Heterointerface engineering for enhancing the electrochemical performance of solid oxide cells

Abstract

Solid oxide cells (SOCs) have the potential to be the most efficient energy storage and conversion systems. To minimize energy loss due to charge and mass transport associated with the operation of SOC systems at intermediate temperatures, electrodes and electrolytes containing different types of heterointerfaces have been designed, fabricated, and tested under various conditions. While heterointerfaces can significantly enhance not only the ionic and/or electronic conductivity but also the electrocatalytic activity and stability of SOC components, as predicted by theoretical calculations and demonstrated by experimental results, the mechanisms of these enhancements are yet to be fully understood. In this review, we start with an overview of the techniques for fabrication of heterointerfaces with controlled composition, structure, and morphology. Then, the latest developments in performance enhancement of SOCs with heterointerfaces are summarized, including boosting the ionic conductivity of heterostructured electrolytes (oxygen ion conductors and proton conductors) and increasing the electrocatalytic activity and durability of heterostructured electrodes (oxygen electrodes and fuel electrodes). Subsequently, we will highlight the unique attributes of heterointerfaces in the enhancement of the SOC performance and provide important insights into the mechanisms of performance enhancement in order to establish the scientific basis for rational design of better electrolyte and electrode materials. Finally, the remaining challenges in design and fabrication of novel materials for advanced solid-state electrochemical systems will be discussed, together with possible strategies to overcome these critical issues, new research directions, and future perspectives.

---

Chenhuan Zhao

Chenhuan Zhao is now studying for a PhD degree from the Institute of Nuclear and New Energy Technology (INET) at Tsinghua University, China. She received her BS degree in the Department of Chemical Engineering from Tsinghua University in 2014. Her research interests include energy storage and conversion using solid oxide cells, especially for heterostructured oxygen electrodes with high activity.

Yifeng Li

Yifeng Li is currently studying for a PhD degree from the Institute of Nuclear and New Energy Technology (INET) at Tsinghua University, China. He received his BS degree in the Department of Chemical Engineering at Tsinghua University in 2015. His research interests include electrochemical energy storage and conversion, especially the research of high temperature electrolysis of CO2/H2O to produce sustainable fuels using solid oxide electrolytic cells (SOECs), as well as the degradation issues of SOEC electrode materials.

Bo Yu

Dr Bo Yu is an Associate Professor and PhD Supervisor of Tsinghua University, China. She received her PhD from Tsinghua University in 2004 and joined Nuclear Science & Engineering at Massachusetts Institute of Technology as a visiting researcher in 2012. Dr Yu has been responsible for the research and development of nuclear hydrogen or syngas production at Institute of Nuclear and New Energy Technology since 2005. She has published more than 100 articles and holds over 30 patents. Her research interests are electrochemical energy storage and conversion with a focus on high temperature electrolysis of CO2/H2O through SOEC technologies.

Jing Chen

Dr Jing Chen, as a professor, works at Institute of Nuclear and New Energy Technology (INET), Tsinghua University, China. He received his BS degree and PhD degree in Chemical Engineering from Tsinghua University in 1992 and 1996, respectively. His research area is chemical engineering for advanced energy systems, especially hydrogen production and utilization via high temperature solid oxide cell systems.

Yan Chen

Yan Chen is a Professor at the School of Environment and Energy at the South China University of Technology. She received her BS and ME from Peking University, and PhD from the Massachusetts Institute of Technology. Prior to her current position, she worked as a postdoctoral fellow at MIT. Her research focuses on rational design of new materials to enable high performance, economical energy devices based on understanding and controlling the surface and interface properties under extreme conditions (electric polarization, reactive environment, radiation etc.). She is also working on the application of ion beam implantation in low dimensional materials for energy and electronics.

Meilin Liu

Meilin Liu is the B. Mifflin Hood Chair, Regents' Professor, and Associate Chair of the School of Materials Science and Engineering at the Georgia Institute of Technology, Atlanta, Georgia, USA. He received his BS from the South China University of Technology and both MS and PhD from the University of California at Berkeley. His research interests include design, fabrication, in situ/operando characterization, and simulation of membranes, thin films, and nanostructured electrodes in devices for energy storage and conversion, aiming at achieving rational design of materials and structures with unique functionalities.

Broader context

Low-cost and efficient energy storage technologies are the enabler for the implementation of intermittent renewable energy technologies. Solid oxide cells (SOCs), when operated in reversible mode, have the potential to be the most efficient option for large-scale energy storage and power generation. However, broader commercialization of SOC technology hinges on breakthroughs in material development for high performance and long operational life. Heterostructured materials have demonstrated significantly enhanced charge and mass transport and electro-catalytic activity, especially at relatively low temperatures. This review provides a timely and comprehensive update on the latest advancements in the development of heterostructured SOC components. The methodology and engineering strategies are valuable not only to SOC technology but also to a wide range of energy applications for a clean and sustainable energy future, including photo/electrochemical water splitting and solar thermal CO₂ reduction.

1. Introduction

1.1 Overview of solid oxide cells

Rapid growth of energy consumption in the fields of industrial production, transportation, and electricity generation has posed unprecedented challenges to sustainable growth. In addition, the use of fossil fuels discharges large amounts of CO₂ and polluting gases that cause global warming and serious environmental pollution. To meet future energy demands and reduce CO₂ emissions, many renewable and environmentally friendly energy sources (such as solar, geothermal, wind, and biomass) have been extensively investigated. However, the deployment of these renewable resources is severely hindered by their intermittent nature, leading to urgent demands for efficient energy storage and CO₂ conversion systems. Due to their high efficiency and the capability to meet the requirement for large scale energy storage and conversion, solid oxide cells (SOCs) have attracted global attention.^1–9Fig. 1 shows the working principle of SOCs for electric energy storage and conversion, together with a number of potential applications such as distributed power generation, electrical vehicles, and smart grids.^10,11

Fig. 1 Working principles and representative applications of SOCs.

A SOC consists of a dense electrolyte, an oxygen electrode, and a fuel electrode. During operation, the oxygen electrode provides active sites for the oxygen reduction or evolution reaction (ORR/OER)^12–14 whereas the fuel electrode provides active sites for oxidation of fuel (e.g., H₂, CH₄, and CO) or production of fuel (e.g., electrolysis of H₂O and CO₂). Both electrodes must provide not only a sufficient number of active sites for the electrode reactions but also proper pathways for rapid transport of the species involved in the electrode reactions (such as ions, electrons, and gas molecules). The electrolyte in an SOC facilitates ion migration (oxygen ions, protons, or both) between the oxygen and the fuel electrodes while blocking the transport of electrons and gas molecules to avoid the short circuiting of the cell. SOCs can operate as a solid oxide fuel cell (SOFC) or as a solid oxide electrolysis cell (SOEC) as needed.^15–18 SOFCs can convert chemical fuels directly to electricity with high energy efficiency, high energy and power densities, and fuel flexibility.^3,4,7,19,20 SOECs can store “excess” electric energy in the form of chemical fuel (H₂, CO, syngas, or other renewable fuels) through the electrolysis of H₂O and/or CO₂^21–24 while maintaining similar advantages such as high energy conversion efficiencies, high production rates, fast power cycling, and low manufacturing costs.

Despite these numerous advantages and broad application prospects, the large-scale application of SOC systems is severely inhibited by the high operating temperature (e.g., 800 to 1000 °C).^18,25 Although such high operating temperatures can significantly accelerate the reaction kinetics and ionic conductivity, they can lead to detrimental interactions between cell components, leading to poor durability and high costs of utilization.^3,26,27 Therefore, to develop more efficient, durable, and economically feasible SOC systems, the SOC operating temperature must be reduced to about 500 °C while maintaining sufficiently high performance.^28–30 As the operating temperature is reduced, however, the cell performance will decrease exponentially due to slower oxygen reduction/evolution (ORR/OER) reaction kinetics on the oxygen electrode,^31–35 lower ionic conductivity of the electrolyte,³⁶ and increased polarization resistance of the fuel electrode.³⁷ Therefore, new materials or novel structures for SOC components need to be designed to enable high ionic conductivity in the electrolytes, high electrocatalytic activity at the electrodes, and good compatibility with other cell components under operating conditions.^38,39

1.2 Heterointerface engineering

Both experimental results and theoretical calculations have shown that heterointerfaces have the potential to significantly enhance the ionic and/or electronic conductivity, electrocatalytic activity, and stability of electrolytes and electrodes in SOCs.^40–45 For example, Su et al.⁴⁶ reported that the ionic conductivity of vertically aligned nanocomposite GDC/YSZ heterostructured electrolytes (0.96 S cm⁻¹) was 2 times higher than that of the YSZ single phase at 600 °C. Furthermore, Sase et al.⁴⁷ used ¹⁸O isotope exchange with secondary ion mass spectrometry (SIMS) to clearly demonstrate ∼10³ times faster oxygen exchange kinetics at the heterointerface of (La,Sr)CoO₃/(La,Sr)₂CoO₄ (LSC₁₁₃/LSC₂₁₄) with PLD-layered films. Crumlin et al.⁴⁸ decorated (La,Sr)CoO₃ thin films with epitaxial growth of (La,Sr)₂CoO₄ using PLD and achieved a ∼1–3 orders of magnitude enhancement in the ORR activity. In addition, Sun et al.⁴⁹ reported that La_0.3Sr_0.6Ce_0.1Ni_0.1Ti_0.9O_3−δ (LSCNT) fuel electrodes with Ni–Ce exsoluted particles presented outstanding activities with a peak power density of 600 mW cm⁻² and were resistant to coking for more than 80 h in a CH₄ atmosphere.

Overall, these results demonstrate that heterointerface engineering is a promising approach to enhancing the SOC component performance at intermediate temperatures. Such enhancements can be attributed to many factors, including lattice strain, dislocation, cation segregation, changes in electronic structure, and expansion of triple phase boundaries (TPBs) (Fig. 2).^4,50,51 For example, research into electrolytes^52,53 and oxygen electrodes^54,55 suggested that tensile strain at heterointerfaces can accelerate ionic transport by decreasing the migration barrier of charge carriers. In addition, results showed that the electronic structure near the heterointerface of LSC₁₁₃/LSC₂₁₄ oxygen electrodes can be different from their single-phase counterpart, leading to strongly enhanced ORR kinetics.⁵⁶ The presence of cation segregation near the interfacial region was reported to provide more active sites for the ORR/OER.^57,58 Furthermore, Ni/metal oxide interfaces of fuel electrodes have been reported to have higher coking resistance by facilitating water adsorption and water-mediated carbon removal.^59,60 To guide the rational design of high performance SOCs, it is imperative to fully understand the role of heterointerfaces in each of the key cell components: the electrolyte, the oxygen electrode, and the fuel electrode.

Fig. 2 The possible mechanisms for the performance enhancement in heterostructured SOC components.

It is noted that the factors that affect the properties of heterostructured SOC components (Fig. 2) are often inter-related. For examples, small lattice mismatches at a heterointerface may be readily accommodated by local lattice strains;^52,53 when the lattice mismatches are sufficiently large, however, the formation of dislocations may become energetically favorable to relieve local strain and stabilize interfacial structures. Furthermore, it has been proven that lattice strain may influence defect formation energies and migration barriers,^54,55 thus altering the distribution of charge carriers near heterointerfaces or in the space charge region (e.g., cation segregation at the interfacial region^57,58). The presence of or a change in lattice strain, defects (e.g., oxygen vacancies or dislocations), and redistribution (accumulation or depletion) of charge carriers in space-charge regions can induce electron transfer and influence the electronic structure⁵⁶ (or the energy landscape) of the interface, which determines the fundamental properties of the heterointerfaces. However, the specific correlations among these factors are yet to be fully understood.

This review aims to provide a comprehensive and in-depth overview of the impacts of heterointerfaces on the ionic transport, catalytic activity, and stability of heterostructured materials, providing important insights into the mechanisms of performance enhancement and establishing a scientific basis for the rational design of better electrolyte and electrode materials. The fundamentals of heterointerfaces are reviewed in Section 2, including basic concepts, structural characteristics, and controlled fabrication processes. Subsequently, applications of heterostructures in electrolytes, oxygen electrodes and fuel electrodes of SOCs are highlighted, along with possible mechanisms for the performance enhancement in those heterostructured SOC components, including lattice strain, dislocation, cation segregation, modified electronic structure, and extended TBPs at heterointerfaces. In addition, new research directions, future perspectives, existing challenges and possible solutions for the rational design of novel materials for advanced solid-state electrochemical systems involving heterostructures will be summarized. Overall, this review will provide a useful guide for the rational design of novel materials with unique functionalities for high-performance SOCs. The knowledge and methodology illustrated in this review are also suitable for the development of novel catalysts for other energy storage and conversation systems.

2. Fundamentals of heterointerfaces for SOCs

This section will provide a brief introduction to the basic concepts of heterointerfaces along with representative structures and synthesis techniques of oxide heterostructures used in SOCs. After a brief discussion of the physical properties of heterojunctions, we will classify heterointerfaces into different categories according to their structures. Then, several approaches for synthesis of heterointerfaces will be discussed, including atomic layer deposition, infiltration, screen printing, etc.

2.1 Basic concepts of heterojunctions

When two dissimilar materials are brought into contact, a heterointerface is formed. The difference in lattice constant of the two dissimilar materials will result in strain near the heterojunction and the difference in Fermi energy of the two materials will lead to electron transfer across the heterointerface, inducing a built-in electric field and a space charge region. Similarly, the difference in mobile ion concentrations of the two materials causes ion transfer across the heterointerface and a space charge region. The unique characteristics of heterojunctions include local accumulation (or depletion) of charge carriers, enhanced (or decreased) charge carrier mobility, quantum effects, and other extraordinary effects not observed in single phase materials.⁶¹

With technological developments and in-depth studies of heterojunctions, the definition and research of heterostructures have largely expanded. The extraordinary properties associated with heterointerfaces include high electrical conductivity,^36,62 enhanced thermoelectric effects, boosted catalytic activity,⁶³ super-wettability,^64,65 and induced magnetization.^66–69 The first study on the enhancement of ionic conductivity in composite materials was based on inorganic salt/Al₂O₃ and AgI/AgBr heterointerfaces in the 1980s.⁷⁰ Subsequently, significantly enhanced ion conduction was also found in multi-layered CaF₂/BaF₂^36,71 and YSZ/STO epitaxial heterostructures.⁷² In addition, high-mobility electron gases at the heterointerface of LaAlO₃/SrTiO₃ were investigated by Ohtomo et al.^73,74 and Nakagawa et al.⁷⁵, which demonstrated extremely high carrier mobility exceeding 10 000 cm² V⁻¹ s⁻¹ as a result of the built-in polarity discontinuity at the interface. These remarkable enhancements in conductivity and activity inspired the application of heterostructured oxides in SOCs.

2.2 Structures of heterointerfaces

Based on the structural characteristics, heterointerfaces in SOCs can be classified into several categories, including (a) multilayers, (b) vertically aligned nanocomposites (VAN), (c) composites, and (d) skeleton/surface coating (Fig. 3). Due to the significantly different structures, these heterointerfaces display large variations in local composition, lattice strain, electronic structure, and other properties.

Fig. 3 Schematics of the heterogeneous structures of component materials in SOCs. (a) Multilayer structure, (b) vertically aligned nanocomposite (VAN) structure, (c) composite structure, (d) skeleton/surface coating structure.

(a) The multilayer structure is a classical model for theoretical investigations in which heterointerfaces between two adjacent layers are parallel to the surface. Here, well-ordered interface structure enables the precise control of interfacial properties such as local compositions, lattice strains, electronic structures, etc. And in particular, the interfacial properties can play a dominate role in determining the functionality of the multilayer when the thickness of each layer approaches several nanometers.

(b) The VAN structure is another widely used model system for the investigation of the role of interfaces in the functionality of materials. Compared with the multilayer structure model, the unique advantage of the VAN structure is the maximum exposure of the active heterointerface. Furthermore, the interfaces in VAN structures are either directly exposed to gas (as electrodes) or along the direction of ion flow (as electrolytes), enabling full utilization of the heterointerfaces in SOCs. In addition, a VAN structure can sustain higher interfacial strain compared with a multilayer structure.

(c) Composite structures are often used in practical solid oxide cells; the heterointerface in this structure is random and the actual contact of the two phases is complicated. It is difficult to quantify the interfacial region between the two phases.

(d) The skeleton/surface coating structure is composed of a porous backbone decorated with a thin-film coating, which is also widely used in practical SOCs. In this structure, the backbones (or the materials serving as the mechanical support) generally possess good ionic or mixed conductivity, whereas the coating materials are generally highly active catalysts.

2.3 Fabrication of heterostructured materials or devices

Tremendous techniques have been developed for fabrication of materials containing heterointerfaces with specific structures and compositioins,⁷⁶ including pulse laser deposition (PLD),^77–81 molecular beam epitaxy (MBE),^36,82,83 atomic layer deposition (ALD),^84–88 electrophoretic deposition (EPD),⁸⁹ infiltration,^90–94 freeze drying tape casting,^92,95–98 screen printing,^99,100etc. In this review, advanced atomic layer deposition techniques are emphasized for the precise fabrication of high-quality oxide heterostructures.

(a) Thin film techniques. To precisely control the structure, composition, and morphology of heterointerfaces, thin film techniques have been extensively utilized, including PLD, ALD, MBE, and EPD. Among them, PLD and MBE have been widely applied in the research of heterostructured SOC components due to the flexibility of material compositions and excellent quality.^101,102 They could be used to produce layer-by-layer smooth interfaces that are typically suitable for investigating the interface properties on atomic scales.

To be specific, PLD can be used to fabricate complex heterostructured thin films with desired stoichiometric ratios, in which a high-power pulsed laser beam is produced by an excimer laser (e.g. KrF) and focused inside a vacuum chamber through a lens to a target (Fig. 4a).¹⁰³ Here, the target material is evaporated as a plasma plume in a high vacuum or in the presence of a background gas (e.g. oxygen of various concentrations) and subsequently deposited onto a substrate. And by optimizing operating conditions such as temperature, oxygen partial pressure, laser energy, etc., thin films with well-ordered crystal lattice units and specific orientations can be prepared. In addition, the use of PLD can produce various heterostructured oxide morphologies including bilayer,¹⁰⁶ ML¹⁰⁷ and VAN⁷⁷ structures. For example, Schichtel et al.¹⁰⁸ fabricated YSZ|Sc₂O₃ multilayers using PLD to study strain effects on the ionic conductivity.

Fig. 4 Schematics of a (a) PLD system,¹⁰³ (b) MBE system¹⁰⁴ and (c) up: tubular ALD reactor, down: layout of a plasma-enhanced ALD cycle for Cu deposition.¹⁰⁵ Reproduced from ref. 103–105 with permission from RSC, copyright 2016; Baiutti et al; license Beilstein-Institute (https://www.beilstein-journals.org/bjnano/articles/5/70), copyright 2014 and American Chemical Society, 2015, respectively.

