21 Ω cm 2 at 600 • C, only one-third of that of the SSC-BSC composite electrode prepared by screen printing.However, many challenges remain with the current infiltration technology, such as reproducibility, the possible particle coarsening, and thermal instability.In addition, the process is time-consuming as it requires several infiltrations with intermediate calcination steps.Lastly, it may seem unsuitable for large-size cell fabrication unless may be semi-or fully automated.Therefore, the practical application of infiltration still requires substantial effort and validation.
The exsolution of catalytic nanoparticles on external or inner surfaces of the electrode is an attractive way to design smart functional materials for renewable energy applications (photocatalysis, SOCs, batteries, etc.) [89,80,90].Unlike the conventional deposition techniques (screen-printing, painting), exsolution produces higher quality and better-distributed nanoparticles.It is also time-and cost-effective.Moreover, the in-situ growth of nanoparticles may avoid catalyst agglomeration as the nanoparticles are socketed, thus increasing the catalyst's lifetime [91].Many conductive ceramics have perovskite or perovskite-related structures as ideally suited for use in PCC electrodes that take advantage of the exsolution concept.Previous studies show that strongly reducing conditions are typically required to bring metals to the surface of these exsolution catalysts [92].Hence, this methodology has primarily been used to exsolve metal nanoparticles from perovskite oxide lattices, such as nickel and ruthenium [93,94], an approach that is suitable for hydrogen electrodes but not amenable to oxygen electrodes.In contrast, Rioja-Monllor et al. [80] used a novel exsolution process to fabricate complex all-oxide nanocomposite oxygen electrodes for PCFCs.The single-phase precursor La 0.3 Ba 07 Zr 0.4 M 0.6 O 2.75 (M is Mn, Fe and/or Co) was synthesized by a modified Pechini route.The precursor was heated at 900 • C in air to achieve a composite consisting of La 1-x Ba x CoO 3-δ -and a BaZrO 3 -based material.The exsolution process results in a highly nanostructured and intimately interconnected percolating network of the two final phases, one proton-conducting BaZrO 3 -based and one mixed oxygen ion and electron conducting (La 1-x Ba x CoO 3-δ ), yielding excellent oxygen electrode performance.In another study, Chen et al. [95] reported a hybrid oxygen electrode based on PrNi 0.5 Mn 0.5 O 3 (PNM) with exsolution fluorite PrO X particles produced in situ through a glycine-nitrate solution combustion process, which shows excellent electrocatalytic activity with R P of 0.052 Ω cm 2 in ambient air at 700 • C under symmetrical cell testing (PNM-PrO X //BZ-CYYb//PNM-PrO X ).Meanwhile, the full cell with a configuration of Ni-BZCYYb//BZCYYb//PNM-PrO X achieved a power density of 650 mW cm − 2 and long-term stability (~ 500 h) at a cell voltage of 0.7 V with 3% humidified H 2 as fuel and ambient air as the oxidant.Although this method has received attention in recent years, a comprehensive understanding of exsolution phenomena in perovskite oxides is still lacking because of difficulties in finding a rational combination of driving forces and perovskite supports.

Thin-film deposition
To reduce the ohmic resistance and improve cell performance, advanced thin film deposition technologies such as vacuum deposition, sputtering, pulsed laser deposition (PLD) and atomic layer deposition (ALD) have been widely used in PCCs for preparing thin, fully dense electrolyte layers (≤ 1 μm) or functional layer between electrolyte and electrode [96,68,11].For example, Choi et al. [68] deposited a thin (100 nm) dense PrBa 0.5 Sr 0.5 Co 1.5 Fe 0.5 O 5+δ (PBSCF) interlayer by PLD between a PBSCF oxygen electrode and a BZCYYb electrolyte to improve the interface contact.A significant reduction in ohmic losses by ∽ 0.1 Ω cm 2 at 600 • C was achieved as compared to cells without the PLD layer.On the other hand, the thin film electrodes were studied extensively as a model system for functional studies such as the electrode reaction process (ORR, OER) or active site.

3D fiber structure
Electrospinning is a simple and effective method to produce nanofibers, which has attracted considerable attention.The fabricated fibers display unique characteristics, such as high porosity, small diameter, excellent pore interconnectivity, and a high surface-to-volume ratio [97,98].These properties make electrospun fibers the optimal candidates for important potential applications, including PCCs.As already mentioned, the PCC performance can be significantly altered by changing the composition and microstructure of the oxygen electrode.And electrospinning is a striking example.Tang et al. [83] prepared 3D nanofiber-structured La 2 NiO 4+δ (LNO) and La 2 Ni 0.6 Fe 0.4 O 4+δ (LNF) oxygen electrodes incorporated into PCFCs with a half-cell Ni-BZ-CY//BZCY, resulting in an encouraging cell peak power density of 508 mW cm − 2 and 551 mW cm − 2 at 700 • C, respectively.Compared with oxygen electrodes made from powders, the nanofiber-structured one shows higher electrochemical performance, suggesting that the construction of nanofiber-structure electrodes provides an alternative and promising route to prepare high-performance oxygen electrodes for PCCs.Only a few results have been published on electrospun PCC electrode materials.Consequently, there are still some fundamental open questions that need to be addressed: How can the stable nanofibers structure be maintained during sintering?How do the nanofibers evolve under long-term operation?How can electrospinning become an easily scalable process?
In general, each fabrication method presents its advantages and disadvantages and application fields in electrode development.Different electrode fabrication methods also result in different electrode/electrolyte interface structures.An ideal interface structure requires sufficient adhesion, well-adjusted chemical compatibility, and well-matched thermal expansion behavior between the two adjacent layers (electrode/ electrolyte, support layer/electrode, functional layer/electrolyte) [99].Interface adhesion is one of the overlooked issues concerning the adjacent layers in the manufacture of PCCs.Compared with the hydrogen electrode, the adhesion between the oxygen electrode and the electrolyte is often weaker because of the lower sintering temperature required to prevent oxygen electrode densification and cation interdiffusion [100].As a result, the oxygen electrode could be easily peeled off from the electrolyte.Consequently, some studies have reported that cell electrochemical performance is mainly dependent on the oxygen electrode/electrolyte interface [101,102,68].Delamination between the electrolyte and the electrode is one of the main reasons for cell degradation.The stronger the interfacial adhesion, the better the interface bonding; thus, the smaller the interfacial contact resistance, and the better the cell performance.In conventional O-SOCs, many advancements have been made in enhancing the interface adhesion, such as inserting a functional layer between the electrode and electrolyte.However, there is little work that exclusively focuses on interface adhesion issues in PCCs.Recently, Choi et al. [68] deposited a dense PBSCF layer between the oxygen electrode and electrolyte to improve the interface contact, dramatically reducing ohmic losses.

Oxygen electrode materials
In the early development of PCCs, the oxygen electrode materials were inspired by those used in O-SOCs due to their operating similarities.However, most conventional oxygen electrode materials in O-SOCs possess poor/no proton conductivity and low ionic/electronic conductivity under intermediate temperatures.PCC oxygen electrode materials require electrochemical activity not only for O 2− and e − , but also for H + .Therefore, increasing efforts have been devoted to enhancing the oxygen electrode material activity by introducing proton conduction.This is realized using either single-phase oxides with O 2− /H + /e − conductivities or composite electrodes made with a mixed ionic/electronic conductor and a proton conductor (typically, the same material as the electrolyte).Furthermore, the ionic/electronic conductivity of oxygen electrode materials under intermediate temperatures must be increased by developing novel materials or composition optimization.
In this section, we present a review of the research process, advancements, and challenges in the development of oxygen electrode materials in PCCs, and various strategies to boost the material WSR/ ORR performance, with a schematic illustration in Fig. 4. Generally speaking, three types of materials can be selected for the PCC oxygen electrodes: (1) mixed ionic (oxygen ion)-electronic conductor (MIEC), (2) mixed protonic-electronic conductor (MPEC), and (3) tripleconducting oxides (TCOs).Fig. 5 shows the potential reaction pathways and reaction sites at these three types of PCCs oxygen electrode materials for both PCFCs and PCECs.Fig. 5a illustrates the possible reaction pathways in the composite oxygen electrode containing a protonconducting phase and an MIEC.The TPB sites exist at the contact points between the MIEC phase, the proton-conducting phase, and the gas/pore phase.There are two potential pathways for the transfer of oxygen ions: the surface and bulk pathways.The contribution of the bulk transfer to the electrode reaction process depends on the oxygen surface exchange coefficient (k) and the bulk oxygen ion diffusivity (D) of the MIEC phase, as well as the electrode microstructure.Fig. 5b illustrates the working mechanism using an MPEC material as the oxygen electrode in both PCEC and PCFC.Here two potential pathways exist for the proton transfer: surface and bulk pathways.The relative contributions of each pathway depend on both the surface and bulk transport properties of the MPEC material and the electrode microstructure.The oxygen electrodes based on TCO can spatially expand the active electrochemical reaction zone, thus maximizing the active electrode area.The two potential pathways for the transfer of oxygen ions and protons (surface and bulk) are depicted in Fig. 5c.
The performance of the PCC oxygen electrodes reported in literature has been summarized in Table 2 for those measured in symmetrical cells, in Table 3 for PCFCs, and in Table 4 for PCECs.From Tables 2-4, it can be seen that most of the studies on PCCs oxygen electrodes have been focused on using two-or multi-phase composite materials.One can observe that the development of oxygen electrodes in PCCs is based on conventional oxygen electrode materials in O-SOCs.

MIEC materials
MIECs are the most popular oxygen electrode materials in PCCs owing to their great achievements and wide application in O-SOCs during the past decades.However, when using an MIEC as the oxygen electrode in PCCs, the reaction sites are limited to the TPBs at the Q.Wang et al. electrode-electrolyte interface due to the lack of proton charge carriers and corresponding transfer pathways.To resolve this issue, the MIEC materials can be mixed with a proton-conducting phase (often the same material as the electrolyte) to form a composite electrode (Fig. 5a).Composites can be prepared using different electrode fabrication techniques such as conventional physical mixing, infiltration, or in-situ exsolution to extend the TPBs.