The main advantage of the MBE method (Fig. 4b) is the excellent control of oxide growth as a result of the single-element fluxes, minimized kinetic energies and slow kinetics.^104,109,110 The difficulty of the MBE method, however, is the inability to maintain desired cationic stoichiometry during growth, which requires cooperation with an in situ monitoring auxiliary system, including a quartz crystal microbalance (QCM), real-time absorption spectroscopy (AAS), reflection high-energy electron diffraction (RHEED) etc. Furthermore, applications of MBE and PLD are also limited by their expensive deposition systems.

In addition, ALD is applied in layer-by-layer structure fabrication, which is based on the gas–solid reaction between the atmosphere and a substrate (Fig. 4c).^105,111 ALD can produce conformal and pinhole-free coatings of porous structures and is usually used in the fabrication of binary or ternary oxides.^112,113 For example, high quality films of many single oxides (e.g., MgO,¹¹⁴ NiO¹¹⁵ and CuO¹¹⁶) can be readily coated onto perovskite LSCF backbones using ALD.

(b) Infiltration. The infiltration method is an important approach to the fabrication of skeleton/surface coating heterostructures¹¹⁷ and typical infiltration processes consist of the following steps: (i) the fabrication of a porous scaffold with adequate mechanical strength, (ii) impregnation of the precursor solution of the surface coating material into the porous scaffold, (iii) drying and subsequent pyrolysis of the precursors to form the target catalyst coating, and (iv) repetition of the above procedure if needed to form a continuous thin-film coating. For example, Ding et al.¹¹⁸ reviewed the enhancement of the SOFC cathode performance through surface modifications using the infiltration method and reported positive impacts of catalyst infiltration on the SOFC cathode performance. In addition, researchers report that some synergetic coupling between the infiltrated layers and the backbones can greatly enhance the performance. For example, enhanced electro-catalytic activities were reported by Liang et al.¹¹⁹ for LSM-infiltrated YSZ and Pd-infiltrated LSM–YSZ.

(c) Screen printing. Screen printing has been extensively utilized in the fabrication of practical composite electrodes for solid oxide cells. It typically involves the following steps: (i) preparation of the powders of the phases to be printed, (ii) preparation of a slurry that contains the phases, (iii) uniform printing of the slurry onto a substrate through a screen, and (iv) drying and firing at high temperatures to form the desired phases and microstructure. Another method for fabrication of a composite electrode using screen printing is to print one material (or phase) at a time and alter the type of material to form heterointerfaces. As an example, Yashiro et al.¹²⁰ were able to prepare bilayer porous LSC_113/214 electrodes using screen printing and obtained much lower area specific resistance (ASR) and higher electrode performance than single phase electrodes.

3. Heterointerfaces in electrolytes of SOC devices

The reduction of operating temperatures increases the need for superior alternatives with high-conductivity electrolytes at intermediate temperatures. Here, heterostructured electrolyte materials with extraordinary ionic conductivities have been shown to be promising in SOC applications at intermediate temperatures. However, the corresponding mechanisms need to be further verified. The impact of heterointerfaces on the ionic conductivity and the mechanisms of such impacts will be introduced in this section.

Electrolytes are key components of fuel cells, which largely determine SOC electrode compositions.^3,88,121,122 To improve the performance of SOCs at intermediate temperatures, the electrolyte materials should meet the following requirements: (1) high ionic conductivity and negligible electronic conductivity under operating conditions; (2) good stability in various atmospheres over a wide range of temperature; (3) a small mismatch in the thermal expansion coefficient with electrodes and other cell components, (4) negligible interactions with electrodes and interconnect materials under operating and processing conditions, and (5) adequate mechanical strength or integrity.

High ionic conductivity of SOC electrolytes requires a sufficient concentration of mobile charge carriers with high mobility. Based on the types of ionic charge carriers, SOC electrolytes can be classified into oxygen ion conductors, proton conductors, and mixed ion conductors. To date, tremendous efforts have been contributed to the development of high-performance electrolyte materials with oxygen ion conductivity,^123,124 including doped zirconia,^125,126 doped ceria,^125,127,128 lanthanum gallate-based oxides,^129,130 and bioxide based materials.^131,132 The activation energy for O²⁻ transport in oxygen ion conductors is usually in the range of 50–60 kJ mol⁻¹,^133,134 indicating significantly lower ionic conductivity at reduced temperatures.

In addition, since the discovery of alkali earth doped cerates (ACeO₃, A = Sr and Ba) with perovskite structures as promising proton conductors,^135,136 researchers have devoted much attention to these materials.^137–141 The general formula of perovskite type proton conductors can be expressed as AB_1−xM_xO_3−δ, in which M denotes a trivalent dopant and δ represents the oxygen deficiency per unit cell.^139,142 Here, the mechanisms of proton conduction include proton transfer, structural reorganization, diffusion motion of extended moieties, etc.^{135,136,139,143}

3.1 Performance of electrolytes with heterointerfaces

Heterostructured electrolytes have attracted great attention due to the enhancement of ionic conductivities by the heterointerfaces, especially at low or intermediate temperatures. Representative mechanisms of enhanced ionic conductivities through heterointerfaces are highlighted in Fig. 5 and more detailed information is summarized in Table 1. Heterogeneous structures of various architectures and morphologies have been designed and fabricated to enhance the ionic conductivity, including multilayers,^{36,72,144–151} vertically aligned nanocomposites,^78,152–154 and composites of randomly distributed phases.^155,156

Fig. 5 Schematic of the functional principles for heterostructured electrolytes. Insets: (left) Schematic view illustrating the distortion of the unit cell under lattice strain;¹⁴⁵ activation energies for vacancy migration in CeO₂ biaxially strained along [100] and [010];¹⁵⁷ oxygen vacancy formation energies at the MgO layer in a TiO₂-terminated SrTiO₃ interface;¹⁵⁸ lateral ionic conductivity versus inverse temperature for YSZ films deposited on different substrates and with film thicknesses in the range 420 to 58 nm;¹⁵⁹ charge distribution in the STO/LAO/STO heterostructure;¹⁶⁰ normalized vacancy concentration as a function of distance from the GB interface at 440 °C^161,162 (right). Reproduced from ref. 145 and 157–161 with permission from RSC, copyright 2017; RSC, copyright 2012; Springer Nature, copyright 2014; WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim, copyright 2010; Springer Nature, copyright 2018 and RSC, copyright 2015, respectively.

Table 1 Summary of electrolytes with heterointerfaces in the literature from 2000 to 2018

Material	Geometry	T (°C)	Relative increase in conductivity	Mechanism	Ref.
GdCeO/MgO	Bilayer	500–800	×3	Heat treatment that releases strain energy and removes dislocation	148
YSZ/MgO	Bilayer	400–800	×10^2.5	Lattice mismatch strain	43
YSZ/MgO YSZ/STO	Bilayer	150–500	×10^3	Combination of mismatched dislocation density and elastic strain	159
Ce0.9Gd0.1O2−δ/STO	Bilayer	450–900	×10–10^2	Compressive strain	163
YSZ/Al2O3	Bilayer	27–377	×10^0.5–1.5	Tensile strain and edge dislocation	164
BZY/NGO	Bilayer	550–600	×10^2	Dislocation	144
Ce0.8Gd0.2O2−δ/Si3N4	Bilayer	300–500	×10	Strain	165
GDC/Si3N4/Si	Bilayer	327	×10^2	Strain	145
YSZ/LAO, NGO, MgO, Al2O3	Bilayer	300–500	×10^−2	Film texture and grain boundary density	166
SDC/STO/BZO/MgO	Bilayer	600	×3	Strain	146
YSZ/Y2O3, Sc2O3, Lu2O3	Multilayer	610	×2–10	Strain	167
YSZ/CZO	Multilayer	650	Almost 0	Strain	168
GDC/ZrO2	Multilayer	377	18 (compared to YSZ single crystal)	Strain and dislocation	169
CSZ/Al2O3	Multilayer	350–700	×10^2	Structural disorder/mismatch	150
YSZ/STO	Multilayer	84–258	×10^8	Atomic reconstruction	72
YSZ/Y2O3	Multilayer	350–700	×10	Dislocation	170
YSZ/STO	Multilayer	357–531	—	p-Type conductivity	171
YSZ/STO	Multilayer	100–500	×10^5 compared to YSZ/Al2O3	Cation interdiffusion	172
			(all electronic cond.)
Ce0.8Sm0.2O2−δ (SDC)/YSZ	Multilayer	400–800	×10 compared to SDC	Tensile strain	173
			×10 compared to YSZ
YSZ/STO	Multilayer	Room T–500	×10^2	Cation interdiffusion	174
Sc2O3/YSZ	Multilayer	420–780	×10^5 total	Strain	108
YSZ/CeO2	Multilayer	400–700	Almost 0	Strain	125
ZrO2/STO	Multilayer	—	Fluorite not stable for ε > 0.05	Strain	151
SrTiO3/YSZ/SrTiO3	Multilayer	227	Almost ×10^3	Defect redistribution	175
YSZ/Gd2Zr2O7	Multilayer	250–475	×10^2	Tensile strain	176
YSZ/Y2O3	Multilayer	520	×10^2	Strain	177
Sc2O3/YSZ	Multilayer	420–780	×10^5 total	Strain	108
CeO2/Y2O3, La2O3, Gd2O3	Multilayer	610–1000	×10	Lattice mismatch strain	149
GDC/YSZ	VAN	600	×2	Strain and fast ionic transport rates	46
YSZ/SDC, STO	VAN	360	×10^2	Strain and structurally mismatched vertical interface	152
SrTiO3/SDC	VAN	400	×10	High crystallinity nanopillars	153
YSZ/STO	Composite	127–727	×10^1.5 @ 1000 K ×10^3.5 @ 400 K	Lattice strain	53
SDC/LiNaCO3	Composite	300–650	×10	Surface properties and electrolyte thickness	156
CeO2/M2O3	Composite	227	×10^4	Strain	157
YSZ/STO	Composite	200–600	×10^−2	Defect formation and transport, p-type conductivity	178
YSZ	Calculation	137–927	×10	Strain	179
BZY/MgO	Calculation	300–700	Almost 0	Segregation tendency, but saturated	180
ZrO2/Cr2O3	Calculation	527–927	—	Defect redistribution	181

---

Many multilayer structured oxides have been reported to show high conductivity and low activation energies due to the presence of heterointerfaces. For example, Azad et al.¹⁶⁹ synthesized a multilayer structure consisting of Gd₂O₃-doped CeO₂ and ZrO₂ films grown on Al₂O₃(0001) using MBE and reported that the conductivity of the 10-layer GDC/ZrO₂ heterostructure was 2.9 × 10⁻⁴ S cm⁻¹ at 650 K in air, which was an ∼18 times enhancement as compared with YSZ single crystals. In another example, Peters et al.¹⁵⁰ found that in CSZ (ZrO₂ + 8.7 mol% CaO)/Al₂O₃ multilayer systems with coherent interfaces, the activation energy for ionic conduction was decreased from 1.47 eV to 0.96 eV as the number of CSZ layers increased from 1 to 5. There was a two orders of magnitude enhancement in the total conductivity when the individual CSZ layers got thinner from 0.78 μm to 40 nm (Fig. 6a and b). Sillassen et al.¹⁵⁹ deposited YSZ thin films onto MgO(110)(111) substrates and observed remarkable conductivity enhancements; the ionic conductivity of the resulting YSZ thin film was increased to 1 S cm⁻¹ at 500 °C, nearly 3.5 orders of magnitude higher than that of bulk YSZ.

Fig. 6 (a) TEM micrograph of a CSZ/Al₂O₃ multilayer system after heat treatment at 800 °C for ∼100 h. (b) Arrhenius plot of the total conductivities (σ_tot) measured on different CSZ/Al₂O₃ multilayer systems. Reprinted with permission from ref. 150. Copyright 2007 Elsevier. (c) High-resolution TEM image of a vertical SDC–STO interface in a cross-sectional view. Scale bar, 10 nm. (d) Temperature dependence of the real part of the ac conductivity (σ_ac′) measured in the nanoscaffold SDC–STO with comparisons with the single phase thin film and literature data.¹⁵³ Reprinted with permission from ref. 153. Copyright 2015 Springer Nature. (e) Z-STEM image of the BZY thin film and its interface with the NGO substrate, inset: Z-dependent intensity profile of the cations taken along the dashed line in the (−1 0 1) direction. (f) Arrhenius plot of the conductance for the whole measured set of film thicknesses.¹⁴⁴ Reprinted with permission from ref. 144. Copyright 2014 AIP publishing.

Compared with multilayer structures, vertically aligned nanocomposite (VAN) heterostructure films can potentially accommodate larger strain near the interface, creating fast ionic diffusion paths along the vertical interface.^78,152 For example, Yang et al.¹⁵³ prepared highly crystalline micrometer-thick vertical nanocolumns on a Nb:STO substrate using PLD (Fig. 6c) and reported that this SDC–STO VAN structure exhibited significantly enhanced oxygen ion conductivities, approaching 3 × 10⁻² S cm⁻¹ at 400 °C as recorded using scanning probe microscopy measurements. In contrast, the conductivity of SDC, YSZ, and STO single phase thin films is on the order of 10⁻³, 10⁻⁴, and 10⁻⁶ S cm⁻¹, respectively, under the same conditions (Fig. 6d). Here, the activation energy for oxygen migration in the SDC–STO VAN structure (0.65 ± 0.02 eV) was also slightly lower than that of the SDC film (0.70 ± 0.04 eV). Lee et al.¹⁵² also demonstrated that a double-layered vertical YSZ–STO/SDC–STO heterostructured film prepared by PLD can provide a high ionic conductivity of ∼10⁻² S cm⁻¹ at 635 K, which was over 2 orders of magnitude higher than that of YSZ films (∼6 × 10⁻⁵ S cm⁻¹). In addition, Su et al.⁴⁶ reported that in GDC/YSZ VAN systems, two-phase strain coupling can improve the ionic conductivity along the vertical heterointerface. The ionic conductivity at 600 °C of the GDC/YSZ VAN thin film (0.96 S cm⁻¹) was about 2 times higher than that of single phase YSZ thin films (0.41 S cm⁻¹) and slightly larger than that of the GDC/YSZ multilayers (0.82 S cm⁻¹), as determined from impedance spectroscopy. As a result, the corresponding single cell (Ni–YSZ|YSZ/GDC/VAN|LSC) exhibited peak power densities (PPD) of 0.488, 0.694 and 0.883 W cm⁻² at 700, 750 and 800 °C, respectively.

Similarly, heterostructured proton conductors have been reported to exhibit higher proton conductivity than those of the corresponding single phases. For example, Foglietti et al.¹⁴⁴ reported that the conductivity of a 10 nm thick film was about 2 orders of magnitude larger than that of a 500 nm thick film (Fig. 6e and f) and the average in-plane conductivity of the 10 nm thick film was 20 S cm⁻¹ at 550–600 °C, as demonstrated by a series of strained BaZr_0.8Y_0.2O_3−δ (BYZ) thin films of different thicknesses deposited on NdGaO₃ (NGO) substrates using PLD. The enhancement in proton conductivity was attributed to fast ionic transport induced by dislocations along the heterointerfaces between dissimilar materials. In another example, Yang et al.¹⁸² characterized the electrochemical activities and transport properties of BaZr_0.8Y_0.2O_3−δ (BZY)/NdGaO₃ (NGO) heterointerfaces using electrochemical strain microscopy (ESM), demonstrating that the mismatch dislocation network reduced the activation energy (0.1 eV for the 20 nm thick film) and induced a 2D transport phenomenon, which will be discussed in detail in Section 3.2.2.

3.2 Impacts of heterointerfaces on the ionic conductivity

Due to the extraordinary effects of heterointerfaces in various electrolyte systems, great efforts have been devoted to gaining a deeper understanding of the detailed mechanisms of ionic conductivity enhancement,¹⁸³ including the effect of strain, defects (such as dislocations), and the space-charge region near the heterointerfaces.

3.2.1 Strain effects. According to the degree of mismatch in the lattice constant, heterointerfaces can be divided into coherent, semi-coherent, and incoherent types. When the mismatch in the lattice constant is sufficiently small, a coherent interface may be formed. In this case, material can accommodate epitaxial strain by local lattice deformations such as oxygen octahedral rotations and distortions, avoiding breaking into structural domains. This elastic strain effect in heterointerfaces has been used to enhance the ionic conductivity of electrolyte materials.^{52,54,167,170,175,184} In addition, the effect of lattice strain on the conductivity is confined to the vicinity of the interface. Zou et al.¹⁸⁵ suggested that ionic conductivity enhancements mainly originate from the first lattice layer near the interface and are more apparent for materials with larger Young's modulus. And based on the distinct impact of structural strain, researchers have made extensive efforts to uncover the origin and quantize strain effects theoretically.^{46,145,166,178,186}

Tensile elastic strain is generally believed to accelerate ionic transport by decreasing the energy barrier of migration or by reducing the activation enthalpy for charge-carrier migration.^{52,53,157,187} For example, Kushima et al.⁵³ investigated changes in oxygen ion migration barriers in YSZ under biaxial lattice strain using DFT and the nudged elastic band (NEB) method. They found that the oxygen migration pathways may diversify and coexist depending on the local distribution of defects such as vacancies and dopant cations along the migration path. Here, the researchers reported that at lower stress levels, tensile strain can expand the migration space, increase the distance between cations (Δ(Zr–Zr)) among the migration path in this configuration and weaken O–C bonds (Fig. 7), all of which can decrease vacancy migration barriers and exponentially increase the oxygen diffusivity. Specifically, under 4% biaxial tensile strain, a maximum 6.8 × 10³ times diffusivity enhancement was obtained at 400 K for YSZ layers.

Fig. 7 (a) Change in the cation–cation distance Δ(Zr–Zr) as a function of tensile strain from 0.00 to 0.08; and (b) simulated bond strengths for O–C in the migration process under biaxial lattice strain. Reprinted with permission from ref. 53. Copyright 2010 the RSC.

Furthermore, Souza et al.¹⁵⁷ used quantitative equations to describe the impacts of stress on the activation enthalpy and ionic conductivity in fluorite structured solid solutions such as ZrO₂–M₂O₃ and CeO₂–M₂O₃. Here, the researchers used static atomistic simulations to calculate the energy change along oxygen vacancy migrations in strained CeO₂ based on empirical pair-potentials (EPP) and evaluated the correlation between conductivity and strain using the equations:

where ΔH_mig is the activation enthalpy of the oxygen vacancy migration volume, ΔV_mig is the activation volume and σ_χχ is the biaxial stress. And through further analysis, the researchers reported that 4% tensile strain can increase the in-plane conductivities at 500 K by as high as four orders of magnitude. Moreover, Schichtel et al.¹⁸⁸ proposed a modified equation that described a linear dependence of the logarithm of the interfacial conductivity on the lattice mismatch at the interface. Experiments on multilayer systems such as YSZ/Y₂O₃, YSZ/Lu₂O₃, and YSZ/Sc₂O₃ also confirmed the critical role of strain in determining the ionic conductivity.