Cubic-type perovskite oxides (ABO 3 )
The perovskite-type oxides (ABO 3 ) are the most intensively investigated oxygen electrode materials [190].In the ABO 3 structure, the large A cation with 12-fold oxygen coordination can be rare earth, alkaline earth, or alkali elements, whereas the smaller B site with 6-fold oxygen coordination can accommodate many transition metals (Mn, Co, Fe, Cu, and Ni).ABO 3 displays a diversity of properties, such as electric, optical, and magnetic properties, because of the feasibility of substituting the Aor B-sites cations.90% of the elements in the periodic table can be accommodated in the perovskite structure [191,192].The B-site cations have a significant influence on electrocatalytic activity because the B-site is often associated with the reaction site.A proper choice of B dopants can modify the electronic conductivity and catalytic properties [193,192].Generally speaking, cation substitution (A-and/or B-sites) in ABO 3 is an effective approach to modifying the properties of the oxide.
Initially, strontium-doped-samarium-cobaltite materials (Sm 0.5 Sr 0.5 CoO 3-δ , SSC) were widely applied as the oxygen electrode material in PCCs because of their high electrical conductivity (up to 10 3 S cm − 1 ), high surface oxygen exchange rate and bulk oxygen ion      between high electronic conductivity and good ionic conductivity (10 2 and 1 × 10 − 2 S cm − 1 at 800 • C, respectively), good catalytic properties (given by the Co cations), and improved stability at high temperature [199,200].LSCF has also been widely used as the oxygen electrode material in PCCs.The LSCF-BZCY composite oxygen electrode in PCFC (LSCF-BZCY//BZCY//Ni-BZCY) showed high catalytic activity toward oxygen reduction, demonstrating a peak power density of 522 mW cm − 2 and total R P of 0.47 Ω cm 2 at 650 • C [122].Ricote et al. [78] studied the electrochemical performance and microstructure of the LSCF-BZCY composite oxygen electrodes in a symmetric cell configuration.The electrodes were fabricated by either infiltration or spray pyrolysis.Analysis of the electrochemical impedance spectra at different temperatures revealed that the charge transfer contribution was lower for the infiltrated electrode, while the oxygen dissociation/adsorption contribution was lower for the spray-pyrolysis one.To further enhance ORR and OER kinetics and stability of the LSCF oxygen electrode, Zhou et al. [169] reported an efficient bifunctional oxygen electrode for reversible protonic ceramic cells (R-PCCs) by infiltrating a catalyst coating of barium cobaltite (BCO) into LSCF surface.R-PCCs with the BCO-LSCF oxygen electrode showed remarkable electrochemical performance and outstanding durability.The polarization resistance for BCO-LSCF was lower than that of the bare LSCF under the same conditions (only accounting for 30% of bare LSCF).A low degradation rate of 0.8%/1000 h (or 10 mV/1000 h) was reported during the 1150 h test in the electrolysis mode.The performance enhancement is mainly attributed to the enhanced kinetics of the surface oxygen exchange process enabled by the BCO catalyst.
Another interesting material known for its oxygen surface exchange and diffusion properties is Ba 0.5 Sr 0.5 Co 0.8 Fe 0.2 O 3-δ (BSCF).The optimized composition of BSCF as the oxygen electrode yields good O-SOC performance when combined with a doped ceria electrolyte.As previous studies have shown that BSCF displays proton conduction in humidified atmospheres [201], it is thus a potential candidate for intermediate-temperature PCCs [96,202].Lin et al. [121] reported a BSCF-BZCY composite oxygen electrode for PCFC, and the single-cell with BSCF-BZCY//BZCY//Ni-BZCY (functional layer)//Ni-BZCY showed a power density of 267 mW cm − 2 at 600 • C. Li and Xie [173] compared LSCF and BSCF as oxygen electrodes in PCEC based on a proton-conducting electrolyte BaCe 0.5 Zr 0.3 Y 0.16 Zn 0.04 O 3− δ (BCZYZ) and a Ni-BCZYZ hydrogen electrode.Their results revealed that the cell with the BSCF oxygen electrode has a lower cell R P of 0.075 Ω cm 2 at 800 • C, as compared to 0.36 Ω cm 2 for the cell with the LSCF oxygen electrode.This illustrates that BSCF outperforms LSCF under the investigated conditions.In addition to PCFC and PCEC applications, BSCF was also proposed as the oxygen electrode material for R-PCCs.Marrony and Dailly [203] reported 4.5 × 4.5 cm 2 R-PCCs (BSCF-BZCY//BZCY-Z-nO//Ni-BZCY) with a BSCF current collector prepared by screen printing.R-PCCs showed promising performances: >250 mW cm − 2 at 0.7 V in  fuel cell mode, 50 mA cm − 2 at 1.3 V in the electrolysis mode, and reliability during more than 3500 h with limited degradation of < 2%/kh at 700 • C. Note that this degradation rate is lower than those achieved for proton-conducting cells and comparable to those of oxide ion-conducting electrolyte-based electrolysis cells.Despite its excellent electrocatalytic activity and durability, the use of BSCF can be problematic since its thermal expansion coefficient (TEC) (25.1 × 10 − 6 K − 1 between 50 and 1000 • C) does not match the TEC of proton-conducting oxides (8 × 10⁻⁶ ~ 11.2 × 10 − 6 K − 1 ) [204,205].Moreover, BSCF, as most of the perovskite oxides containing alkaline earth elements, is susceptible to CO 2 poisoning, especially at temperatures below 550 • C, where poisoning effects are irreversible.Improving the stability of BSCF in CO 2 -containing environments and the TEC matching degree between BSCF and electrolyte were tried out by partially replacing Fe in BSCF with Ta.There are reports showing that Ta doping can reduce the thermal expansion coefficient and increase the CO 2 tolerance of the oxygen material in SOCs.The TEC of Ta-doped BSCF, Ba 0.5 Sr 0.5 Co 0.8 Fe 0.1 Ta 0.1 O 3-δ (BSCFT), was decreased to 17.9 × 10 − 6 K − 1 .Theoretical calculations by density functional theory show that the CO 2 adsorption energy of BSCFT (− 0.78 V) is 0.12 eV higher than that of BSCF (− 0.9 V), indicating that BSCFT exhibited improved resistance to CO 2 corrosion compared with BSCF [204].The hydration properties and proton migration of BSCF can also be further optimized by partially replacing barium in BSCF with potassium.Xu et al. [144] designed a novel oxygen electrode material Ba 0.4 K 0.1 Sr 0.5 Co 0.8 Fe 0.2 O 3-δ (KBSCF) and found that K-doping enhances the electrochemical performance of the BSCF electrode and reduces the R p .The cell (KBSCF-BZCY//BZ-CY//Ni-BZCY) reached a peak power density of 441 and 1275 mW cm − 2 at 600 and 700 • C, respectively, which are significantly higher than those of the cell using the BSCF oxygen electrode (289 and 904 mW cm − 2 at 600 and 700 • C, respectively).Meanwhile, the cell R p at 700 • C also decreased from 0.088 (without K-doping) to 0.054 Ω cm 2 .While cubic Co-containing SSC, LSCF, and BSCF perovskites have been developed and applied in PCCs, the flexible Co(III)/Co(IV) redox behavior leads to reduced chemical stability and high TEC [206].In addition, the cost and availability of cobalt are an issue.Accordingly, Co-free oxygen electrodes based on SrFeO x are getting increasing attention due to their low TEC, low cost, and less flexible redox behavior of Fe (thus enhanced stability).Three types of SrFeO x -based oxygen electrode materials have been explored to enhance electrochemical performance.Type-A stands for A-site substitution, such as La 0.5 Sr 0.5 FeO 3-δ (LSF), Pr 0.5 Sr 0.5 FeO 3-δ (PSF), La 0.25 Pr 0.25 Sr 0.5 FeO 3-δ (LPSF) [165], Ba 0.5 Sr 0.5 FeO 3-δ (BSF) [124], and Sm 0.5 Sr 0.5 FeO 3-δ (SSF) [127]; Type-B stands for B-site substitution, such as SrFe 0.95 Nb 0.05 O 3-δ (SFN) [77], while type-AB denotes both A and B site substitution, such as LaNi 0.6 Fe 0.4 O 3-δ (LNF) [139]; La 0.6 Sr 0.4 Fe 0.9 Mo 0.1 O 3-δ (LSFMo) [167], La 0.6 Sr 0.4 Fe 0.8 Nb 0.1 Cu 0.1 O 3-δ (LSFNC) [65].Among these oxygen electrode materials, PCFC single cells with the cobalt-free LSFMo-BZCY composite oxygen electrode showed excellent performance, demonstrating a peak power density of 1174 mW cm − 2 at 700 • C [167].