Based on the model described by eqn (1) and (2), Korte et al.¹⁶⁷ further established a computational model by simulating elastically deformable crystallites. And based on grain boundary dependent strain relaxation in solid electrolyte films, their equation can adequately approximate the relationship between the total conductivity and structural parameters when the lattice mismatches are below 10%:

Subsequent experimental measurements were also consistent with these theoretical models and highlighted the dependence of ionic conductivity on lattice strain.^108,159,189 Schichtel et al.^108,189 systematically investigated the influence of interfacial stain effects at semi-coherent and coherent heterointerfaces such as YSZ(111)/RE₂O₃(111)/Al₂O₃(0001) and Al₂O₃(0001)/Sc₂O₃(111)/YSZ(111) on ionic transport along the interface. In these studies, the researchers reported that the conductivity decreased with increasing interfacial density and attributed this to the compressive strain in YSZ at the interface. This was also in agreement with model (3). Furthermore, Sillassen et al.¹⁵⁹ suggested that eqn (3) more accurately explained strain effects on continuous, flat and coherent interfaces (or with slight dislocations).

Overall, interfacial tensile strain generally increases the ionic conductivity and decreases ionic migration barriers whereas compressive strain has the opposite effect. However, lattice mismatch for heterointerfaces can lead to tensile strain on one side and compressive strain on the other, in which heterointerfaces coupled with two opposite effects simultaneously can face uncertainty and complexity. Therefore, the aim of heterointerface engineering is to maximize positive effects through the optimal design of the morphology and selection of material systems.

3.2.2 Dislocations. Large lattice mismatch in interfacial regions cannot solely compensate for elastic strain; therefore, dislocations can be introduced to relieve local strain. This local lattice relaxation can reduce residual elastic strain and stabilize interfacial structures. For example, dislocations can compensate for the large lattice mismatch at the YSZ/Sc₂O₃ interface (Fig. 8a).¹⁰⁸ In addition, Petrisor et al.¹⁹⁰ combined TEM characterization with Monte Carlo algorithms and proposed that dislocations were distributed periodically along the YSZ/GDC interface and provided dislocation spacing distribution configurations. Furthermore, it is reported that an interface can change from a coherent to an incoherent system with increased lattice mismatch and dislocations.¹⁷⁰ Therefore, dislocations can impact heterointerface behaviors together with lattice strain in semi-coherent systems and dominate behaviors in incoherent systems.

Fig. 8 (a) HRTEM micrograph of a Sc₂O₃/YSZ heterointerface with a regular network of dislocations (white marks).¹⁰⁸ Reprinted from ref. 108 with permission from RSC, copyright 2013. (b) HRTEM micrograph of a YSZ/MgO interface with mismatch dislocations marked by T symbols. The dotted rectangle exhibits a Burgers circuit in which the difference between atomic steps on each side represents the quantity of dislocations in each of the two orientations.¹⁵⁹ Reprinted from ref. 159 with permission from WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim, copyright 2010.

Recent studies also suggested that dislocations can critically impact the ionic transport properties.^{158,159,182,191–193} For example, Sillassen et al.¹⁵⁹ found that the conductivity of YSZ/MgO systems with 58 nm-thick YSZ films can be enhanced by almost 3.5 orders of magnitude with a large lattice mismatch of −18.01% between the YSZ and MgO lattices. They assumed that the dislocation network along the semi-coherent YSZ/MgO interface (Fig. 8b) may play a role of fast transport pathways for oxygen diffusion.

In another study on the MgO/SiTiO₃ system, Dholabhai et al.¹⁵⁸ used atomistic simulations with LAMMPS and proposed that dislocation networks in the TiO₂-terminated interface can reduce oxygen vacancy formation energies and accelerate ionic transport. Here, the researchers reported that oxygen vacancies preferred to form along dislocations lines with lower formation energies, which was similar to those in metallic nanocomposites.¹⁹⁴ They also suggested that TiO₂-terminated interfaces can exhibit analogous pipe diffusion mechanisms to metallic nanocomposites,¹⁹⁵ which were responsible for remarkable ionic conductivities. In comparison, SrO-terminated interfaces possessed oxygen vacancies that were concentrated on terraces and were inhibited from migrating.

Based on TEM analysis and calculation, Foglietti et al.¹⁴⁴ suggested that the enhanced in-plane proton conductivity was due likely to the high densities of defects, particularly dislocations, at the interface of the film. Similarly, Yang et al.¹⁸² proposed that the 2D dislocation network near the interfacial region can facilitate the proton conductivity by lowering the activation energies (Fig. 9a). STEM characterization identified mismatch dislocations spaced −10 unit cells apart at the BZY–NGO interface. Here, the atomic arrangements were highly disturbed with more oxygen vacancies along the dislocation network, allowing the redistribution of defects to induce charge imbalance and attract proton incorporation near the interface. In addition, they reported that the local distortion of oxygen octahedra can reduce the trapping effect, allowing the 2D dislocation network at the BZY/NGO interface to provide easily accessible sites to anchor H⁺ and decrease activation energies.

Fig. 9 (a) Schematic of the dislocation network at the interface of a BZY thin film and NGO substrate based on STEM characterization results.¹⁸² (b) Map of oxygen vacancy formation energies around one dislocation core.¹⁹⁷ Reprinted with permission from ref. 182 and 197, both with permission from American Chemical Society, 2015.

Despite many studies reporting positive impacts of dislocations on the ionic conductivity, some have also reported negative impacts. For example, Sun et al.¹⁹⁶ used atomic simulations and found that strain fields can enrich the concentration of trivalent dopant cations and oxygen defects near the 1/2 〈110〉 {100} edge dislocations in CeO₂ with depletion at the other edge. In the enriched zone, strong dopant–vacancy and vacancy–vacancy interactions can diminish oxide ion diffusion, whereas in the depleted zone, the lack of charge carriers can decrease the conductivity. In another study, Marrocchelli et al.¹⁹⁷ found that the redistribution of defects resulted in overlapping electrostatic fields surrounding the dislocation cores with positive charge, which further influenced the electrical and ionic conductivity. And although the oxygen vacancy formation energies were lower than 2 eV at sites close to the dislocation cores of 〈100〉 {011} as compared with the SrTiO₃ bulk (Fig. 9b), these researchers did not find any significant enhancement in the oxide ion mobility. Similarly, Schichtel et al.¹⁰⁸ reported the negligible role of dislocations in the ionic conductivity along the heterointerface of Al₂O₃(0001)/Sc₂O₃(111)/YSZ(111). Yep et al.¹⁹⁸ also reported that edge dislocations induced by kinking on the (220) slip plane near the YSZ/quartz interface can inhibit oxygen ion migration along the interfacial region. Overall, the complex role of dislocations in the ionic diffusion kinetics is still not well understood and requires further systematic research.

3.2.3 Space-charge effects. The dissimilarities between the two materials across a heterointerface lead to redistribution of charge carriers and defects.^199,200 This accumulation or depletion of charge carriers can further break electrical neutrality and create local electric fields around space-charge layers.¹⁶⁰ The effective range of the space-charge width, known as the Debye length (L_D), is inversely proportional to the square root of the defect concentration in bulk materials.⁵² And for doped oxides, L_D is even smaller. For example, the space-charge region of 8 mol%-doped YSZ can be as narrow as 1 Å at 500 °C according to the Gouy–Chapman conditions. The space-charge effect in undoped conductors or conductors with low dopant concentrations can be described using the Poisson–Boltzmann equation:²⁰¹which describes the space-charge potential (Ψ) as a function of the distance from the interface (x), defect charge number (z), concentration of each defect species (c), permittivity of free space (ε₀), relative permittivity of the material (ε_r) and temperature in Kelvin (T). In addition, space-charge effects can lead to apparent changes in transport properties for charge carriers. For example, Korte et al.¹⁶⁷ demonstrated that for intrinsic conductors (undoped materials), the space-charge effect usually leads to superior conductivity enhancements. Many efforts have been devoted to studying the influence of space-charge effects on conductivity in oxide conductors.^202,203

The positive impact of space-charge effects on the ionic conductivity at interfacial regions has been highlighted in several studies.^43,204,205 For example, Kosacki et al.⁴³ proposed that due to space-charge effects, the ionic conductivity of 1.6 nm-thick YSZ/MgO interface layers can reach ∼2 S cm⁻¹ at 700 °C, which was 3 orders of magnitude larger than that of the lattice. In another example, Karthikeyan et al.²⁰⁵ reported that in YSZ/MgO, the ionic conductivity increased as the film thickness decreased. They also found that the high activation energy indicated that the space-charge effect at the interface and grain boundary dominated the conductivity enhancements. Despite these promising results, however, a straightforward relationship between the space-charge effect and ionic enhancement still needs to be further investigated. Thus far, there were only a few reports about tuning the space-charge potential and defining the vacancy accumulation regions clearly.^206–208 For example, Gäbel et al.²⁰⁷ reported that a nanocrystalline GDC thin film on a MgO substrate can exhibit relatively high ionic conductivities (2.7 × 10⁻⁴ S m⁻¹ S at 700 °C in 10⁻⁵ bar O₂) with a low space-charge potential of 0.19 ± 0.05 V, in which the researchers attributed this to the low depletion of oxygen vacancies at the grain boundaries.

Alternatively, there were studies demonstrating that the impacts of space-charge effects on conductivities are complicated. Some researchers found that the positively charged grain boundary cores can block the migration of oxygen vacancies, significantly influencing the electronic and ionic conductivities.^209,210 For example, Adepalli et al.²¹¹ reported that dislocation-induced space-charge fields can enhance the n-type conductivity of SrTiO₃ by 50 times at 10⁻⁵ atm p(O₂) at 650 °C and decrease the p-type conductivity by 50 times and reduce the oxygen diffusion coefficient by three orders of magnitude at high oxygen pressures. In another study, Lupetin et al.²¹² reported that in SrTiO₃, a reduction of the grain size to 30 nm can bring out a 3 orders of magnitude enhancement of the n-type conductivity and a significant decrease in the p-type conductivity and oxygen vacancy conductivity. The researchers attributed such decreases to increased space-charge effects near the interface as the particle size decreases. Similar effects have also been found for nanocrystalline ceria.²¹³

Furthermore, Mebane et al.¹⁶¹ combined the Poisson–Boltzmann with the Cahn–Hilliard model to estimate space-charge effects in solid solutions with wide dopant levels. The ‘Poisson–Cahn’ theory is based on local and non-local chemical interactions and can yield activity coefficients for point defects. And in the case of heterostructured CeO₂–Gd₂O₃, the model successfully predicted defect behaviors near the grain boundary over the entire concentration range. In addition, the model, included existing conductivity considering local dopant concentrations, covered various conditions of solid solutions, and finally gave a comprehensive prediction of both bulk and grain boundary conductivities.

In this section, we have discussed the role of heterointerfaces in electrolyte materials for SOCs. In summary, heterointerfaces can greatly influence the transfer properties of charge carriers (such as oxygen ions and protons) via various mechanisms including lattice strain, dislocation and space-charge effects. Actually, the evolution of the local lattice structure, cation arrangement, and electronic structure induced by heterointerfaces may not only affect the mass transfer, but also have a more profound influence on the primitive steps of electrode reactions, which will be discussed in detail in the next sections.

4. Heterointerfaces in oxygen electrodes of SOC devices

Due to sluggish ORR/OER activities at intermediate temperatures, the development of oxygen electrodes with high catalytic activity and stability is essential for high performance SOCs.^214–218 To address this, heterointerface engineering has recently emerged as a promising approach to optimize the oxygen electrode performance. This section will provide an overview of oxygen electrodes with heterointerfaces and the mechanisms behind the impact of the interface on the ORR/OER kinetics.

ORR/OER processes on mixed ionic and electronic conductor (MIEC) electrode materials involve several elementary steps. ²¹⁵ The detailed ORR process includes the following steps (Fig. 10):

Fig. 10 Schematic of the elementary steps of an oxygen reduction reaction process on a heterostructured oxygen electrode.

① Gas diffusion, O_2(gas) → O_{2(near surface)};

② Adsorption, O_{2(near surface)} → O_2(ads);

③ Dissociation, O_2(ads) → 2O_(ads);

④ Electron transport, O_2(ads) + e⁻ → O₂⁻_(ads), O₂⁻_(ads) + e⁻ → O₂²⁻_(ads), O₂²⁻_(ads) → 2O⁻_(ads), O_(ads) + e⁻ → O⁻_(ads), O⁻_(ads) + e⁻ → O²⁻_(ads);

⑤

⑥ Diffusion of O²⁻ to the electrode/electrolyte interface;

⑦ O²⁻ transport across the electrode/electrolyte interface.

Many of the elementary reaction steps above involve the transportation of ionic and electronic species. Therefore, optimal ionic and electronic conductivities as well as surface exchange coefficients are desirable for oxygen electrodes. Furthermore, oxygen electrode materials should meet additional requirements such as high chemical stability, compatibility with electrolyte materials, thermodynamic properties matching other materials, etc.

4.1 Performance of SOC oxygen electrodes with heterointerfaces

Many theoretical and experimental studies have demonstrated that heterogeneous structures can exhibit much faster oxygen surface exchange and diffusion kinetics compared with the single-phase counterparts (Fig. 10). To be specific, oxygen electrodes with heterointerfaces that exhibit high performance include (La,Sr)MnO₃(LSM)/LaNiO₃,²¹⁸ LSM/LaAlO₃,¹⁹² Er_0.4Bi_0.6O_3−δ/La_0.8Sr_0.2MnO_3−δ,²¹⁹ La_0.85Sr_0.15MnO₃/La_0.6Sr_0.4Co_0.2Fe_0.8O_3−δ,²²⁰ Ba_0.5Sr_0.5TiO₃/La_0.67Sr_0.33MnO₃,²²¹ LSC₁₁₃/LSC₂₁₄,^48,222 Nd_0.8Sr_1.2CoO_4±δ (NSC₂₁₄)/Nd_0.5Sr_0.5CoO_3−δ (NSC₁₁₃),²²³ SrCo_0.2Fe_0.6Ni_0.2O₃ (SCFN)/La_0.8Sr_0.2Co_0.2Fe_0.8O_3−δ (LSCF),²²⁴ La_0.6Sr_0.4Co_0.2Fe_0.8O_3−δ/CeO₂,²²⁵etc. and are summarized in Table 2.

Table 2 Summary of oxygen electrodes with heterointerfaces reported in the literature from 2008 to 2018

Material	Geometry	T (°C)	Performance	Mechanism	Ref.
La0.8Sr0.2CoO3−δ/LaSrCoO4+δ	Multilayer	550	k ^q (from EIS) ∼10^3–10^4 enhancement	Strain, space-charge effects and an increase in electronic structure or oxygen vacancies	48
La0.8Sr0.2CoO3−δ/YSZ	Multilayer	510	Surface exchange coefficient ∼10 times	Sr decoration	226
(La0.5Sr0.5)2CoO4±δ/La0.6Sr0.4CoO3−δ	Multilayer	550	k ^q ∼ 3 orders of magnitude enhancement	Heterointerface	222
SrTi1−xFeO3/YSZ	Multilayer	345	Band gap increased from 2.5 ± 0.5 eV to 3.6 ± 0.6 eV	Surface electronic structure	227
La1−xSrxCoO3−δ/SrTiO3	Multilayer	280–475	Tensile strain with faster surface exchange (∼4 times) and diffusion (∼10 times)	Strain	228
La1−xSrxCoO3−δ/LaAlO3
La0.8Sr0.2CoO3−δ/La0.8Sr0.2MnO3−δ	Multilayer	550	k ^q (cm s^−1) (from EIS) ∼10–10^2 enhancement	Mn substitution (cation diffusion)	79
La0.8Sr0.2CoO3/(La0.5Sr0.5)2CoO4	Multilayer	250	Energy gap decreased from 1.4 eV (LSC113) and 2.6 eV (LSC214) to 1.0 eV (LSC113/214)	Electron donation and transfer of oxygen vacancies across the heterointerface	229
(La,Sr)2CoO4±δ/La1−xSrxMO3−δ	Multilayer	550	Surface exchange coefficients enhanced by 10^2 times	Sr segregation	230
SrCoOx/(LaAlO3–(SrAl0.5Ta0.5O3)0.7) or SrTiO3 or DyScO3 or GdScO3 or KTaO3)	Multilayer computation	300	Oxygen activation energy barriers decreased by ∼30%	Strain	231
La0.6Sr0.4CoO3−δ/Ce0.9Gd0.1O1.95	Multilayer	700	Slight increase in ^18O ratio	Extended TPB	232
La0.8Sr0.2MnO3/SrTiO3	Multilayer	600	Oxygen ion diffusion enhanced by 10^3	Dislocation and strain	192
La0.8Sr0.2MnO3/LaAlO3
La0.8Sr0.2CoO3−δ/Nd2NiO4+δ	Multilayer	300–415	Oxygen exchange kinetics are much slower	Charge transfer	233
La0.6Sr0.4Co0.2Fe0.8O3/PrxCe1−xO2	Multilayer	600	PrO2/LSCF: nearly 6 times faster	Inducing higher oxygen vacancy concentrations	234
			CeO2/LSC: slight enhancement
La0.6Sr0.4CoO3−δ/LaSrCoO4±δ	Multilayer	600–850	k ^q ∼ 5–10 times that of the single phase	Sr enrichment at the interfacial region and stabilized against detrimental Sr segregation	106
Nd0.5Sr0.5CoO3−δ/Nd0.8Sr1.2CoO4±δ	Multilayer	500	k ^q ∼ 10^2–10^3 times that of the single phase	Enriched Sr concentration and decrease in the valence state of Co	107
La0.65Sr0.35MnO3/SrTi0.2Fe0.8O3	Multilayer computation	—	Bandgap values (1 and 2 eV for LSM and STF) decreased to 0.3 and 1 eV for LSM and STF	Electronic structure, strain, defects	235
La0.8Sr0.2CoO3/(La0.5Sr0.5)2CoO4	Multilayer	300	10^3 times increase in charge transfer rates to oxygen	Electronic structure	56
La0.8Sr0.2CoO3/HfO2 decoration	Multilayer	530	k ^q ∼ 30 times enhancement	Optimum surface oxygen vacancy concentration	236
La0.6Sr0.4CoO3−δ/Co2O3 decoration	Multilayer	450	Surface exchange resistance reduction by 13% per laser pulse	Reactivation of LSC surface
La0.8Sr0.2CoO3−δ/LaSrCoO4+δ	VAN	320–400	k ^q ∼ 10 times enhancement	Electronic activation and more stable cation composition at the interface	77
La0.6Sr0.4CoO3−δ/LaSrCoO4+δ	Composite	500	Exchange coefficient k* (cm s^−1) (from SIMS) enhanced by 10^3	Fast oxygen incorporation paths along the heterointerface boundary	216
La0.6Sr0.4CoO3−δ/LaSrCoO4+δ	Composite	500	Exchange coefficient σe (S cm^−2) (from EIS) 10–10^1.5	Heterointerface enhanced the oxygen catalytic reaction	237
Fe2O3–LaCePrOx	Composite	400–600	Ionic conductivity increased by 10 times	Heterogeneity and interfacial conduction effects	238
SP Ba0.5Sr0.5(Co0.7Fe0.3)0.6875W0.3125/DP Ba0.5Sr0.5(Co0.7Fe0.3)0.6875W0.3125	Composite	550–700	ASR ∼3 times lower	Cooperative DP/SP effects	239
Er0.4Bi1.6O3−δB/La0.8Sr0.2MnO3−δ	Composite	550–650	R p/5	Particle sizes	219
La0.6Sr0.4Fe0.8Co0.2O3−δ/Gd0.2Ce0.8O1.9−δ–ZrO2	Skeleton/surface coating	800	∼25% degradation rates after overcoating in 1100 hours of operation	Synergistic function of porosity, mixed conductivity and suppressed Sr enrichment	240
La0.6Sr0.4CoO3−δ/ZrO2	Skeleton/surface coating	700	Lower polarization resistances and 19 times slower degradation rates over 4000 h	Porosity, mixed conductivity and suppressed Sr enrichment	241
La0.58Sr0.4Co0.2Fe0.8O3−δ/GDC	Skeleton/surface coating	900	k chem ∼ 6 times higher than pure LSCF	Facilitating dissociative adsorption of O2	242
La0.6Sr0.4CoO3−δ/Al2O3	Skeleton/surface coating	800	Power output reduced from 1.62 W cm^2 to 0.742 W cm^2	Blocking effect of excessive Al2O3	243
La0.6Sr0.4Co0.2Fe0.8O3−δ/MgO	Skeleton/surface coating	650	ASR reducing from 0.49 Ω cm^2 to 0.31 Ω cm^2	Promotion of charge transfer process	114
La0.6Sr0.4Co0.2Fe0.8O3−δ/CuO	Skeleton/surface coating	750	k chem ∼ 9.3 × 10^−5 cm s^−1, 3.5 times enhancement	Additional reactions on both the CuO surface and the LSCF–CuO-gas boundaries (3PBs)	116
La0.6Sr0.4Co0.2Fe0.8O3−δ/NiO	Skeleton/surface coating	800	k ^q ∼ 6.9 × 10^−4 cm s^−1, 20 times enhancement	Promotion of oxygen incorporation process	115
(La,Sr)2FeO4−δ/La0.8Sr0.2FeO3−δ	Skeleton/surface coating	650–800	ORR activity enhanced by 10 times	Higher catalytic activity of LSF214 and mismatch between LSF214 and LSF113	244
La0.8Sr0.2MnO3/Ba0.5Sr0.5Co0.8Fe0.2O3−δ	Skeleton/surface coating	650–750	Electrode resistances of ∼0.101, 0.049 and 0.023 Ω cm^2 at 650, 700 and 750 °C respectively	Material intrinsic properties and fabrication method	245
La0.84Sr0.16MnO3−δ/Bi1.4Er0.6O3	Skeleton/surface coating	750	10–10^2 ionic conductivity	Effective TPB extension	246
PrSrCoMnO6−δ/La0.6Sr0.4Co0.2Fe0.8O3−δ	Skeleton/surface coating	750	R p of 0.093 Ω cm^2 after running for 500 h with a performance enhancement of 61.7% as compared with blank LSCF	Material intrinsic properties and fabrication method	247
La0.6Sr0.4Co0.2Fe0.8O3−δ/La0.85Sr0.15MnO3±δ	Skeleton/surface coating	700	Power density of 655 mW cm^−2 at 700 °C	Formation of LSM(C) phase	248
(La0.6Sr0.4)0.995Co0.2Fe0.8O3−δ/La2NiO4+δ	Skeleton/surface coating	750	67% increase in peak power density, a low degradation rate of 0.39% for ∼500 h	Cation segregation of LSCF and favorable acceptance by LNO of Sr/Co doping.	249
La2NiO4+δ/Ba0.5Sr0.5Co0.8Fe0.2O3−δ	Skeleton/surface coating	600	ASR decreased to 0.078 Ω cm^2	Shell increased TPB	250
			Complete resistivity to CO2 attacks at IT
Er0.4Bi1.6O3/La0.76Sr0.19MnO3+δ	Skeleton/surface coating	550–750	Power density of 162 W cm^−2; excellent stability in SOFC, SOEC and reversible SOC operating models for over 200 h	Fast oxygen ion migration at the heterointerface	6
(La,Sr)CoO3−δ/(La,Sr)2CoO4+δ	Computation	500	400 times faster oxygen incorporation kinetics	Anisotropy and strain	51
La0.8Sr0.2CoO3−δ/(La0.5Sr0.5)2CoO4−δ	Computation	500–600	Oxygen vacancy concentrations enhanced by 10^2–10^2.5	Sr segregation	223
(La1−ySry)2CoO4±δ/La1−xSrxCoO3−δ	Computation	—	Anomalous Sr segregation	Cation diffusion	57
LaCoO3/LaAlO3	Computation	—	(100) oriented film possessed better OER performance	Spin-state transition of Co	40

---

Based on the compositions, heterostructured oxygen electrodes can be sorted into four categories: (1) perovskite/Ruddlesden–Popper (ABO₃/A₂BO₄), (2) perovskites with different cations (ABO₃/A′B′O₃), (3) perovskite/simple metal oxides, and (4) others, which will be discussed respectively as follows.

4.1.1 Perovskite/Ruddlesden–Popper (ABO₃/A₂BO₄). The ABO₃/A₂BO₄ heterostructure is one of the most intensively investigated oxygen electrodes.^47,48,251 For example, Sase et al.^47,216 fabricated a (La_0.6Sr_0.4)CoO₃/(La,Sr)₂CoO₄ (LSC₁₁₃/LSC₂₁₄) bilayer structure using PLD and conducted oxygen isotope exchange experiments at 773 K in ¹⁸O-rich oxygen at 0.2 × 10⁵ Pa. Using the SIMS imaging profile, they found distinct ¹⁸O enrichment at the interfacial region. The calculated oxygen exchange coefficient was 8 × 10⁻⁶ cm s⁻¹, which was three orders of magnitude higher than that of the LSC₁₁₃ single phase (Fig. 11a).^48,237 Yashiro et al.²³⁷ also reported that the impedance resistance of porous LSC_113/214 electrodes fabricated by screen printing was less than half the impedance resistance for LSC₁₁₃ single-phase electrodes at 873 K and in 10% O₂.

Fig. 11 (a) Mapping profile of ¹⁸O signals by SIMS at the surface above LSC₁₁₃/LSC₂₁₄/CGO after oxygen isotope exchange.⁴⁷ (b) HAAADF STEM micrograph of the LSC₁₁₃/LSC₂₁₄ interface;⁴⁸ (c) EIS results of the microelectrodes (∼200 μm) for LSC₁₁₃ films with ∼0.1, ∼0.8, ∼5 and ∼15 nm LSC₂₁₄ surface decoration and the LSC₁₁₃ reference at 550 °C at 1% p(O₂).⁴⁸ (d) TEM image and selected area electron diffraction of a LSC₁₁₃/LSF₂₁₄ particle.²⁴⁴ Reproduced from ref. 47, 48 and 244 with permission from the Electrochemical Society, copyright 2008; and American Chemical Society publications, copyright 2010 and 2017, respectively.

The ultra-fast oxygen exchange kinetics near the interface of LSC_113/214 was confirmed by Crumlin et al.⁴⁸ using epitaxial La_0.8Sr_0.2CoO_3−δ (LSC₁₁₃) thin films with partial decoration of (La_0.5Sr_0.5)₂CoO_4±δ (LSC₂₁₄) (Fig. 11b). Here, the researchers reported that the LSC₂₁₄ decorated LSC₁₁₃ presented a 3–4 orders of magnitude enhancement in the surface exchange coefficient (k^q) compared with the LSC₁₁₃ single phase reference, reaching as high as ∼3 × 10⁻⁶ cm s⁻¹ in 1 atm O₂ at 550 °C (Fig. 11c). Furthermore, Lee et al.²³⁰ compared the exchange coefficients for LSC₁₁₃/LSC₂₁₄ and LSCF₁₁₃/LSC₂₁₄ heterostructures and reported that in 1 atm O₂ at 550 °C, both LSCF₁₁₃/LSC₂₁₄ (∼7 × 10⁻⁸ cm s⁻¹) and LSC₁₁₃/LSC₂₁₄ (∼2 × 10⁻⁷ cm s⁻¹) exhibited significantly enhanced k^q values relative to the LSCF₁₁₃ reference (∼3 × 10⁻⁸ cm s⁻¹) and the LSC₁₁₃ reference (∼5 × 10⁻⁹ cm s⁻¹), respectively.

Recently, Zhao et al.²¹⁴ also reported that bilayered LSC₁₁₃/LSC₂₁₄ was able to maintain superior ORR activities at 873–1173 K relative to single-phase LSC₁₁₃. At 973 K in air, k^q of LSC₁₁₃/LSC₂₁₄ was 2.5 × 10⁻⁵ cm s⁻¹, which was 10 times higher than that of the LSC₁₁₃ reference. Moreover, in another study conducted by the same group, Zheng et al.¹¹ fabricated a series of Nd_0.5Sr_0.5CoO_3−δ/Nd_0.8Sr_1.2CoO_4±δ multilayers with the same total thickness and different numbers of interfaces. By conducting ¹⁸O isotope exchange experiments (97.1% ¹⁸O isotope enriched, 773 K, 15 s) and utilizing SIMS, the researchers found that the ¹⁸O⁻ signal strongly concentrated along the NSC₂₁₄/NSC₁₁₃ interface. With the increased layers of the heterostructure, k^q showed a significant enhancement. Specifically, the bilayer NSC_214/113 film (∼5.5 × 10⁻⁸ cm s⁻¹) exhibited a ∼41 times higher k^q value than that of the NSC₁₁₃ reference (∼1.3 × 10⁻⁹ cm s⁻¹), whereas the 18-layered NSC_214/113 film (∼2.9 × 10⁻⁷ cm s⁻¹) exhibited a 165 times enhancement.

Hong et al.²⁴⁴ used the infiltration method to fabricate a heterostructured (La,Sr)₂FeO_4−δ (LSF₂₁₄)–La_0.8Sr_0.2FeO_3−δ (LSF₁₁₃) electrode (Fig. 11d). After repeated infiltration of LSF₂₁₄ onto a LSF₁₁₃ skeleton (4 times), numerous LSF₂₁₄ nano-islands were generated, and its chemical surface exchange coefficient (k_chem) increased from 1 × 10⁻⁵ cm s⁻¹ to 4 × 10⁻⁴ cm s⁻¹ accordingly, as calculated by ECR experiments at 750 °C.

Aside from bilayer or multilayer structures, Ma et al.⁷⁷ also fabricated VAN structured LSC₁₁₃/LSC₂₁₄ oxygen electrodes that were coherently grown onto STO(001) and YSZ/GDC(001) substrates using PLD. Results from high resolution STEM-EDX verified the separation of LSC₁₁₃ and LSC₂₁₄ columns in near-vertical orientation with a ∼300 nm average size of each column. Here, the ASR of the VAN thin-film electrode (∼2 × 10⁴ Ω cm²) was only ∼10% of LSC₁₁₃ (∼1.6 × 10⁵ Ω cm²) and LSC₂₁₄ (∼3.5 × 10⁵ Ω cm²) single phase electrodes in air at 320 °C.

ABO₃/A₂BO₄ electrodes with different elemental composition of the ABO₃ and A₂BO₄ components have also been reported to exhibit enhanced performance. For example, Zhang et al.²⁴⁹ demonstrated that the infiltration of La₂NiO_4+δ (LNO) into (La_0.6Sr_0.4)_0.995Co_0.2Fe_0.8O_3−δ (LSCF) can decrease the ASR value from 1.34 Ω cm² to 0.042 Ω cm² at 750 °C with the activation energy changing from 1.38 eV to 1.06 eV. Here, the resulting cell with the LSCF–LNO cathode demonstrated a 67% increase in peak power density (697 mW cm⁻²) relative to the cell with LSCF at 750 °C. And after long-term operation at a constant current density of 250 mA cm⁻², the degradation rate was reported to be as low as 0.39% after 500 hours’ testing.

4.1.2 Perovskites with different cations (ABO₃/A′B′O₃). Heterostructured ABO₃/A′B′O₃ have been reported to demonstrate considerable performance and durability. For example, Lee et al.⁷⁹ reported that La_0.8Sr_0.2CoO_3−δ (LSC82) thin films with an extremely thin LSM layer covering exhibited significantly accelerated surface exchange kinetics and stabilized surface chemistry. For instance, k^q of the LSC82/LSM (∼0.3 nm) bilayer was measured to be ∼3 × 10⁻⁸ cm s⁻¹ in 10% O₂ at 550 °C, which was nearly 1 order of magnitude higher than that of the LSC82 base film. In addition, the decorated film demonstrated no obvious performance degradation for over 70 h in 1 atm O₂ at 550 °C, whereas the activity of the bare LSC82 film rapidly degraded to ∼1% after 20 hours’ annealing (Fig. 12a).

Fig. 12 (a) k^q of LSC82 with a thin LSM82 coating (0.9 nm) and the LSC82 reference as a function of annealing time.⁷⁹ (b) Comparison of the electrochemical activities of LSCF decorated with CeO₂, PCO and PrO₂.²³⁴ (c) Schematics of a LSCF backbone and conformal, dense PNM decoration as well as exsoluted PrO_x nanoparticles.²⁵⁶ (d) Typical I–V–P curves and (e) long-term stability performance (at a constant cell voltage of 0.7 V) for Ni–YSZ anode supported cells with different cathodes: bare LSCF, PNM-, PrO_x- and hybrid PNM–PrO_x catalyst-decorated LSCFs with 3% humidified H₂ as the fuel and air as the oxidant at 750 °C.²⁵⁶ Reproduced from ref. 79, 234 and 256 with permission from American Chemical Society, copyright 2014; American Chemical Society, copyright 2018 and RSC, copyright 2017, respectively.

Lynch et al.²⁴⁸ also reported that the electrochemical properties of porous La_0.6Sr_0.4Co_0.2Fe_0.8O_3−δ (LSCF) cathodes can be enhanced through a uniform La_0.85Sr_0.15MnO_3±δ (LSM) decorating layer using the infiltration method. The current densities improved from 0.7 A cm⁻² to 1.1 A cm⁻², resulting in enhanced peak power densities (655 mW cm⁻² at 700 °C) in 200 hours’ operation. Results from TEM and convergent beam electron diffraction (CBED) for the aged oxygen electrodes showed that the LSCF skeleton retained its perovskite structure after operation, and the LSM surface coatings exhibited a loss in crystallinity. In another example, Zhu et al.²⁴⁵ fabricated a (Ba,Sr)(Co,Fe)O₃ porous cathode and decorated it with a (La,Sr)MnO₃ nanofilm through solution impregnation. The polarization resistance of the LSM-decorated BSCF was found to be 0.023 Ω cm² at 750 °C in air, which was ∼20% lower than that of the undecorated cathode. The researchers also reported that the LSMO layer on the BSCF backbone prevented inherent electrode poisoning by CO₂, H₂O and SO₂ in air, resulting in considerable stability enhancements. Moreover, Wang et al.²⁵² reported that the infiltration of perovskite SrCo_0.9Nb_0.1O_3−δ (SCN10) onto LSCF can lead to an 89.4% reduction in ASR, indicating enhanced ORR/OER kinetics.

4.1.3 Perovskites/simple metal oxides. The combination of perovskites and simple metal oxides in heterostructures can enhance the electrocatalytic activity and electrode stability in SOC operation. For example, Pr-based oxides were reported to be able to decrease the formation energy of oxygen vacancies, whereas Ce-based oxides could stabilize the electrode chemistry and structure. In one study, Chen et al.²³⁴ fabricated LSCF/Pr_xCe_1−xO₂ bilayers using PLD and reported significant ORR activity enhancements (Fig. 12b). Here, the researchers proposed that the introduction of PrO₂ with high oxygen vacancy concentrations onto LSCF can enhance the ORR process by as much as 6 times compared with bare LSCF, reaching ∼10⁻⁷ cm s⁻¹ at 600 °C in air. The researchers also suggested that Pr_0.2Ce_0.8O₂ and CeO₂ coatings with low oxygen vacancy densities can prevent Sr enrichment and segregation at the surface, resulting in stable operation at 600 °C for ∼60 h in air.

Tsvetkov et al.²³⁶ conducted systematic research into the decoration of less reducible cations on perovskite LSC, including V₂O₅, Nb₂O₅, TiO₂, ZrO₂, HfO₂ and Al₂O₃, and concluded that related metal oxide decoration can improve the surface exchange kinetics with the exception of V₂O₅. A ‘volcano’ relation between k^q and the oxygen vacancy formation enthalpy (ΔH^V_f(MeO_x)) was proposed in which optimal k^q enhancement (∼30 times) was achieved through HfO₂ addition after 54 h in air at 530 °C.

Furthermore, Rupp et al.²⁵³ used in situ impedance spectroscopy during pulsed laser deposition (IPLD) to investigate the electrochemical activity of LSC during the deposition of well-defined monolayer fractions of Sr-, Co- and La-oxides. Here, the researchers reported that Co₂O₃ decoration using one laser pulse reduced the surface exchange resistance (R_surf exch) by 13%, whereas SrO decoration using one laser pulse (∼13 pmol Sr) increased R_surf exch by 42 ± 6%. The researchers in this study attributed this difference to the fact that Co decoration can induce more active sites whereas Sr decoration can deactivate LSC surfaces.

The decoration of Zr-based and Ce-based oxides onto perovskite surfaces has also been reported to improve the reaction activity and operation stability of oxygen electrodes.^240–242 For example, Gong et al.²⁴¹ decorated LSC surfaces with less electrocatalytically active ZrO₂ using atomic layer deposition (ALD) and reported that the ZrO₂ overcoating gradually became porous and mixed conducting after thermal exposure, resulting in elevated ORR properties. In addition, the researchers reported that the ZrO₂ coating prevented LSC nanoparticles from agglomerating and suppressed Sr enrichment at the surface during long-term operation. As a result, the nanostructured LSC–ZrO₂ electrode exhibited significantly lower polarization resistance and a 19 times slower degradation rate over 4000 h at 700 °C in air as compared with pristine LSC. In another study, Gong et al.²⁴⁰ reported that ZrO₂ coatings can reduce the degradation rate of porous LSCF–GDC to ∼25% as compared with undecorated samples during 1100 hours of operation at 800 °C. Saher et al.²⁴² also reported that LSCF skeletons with a porous nanoparticulate GDC layer can exhibit a high surface exchange coefficient (k_chem) of ∼3 × 10⁻⁴ m s⁻¹ at 900 °C, which was 6 times higher than that of pure LSCF. Here, the researchers suggested that the decoration of GDC particles facilitated the dissociative adsorption of O₂, leading to enhanced activities for the heterostructure.