Double-perovskite oxides (AA'B 2 O 6 )
Another class of materials derived from the perovskite structure is the oxide materials known as double perovskites with the general formula AA'B 2 O 6 (A = rare earth, A' = alkaline earth, and B = Co or Mn).Barium is the most commonly used alkaline earth cation because cation ordering is induced by the size mismatch between the rare-earth and alkaline earth metal radii.The materials of this family are typically oxygen-deficient systems with oxygen vacancies mainly located in the rare-earth layer.Recently, double perovskite oxides have also been developed as the next-generation electrode materials.One of the promising properties of double perovskites is their electronic conductivity (up to 10 3 S cm − 1 ) in the temperature range of 200-900 • C [207].Moreover, they show rapid surface exchange rates as well as bulk diffusion rates due to the large amounts of oxygen vacancies and anisotropic oxygen-diffusion behavior [208,209].
Most of the studies on double perovskite have focused on pure or partially substituted LnBaCo 2 O 5+δ ((Ln = rare earth and Y) oxides.Systematic studies of LnBaCo 2 O 5+δ oxides have shown that with decreasing the Ln 3+ ionic radius (Pr 3+ > Sm 3+ > Gd 3+ > Y 3+ ), the TEC, electrical conductivity, and ionic conductivity decrease due to the increase in the oxygen ion vacancy concentration and bending of the O-Co-O bonds.Single-phase GdBaCo 2 O 5+δ (GBCO), SmBaCo 2 O 5+δ (SBCO), and PrBaCo 2 O 5+δ (PBCO) have been applied as the oxygen electrode on a BZCY (10-25 µm) electrolyte.The cell R P at 700 • C, which is measured on cell configuration LnBaCo 2 O 5+δ //BZCY//Ni-BZCY, was reported as 0.15, 0.16, and 0.06 Ω cm 2 for oxygen electrode SBCO [210] GBCO [211] and PBCO [125], respectively.The performance of PBCO is superior.This may result from the unusually rapid oxygen ion diffusion rate and surface exchange kinetics for PBCO compared to those for the LnBaCo 2 O 5+δ family.Similar to the other perovskite family, there have been significant efforts to explore the compositional variations, which confirm the strong relationship among cation substitutions, crystal structure, and oxygen contents.For example, PBCO was further modified by partially substituting Ba with Sr, and substituting Co with Fe.In particular, the PrBa 0.5 Sr 0.5 Co 1.5 Fe 0.5 O 5+δ (PBSCF) oxygen electrode doped with Sr and Fe showed good performance.A PCFC of PBSCF//BZCYYb//Ni-BZCYYb has achieved a peak power density exceeding 500 mW cm − 2 at 500 • C [68].Liu et al. [212] adopted partial substitutions of Ca for Ba and Zn for Co to improve the electrochemical performance and structural stability of PBCO in PCFC.PrBa 0.9 Ca 0.1 Co 2-x Zn x O 5+δ (PBCCZx, x = 0, 0.05, 0.10, 0.15) were prepared and investigated in the aspects of crystal structure, thermal expansion behavior, defect chemistry, and electrochemical  performance.With the increase in the content of doped Zn, the TEC of PBCCZx remains unchanged essentially.But the bulk and surface concentrations of oxygen vacancy are increased, and the maximum power density of the PCFCs at 750 • C with PBCCZx-BZCYYb composite oxygen electrode is improved from 335 (x = 0) to 876 (x = 0.15) mW cm − 2 .These examples show the opportunity to find the optimal composition by fine-tuning chemical compositions (e.g., multiple cationic substitutions for both A-and B-sites).

Ruddlesden-Popper (RP) phase oxides (A n+1 B n O 3n+1 )
In addition to the above type perovskite oxides, the Ruddlesden-Popper (RP) series of oxide A n+1 B n O 3n+1 has also received extensive attention [213,214].A n+1 B n O 3n+1 is comprised of n consecutive perovskite layers (ABO 3 ) alternating with a rock-salt layer (AO) along the crystallographic c-axis direction.They possess considerably higher oxygen exchange kinetics, leading to fast ion conduction and potential application as oxygen electrodes at intermediate temperatures [215].Of particular interest for PCCs are A 2 BO 4 and A 3 B 2 O 7 (n = 1, 2).The A-site cations are usually occupied by rare earth metal ions (La 3+ , Pr 3+ , Gd 3+ , Nd 3+ ), whereas the B-site is often filled with transition metal ions (Ni 2+ , Co 2+ , Cu 2+ ).To increase the electronic/ionic conductivity, the A-and B-sites can be tailored by partial substitution with alkaline earth metal ions such as Ba 2+ , Sr 2+ , and Ca 2+ .
Among the various RP oxides, Ln 2 NiO 4+δ -based oxides, such as La 2 NiO 4+δ (LNO) [216,217] and Pr 2 NiO 4+δ (PNO)-based [182,218] are the most widely studied owing to their electrocatalytic properties, relatively low TEC [71], superior oxygen ion transport and excellent chemical compatibility with the BaCeO 3 -and BaZrO 3 -based electrolyte materials [219][220][221][222]. Furthermore, avoiding Co, which generally requires special recycling, is considered advantageous.The electrical conductivity of Ln 2 NiO 4+δ and its derivatives is predominantly p-type electronic.As expected for a layered structure, the electrical properties of Ln 2 NiO 4+δ nickelates are highly anisotropic.It has been demonstrated that the electrical conductivity of LNO single crystals along the a-b plane is more than three orders of magnitude higher than that along the c-axis [223].Additionally, the oxygen-ionic transport displays a strong anisotropy (oxygen diffusivity along the a-b plane is about three orders of magnitude faster than along c axis) [219,224].The RP oxides may have phase stability issues, such as the decomposition of PNO to Pr 4 Ni 3 O 10 (n = 3) and Pr 6 O 11 at above 850 • C in pure oxygen .Note that the segregation of praseodymium oxide, having high catalytic activity in electrochemical reactions involving gaseous oxygen, may be advantageous for the PCCs oxygen electrode [142].
LNO was mixed with a new proton-conducting electrolyte material, BaCe 0.5 Zr 0.3 Dy 0.2 O 3-δ (BCZD), as a composite oxygen electrode for R-PCCs (LNO-BZCD//BCZD//Ni-BCZD//Ni-BCZD) and demonstrated reasonable cell performance in both fuel cell and electrolysis cell modes [175].Li et al. [182] used PNO-BZCY as the oxygen electrode in PCECs.The corresponding cell (PNO-BZCY//BZCY//Ni-BZCY) achieved a current density of 977 mA cm − 2 at 700 • C with a 1.3 V electrolyzing potential, i.e., a 0.37 V overpotential, which is one of the best PCEC performances reported so far.Furthermore, the PNO-BZCY oxygen electrode accounts for only 16% of the overall cell R p at 700 • C.However, the property limiting the electrochemical performance of these RP phases seems to be their moderately low electronic conductivity (∽ 50-100 S cm − 1 ) in the intermediate temperature range.Research on the substitution of A element has been conducted to improve its performance.With the low-valence dopants, Sr 2+ or Ca 2+ , replace La 3+ , such as in La 2-x Sr x NiO 4+δ (LSN) [142] and La 2-x Ca x NiO 4+δ (LCN) [164], extra electronic holes can be produced (localized on the Ni ions and presented as Ni 3+ ) as charge compensation [225].For example, Inprasit et al. [226] reported the effects of substituting Sr for La on the phase, sintering, and conducting behaviors of LSN (x = 0, 0.2, 0.4, 0.6, 0.8).They found that the Sr-substitution can improve the stability and electrical conductivity of LNO.The La 1.2 Sr 0.8 NiO 4 composition was synthesized via the sol-gel process and showed the highest electrical conductivity (160 S cm − 1 ) at 500 • C, with a TEC value of 13 × 10 − 6 K − 1 between 400 and 700 • C. The introduction of Ca leads to the elongation of the La-O (2) bond length, which provides more space for the migration of oxygen ions in the La 2 O 2 rock salt layers.The substitution of Ca remarkably increases the electronic conductivity of LCN [227].
Higher-order RP phases of Sr 3 Fe 2 O 7-δ (SFO, n = 2)-based materials are also considered promising oxygen electrode materials, especially in the electrolysis mode, owing to their unique ion transporting properties, low proton formation energy (~ 0.23 eV), good stability in the high steam content atmosphere, and fast oxygen surface exchange kinetics [136].Huan et al. [141] doped SFO with Eu and Co to form SrEu 2 Fe 1.8 Co 0.2 O 7-δ (SEFC) as a robust, stable, and efficient oxygen electrode for R-PCCs.Sr atoms occupy the center of perovskite slabs, whereas Eu atoms form an ordered arrangement in the rock-salt layer, as illustrated in Fig. 6a.The structural stability of SEFC was evaluated under a high steam atmosphere (10% H 2 O− air).The XRD patterns in Fig. 6b show SEFC is structurally stable after heating at 600 • C for 100 h.The button cell with SEFC oxygen electrode exhibits stable performance: over the 135 h FC-/EC switching tests.In addition, R S remains stable, whereas regular fluctuations are observed in R P due to the different species adsorbed on the electrode surface in the EC and FC operating modes within the 135 h test.In another study, Shi et al. [17] used the SEFC oxygen electrode in PCEC (SEFC-BZY//BZY//Ni-BZY) to achieve sustainable carbon cycling.In their concept, CO 2 could be efficiently and controllably converted into CO and CH 4 .A high current density of 1730 mA cm − 2 at 1.5 V was reached, and the ohmic resistance and R P were measured to be 0.69 and 0.84 Ω cm 2 at 600 • C with a 20% CO 2 -80% H 2 gas mixture to the hydrogen electrode.Meanwhile, the cell exhibited steady operation for almost 100 h at a current density of 800 mA cm − 2 for CO 2 conversion at 550 • C.