Dispersed MgO,¹¹⁴ NiO¹¹⁵ and CuO¹¹⁶ particles have also been highlighted as promising synergistic catalysts of LSCF electrodes. For example, Yang et al.¹¹⁴ reported that by infiltrating MgO particles onto LSCF skeletons, the ASR of the oxygen electrode can be reduced from 0.49 Ω cm² to 0.31 Ω cm² at 650 °C. In addition, researchers reported that the introduction of CuO can greatly improve the surface exchange kinetics of LSCF with k_chem rising from 2.6 × 10⁻⁵ cm s⁻¹ to 9.3 × 10⁻⁵ cm s⁻¹ at 750 °C.¹¹⁶ Furthermore, coatings on cathodes can also enhance the cell performance, in which cells with Ni–YSZ|YSZ/SDC|LSCF–NiO can exhibit a high peak power density of 1.031 W cm⁻² at 800 °C, which is ∼1.5 times higher than that of single cells with bare LSCF as the cathode.¹¹⁵

Despite these promising results, however, there are also studies highlighting the fact that surface decorations with simple metal oxides may reduce the electrochemical performance. For example, Kim et al.²⁴³ modified La_0.6Sr_0.4CoO₃-based electrodes with a Al₂O₃ layer using ALD and reported that excessive Al₂O₃ coating on LSC-based electrodes can negatively affect the power output (reduced from 1.62 to 0.742 W cm² at 800 °C). In addition, there are also controversial evaluations concerning CoO_x and SrO coatings,^254,255 in which the results suggest that the surface modification process needs to be rationally controlled. Here, proper amounts of decoration can activate electrode surfaces whereas excessive coating may further block active sites and degrade the performance.

4.1.4 Other types of heterostructured electrodes. In terms of other types of heterostructured electrodes, Ai et al.⁶ fabricated a special core–shell structured Er_0.4Bi_1.6O₃ (40 wt%)–La_0.76Sr_0.19MnO_3+δ (ESB–LSM) oxygen electrode with high activity and durability using a new gelation and direct assembly method. The single cell with ESB–LSM as the oxygen electrode exhibited a peak power density of 1.62 W cm⁻² at 750 °C, which was obviously larger than that of a full cell with the LSM oxygen electrode (0.48 W cm⁻²). Here, the researchers reported that the uniform coverage of bismuth deposited onto the LSM surface inhibited the grain growth of LSM through thermal ripening and the cell also demonstrated good durability in all fuel cell, electrolysis and reversible SOC modes for over 200 h at 500 mA cm⁻² and 750 °C.

Furthermore, Chen et al.²⁵⁶ demonstrated that a hybrid catalyst coating consisting of a conformal PrNi_0.5Mn_0.5O₃ (PNM) thin film and exsoluted PrO_x nanoparticles on LSCF obtained using a one-step infiltration process can significantly improve the ORR kinetics and durability (Fig. 12c). The heterostructured cathode reduced the polarization resistance to ∼0.022 Ω cm² at 750 °C, which was 1/6 that of bare LSCF cathodes. Here, the researchers reported that in full cell operation, anode-supported cells with the hybrid catalyst-coated LSCF cathode exhibited an extraordinary peak power density of ∼1.21 W cm⁻² (Fig. 12d) and excellent durability at 0.7 V for 500 h (Fig. 12e). And if applied in BaZr_0.1Ce_0.7Y_0.1Yb_0.1O₃-based protonic fuel cells, electrodes combining PNM with exsoluted PrO_x particles can exhibit excellent electrocatalytic activity with ASRs as low as 0.052 Ω cm² at 700 °C and operation stability under 0.7 V for ∼500 h.²⁵⁷ In another study, a PrBa_0.5Sr_0.5Co_1.5Fe_0.5O_5+δ (PBSCF) nanofibre-based cathode with PrO_x nanoparticles exhibited high activity for oxygen reduction. The PPD of PBSCF–PrO_x|SDC|Ni–BZCYYb reached as high as 0.37 W cm⁻² at 500 °C with nearly dry methane as the fuel and air as the oxidant.²⁵⁸

4.2 The role of heterointerfaces in oxygen electrodes

Heterostructured oxygen electrodes demonstrate strongly enhanced performance and in the following section the mechanisms proposed in the literature regarding the effects of heterointerfaces on the oxygen electrode performance are summarized,^48,229 such as strain effects,^48,51,231 electronic structure reconstructions,^{40,56,77,200,259} cation diffusion^{79,223,230,260–262} and extended TPBs.^{4,48,232,235,263,264} These mechanisms will be discussed in detail respectively in the following sections.

4.2.1 Strain effects. Due to lattice parameter mismatches, crystal lattice distortions are commonly observed near heterointerfaces, leading to lattice strain in heterostructured materials. Since the lattice strain normally presents near the heterointerfacial region (normally several layers of atoms),^51,53 strain effects are more pronounced in epitaxial thin films than conventional electrodes. In addition, epitaxial thin films of well-ordered structure are ideal model systems for investigation of the strain effect,^265–271 while it's difficult to quantify the strain effect in conventional electrodes. For oxygen electrodes of SOCs, such lattice strain can have a profound influence on the ORR/OER kinetics by not only changing vacancy formation energies and migration barriers but also modifying electronic structures near heterointerfaces.^{231,265–267}

Researchers have proposed that tensile strain can generally increase the free spacing in crystal lattices and decrease bond energies and weaken interactions between ions,⁵³ thus facilitating the formation of oxygen vacancies.^231,267 For example, Petrie et al.²³¹ reported that the amount of oxygen deficiency in strained SrCoO_3−x can increase with x changing from 0 to 0.25 as the tensile strain increases from +1% to +4.2% (Fig. 13a). Here, the researchers used DFT and proposed that tensile strain can decrease the electronic hybridization between Co 3d and O 2p and further weaken Co–O bonds, leading to a greater intercalation formation enthalpy (H_i) and more thermodynamic oxygen instability in the tensile strained SrCoO_x. The researchers also reported that a 2% tensile strain can reduce the activation energy barrier (E_a) by ∼30% and facilitate the process of oxygen diffusion. Such an increase in H_i and decrease in E_a will both result in an oxygen deficient state and further influence the electrochemical properties.

Fig. 13 (a) Oxygen deficiency in tensile strained P-SCO determined by XRD and O-K and Co-L₂ edge XAS measurements.²³¹ (b) Vacancy formation energies of LSM as a function of strain using an oxygen chemical potential for 1173 K and 0.1 atm pO₂.²⁶⁷ (c) Comparison of calculated O_vacancy formation energies for LSC/STO and LSC/LAO with different oxygen vacancy concentrations. The insets exhibit the smaller oxygen vacancy formation energies for LSC/STO.²⁶⁸ (d) Depth profiles of ¹⁸O⁻ distributions for tensile strained (100) LSC82 thin films on STO and compressive strained LSC82 on LAO after oxygen isotope exchange at 400 °C for 5 min. Reproduced from ref. 228, 231, 267 and 268 with permission from WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim, copyright 2016; Elsevier, copyright 2017; WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim, copyright 2019 and American Chemical Society, copyright 2013 (https://doi.org/10.1021/nn305987x), respectively.

In another example, Morgan et al.²⁶⁷ used ab initio calculations to develop a quantitative model for the evaluation of strain impacts on vacancy formation energies (Fig. 13b) and proposed that tensile strain can increase lattice volumes, which caused decreased vacancy formation energies. By systemically studying strained LaMnO₃, La_0.75Sr_0.25MnO₃, LaFeO₃ and La_0.75Sr_0.25FeO₃, the researchers inferred that the vacancy formation energies for most oxides generally declined under tensile strain; a 1% increase in the biaxial strain was estimated to correspond to a ∼30–100 meV decrease in the vacancy formation energy.

Recently, Chen et al.^268,269 investigated the critical impact of lattice strain on the oxygen defect chemistry and OER activity of LSC in liquid solutions through a combination of DFT calculations and experimental measurements. Here, the DFT results showed that the oxygen vacancy formation energy of tensile LSC/STO was smaller than that of LSC/LAO (Fig. 13c). As a result, LSC/STO contained more lattice oxygen vacancies than LSC/LAO. These calculations were also consistent with the larger δ values observed in HRXRD and the increased Co²⁺ signals in XPS characterization of LSC/STO.

Han et al.⁵¹ suggested that strain effects can influence not only the formation of oxygen vacancies, but also the incorporation and transportation of oxygen defects. By using DFT+U calculations, the researchers proved that lattice strain near the LSC₁₁₃/LSC₂₁₄ heterointerfacial region can enhance the concentration and mobility of oxygen vacancies, leading to highly accelerated oxygen dissociation and incorporation kinetics.^265,270 Here, the researchers reported that the tensile strain in LSC₁₁₃(001) (+1.9%) reduced elastic interactions along the oxygen migration paths and facilitated both oxygen adsorption and oxygen vacancy formation in neighboring regions. In addition, the researchers also reported that the tensile strain on LSC₁₁₃ in the (110) direction can lead to more available spacing among crystal lattices for oxygen migration, minimizing energy barriers. Furthermore, calculations in this study demonstrated that lattice strain can contribute to 3 × 10² times of conductivity enhancement at 500 °C (considered both in LSC₁₁₃ and LSC₂₁₄ together).

There were also some studies indicating that strain effects on electronic structures can greatly influence the charge transfer kinetics in ORR/OER processes. For example, Cai et al.²⁵⁹ studied the dependence of surface electronic structures on strain states in epitaxial La_1−xSr_xCoO_3−δ (LSC) thin films with STO and LAO substrates. Results from in situ STM/STS revealed an electronic structural transformation from a semiconducting state with an energy gap of 0.8–1.5 eV at room temperature to a metallic-like state with no energy gaps at 200–300 °C. The researchers also found that in tensile strained LSC, the electronic density around the Fermi level was strengthened following this transition, facilitating charge transfer in the ORR. And by collaborating with Kubicek et al.,²²⁸ these researchers further demonstrated that the surface exchange coefficient (k*) and diffusion coefficient (D*) of the tensile strain of LSC–STO were ∼4 and ∼10 times larger than that of the compressive strain of LSC–LAO, respectively (Fig. 13d).

In a recent review article, Hwang et al.²⁷¹ discussed the impacts of lattice strain on the electronic structure and electrocatalytic activity of perovskite oxides. They proposed that lattice strain can change the level of octahedral rotation and structural deviation as well as the bond length of metal-oxides, thus modifying the oxide electronic structures through 3d orbital occupancy and M 3d–O 2p orbital overlap.^272,273 As the energy gap between M 3d and O 2p centers was considered as a quantitative descriptor of the electronic changes, the lattice strain may therefore have broad effects on not only the surface reactivity such as the oxygen adsorption and dissociation energy, but also the bulk defect energetics such as the oxygen vacancy formation and migration energy.

4.2.2 Cation segregation. Driven by a combination of electrostatic and elastic forces, cations inside lattices tend to segregate to the surface or interface to minimize the total energy and stabilize electrode systems.^{227,262,274–276} Such cation rearrangement typically occurs on the nanoscale yet has profound impacts on the properties and functionality of heterointerfaces.²⁷⁷

Zhang et al.²⁴⁹ used TEM/EDAX to observe Sr, Fe and Co signal peaks on LNO nanoparticles in LNO-infiltrated LSCF cathodes after long term testing, revealing the occurrence of the cation segregation phenomenon (Fig. 14a). Here, the researchers reported that the segregation of highly-active cobaltic oxides accelerated oxygen surface exchange and improved the electrocatalytic activity. Furthermore, the Sr diffusion increased the concentration of electron holes and enhanced the charge transport process. These resulted in a 67% increase in power output over long-term operation.

Fig. 14 (a) TEM image and EDAX spectra of LNO-infiltrated LSCF particles after 1000 h operation.²⁴⁹ (b) HADDF-STEM micrograph of the NSC₁₁₃/NSC₂₁₄ interface of the NSC_214/113-17L sample and ratios of Sr/(Sr + Nd) and (Sr + Nd)/Co along the (001) direction.¹⁰⁷ Reproduced from ref. 107 and 249 with permission from Elsevier, copyright 2014 and 2018, respectively.

Researchers have attempted to uncover the driving forces of cation segregation at heterointerfaces. For example, Zheng et al.¹⁰⁷ used HADDF-STEM to obtain direct evidence of cation segregation in the interfacial region of NSC_113/214. According to their results, the stacking sequence of atoms along the c-axis was in the order of perovskite (Nd_0.5Sr_0.5CoO_3−δ) and Ruddlesden–Popper (Nd_0.8Sr_1.2CoO_4±δ), and the (Sr + Nd)/Co ratio conformed to the stoichiometry, respectively. However, the Sr/(Sr + Nd) ratio was found to be enhanced obviously at the interfacial region, revealing that Sr strongly segregated near the NSC_113/214 interface (Fig. 14b). The researchers also used DFT calculations to reveal that E(Sr_Nd) (the energy required to replace a surface Nd³⁺ cation with a lattice Sr²⁺ cation) decreased significantly at the interfacial region (−0.53 eV), indicating a strong thermodynamic trend that can drive Sr from the perovskite bulk to the interface.

Gadre et al.²²³ used DFT simulations to calculate cation diffusion at the LSC₁₁₃/LSC₂₁₄ heterointerface and reported that Sr tended to migrate to the interfacial region thermodynamically. Moreover, the heterointerface can stabilize high dopant Sr concentrations at the interface and suppress the formation of secondary passive particles (such as SrO). In addition, the researchers reported that the increased concentration of Sr can induce the formation of oxygen vacancies, leading to enhanced electrocatalytic activities. In addition, combining COBRA with ab initio calculations, Feng et al.⁵⁷ provided atomic-scale structural information on the LSC₁₁₃/LSC₂₁₄ heterointerface, in which the signals at (0,0,0) and (0.5,0.5,0) (Fig. 15a and b) exhibited apparent Sr segregation at the interface and near the surface of LSC₂₁₄. Here, the low Sr_La substitution energy in the heterogeneous region was believed to be responsible for Sr segregation. Using ab initio calculations, Lee et al.²³⁰ found that the Sr_La substitution energy in the LSCF₁₁₃/LSC₂₁₄ heterogeneous region was much lower than that of LSC_113/214 (Fig. 15c), which was consistent with the weaker Sr interdiffusion trend of LSCF₁₁₃/LSC₂₁₄ as observed by AES.²⁷⁸

Fig. 15 Electron densities of as-deposited LSCO₂₁₄/LSCO₁₁₃/STO thin films along: (a) the (0 0 Z) and (b) (0.5, 0.5, Z) lines using COBRA.⁵⁷ (c) Up: Sr_La substitution energies in bulk LSC₂₁₄, LSC₁₁₃ and LSCF₁₁₃ calculated by DFT simulations. Middle: heterostructure model used in DFT simulations. La/Sr (dark blue), Fe/Co (light blue, center of the octahedra) and O (red). Bottom: Sr_La substitution energies of each layer near the heterointerfacial region.²³⁰ Reproduced from ref. 57 and 230 with permission from American Chemical Society, copyright 2014 (https://pubs.acs.org/doi/10.1021/jz500293d) and RSC, copyright 2015, respectively.

In addition, Tsvetkov et al.²³⁶ proposed that the phase separation of Sr-rich particles was detrimental to the ORR kinetics and that the decoration of less reducible cations (, Me = V⁵⁺, Nb⁵⁺, Ti⁴⁺, Zr⁴⁺, Hf⁴⁺ and Al³⁺) onto perovskite LSC can balance Sr segregation with the ORR kinetics. Here, the researchers suggested that decoration can reduce the oxygen vacancy concentration acting as reactive sites, while the coatings can increase the Co oxidation state and decrease the electrostatic attraction of Sr enrichment onto the surface, further suppressing harmful phase separation and stabilizing the surface composition. Based on the results of APXPS/XAS, the researchers summarized that these two factors resulted in a ‘volcano’ relationship between k^q and ΔH^V_f(MeO_x). In another study, Rupp et al.²⁵³ reported that SrO decoration onto LSC severely deactivated the LSC thin film, whereas deposited Co₃O₄ re-activated the surface. Here, the researchers used IPLD and found that deposited Sr cations can migrate and attach to highly active sites of inhomogeneous surfaces, hindering ionic and electronic transport in the ORR process,²⁷⁹ leading to the deactivation of active sites. Alternatively, Co₃O₄ decoration can induce highly active sites on surfaces.²⁸⁰

Surface Sr segregation and separation are major causes of the degradation of Sr containing electrodes. It has been demonstrated that the deposition of a secondary phase to form a heterostructure can effectively suppress the phase separation of Sr. For example, Wen et al.²⁸¹ proposed that ZrO₂ coatings fabricated using ALD can suppress excess Sr segregation in LSC epitaxial films and improve the stability of oxygen electrodes. Here, the researchers suggested that the existence of cation exchange between Zr and Co near the interface can reduce the electrostatic force of surface Sr segregation, thus improving the thermal and electrochemical stability. In another study, Gong et al.²⁴¹ confirmed that ZrO₂ overcoats can optimize nanostructured LSC surfaces by preventing agglomeration and suppressing Sr segregation in long term operation. Similarly, the suppression of surface Sr segregation using surface coatings was also reported for LSM-decorated LSCF,⁴¹ Pr_0.2Ce_0.8O₂/CeO₂-decorated LSCF,^234,282 PSM/PSCM-decorated LSCF,^247,283etc.

4.2.3 Electronic structure. The electronic structure has a crucial influence on the electrochemical reaction activity of oxygen electrodes. For example, Jin et al.²⁸⁴ have reported a volcano-shaped relationship between the OER activity and the occupancy of the 3d orbital with an e_g symmetry of the surface transition cation. At the heterogeneous interfacial region, the electronic structure of materials can differ significantly from the single phase counterparts. For example, Kuru et al.²³⁵ prepared La_0.65Sr_0.35MnO₃ (LSM)/SrTi_0.2Fe_0.8O₃ (STF) multilayers using PLD and demonstrated that the bandgap values decreased to 0.3 and 1 eV for LSM and STF respectively, which were considerably smaller than the bulk counterparts (∼1 eV and ∼2 eV for LSM and STF). Furthermore, Dagotto et al.²⁰⁰ proposed that the displacement of interfacial ions can lead to electronic reconstruction at the interface, which was responsible for the unique electrochemical properties. These changes in electronic structure can also strongly impact charge transfer processes from the electrode to adsorptive oxygen.^285,286

By combining focused ion beam (FIB) milling²³⁵ with in situ scanning tunneling microscopy and spectroscopy (STM/STS), Chen et al.⁵⁶ discovered that LSC₂₁₄ in a LSC_113/214 multilayer (ML) can be electronically activated. This electronic activation was highlighted by the smaller bandgap and a reversible transition from a semiconductor state to a metallic-like state with the temperature increasing to ∼250 °C (Fig. 16a). Here, the researchers reported that in the LSC_113/214 interface, electron injection from LSC₁₁₃ to LSC₂₁₄ can result in band alignment and a decreased distance between the Fermi level and the conducting band of LSC₂₁₄. The reduced distance can exponentially increase the quantity of available electrons and accelerate the kinetics of electron transfer from the electrode surface to adsorptive oxygen (Fig. 16b). Furthermore, Ma et al.⁷⁷ reported that VAN-structured LSC_113/214 exhibited similar electronic structure changes to ML LSC_113/214. The energy bandgap of VAN was 1.5 ± 0.2 eV at room temperature, while metallic-like behavior with no energy gap was found above 250 °C. Here, the researchers proposed that electronic activation and stable surface composition can lead to enhanced ORR activity. In a recent study, Chen et al.⁸⁰ used HRXRD (high resolution X-ray diffraction) and HAXPES (hard X-ray photoelectron spectroscopy) to probe the distribution of oxygen defects across the heterointerfaces of LSC₁₁₃/LSC₂₁₄ and LSC₁₁₃/LNO₂₁₄. Here, larger concentrations of oxygen vacancies were found in the LSC₁₁₃ of LSC₁₁₃/LSC₂₁₄ than in the LSC₁₁₃ of LSC₁₁₃/LNO₂₁₄ in a reducing environment, while both the LSC₂₁₄ and LNO₂₁₄ layers in the heterostructured samples were found to be more reduced compared with the respective single counterparts (Fig. 16c).

Fig. 16 (a) Transformation of apparent energy gaps in single phase LSC₁₁₃ and LSC₂₁₄ as well as multilayered LSC₁₁₃/LSC₂₁₄ at different temperatures.⁵⁶ (b) Origin of the enhanced ORR activity at heterostructured LSC₁₁₃/LSC₂₁₄ layers.⁵⁶ (c) Illustrations of how the LSC₁₁₃/LSC₂₁₄ and LSC₁₁₃/LNO₂₁₄ interfaces induce the formation of local electric fields and influence local oxygen defect equilibria; (left) the misalignment of Fermi levels in LSC₁₁₃ and LSC₂₁₄ (LNO₂₁₄) single phases after the loss of oxygen through annealing at high temperatures, (right) the alignment of Fermi levels in LSC₁₁₃ and LSC₂₁₄ (LNO₂₁₄) heterostructures after annealing through the creation of local electric fields.⁸⁰ Reproduced from ref. 56 and 80 with permission from WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim, copyright 2013 and American Chemical Society, copyright 2018, respectively.

4.2.4 Extension of TPBs. It is accepted that the pathways of gas-phase molecules, electrons and oxygen ions (or vacancies) are essential for electrochemical reactions, thus a triple-phase boundary is needed.^4,264 For EC materials such as LSM and MIEC with poor ionic conductivity, the ORR/OER mainly occurs near the interface of the electrode/electrolyte. Therefore, the introduction of ionic conductors is particularly important in these systems. For example, in LSM–GDC composite electrodes, LSM can provide high electronic conductivity and catalytic capacity, while GDC can act as a highway for oxygen transport,²⁸⁷ as illustrated in Fig. 17(a).

Fig. 17 (a) Illustration of an idealized model of an integrated miniature half-cell with sufficient TPB; (b) TEM image and schematic of a LSC₁₁₃ thin film with LSC₂₁₄ particle decoration.⁴⁸ Reproduced from ref. 48 with permission from American Chemical Society, copyright 2010.

Crumlin et al.⁴⁸ reported high ORR activities on the heterointerface of nano-island LSC₂₁₄ (La_0.5Sr_0.5)₂CoO_4±δ decorated LSC₁₁₃ (La_0.8Sr_0.2CoO_3−δ) films, in which they proposed that the dispersive nano-island LSC₂₁₄ can extend the TPB, leading to excellent ORR activity (Fig. 17b). Wan et al.²⁸⁸ also proposed that the oxygen reduction kinetics of PrBaCo₂O_5+δ–Sm_0.2Ce_0.8O_1.9 (PBC–SDC) composites can be enhanced through the introduction of 3PB (PBC–SDC-gas), in which the polarization resistances of the oxygen electrode decreased from 0.281 Ω cm² (for bare PBC) to 0.107 Ω cm² (for the composite with 60 wt% SDC) at 700 °C. Here, the researchers suggested that the 3PB accounted for more than 80% of active sites in the oxygen incorporation processes.

5. Heterointerfaces in fuel electrode of SOC devices

SOC fuel electrodes can provide active sites for fuel oxidation or CO₂/H₂O decomposition depending on the operating modes. High ionic conductivity, electronic conductivity, and catalytic activity are essential characteristics. In addition, high resistance to sulfur poisoning and carbon coking are also necessary because SOC fuel electrodes may be fed with H₂, H₂S, CO and hydrocarbon containing fuels such as syngas, biogas and gasification reformate from hydrocarbons.^289–291 Currently, nickel/yttria-stabilized zirconia (Ni–YSZ) has been the most widely used fuel electrode material for SOCs due to its high catalytic activity, operational reliability and acceptable cost.²⁹² However, the long-term durability of Ni–YSZ is inadequate due to inherent redox instability and serious carbon deposition during operation. To address this, studies have recently highlighted the introduction of secondary phases into Ni–YSZ electrodes to form heterostructures as a potential approach to enhance the electrochemical performance and suppress surface coke deposition.^59,293,294 In addition, alternatives to Ni–YSZ electrode materials, including various perovskite-based heterostructured oxides, have been developed and investigated, many of which have shown considerable catalytic activity and stability.^258,295,296 Furthermore, studies have shown that treating these perovskite-based electrode materials under reducing conditions can further lead to the generation of metallic nanoparticles (known as in situ exsolution) on oxide matrixes and the formation of unique oxide–metal heterostructures. Different from the skeleton/surface coating structure fabricated by conventional approaches (such as infiltration), the concentration of oxygen vacancies can promote the catalytic activity of CO₂ electrolysis and ionic conductivity, while the strong bonding of the exsolved metal particles to the parent phase can greatly enhance the stability of the heterostructures under fuel cell operating conditions (Fig. 18).^4,297,298

Fig. 18 Comparison of heterostructured fuel electrodes fabricated by infiltration and exsolution.

Overall, the development of high-performance heterointerfaces is not only a promising strategy to enhance the electrode performance but can also greatly extend the scope of selection for new electrode materials. Therefore, recent progress in the utilization of heterostructured oxides as high performance fuel electrodes will be summarized in this section to provide direction for the research of more efficient and economical fuel electrode materials to further improve the SOC performance and stability.

5.1 Ni/oxide heterostructures in Ni/YSZ fuel electrodes

Conventional Ni/YSZ electrodes are troubled by surface coking in carbon-containing fuels. To address this, efforts have been made to modify the surface chemistry in which researchers report that extrinsic decoration with simple oxides (BaO, CaO, MgO, etc.) can impact the fuel electrode performance.^{59,293,294,299,300} For example, finely distributed nanoparticle oxide coatings have been recently deposited onto anode surfaces to generate oxide/metal heterostructures through various surface engineering techniques, including infiltration,³⁰¹ evaporation deposition,²⁹³ carbonate decomposition,⁵⁹etc. They have shown promise in the efficient enhancement of the electrochemical catalytic properties and the simultaneous extension of lifespans for Ni/YSZ electrodes if applied in hydrocarbon atmospheres.

In one example, Yang et al.²⁹³ synthesized a novel BaO–Ni/YSZ anode with a finely distributed Ni/BaO heterointerface using vapor deposition and subsequent reduction and tested a single cell (BaO–Ni/YSZ|YSZ/SDC|LSCF) in CO. Here the researchers achieved a typical PPD of ∼0.70 W cm⁻² at 750 °C, which was more than twice that of conventional Ni–YSZ anodes.^301,302 Furthermore, the cell in this study provided stable power outputs for ∼100 h with 500 mA cm⁻² in wet CO, demonstrating superior coking resistance. In another example, Tao et al.⁶⁰ found that the BaO/Ni interface can stabilize cathode surfaces against coking in hydrocarbon fuels, in which the peak power density of their oxide cell (Ni–BaZr_0.1Ce_0.7Y_0.2O_3−δ (BZCY)–SDC|SDC|LSCF–SDC) reached 570 mA cm⁻² at 700 °C in methane in 150 minutes of operation with a low electrode resistance of 0.136 Ω cm². Likewise, Islam et al.³⁰³ reported that 5 wt% BaO loading on Ni/YSZ through infiltration increased the peak power density by ∼40% at 1073 K in 100% CH₄.

Researchers have studied the mechanisms of coking resistance properties provided by extrinsic metal oxide decoration.^59,60,293 For example, Yang et al.²⁹³ proposed that under operating conditions, carbon can be reversibly generated and removed at Ni surfaces through 5 primitive steps, including: (1) CO adsorption onto Ni due to a high adsorption energy (−2.6 eV) and the water-gas shift reaction; (2) carbon deposition on the surface of Ni catalysts due to the dehydrogenation of hydrocarbon fuels or CO disproportionation; (3) strong H₂O adsorption on the BaO sites of BaO/Ni(111), leading to barrierless O–H bond cleavage surrounding BaO; (4) the reaction between dissociated OH and adsorbed C on Ni surfaces to form an intermediate COH, which dissociates to CO and H; and (5) CO and H diffusion to the TPB and electrochemical oxidation to CO₂ and H₂O.

Alternatively, Li et al.⁵⁹ have studied the origin of enhanced coking resistance through BaO decoration on Ni surfaces using in situ Raman spectroscopy. They found the formation of a thin Ba(OH)₂ layer on Ni surfaces during operation in wet hydrocarbon fuels, which further enabled the conversion of deposited coke to CO or CO₂ rapidly. Here, the possible chemical reactions occurring near the interface can be summarized as follows:

BaO + H₂O ⇔ Ba(OH)₂; C_ad + Ba(OH)₂ → BaO + CO + H₂; CO + Ba(OH)₂ → BaO + CO₂ + H₂; BaO + CO₂ → BaCO₃.

This mechanism was supported by the rapid decrease in the –OH intensity accompanied by the gradually increasing –CO₃ signal as measured by in situ Raman spectroscopy (Fig. 19).⁵⁹ Likewise, Tao et al.⁶⁰ proposed that the BaO/Ni interface can accelerate water adsorption and enhance the reaction of water-mediated carbon removal, whereas fast oxygen ion transportation and generated steam at the anode can promote hydrocarbon reformation to enhance the performance and stability.

Fig. 19 (a) Integrated peak intensities of the carbon G-bands for bare Ni and BaO–Ni as functions of exposure time in wet C₃H₈ at 450 °C. (b) Collection of time-resolved Raman spectra on Ni powders and (c) on BaO modified Ni powders. Reproduced from ref. 59 with permission from American Chemical Society, copyright 2015.

Despite enhanced surface stability, however, surface decoration needs to be precisely controlled to avoid negative effects such as the formation of insulating alkaline oxides that can block active sites. For instance, the covering of TPB active regions with BaO particles can increase electrode polarization resistances and partly lower power densities.^{59,294,304,305} To address this, Islam et al.³⁰³ recently developed a novel fabrication technique combining microwave irradiation with vapor deposition to control the distribution and quantity of deposited nanosized BaO particles to ensure that additive components were only deposited separately on Ni surfaces.

In addition to BaO, decoration with alkaline earth metal oxides such as MgO and CaO has also been reported to enhance the electrode stability through similar mechanisms. For example, Qu et al.²⁹⁴ modified Ni–SDC anodes using CaO nanoparticles and achieved only slight degradations in performance during operation in a wet CH₄ atmosphere for 70 h at 650 °C whereas an unmodified Ni–SDC-based cell failed to generate power within 1.2 h in the same conditions. Hua et al.²⁹⁹ also infiltrated Ni–YSZ cermet anodes with MgO additives and reported an R_p of 0.93 Ω cm² after 48 h operation, which was 31% lower than that of a NiO infiltrated anode. Furthermore, Lay et al.³⁰⁰ proposed that a moderate amount of MgO decoration on the Ni–YSZ surface can lead to accelerated steam gasification reactions and enable less generation of surface carbon fibers in CH₄ atmospheres.

5.2 Fluorite/perovskite heterostructures in Ni/YSZ fuel electrodes

Many perovskite structured oxides have been shown to possess considerable conductivities and superior coking tolerances in application as electrolytes or oxygen electrodes. However, these materials possess limited application as fuel electrodes due to inherent structural instabilities in reducing atmospheres. Despite this, some nanosized perovskites have been shown to remain stable under reducing conditions and currently investigations aimed at introducing such perovskite nanoparticles onto SOC fuel electrode surfaces to form fluorite/perovskite heterostructures and enhance the stability against surface side reactions are being conducted.^{258,295,296,306}

Vincent et al.²⁹⁵ fabricated a YSZ/La_0.4Sr_0.5Ba_0.1TiO₃ anode through infiltration and achieved a peak power density in the single cell of 200 mW cm⁻² at 850 °C in H₂/H₂S (5000 ppm). Moreover, Xu et al.²⁹⁶ fabricated a CeO₂/La_0.3Sr_0.7Ti_0.3Fe_0.7O_3−δ composite anode and reported polarization resistances as low as 0.151 Ω cm² at 850 °C in a CO atmosphere. More recently, Chen et al.²⁵⁸ designed a Ni–BaZr_0.1Ce_0.7Y_0.1Yb_0.1O_3−δ (BZCYYb)/SDC hybrid SOFC anode (Fig. 20) and reported that the corresponding cell was able to operate in nearly dry CH₄ fuel (with ∼3.5% H₂O) for more than 550 h at only 500 °C without coke formation.

Fig. 20 (a) Schematic of a single cell with yellow and grey grains representing BZCYYb and Ni phases. (b) SEM of the anode after long term operation for ∼500 h in nearly dry CH₄, indicating excellent coking tolerance. Reproduced from ref. 258 with permission from Springer Nature, copyright 2018.

In another example, Hua et al.³⁰⁷ reported that decorating Ni/YSZ anodes with PrBaMn₂O_5+δ nanoparticles can significantly enhance the catalytic activity and durability, leading to a 1.35 W cm⁻² peak power density in a biogas atmosphere (containing CH₄, CO₂ and small amounts of H₂S) at 800 °C as well as negligible performance degradation after 100 h of testing. Here, the researchers suggested that through infiltrating the Ni/YSZ surfaces with perovskite oxides, the finely distributed perovskites with high catalytic activity can act as micro-anodes that can increase the TPB area largely and enhance the processing capacity of the anode for converting hydrocarbon fuels (Fig. 21). Furthermore, Irvine et al.⁴ suggested that the heterogeneous and nanosized decoration of perovskite oxides such as SrTiO₃ and LaCrO₃ on fuel electrode surfaces can improve the long-term stability, resulting from the superior stability over a wide range of environments and moderate electronic conductivities. This mechanism appears to also be applicable in perovskite-modified Ni/YSZ electrode systems as well.

Fig. 21 Comparison of Ni/YSZ anodes before (a) and after (b) PrBaMn₂O_5+δ infiltration. The infiltrated perovskite nanoparticles resisted carbon species deposition on the Ni/YSZ surfaces and enhanced carbon removal. Reproduced from ref. 307 with permission from Elsevier, copyright 2017.

5.3 In situ assembly of oxide–metal heterostructures through exsolution

To improve the stability of Ni-based electrodes in carbon-containing environments, many perovskite-structured materials have been studied as potential alternatives to conventional Ni/YSZ fuel electrodes. B-site cations in perovskite lattices can be partly reduced in reducing conditions and generate numerous finely distributed metallic nanoparticles, thus forming numerous perovskite–metal heterointerfaces, which is defined as B-cation exsolution.^4,7 These in situ exsolved metal normally are strong bonded to the oxide skeleton, leading to facilitated charge and mass transfer processes and high stability of the materials during operation.³⁰⁸

Self-assembled metallic nanoparticles can normally exhibit higher conductivities and catalytic activities than perovskite bulk phases, leading to significant enhancements in fuel electrode electrochemical activity.^4,309–312 For example, Yang et al.³¹¹ proposed that exsolved NiO particles on SrFe_0.85Ti_0.1Ni_0.05O_3−δ (STFNi) electrode surfaces can significantly reduce the ASR from ∼0.4 to ∼0.15 Ω cm² at 700 °C, in which the anode-supported cell with NiO–STFNi as the fuel electrode can exhibit high peak power densities of 1650, 1420 and 1100 mW cm⁻² at 750, 700 and 650 °C respectively. Tsekouras et al.³¹² further compared La_0.4Sr_0.4Ni_0.06Ti_0.94O_2.94 (LSNT) with Ni⁰ exsolution, La_0.4Sr_0.4Fe_0.06Ti_0.94O_2.97 (LSFT) with Fe⁰ exsolution and La_0.4Sr_0.4TiO₃ (LST) and found that the exsolution trend of Ni⁰ was generally larger than Fe⁰ exsolution. Here, electrolysis cells operating in 47% H₂O/53% N₂ at 900 °C at −1.5 V with the above cathodes showed reduced HTSE onset potentials (E_HTSE) from −1.19 V for the LST cathode to 0.98 V and 0.63 V for the Fe⁰ and Ni⁰ exsoluted cathodes, respectively, with R_p decreasing from 1.7 to 1.5 and 1.2 Ω cm². Likewise, Li et al.³⁰⁹ fabricated a perovskite-structured (La_0.2Sr_0.8)(Ti_0.9Mn_0.1)O_3−δ (LSTM) skeleton with exsolved Ni nanocrystals on the surface and reported a ∼50% enhancement in catalytic activity for CO₂ electrolysis as compared with a bare electrode. The high catalytic activity for CO₂ electrolysis is likely associated with the high oxygen vacancy concentration.²⁹⁸ The exsolution of metal particles is often accompanied by formation and aggregation of oxygen vacancies near the interface. Theoretical calculations indicated that the exsolved Ni clusters and oxygen vacancies could synergistically weaken the C–O1 bond, thus activating the adsorbed CO₂ molecule and enhancing the CO₂ reduction. In addition, the increase in the oxygen vacancy concentration enhances the ionic conductivity of the fuel electrode through a vacancy-mediated mechanism.

Many studies have also confirmed that the surface composition and electrochemical performance of these exsolved particle systems experience no significant decline after many redox cycles, demonstrating considerable operational stability.^7,314–317 For example, Liu et al.³¹⁸ used a double-layered perovskite ((Pr_0.4Sr_0.6)₃(Fe_0.85Mo_0.15)₂O₇) with exsolved Co/Fe metallic nanoparticles and found that no graphite was generated during polarization for 100 h of CO₂ electrolysis, indicating excellent anti-coking properties. Furthermore, Song et al.³¹⁹ fabricated a self-assembled dual-phase structure (R-LS_0.6FCN) consisting of both SP and RP (SLF₂₁₄ + LFO₁₁₃) layers with exsolved Fe–Cu bimetal nanofibers and applied it as the fuel electrode of a SOFC. Here, the researchers reported that the peak power density of the solid oxide cell using R-LS_0.6FCN as the anode yielded 1.03 W cm⁻² at 850 °C in H₂/3% H₂O, which was higher than that of a similar single cell with Ni–Fe/SDC cermet as the anode. These researchers also reported that there was no obvious performance degradation after operation in H₂/50 ppm H₂S or syngas at 700 °C for 100 h, demonstrating high durability against coking/sulfur deposition. Here, such high stabilities were attributed to the formation of nanofiber networks without agglomeration and to the construction of an SP/RP heterointerface at the anode. More studies about the coking resistance capacity of perovskite-based fuel electrodes with exsolved metals are summarized in Table 3.