MPEC materials
It is highly desirable for the oxygen electrode materials in PCCs to have proton conductivity, thus utilizing a larger part of the electrode surface.As discussed in MIEC materials, an effective oxygen electrode can be made by mixing a conventional MIEC with a proton conductor.Another strategy is to design and synthesize single-phase oxides with highly active catalytic surfaces and mixed protonic and electronic conduction (MPEC).The MPEC oxides must exhibit sufficient electronic conductivity (typically > 1 S cm − 1 ) to ensure adequate current collection and fast charge transfer.
However, MPEC oxides with high catalytic activity and stability are notoriously elusive.Only a few studies have been published concerning the hydration, hydrogenation, and proton uptake of perovskite-related MPEC oxides.Quantifying the proton conductivity and migration of a mixed-conducting oxide is not trivial because complications arise from the fact that the protons can be combined with some electronegative atoms/ions and can be shared between two electronegative atoms (e.g., O − H⋯O) [228].In addition, there is very little direct experimental evidence regarding the effect and contribution of protons on the electrode reaction processes.In 2015, Strandbakke et al. [229] investigated and analyzed the hydration properties of the Ba 0.5 Gd 0.8 La 0.7 Co 2 O 6-δ (BGLC) double-perovskite using TGA.The BGLC exhibits mixed proton (proton concentration 3 mol%) and electron conductions.As a result, it may be considered a possible MPEC material for use in oxygen electrodes for PCCs.Vøllestad et al. [184] prepared a fully operational BaZrO 3 -based tubular PCEC with a BGLC oxygen electrode for steam splitting.The BGLC-BZCY//BZCY//Ni-BZCY cell exhibited a total R P below 1 Ω cm 2 at 600 • C and a Faradaic efficiency close to 100% in a high steam atmosphere.In addition, they reported that BGLC is chemically stable in steam (1.5 bar) for 100 h at 600 • C, demonstrating it is suitable as the oxygen electrode material for PCEC.Leonard et al. [230] also investigated and compared the cell R P of Ba 0.5 La 0.5 CoO 3− δ (BLC) and BGLC oxygen electrodes on PCECs based on the SrZr 0.5 Ce 0.4 Y 0.1 O 3− δ (SZCY) electrolyte.The cell terminal voltages of 1.74 and 1.93 V were Q.Wang et al. observed at 0.5 A cm − 2 for both BLC and BGLC, corresponding to an R P of 1.9 and 2.3 Ω cm 2 , respectively.Hydrogen tracer diffusion studies using time-of-flight secondary ion mass spectrometry depth profiling confirmed that protons were incorporated into the bulk of BGLC relative to BLC.However, the present steam electrolysis results show a better performance for BLC at 600 • C. In a recent study, Wang et al. [231] reported a BGLC infiltrated BZY composite oxygen electrode for PCCs, which achieved a low R P of 0.44 Ω cm 2 at 600 • C.During the durability test at 600 • C in 10% humidified synthetic air, the cell (BGLC-BZY//B-ZY//BGLC-BZY) showed no degradation in the first 350 h but degraded afterward (the resistance increased from 0.63 to 0.88 Ω cm 2 over 130 h).XRD and TEM analyses of the tested electrode demonstrated that the formation of a secondary BaCO 3 phase and the accompanying changes to the parent BGLC phase are likely detrimental.

Triple conducting oxide
The oxygen electrode reaction processes in PCCs involve oxygen ions (O 2− ), protons (H + ), and electrons (e − ).It is thus speculated that materials conducting simultaneously these charge carriers would be a good candidate for the oxygen electrode.Most research efforts on singlephase TCOs have focused on enhancing the proton conductivity of a conventional MIEC to increase its triple-conducting potential by fostering hydration/protonation (H [132].It has been estimated that the proton conductivity in TCO oxygen electrode materials must exceed 10 − 5 S cm − 1 [232]. TCO material BaCo 0.4 Fe 0.4 Zr 0.1 Y 0.1 O 3-δ (BCFZY) was developed as a promising oxygen electrode material for PCCs by substituting Y 3+ for Zr 4+ in BaCo 0.4 Fe 0.4 Zr 0.2 O 3-δ (BCFY) [233].Y doping affects the proton conductivity, lattice volume, and electrochemical activity of BCFZ.Additionally, substituting Zr 4+ with the larger Y 3+ cations reduces the distortion from the ideal cubic perovskite structure [20].Chen et al. [234] investigated the proton uptake kinetics of BCFZY using the electrical conductivity relaxation method and reported the proton surface exchange coefficient as 3.85 × 10 − 6 cm − 1 at 500 • C. The chemical compatibility between BCFZY and proton-conducting BZY was examined by conducting a postmortem analysis of the oxygen electrode-electrolyte interface in a PCFC cell after 8000 h operation with methanol fuel.Energy-dispersive X-ray spectroscopy mapping showed neither chemical reaction nor obvious elemental segregation in the oxygen electrode, indicating the stability of the BCFZY-BZY interface [36].Duan et al. [39]  A peak power density of 410 mW cm − 2 in PCFC mode was achieved under air/H 2 gasses, and a current density of 370 mA cm − 2 was obtained at 1.3 V and 600 • C in PCEC mode using 12 vol.%H 2 O humidified air as the gas to the oxygen electrode and wet H 2 to the hydrogen electrode.Moreover, Ren et al. [236] attempted to tune the properties of the BCFZY perovskite oxides via defect engineering at the A-site to stimulate the generation of oxygen vacancies and oxygen-ion bulk diffusion and proton hydration kinetics.A peak power density of 668.64 mW cm − 2 at 600 • C was obtained with a cell incorporating Ba 0.9 Co 0.4 Fe 0.4 Zr 0.1 Y 0.1 O 3-δ as the single-phase oxygen electrode under wet H 2 (3 vol.% steam)/air.The performance improvement demonstrates that the introduction of an A-site deficiency is an effective strategy to enhance the oxygen-ion bulk diffusion and proton hydration kinetics of the TCO oxides.
To further improve the performance of BCFZY at reduced temperatures, Liang et al. [166] systematically investigated partial doping of the B sites with different metal elements, including Mn, Ni, Cu, and Zn, at a fixed content of 5%.Among them, Ba(Co 0.4 Fe 0.4 Zr 0.1 Y 0.1 ) 0.95 Ni 0.05 O 3-δ (BCFZYN) exhibited the lowest polarization resistance in both oxygen ion and proton-conducting fuel cells.They found that nickel doping improves oxygen mobility, surface exchange kinetics, and bulk oxygen ion and proton conductivity.Furthermore, the cells with the BCFZYN electrode showed excellent durability (200 h of operation in a symmetrical cell (BCFZYN//BZCYYb//BCFZYN) and 1000 h of operation in a single cell (BCFZYN//BZCYYb//Ni-BZCYYb) at 550 • C).BCFZY is, however, not chemically stable in a carbon-containing atmosphere.To enhance the CO 2 tolerance, a Ba 0.95 Ca 0.05 Co 0.4 Fe 0.1 Zr 0.1 Y 0.1 O 3-δ (BCaCFZY) oxygen electrode material was proposed, and the stability of BCaCFZY and BCFZY in CO 2 -containing atmospheres was compared.After being exposed to 1% and 10% CO 2 at 700 • C for 120 h, BCaCFZY was found to be more stable than BCFZY, with the surface of the BCaCFZY oxygen electrode free of carbonate particles.The improved stability by the Ca doping is due to reduced Ba segregation on the surface [237].These results indicate that doping is a good solution to alter the lattice diffusion and surface exchange properties of perovskites and to tailor the catalytic performances of various electrode materials.
Another TCO specifically developed for PCC oxygen electrode is NdBa 0.5 Sr 0.5 Co 1.5 Fe 0.5 O 5-δ (NBSCF), showing fast ORR kinetics and excellent stability [132].A PCFC with NBSCF//BZCYYb//Ni-BZCYYb demonstrated high power densities of 690 and 1370 mW cm − 2 at 600 and 700 • C, respectively.Choi et al. [143] further developed a TCO oxide PrBa 0.5 Sr 0.5 Co 1.5 Fe 0.5 O 5+δ (PBSCF) based on NBSCF, which dramatically enhances oxygen ion diffusion and surface oxygen exchange, showing higher electronic conductivity than NBSCF, and excellent compatibility with the BZCYYb electrolyte and chemical stability under operating conditions.Hence, PCC cells incorporating the PBSCF oxygen electrode can potentially deliver enhanced performance.Seong et al. [238] applied PBSCF as the oxygen electrode in a cell configuration of PBSCF//BZCYYb//Ni-BZCYYb, which reached a peak power density of 1620 mW cm − 2 at 650 • C. Wu et al. [84] fabricated a 3D hollow PBSCF fiber oxygen electrode for steam splitting reaction with the same cell configuration.At 600 • C, the cell achieved an electrolysis current density of 2020 mA cm − 2 at 1.6 V, and good stability without degradation at 1.6 V and 500 • C for over 75 h.Choi et al. [180] also applied PBSCF//BZCYYb//Ni-BZCYYb in R-PCCs, achieving excellent performance, with no performance degradation detected after 12 cycles (50 h) between 1.3 V (electrolysis mode) and 0.7 V (fuel cell mode) at 550 • C and good stability at 1.3 V and 550 • C for over 500 h.The exceptional performance and stability of PBSCF shown in various PCC applications prove PBSCF to be a promising oxygen electrode material.