Table 3 Recent research about improving the coking resistance of fuel electrode via exsolution

Exsolved metal	Perovskite precursor	Temperature (°C)	Atmosphere	Cell configuration	Performance	Ref.
Fe	La0.6Sr0.3Cr0.85Fe0.15O3−δ (LSCFe)	800	CH4	LSCFe|YSZ–YSZ–LSM|YSZ	PPD ∼ 330 mW cm^−2	320
Co	Pr0.5Ba0.5Mn0.9Co0.1O2.89 (PBMCo)	900	Syngas	PBMCo|GDC–GDC–YSZ–GDC–LSM|GDC	PPD ∼ 900 mW cm^−2	321
Ni	La0.52Sr0.28Ni0.06Ti0.94O3 (LSNT)	800	20% CH4/H2		Slight carbon fiber growth after 4 h exposure	313
	La0.9Mn0.8Ni0.2O3 (LMN)	700	CH4/CO2		Negligible change for 24 h	322
Cu	La0.5Sr0.5Fe0.8Cu0.15Nb0.05O3−δ (LSFCNb)	800	Wet syngas	LSFCNb–SDC–ScSZ–SDC–LSFCNb	PPD ∼ 640 mW cm^−2; 9.4% degradation rate for 100 h at 0.7 V	323
Co–Fe alloys	(Pr0.4Sr0.6)3(Fe0.85Nb0.15)2O7 (RP-PSFN)	800	Wet C3H8	RP-PSFN–CFA|LSGM|BSCF	Stable operation for 100 h under 400 mA cm^−2	324
	Pr0.8Sr1.2(Co,Fe)0.8Nb0.2O4+δ (K-PSCFN)	850	C3H8	K-PSCFN–CFA|LSGM|P-PSCFN	PPD ∼ 940 mW cm^−2; negligible degradation during 26 cycles of redox testing	325
	La0.5Ba0.5Mn0.8Fe0.1Co0.1O3−δ	850	Wet CH4	CoFe–LBMF|LDC–LSGM|BSCF	PPD ∼ 503 mW cm^−2; stable power output for 200 h	326
	La0.3Sr0.7Cr0.3Fe0.6Co0.1O3−δ (LSCrFeCo10)	850	C3H8/H2O	LSCrFeCo10–GDC|LDC–LSGM|LSCF–GDC	PPD ∼ 700 mW cm^−2	327
	(Pr0.4Sr0.6)0.3(Fe0.85Mo0.15)2O7 (DLP-PSFM)	750	C2H6		Stable operation for 100 h under 650 mA cm^−2	328
Fe–Ni alloys	Sr2FeMo0.65Ni0.35O6−δ (SFMNi)	850	Wet CH4	SFMNi–LDC|LSGM|LSCF	PPD ∼ 500 mW cm^−2, 24 h of stability under 200 mA cm^−2	329
	(La0.7Sr0.3)(Cr0.85Ni0.1125Fe0.0375)O3−δ (LSCNi-Fe)	850	5000 ppm H2S-syngas	LSCNi-Fe|YSZ|YSZ–LSM/YSZ	PPD ∼ 400 mW cm^−2 and a 1300 mA cm^−2 maximum current density	330
	La0.5Sr0.5Fe0.8Ni0.2O3−δ (LSFN)	725	Wet syngas	LSFN–SDC|ScSZ|SDC–LSFCNb	Stable operation for 20 h at 0.7 V	331
Fe–Pb alloys	La0.8Sr0.2Fe0.9Nb0.1Pd0.04O3−δ (LSFNP)	800	Dry C3H8	LSFNP|SDC|NBCCo–SDC	25 h operation under 80 mA cm^−2	332
Fe–Re alloys	LaRexFe0.8O-La3ReO8 (x < 0.2) (LRF)	860	CH4/CO2		No significant carbon deposition after 70 h operation	333
Fe–Cu alloys	(La0.75Sr0.25)0.9(Cr0.5Mn0.5)0.9(Cu1−xFex)0.1O3−δ	850	Dry CH4	Cu0.5Fe0.5–LSCM|LSGM |LSM	PPD ∼ 40 mW cm^−2, negligible degradation after 100 h operation and 10 redox cycles	334
Ni–Ce alloys	La0.3Sr0.6Ce0.1Ni0.1Ti0.9O3−δ (LSCNT)	900	CH4	LSCNT–YSZ|YSZ|LSM–YSZ	PPD ∼ 600 mW cm^−2, no degradation for 80 h under 0.5 V	49
Ni–Cu alloys	(La0.75Sr0.25)0.9(Cr0.5Mn0.5)0.9(Ni1−xCux)0.1O3−δ (LSCM-Ni1−xCux)	800	CH4/CO2	Symmetric cell	300 h stability at 0.9 V	335

---

Alternatively, Neagu et al.³¹³ proposed that the high stability of perovskite–metal heterosystems resulted from a combination of size effects and formation of robust interfaces, in which for a specific La_0.52Sr_0.28Ni_0.06Ti_0.94O₃ system with in situ exsolved Ni metal (Fig. 22a), after operation in a 20% CH₄/H₂ environment at 800 °C for 4 h, there was nearly no carbon fiber on the exsolved Ni particles with sizes of 25–60 nm, while carbon growth was found on exsolved Ni particles of ∼80 nm in size. For comparison, there was obvious carbon fiber accumulation on La_0.4Sr_0.4TiO₃ anodes infiltrated by 20 nm Ni particles. TEM micrographs indicated that exsolved nanoparticles were embedded at a considerable depth in the parent perovskite oxide support (Fig. 22b), resulting in a robust interface between the perovskite oxide and the exsolved metal particles. The TEM micrographs and electron diffraction revealed that the exsolved metallic particles could grow epitaxially from the parent perovskite oxide in a specific orientation (perovskite(111), Ni(111)), forming a well-ordered structure across the interface with good stability (Fig. 22c and d). In addition, the cation diffusion across the interface between the metal particles and the parent oxide greatly increases the adhesion of the metal particles to the oxide phases. The strong bonding between the exsolved metal particles and the parent perovskite oxide plays a vital role in preventing particle uplifting and subsequent carbon fiber growth, resulting in high coking resistance and stable catalytic activity in a carbon-containing atmosphere.

Fig. 22 (a) Schematic of the possible mechanism of B-site exsolution in a perovskite-structured La_0.52Sr_0.28Ni_0.06Ti_0.94O₃ model system. (b) HRTEM image of the interfacial region between the perovskite matrix and exsolved Ni nanoparticles. (c) HRTEM micrographs at the perovskite–metal interface, demonstrating the epitaxic atomic planes and orientations. (d) Atomic arrangement at the Ni–La_0.52Sr_0.28Ni_0.06Ti_0.94O₃ interface according to structure characterization. Reproduced with permission from ref. 313, copyright 2015 Macmillan Publishers Limited.

The driving force for B-site cation exsolution was also investigated by several groups.^{4,310,336–339} According to Irvine et al.,^4,310,313 the exsolution of B-site cations is dynamically driven through a combination of chemical and electrochemical reduction and generally initiates at surface defect sites. Specifically, in lattices experiencing reducing conditions, the reduction of B-site transition metals and the formation of surface defects (i.e. oxygen vacancies) become more favorable, collectively leading to metal nucleation at the surface and the lowering of nucleation barriers through crystal defects.³⁰ More recently, Han et al.³³⁶ successfully controlled B-site exsolution in La_0.2Sr_0.7Ni_0.1Ti_0.9O_3−δ (LSNT) thin films through lattice strain control, resulting in a high degree of Ni exsolution with more than 1100 particles μm⁻² and tiny particle sizes of ∼5 nm in diameter. Here, thermodynamic analysis indicated that the small atomic diffusion coefficients of the exsoluted Ni under both tensile and compressive strain were much lower compared to conventionally deposited Ni, indicating a small thermodynamic potential of particle coarsening and agglomeration. Furthermore, the researchers suggested that due to the small size of the exsoluted Ni and strong interactions between the embedded Ni and perovskites, carbon fibers grew in “base growth” without particle growth. The key factor was deduced to be the mismatch-strain relaxation energy (ΔG_mr), which can reduce nuclei sizes and nucleation barriers and simultaneously quicken nucleation rates, leading to smaller exsolution particles with higher densities.

Alternatively, Kwon et al.^337,338 explained B-site exsolution in terms of thermodynamics, in which, based on density functional simulations, the researchers proposed that in a PrBa(Mn,T)₂O_5+δ (T = Fe, Co and/or Ni) double-perovskite model system, substituted cations such as Co²⁺ and Ni²⁺ possess lower exsolution energies than the host cation Mn²⁺ (−0.55 and −0.50 eV vs. −0.47 eV) and demonstrate more favorable thermodynamic trends. In addition, Co/Ni doped PrBa(Mn,T)₂O_5+δ exhibits a lower surface oxygen vacancy-formation energy than undoped samples (2.46 and 2.85 eV vs. 2.97 eV), which can also assist in stabilizing the self-grown metallic phase.

6. Summary, challenges and possible research directions

SOCs are promising and environmentally friendly devices that can achieve efficient conversion between electricity and chemical energy. Tremendous efforts involving the understanding and control of heterointerfaces have been undertaken to improve the performance of SOC materials at intermediate temperatures. For heterostructured electrolytes, significantly enhanced ionic conductivities are attributed to lattice strain, dislocation and space-charge effects. Due to the presence of lattice strain, cation segregation, modified electronic structures and expanded TBPs, many heterostructured oxygen electrodes were reported to show outstanding ORR/OER activities. Decorating suitable secondary oxide decoration on the surface presents an effective approach to improve both the reaction activity and long-term stability of oxygen electrodes. For fuel electrodes, different types of heterostructure, particularly oxide skeletons with exsolved metal nanoparticles, have been reported to exhibit good fuel oxidation activity and high resistance to sulfur poisoning and carbon coking. Despite many successful studies about improving the performance of SOC components through heterointerface engineering, there are still many challenges to overcome, which require further systematic investigation.

First, precise control of the composition and microstructure of heterointerfaces is necessary for the rational design of novel electrode or electrolyte materials via interface engineering. It has been reported that the atomic structure of the heterointerface such as local strain and dislocations can profoundly impact the ionic conductivity and/or the electrocatalytic activity of the heterostructure. The use of thin film techniques such as PLD, ALD and MBE presents unique advantages in controlling the atomic structure of the heterointerfaces, which makes them idea tools for synthesis of novel heterostructured SOC components. The challenges that need to be overcome are the high price of the equipment and the slow growth rates of these thin film techniques.

Secondly, advanced characterization techniques, particular the ones with in situ measurement capability, are needed to further reveal the role of heterointerfaces in determining the ionic conductivity and electrocatalytic activity of the heterostructure. Especially, it is necessary to probe how the interface evolves under the working conditions of SOCs, such as high temperature, a reactive gas environment and electrical bias. There have been several successful attempts to use in situ STM, XPS, and SIMS to investigate the evolution of the electronic structure, cation valence and ion distribution at the interfacial region at elevated temperature or under bias.^11,56,233

Thirdly, although several factors have been proposed to explain the impact of heterointerfaces, an overall theoretical model that describes the role of heterointerfaces in determining the transport properties and catalytic activity of SOC components is still lacking. Advanced theoretical modelling and simulation need to be carried out to be combined with experimental results to uncover the effects of heterointerfaces, hence providing guidance for the rational design of novel materials.

Finally, besides improving the performance of each component of SOCs, system level optimization is also necessary. The compatibility of SOC materials with other parts of the system such as interconnection materials needs to be tested. The solid oxide cell and stack design and the thermal management of the whole system are also critical for the long stability of devices.

While SOC components are taken as examples in this review to introduce the influence of heterointerfaces on the conductivity, electrochemical activity and stability, the mechanism of the performance improvement and methodology for controllable fabrication of interfaces are applicable to other energy conversion systems for a wide range of applications such as photo/electrochemical water splitting, solar thermal CO₂ reduction, and lithium ion batteries.

Nomenclature and acronyms

ALD	Atomic layer deposition
ASR	Area surface resistance
BSCF	Ba1−xSrxCo1−yFeyO3−δ
BZCY	BaZr0.4Ce0.4Y0.2O3
BZO	BaZrO3
BZY	BaZrxY1−xO3−y
CFA	Co–Fe bi-metallic alloy
COBRA	Coherent Bragg rod analysis
CSZ	ZrO2 + CaO
DFT	Density functional theory
DP-BSCFW	Double perovskite Ba0.5Sr0.5(Co0.7Fe0.3)0.687W0.312O3−δ
EC	Electron conduction
EELS	Electron energy loss spectroscopy
EPD	Electrophoretic deposition
HAXPES	Hard X-ray photoelectron spectroscopy
HRXRD	High resolution X-ray diffraction
HRTEM	High resolution transmission electron microscope
HTPC	High-temperature proton conductor
IC	Ionic conduction
LAO	LaAlO3
LCP	LaCePrOx
LNO	LaNiO3±δ
LSC113	La1−xSrxCoO3±δ
LSC214	La2−xSrxCoO4±δ
LSCF	La1−xSrxCoyFe1−yO3±δ
LSCM	La1−xSrxCoyMn1−yO3±δ
LSCMC	(La0.75Sr0.25)0.9(Cr0.5Mn0.5)0.9Cu0.1O3−δ
LSCMN	(La0.75Sr0.25)0.9(Cr0.5Mn0.5)0.9Ni0.1O3−δ
LSGM	LaxSr1−xGayMg1−yO3−δ
LSF	La1−xSrxFeO3±δ
LSM	La1−xSrxMnO3±δ
LSZG	Li13.9Sr0.1Zn(GeO4)4
MBE	Molecular beam epitaxy
MIEC	Mixed ionic and electronic conductor
ML	Multilayer
NEXAFS	Near edge X-ray absorption fine structure
NGO	NdGaO3
NSC113	Nd1−xSrxCoO3±δ
OER	Oxygen evolution reaction
ORR	Oxygen reduction reaction
PLD	Pulsed laser deposition
PPD	Peak power density
PSCM	PrSrCoMnO6−δ
RP	Ruddlesden–Popper
RP-PSFN	RP type layered (Pr0.4Sr0.6)3(Fe0.85Nb0.15)2O7
SCDC	Sm/Ca co-doped ceria
SCFN	SrCo0.2Fe0.6Ni0.2O3
SDC	Samarium doped ceria
SIMS	Secondary-ion mass spectrometry
SOC	Solid oxide cell
SOEC	Solid oxide electrolysis cell
SOFC	Solid oxide fuel cell
SP-BSCFW	Single perovskite Ba0.5Sr0.5(Co0.7Fe0.3)0.6875W0.3125O3−δ
SSC–GDC	Sm0.5Sr0.5CoO3–Gd0.1Ce0.9O2−δ
STEM	Scanning transmission electron microscope
STF	SrTi0.2Fe0.8O3
STM/STS	Scanning tunnelling microscopy/spectroscopy
STO	SrTiO3
TEM	Transmission electron microscope
TOF-SIMS	Time of flight secondary ion mass spectroscopy
TPB/3PB	Triple phase boundaries
VAN	Vertical aligned nanocomposite
XANES	X-ray absorption near-edge structure
XAS	X-ray absorption spectroscopy
XPS	X-ray photoemission spectroscopy
YBCO	(Y,Ca)Ba2Cu3O7
YSZ	Yttira-stabilized zirconia

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (No. 91645126, 21273128, 11975102 and 21677162), the ‘Tsinghua University Initiative Scientific Research Program’ (2018Z05JZY010), the ‘Program for Changjiang Scholars and Innovative Research Team in University’ (IRT13026), Guangdong Innovative and Entrepreneurial Research Team Program (No. 2014ZT05N200), Phillips 66 Research Center, and the ‘Tsinghua-MIT-Cambridge’ Low Carbon Energy University Alliance Seed Fund Program (201LC004). All of the above financial support is gratefully acknowledged.