Co-free materials
As discussed above, most of the recently developed TCO oxides contain cobalt.This has restricted their further development and applications due to the high cost of cobalt, thermal expansion mismatch, and cobalt evaporation [239].Recently, some Co-free TCOs based on BaFeO 3 have been developed, showing good performance.Zohourian et al. [240] measured the proton uptake of 18 TCO compositions belonging to the perovskite family (Ba, Sr, La) (Fe, Co, Zn, Y)O 3-δ by TGA.The partial substitution of Zn on the B-site strongly enhances the proton uptake.A compromise has to be found about the Co content since a high Co content is beneficial for the catalytic activity, but it decreases the proton uptake.The Ba 0.95 La 0.05 Fe 0.8 Zn 0.2 O 3-δ (BLFZ) composition exhibits the highest proton concentration: 10 mol% at 250 • C, 5.5 mol% at 400 • C, and 2.3 mol% at 500 • C. Wang et al. [115] reported that BLFZ shows good chemical compatibility with BZCYYb at temperatures below 1100 • C. At the same time, the average TEC of BLFZ (20.4 × 10 − 6 K − 1 at 30 • C-1000 • C) is larger than that of BZCYYb (9.8 × 10 − 6 K − 1 at 25 • C-1200 • C).By compositing with BZCYYb, the average TEC value of BLFZ-BZCYYb (70:30) can be reduced to 17.7 × 10 − 6 K − 1 .This result showed that an improved thermal match could be achieved by compositing the oxygen electrode and the electrolyte materials.
In another study, the Bi and Sn co-doped BaFeO In general, the development of MIEC electrode materials for PCCs is mature, benefiting to a large extent from the R&D work for O-SOCs.The absence of proton conduction in the MIEC can be compensated by introducing another proton-conducting phase into the electrode using various ceramic processing techniques (physical mixing, infiltration, and in-situ exsolution).Concerning MIEC itself, developing highperformance MIEC electrode materials at lower temperatures is still needed.In addition, MPEC and TCO single-phase oxides with proton conduction are increasingly attracting attention.Though with some success, the research on MPEC and TCO oxygen electrodes is still considered new, with only a few reported materials.Firstly, because of the simultaneous transporting of ionic species (O 2− and/or H + ) and electrons, their properties, reaction mechanism, and performance are still unclear and require further investigation.Secondly, the lack of efficient and direct characterization techniques to separate the contributions of the different charge carriers (H + /e − /O 2− ) is one of the main challenges in MPEC and TCO material development.
An overview of the currently available methods for characterizing mixed-conducting oxides is presented in Fig. 7.Each characterization technique has its advantages and limitations.For example, TGA can measure the protonation of specific electrode material, but it cannot determine the proton migration ability.Electrochemical impedance spectroscopy (EIS) is a powerful technique for material characterization.It enables the separation of different phenomena such as bulk diffusion and surface reaction that occur simultaneously in the SOCs.In recent years, DFT calculations have been applied to oxygen electrode materials in SOCs as a powerful material optimization and performance prediction tool.Most research groups either combine various characterization methods or use a single method under various limiting conditions that enable only one carrier to dominate conduction.Contributions from different carriers can then be separated analytically.As an example of previously adopted approaches, EIS is employed to measure the total conductivity of the TCO electrode material, and proton uptake is determined by TGA.This requires collecting EIS spectra under different measurement conditions (temperature, PO 2 , and PH 2 O) and fitting using equivalent circuit modeling.Such an approach requires multiple steps and may not be able to accurately separate oxygen ion and electronic conductivities.Consequently, it is very important to develop more efficient and direct characterization techniques to study mixed-conducting materials.It is also necessary to compare theoretical calculations with experimental studies.DFT calculations offer the experimentalists a new way to clarify the oxygen electrode reaction steps, a better understanding of the oxygen electrode reaction mechanisms, and a tool to design new oxygen electrode materials.

Summary and perspectives
To date, many novel oxygen electrode materials/structures have been developed to improve the performance of PCCs under operating conditions (see Tables 2-4).Decreasing the high polarization resistance (R P ) and maintaining good long-term stability are still essential requirements of a suitable oxygen electrode.Among the reported electrode materials, some show high power density and low R P when incorporated into a PCC cell.Specifically, strategies to boost the electrochemical performance of oxygen electrodes have been demonstrated by using triple-conducting materials or multi-phase composite electrodes.In addition to the electrode composition, the performance could be improved by optimizing the electrode microstructure.The effective strategies of electrode design introduced in this review include doping, defect engineering, phase-structure engineering, nanostructured morphology engineering, and composition optimization (Fig. 4).Fig. 8 summarizes the composition/doping strategies for perovskites, double perovskites, and Ruddlenden-Popper structures.It can be used as a guide for tailoring the composition of the oxygen electrode.
Although remarkable advances have been made in this field, several challenges remain.Future research efforts could focus on the following aspects: Fig. 7. Selection of available techniques for studying mixed-conducting oxides.The following techniques are shown (clockwise): Secondary Ion Mass Spectrometry (SIMS).Reproduced with permission [241].Copyright 2012, Elsevier.Density functional theory (DFT).Reproduced with permission [242].Copyright 2019, Chinese Materials Research Society.X-ray absorption spectroscopy (XAS).Reproduced with permission [243].Copyright 2020, Elsevier.Thermogravimetric analysis (TGA).Reproduced with permission [146].Copyright 2020, Elsevier.Electrical conductivity relaxation.Reproduced with permission [244].Copyright 2016, Elsevier.Four-point probe.Reproduced with permission [245].Copyright 2018, Elsevier.Electrochemical impedance spectroscopy (EIS).Reproduced with Permission [246].Copyright 2017, Elsevier.(1) Defect engineering: To optimize electrode material and tailor the systems' functionality, it is necessary to gain a fundamental knowledge of the defects in the material system, their concentrations under operating conditions, and their effects on the material properties.Studies in this field include experimental and theoretical work, such as DFT calculations, to provide an accurate description of the electronic structure of oxides systems.(2) Co-free materials: Co-free materials are attracting more attention because of their better TEC match with the electrolyte material and low cost.Despite the promising Co-free electrode materials developed so far, their conductivity and catalytic properties are still not satisfactory, with very few of them reaching the same performance as the Co-containing materials.Exploring new Cofree oxygen electrode materials with better performance remains a topic of future research.Additional studies on material composition and process optimization shall be pursued.(3) Proton uptake: The ideal electrode materials should possess a certain proton conductivity during the oxygen electrode reaction process in PCCs (>10 − 5 S cm − 1 ).Some single-phase/composite oxygen electrode materials involving proton conduction have been proposed and studied, but techniques for the proper design and precise measurements of proton uptake in a mixedconducting material are required.DFT calculations are becoming a powerful tool for identifying the potential proton conduction and proton migration ability of electrode materials.(4) Catalytic activity toward ORR/OER/WSR reactions: Appropriate catalysts with specific activity and stability should be further developed by optimizing the composition (dopants, defect), microstructural properties, and material processing techniques (physical mixing, infiltration, in-situ exsolution, etc.…).However, the lack of understanding of the improvement mechanisms of individual substitutions remains a challenge.In addition, the cost of the catalysts should be considered, as the high cost will increase the overall price of PCCs, thus impeding the commercialization of the relevant fabrication process.(5) (Phase/Structure) stability: Considering the excellent electrode/ cell performance reported so far, the research focus shall be shifted more towards the stability and degradation mechanisms of oxygen electrodes.The electrodes must be chemically (phase) compatible with the electrolyte and the current collector at the operating temperature.They must have stable microstructures under operating conditions concerning both porosity and surface area.To date, very limited research has been devoted to studying the degradation behavior of the oxygen electrode in PCCs.(6) Bonding of the electrode/electrolyte interface: The adhesion of the oxygen electrode to the electrolyte is also an overlooked issue.As mentioned previously, the oxygen-electrode sintering temperature needs to be kept low to prevent electrode sintering and reaction with the electrolyte.Alternatives to improve the adhesion are using an intermediate layer, a composite electrode, or infiltration.(7) Nanostructured engineering: Nano-engineering the oxygen electrode to form a nano-network has been shown to significantly accelerate the ORR/OER/WSR kinetics by introducing more active sites on the electrode surface, resulting in increased PCC cell performance.To maximize the TPBs, a nanostructured oxygen electrode with a higher specific surface area can be achieved by various techniques, such as infiltration or in-situ exsolution.
In summary, recent literature work has reported promising oxygen electrodes in PCCs.Despite several challenges ahead, pursuing research in this area will contribute to the development of low-cost, nanostructured, high-performing, and durable oxygen electrodes for PCEC and PCFC applications.

Declaration of Competing Interest
The authors declare no conflict of interest.

Fig. 4 .
Fig. 4. Summary of the property requirements, challenges, and design strategies for the development of oxygen electrode materials in PCC.

Fig. 5 .
Fig. 5. Potential reaction pathways and reaction sites at the three types of PCCs oxygen electrode materials (a) A proton-conducting phase and an MIEC; (b) A proton-conducting phase and an MPEC and (d) A TCO for PCFC (top line) and PCEC (bottom line), respectively.
applied BCFZY as infiltrates to produce a BCFZY-BaCe 0.6 Zr 0.3 Y 0.1 O 3-δ (BCZY63) composite oxygen electrode to enhance the oxygen reduction reaction kinetics at intermediate temperatures.The cell with a configuration of BCFZY-BZCY//BZCYYb//Ni-BZCYYb achieved a power density of 455 mW cm − 2 at 500 • C under the H 2 /Air conditions.Such cells can be operated at low temperatures down to 350 • C. Meng et al. [235] also used the same oxygen electrode (BCFZY-BCZY63) on a proton-conducting BaCe 0.7 Zr 0.1 Y 0.1 Sm 0.1 O 3-δ (BCZYSm) electrolyte in R-PCCs device.

Fig. 6 .
Fig. 6.(a) Design sketch of the new material: from Sr 3 Fe 2 O 7− δ (SFO) to SrEu 2 Fe 1.8 Co 0.2 O 7-δ (SEFC); (b) XRD patterns of the SEFC powders before and after heating at 600 • C in a 10% H 2 O− air atmosphere for 100 h and of SEFC− BZCY mixtures; (c) Stability of the current densities (CDs), Ohmic and polarization resistance (Rs, Rp) of the button cell over a 135 h reversible cycling test, switching between 0.7 V (FC mode) and 1.3 V (EC mode) measured at 600 • C. Reproduced with permission [141].

Fig. 8 .
Fig. 8. Effect of substitution/doping on various properties for perovskites, double perovskites, and Ruddlenden-Popper structures.TEC and σ stand for the thermal expansion coefficient and conductivity, respectively.

Q
.Wang et al.

Table 1
Comparison of oxygen electrodes in PCEC and PCFC.

Table 2
Performance of the oxygen electrode obtained from symmetrical PCC testing.R p represents the polarization resistance of the two identical electrodes in the symmetrical cell.

Table 3
Performance of the oxygen electrode obtained from PCFC testing.R p represents the cell polarization resistance, including contributions from both the hydrogen and oxygen electrodes.

Table 3
[198]19]ed ) 0.21 Ω cm 2 ) was only one-third of the one made by screen printing (0.66 Ω cm 2 ).The full cells (SSC-BCS//BCS//Ni-BCS) incorporating these two types of oxygen electrodes delivered peak power densities of 222 and 150 mW cm − 2 at 600 • C, respectively[88,119]. Su et al.[197]infiltrated an SSC solution into a porous SSC-Sm 0.2 Ce 0.8 O 2-δ (SDC) composite electrode backbone.The R p of the infiltrated electrode at 700 • C (measured on symmetrical cells) was five times lower than the one without infiltration: 1.14 Ω cm 2 for the blank SSC-SDC composite electrode and 0.23 Ω cm 2 for the 21.2 wt.% SSC infiltrated SSC-SDC electrode.In addition, Zhao et al.[198]also studied the stability of a PCFC with SSC infiltrated Ce 0.9 Gd 0.1 O 2− δ (GDC) composite oxygen electrode based on a BaZr 0.1 Ce 0.7 Y 0.1 Yb 0.1 O 3− δ (BZCYYb) electrolyte, which shows good stability for 120 h at a constant current of 0.5 A cm − 2 using humidified H 2 and humidified H 2 -CO as fuels under 600 • C. One of the most popular and representative MIEC materials used for oxygen electrodes is lanthanum strontium cobalt ferrite (La 0.6 Sr 0.4 Co 0.2 Fe 0.8 O 3− δ , LSCF) for O-SOCs operated at 700-900 • C. It exhibits a good compromise [194][195][196]icient, and good thermal and chemical compatibility with the other cell components[194][195][196].Wuet al. prepared two types of SSC-BaCe 0.8 Sm 0.2 O 2.9 (BCS) composite oxygen electrodes, one by infiltration and the other by screen printing.The R p of infiltrated SSC-BCS composite electrode (

Table 4
Performance of the oxygen electrode obtained from PCEC testing.R p represents the cell polarization resistance, including contributions from both the hydrogen and oxygen electrodes.

Table 4
(continued ) [162] density of 1277 and 841 mW cm − 2 at 700 and 600 • C, respectively, showing record-high performance for PCFCs with a cobalt-free oxygen electrode.The cell showed no detectable degradation during the 300 h durability testing.Wang et al.[162]also reported that La 0.7 Sr 0.3 Mn 0.7 Ni 0.3 O 3-δ (LSMN) has sufficient H + /e − /O 2− triple conductivity as the PCFC oxygen electrode material.Based on in-situ extended X-ray absorption fine structure analysis, they concluded that the hydration reaction occurs via the association between H 2 O and oxygen vacancies, coupled with the redox of Mn and O atoms.LSMN undergoes thermochemical hydration at approximately 400 • C in air by gaining 0.2 molar fractions of proton carriers.