References

1. S. Park, J. M. Vohs and R. J. Gorte, Nature, 2000, 404, 265–267.
2. Z. Shao, S. M. Haile, J. Ahn, P. D. Ronney, Z. Zhan and S. A. Barnett, Nature, 2005, 435, 795–798.
3. E. D. Wachsman and K. T. Lee, Science, 2011, 334, 935–939.
4. J. T. S. Irvine, D. Neagu, M. C. Verbraeken, C. Chatzichristodoulou, C. Graves and M. B. Mogensen, Nat. Energy, 2016, 1, 15014.
5. Y. Zheng, J. Wang, B. Yu, W. Zhang, J. Chen, J. Qiao and J. Zhang, Chem. Soc. Rev., 2017, 46, 1427–1463.
6. Z. P. Cano, D. Banham, S. Ye, A. Hintennach, J. Lu, M. Fowler and Z. Chen, Nat. Energy, 2018, 3, 279–289.
7. J. Myung, D. Neagu, D. N. Miller and J. T. S. Irvine, Nature, 2016, 537, 528–531.
8. D. Burnat, R. Kontic, L. Holzer, P. Steiger, D. Ferri and A. Heel, J. Mater. Chem. A, 2016, 4, 11939–11948.
9. Y. Zhang, R. Knibbe, J. Sunarso, Y. Zhong, W. Zhou, Z. Shao and Z. Zhu, Adv. Mater., 2017, 29, 1700132.
10. S. H. Jensen, C. Graves, M. Mogensen, C. Wendel, R. Braun, G. Hughes, Z. Gao and S. A. Barnett, Energy Environ. Sci., 2015, 8, 2471–2479.
11. Y. Zheng, W. Zhang, Y. Li, J. Chen, B. Yu, J. Wang, L. Zhang and J. Zhang, Nano Energy, 2017, 40, 512–539.
12. Y. Chen, S. Yoo, Y. Choi, J. H. Kim, Y. Ding, K. Pei, R. Murphy, Y. Zhang, B. Zhao, W. Zhang, H. Chen, Y. Chen, W. Yuan, C. Yang and M. Liu, Energy Environ. Sci., 2018, 11, 2458–2466.
13. Y. Zhu, X. Liu, S. Jin, H. Chen, W. Lee, M. Liu and Y. Chen, J. Mater. Chem. A, 2019, 7, 5875–5897.
14. A. M. Saranya, A. Morata, D. Pla, M. Burriel, F. Chiabrera, I. Garbayo, A. Hornés, J. A. Kilner and A. Tarancón, Chem. Mater., 2018, 3, 5621–5629.
15. M. A. Laguna-Bercero, H. Monzón, A. Larrea and V. M. Orera, J. Mater. Chem. A, 2016, 4, 1446–1453.
16. J. T. S. Irvine and S. Tao, Nat. Mater., 2003, 2, 320–323.
17. D. M. Bastidas, S. Tao and J. T. S. Irvine, J. Mater. Chem., 2006, 16, 1603–1605.
18. N. Ai, N. Li, S. He, Y. Cheng, M. Saunders, K. Chen, T. Zhang and S. P. Jiang, J. Mater. Chem. A, 2017, 5, 12149–12157.
19. T. Li, M. F. Rabuni, L. Kleiminger, B. Wang, G. H. Kelsall, U. W. Hartley and K. Li, Energy Environ. Sci., 2016, 9, 3682–3686.
20. C. C. Duan, D. Hook, Y. C. Chen, J. H. Tong and R. O'Hayre, Energy Environ. Sci., 2017, 10, 176–182.
21. W. Zhang, B. Yu and J. Xu, Int. J. Hydrogen Energy, 2012, 37, 837–842.
22. B. Yu, W. Zhang, J. Xu, J. Chen, X. Luo and K. Stephan, Int. J. Hydrogen Energy, 2012, 37, 12074–12080.
23. L. Chen, F. Chen and C. Xia, Energy Environ. Sci., 2014, 7, 4018–4022.
24. Y. Li, Y. Li, Y. Wan, Y. Xie, J. Zhu, H. Pan, X. Zheng and C. Xia, Adv. Energy Mater., 2019, 9, 1803156.
25. J. Tallgren, O. Himanen and M. Noponen, ECS Trans., 2017, 78, 3103–3111.
26. M. Kramer, I. H. Stairs, R. N. Manchester, M. A. McLaughlin, A. G. Lyne, R. D. Ferdman, M. Burgay, D. R. Lorimer, A. Possenti, N. D’Amico, J. M. Sarkissian, G. B. Hobbs, J. E. Reynolds, P. C. C. Freire and F. Camilo, Science, 2006, 314, 97–102.
27. T. Hibino, A. Hashimoto, T. Inoue, J. Tokuno, S. Yoshida and M. Sano, Science, 2000, 288, 2031–2033.
28. Y. Zhou, H. Wu, T. Luo, J. Wang, Y. Shi, C. Xia, S. Wang and Z. Zhan, Adv. Energy Mater., 2015, 5, 1500375.
29. Z. Gao, L. V. Mogni, E. C. Miller, J. G. Railsback and S. A. Barnett, Energy Environ. Sci., 2016, 9, 162–1644.
30. Y. Li, W. Zhang, Y. Zheng, J. Chen, B. Yu, Y. Chen and M. Liu, Chem. Soc. Rev., 2017, 46, 6345–6378.
31. J. Druce, H. Téllez, M. Burriel, M. D. Sharp, L. J. Fawcett, S. N. Cook, D. S. Mcphail, T. Ishihara, H. H. Brongersma and J. A. Kilner, Energy Environ. Sci., 2014, 7, 3593–3599.
32. J. Fleig, Annu. Rev. Mater. Res., 2003, 33, 361–382.
33. F. Santoni, D. M. Silva Mosqueda, D. Pumiglia, E. Viceconti, B. Conti, C. Boigues Muñoz, B. Bosio, S. Ulgiati and S. J. McPhail, J. Power Sources, 2017, 370, 36–44.
34. D. Heidari, S. Javadpour and S. H. Chan, Energy Convers. Manage., 2017, 136, 78–84.
35. C. Zhang and K. Huang, J. Power Sources, 2017, 342, 419–426.
36. N. Sata, K. Eberman, K. Eberl and J. Maier, Nature, 2000, 408, 946–949.
37. P. Boldrin, E. Ruiz-Trejo, J. Mermelstein, J. M. Bermúdez Menéndez, T. Ramírez Reina and N. P. Brandon, Chem. Rev., 2016, 116, 13633–13684.
38. S. Zhang, H. Wang, M. Y. Lu, A. Zhang, L. V. Mogni, Q. Liu, C. Li, C. Li and S. A. Barnett, Energy Environ. Sci., 2018, 11, 1870–1879.
39. Y. Lin, S. Fang, D. Su, K. S. Brinkman and F. Chen, Nat. Commun., 2015, 6, 6824.
40. Y. Tong, Y. Guo, P. Chen, H. Liu, M. Zhang, L. Zhang, W. Yan, W. Chu, C. Wu and Y. Xie, Chem., 2017, 3, 812–821.
41. M. Först, A. D. Caviglia, R. Scherwitzl, R. Mankowsky, P. Zubko, V. Khanna, H. Bromberger, S. B. Wilkins, Y. D. Chuang, W. S. Lee, W. F. Schlotter, J. J. Turner, G. L. Dakovski, M. P. Minitti, J. Robinson, S. R. Clark, D. Jaksch, J. M. Triscone, J. P. Hill, S. S. Dhesi and A. Cavalleri, Nat. Mater., 2015, 14, 883–888.
42. R. Yoshimi, A. Tsukazaki, K. Kikutake, J. G. Checkelsky, K. S. Takahashi, M. Kawasaki and Y. Tokura, Nat. Mater., 2014, 13, 253–257.
43. I. Kosacki, C. Rouleau, P. Becher, J. Bentley and D. Lowndes, Solid State Ionics, 2005, 176, 1319–1326.
44. Z. Gao, V. Y. Zenou, D. Kennouche, L. Marks and S. A. Barnett, J. Mater. Chem. A, 2015, 3, 9955–9964.
45. Y. Ling, L. Chen, B. Lin, W. Yu, T. T. Isimjan, L. Zhao and X. Liu, RSC Adv., 2015, 5, 17000–17006.
46. Q. Su, D. Yoon, A. Chen, F. Khatkhatay, A. Manthiram and H. Wang, J. Power Sources, 2013, 242, 455–463.
47. M. Sase, F. Hermes, K. Yashiro, K. Sato, J. Mizusaki, T. Kawada, N. Sakai and H. Yokokawa, J. Electrochem. Soc., 2008, 155, B793–B797.
48. E. J. Crumlin, E. Mutoro, S. Ahn, G. J. aO, D. N. Leonard, A. Borisevich, M. D. Biegalski, H. M. Christen and Y. Shao-Horn, J. Phys. Chem. Lett., 2010, 1, 3149–3155.
49. Y. Sun, X. Zhou, Y. Zeng, B. S. Amirkhiz, M. Wang, L. Zhang, B. Hua, J. Li, J. Li and J. Luo, J. Mater. Chem. A, 2015, 3, 22830–22838.
50. Z. Cai, Y. Kuru, J. W. Han, Y. Chen and B. Yildiz, J. Am. Chem. Soc., 2011, 133, 17696–17704.
51. J. W. Han and B. Yildiz, Energy Environ. Sci., 2012, 5, 8598–8607.
52. C. Leon, J. Santamaria and B. A. Boukamp, MRS Bull., 2013, 38, 1056–1063.
53. A. Kushima and B. Yildiz, J. Mater. Chem., 2010, 20, 4809–4819.
54. B. Yildiz, MRS Bull., 2014, 39, 147–156.
55. K. Wen, W. Lv and W. He, J. Mater. Chem. A, 2015, 3, 20031–20050.
56. Y. Chen, Z. Cai, Y. Kuru, W. Ma, H. L. Tuller and B. Yildiz, Adv. Energy Mater., 2013, 3, 1221–1229.
57. Z. Feng, Y. Yacoby, M. J. Gadre, Y. Lee, W. T. Hong, H. Zhou, M. D. Biegalski, H. M. Christen, S. B. Adler, D. Morgan and Y. Shao-Horn, J. Phys. Chem. Lett., 2014, 5, 1027–1034.
58. G. J. aO, R. F. Savinell and Y. Shao-Horn, J. Electrochem. Soc., 2009, 156, B771–B781.
59. X. Li, M. Liu, S. Y. Lai, D. Ding, M. Gong, J. Lee, K. S. Blinn, Y. Bu, Z. Wang, L. A. Bottomley, F. M. Alamgir and M. Liu, Chem. Mater., 2015, 27, 822–828.
60. Z. Tao, G. Hou, N. Xu, Q. Zhang and H. Ding, Electrochim. Acta, 2014, 150, 55–61.
61. Y. Chen, S. Huang, X. Ji, K. Adepalli, K. Yin, X. Ling, X. Wang, J. Xue, M. Dresselhaus, J. Kong and B. Yildiz, ACS Nano, 2018, 12, 2569–2579.
62. J. Garcia-Barriocanal, F. Y. Bruno, A. Rivera-Calzada, Z. Sefrioui, N. M. Nemes, M. Garcia-Hernández, J. Rubio-Zuazo, G. R. Castro, M. Varela, S. J. Pennycook, C. Leon and J. Santamaria, Adv. Mater., 2010, 5, 627–632.
63. D. O. Scanlon, C. W. Dunnill, J. Buckeridge, S. A. Shevlin, A. J. Logsdail, S. M. Woodley, C. R. A. Catlow, M. J. Powell, R. G. Palgrave, I. P. Parkin, G. W. Watson, T. W. Keal, P. Sherwood, A. Walsh and A. A. Sokol, Nat. Mater., 2013, 12, 798–801.
64. X. Feng, L. Feng, M. Jin, J. Zhai and D. Zhu, J. Am. Chem. Soc., 2004, 126, 62–63.
65. H. S. Lim, D. Kwak, D. Y. Lee, S. G. Lee and K. Cho, J. Am. Chem. Soc., 2007, 129, 4128–4129.
66. S. Singh, J. T. Haraldsen, J. Xiong, E. M. Choi, P. Lu, D. Yi, X. Wen, J. Liu, H. Wang, Z. Bi, P. Yu, M. R. Fitzsimmons, J. L. MacManus-Driscoll, R. Ramesh, A. V. Balatsky, J.-X. Zhu and Q. X. Jia, Phys. Rev. Lett., 2014, 113, 047204.
67. C. Ran, X. Jiang, N. Qian and B. Arne, Phys. Rev. Lett., 2014, 113, 057601.
68. M. Li, Q. Song, W. Zhao, J. A. Garlow, T. Liu, L. Wu, Y. Zhu, J. S. Moodera, M. H. W. Chan, G. Chen and C. Chang, Phys. Rev. B, 2017, 96, 201301.
69. M. Liu and L. Jiang, Sci. China Mater., 2016, 59, 239–246.
70. M. Joachim, J. Phys. Chem. Solids, 1985, 46, 309–320.
71. X. X. Guo, I. Matei, J. S. Lee and J. Maier, Appl. Phys. Lett., 2007, 91, 103102.
72. J. Garcia-Barriocanal, A. Rivera-Calzada, M. Varela, Z. Sefrioui, E. Iborra, C. Leon, S. J. Pennycook and J. Santamaria, Science, 2008, 321, 676–680.
73. A. Ohtomo, Nature, 2002, 419, 378–380.
74. A. Ohtomo and H. Y. Hwang, Nature, 2004, 441, 423.
75. N. Nakagawa, H. Y. Hwang and D. A. Muller, Nat. Mater., 2012, 5, 204–209.
76. L. S. Mahmud, A. Muchtar and M. R. Somalu, Renewable Sustainable Energy Rev., 2017, 72, 105–116.
77. W. Ma, J. J. Kim, N. Tsvetkov, T. Daio, Y. Kuru, Z. Cai, Y. Chen, K. Sasaki, H. L. Tuller and B. Yildiz, J. Mater. Chem. A, 2015, 3, 207–219.
78. M. G. Blamire, H. Yang, R. Yu, J. L. MacManus-Driscoll, H. Wang, Q. Jia, P. Zerrer, A. Fouchet and J. Yoon, Nat. Mater., 2008, 7, 314–320.
79. D. Lee, Y. L. Lee, A. Grimaud, W. T. Hong, M. D. Biegalski, D. Morgan and Y. Shao-Horn, J. Phys. Chem. C, 2014, 118, 14326–14334.
80. Y. Chen, D. D. Fong, F. W. Herbert, J. Rault, J. P. Rueff, T. Nikolai and B. Yildiz, Chem. Mater., 2018, 30, 3359–3371.
81. A. Evans, A. Bieberle-Hutter, J. L. M. Rupp and L. J. Gauckler, J. Power Sources, 2009, 194, 119–129.
82. D. Kuo, C. J. Eom, J. K. Kawasaki, G. Petretto, J. N. Nelson, G. Hautier, E. J. Crumlin, K. M. Shen, D. G. Schlom and J. Suntivich, J. Phys. Chem. C, 2018, 122, 4359–4364.
83. K. H. L. Zhang, Y. Du, P. V. Sushko, M. E. Bowden, V. Shutthanandan, S. Sallis, L. F. J. Piper and S. A. Chambers, Phys. Rev. B: Condens. Matter Mater. Phys., 2015, 91, 155129.
84. H. Kim, H.-B.-R. Lee and W. J. Maeng, Thin Solid Films, 2009, 517, 2563–2580.
85. N. Cheng, Y. Shao, J. Liu and X. Sun, Nano Energy, 2016, 29, 220–242.
86. Y. Chen, J. Wang, X. Meng, Y. Zhong, R. Li, X. Sun, S. Ye and S. Knights, Int. J. Hydrogen Energy, 2011, 36, 11092–11805.
87. G. Y. Cho, Y. Kim, S. W. Hong, W. Yu, Y. Kim and S. W. Cha, Nanotechnology, 2018, 29, 345401.
88. P. Su, C. Chao, J. H. Shim, R. Fasching and F. B. Prinz, Nano Lett., 2008, 8, 2289–2292.
89. L. Besra and M. Liu, Prog. Mater. Sci., 2007, 52, 1–61.
90. A. J. Samson, M. Sogaard, P. Hjalmarsson, J. Hjelm, N. Bonanos, S. P. V. Foghmoes and T. Ramos, Fuel Cells, 2013, 13, 511–519.
91. Y. Xiang, S. Lu and S. P. Jiang, Chem. Soc. Rev., 2012, 41, 7291–7321.
92. Y. Chen, Q. Liu, Z. Yang, F. Chen and M. Han, RSC Adv., 2012, 2, 12118–12121.
93. Y. Choi, S. Choi, H. Y. Jeong, M. Liu, B. Kim and G. Kim, ACS Appl. Mater. Interfaces, 2014, 6, 17352–17357.
94. T. E. Burye and J. D. Nicholas, J. Power Sources, 2015, 276, 54–61.
95. Y. Chen, Y. Lin, Y. Zhang, S. Wang, D. Su, Z. Yang, M. Han and F. Chen, Nano Energy, 2014, 8, 25–33.
96. Y. Chen, Y. Zhang, G. Xiao, Z. Yang, M. Han and F. Chen, ChemElectroChem, 2015, 2, 672–678.
97. Y. Chen, Y. Zhang, J. Baker, P. Majumdar, Z. Yang, M. Han and F. Chen, ACS Appl. Mater. Interfaces, 2014, 6, 5130–5136.
98. R. Yan, Y. Chen, Y. Lin and F. Chen, Electrochim. Acta, 2016, 217, 187–194.
99. D. Dang, B. Zhao, D. Chen, M. D. Ben, C. Qu, S. Dai, X. Zeng, J. Liu, Y. Lu, S. Liao and M. Liu, Energy Storage Mater., 2018, 12, 79–84.
100. J. Piao, K. Sun, N. Zhang and S. Xu, J. Power Sources, 2008, 175, 288–295.
101. I. Garbayo, F. Baiutti, A. Morata and A. Tarancón, J. Eur. Ceram. Soc., 2019, 39, 101–114.
102. Z. Huang, Ariando, W. X. Renshaw, A. Rusydi, J. Chen, H. Yang and T. Venkatesan, Adv. Mater., 2018, 30, 1802439.
103. Z. Yang and J. Hao, J. Mater. Chem. C, 2016, 4, 8859–8878.
104. F. Baiutti, G. Christiani and G. Logvenov, Beilstein J. Nanotechnol., 2014, 5, 596–602.
105. Z. Guo, H. Li, Q. Chen, L. Sang, L. Yang, Z. Liu and X. Wang, Chem. Mater., 2015, 27, 5988–5996.
106. C. Zhao, X. Liu, W. Zhang, Y. Zheng, Y. Li, B. Yu, J. Wang and J. Chen, Int. J. Hydrogen Energy, 2019, 44, 19102–19112.
107. Y. Zheng, Y. Li, T. Wu, W. Zhang, J. Zhu, Z. Li, J. Chen, B. Yu, J. Wang and J. Zhang, Nano Energy, 2018, 51, 711–720.
108. N. Schichtel, C. Korte, D. Hesse, N. Zakharov, B. Butz, D. Gerthsen and J. Janek, Phys. Chem. Chem. Phys., 2010, 12, 14596.
109. M. Brahlek, A. S. Gupta, J. Lapano, J. Roth, H. T. Zhang, L. Zhang, R. Haislmaier and R. Engel-Herbert, Adv. Funct. Mater., 2017, 28, 1702772.
110. X. Yuan, L. Tang, S. Liu, P. Wang, Z. Chen, C. Zhang, Y. Liu, W. Wang, Y. Zou and C. Liu, Nano Lett., 2015, 15, 3571.
111. R. Zhao and X. Wang, Chem. Mater., 2018, 31, 445–453.
112. X. Meng, X. Wang, D. Geng, C. Ozgit-Akgun, N. Schneider and J. W. Elam, Mater. Horiz., 2017, 4, 133–154.
113. J. Wang, Z. Guo, X. Wei and X. Wang, Chem. – Eur. J., 2018 DOI:10.1002/chem.201803327.
114. Y. Yang, M. Li, Y. Ren, Y. Li and C. Xia, Int. J. Hydrogen Energy, 2018, 43, 3797–3802.
115. M. Nadeem, B. Hu and C. Xia, Int. J. Hydrogen Energy, 2018, 43, 8079–8087.
116. T. Hong, K. Brinkman and C. Xia, J. Power Sources, 2016, 329, 281–289.
117. T. Wu, W. Zhang, Y. Li, Y. Zheng, B. Yu, J. Chen and X. Sun, Adv. Energy Mater., 2018, 8, 1802203.
118. D. Ding, X. Li, S. Y. Lai, K. Gerdes and M. Liu, Energy Environ. Sci., 2014, 7, 552–575.
119. F. Liang, J. Chen, S. P. Jiang, B. Chi, J. Pu and L. Jian, Electrochem. Commun., 2009, 11, 1048–1051.
120. K. Yashiro, T. Takamura, M. Sase, F. Hermes, K. Sato, T. Kawada and J. Mizusaki, ECS Trans., 2007, 7, 1287–1292.
121. M. Tsuchiya, B. Lai and S. Ramanathan, Nat. Nanotechnol., 2011, 6, 282–286.
122. S. M. Haile, D. A. Boysen, C. R. I. Chisholm and R. B. Merle, Nature, 2001, 910–913.
123. V. Kharton, F. Marques and A. Atkinson, Solid State Ionics, 2004, 174, 135–149.
124. T. Ishihara, Bull. Chem. Soc. Jpn., 2006, 79, 1155–1166.
125. D. Pergolesi, E. Fabbri, S. N. Cook, V. Roddatis, E. Traversa and J. A. Kilner, ACS Nano, 2012, 6, 10524–10534.
126. I. Kosacki, C. M. Rouleau, P. F. Becher, J. Bentley and D. H. Lowndes, Electrochem. Solid-State Lett., 2004, 12, A459–A461.
127. L. Fan, C. Wang, M. Chen and B. Zhu, J. Power Sources, 2013, 234, 154–174.
128. K. Lee, K. Ahn, Y. Chung, J. Lee and H. Yoo, Solid State Ionics, 2012, 229, 45–53.
129. Y. Wu and S. Chen, Int. J. Hydrogen Energy, 2017, 42, 27284–27297.
130. F. Bozza, R. Polini and E. Traversa, Electrochem. Commun., 2009, 11, 1680–1683.
131. N. M. Sammes, G. A. Tomsett, H. Nafe and F. Aldinger, J. Eur. Ceram. Soc., 1999, 19, 1801–1826.
132. Y. Mishima, H. Mitsuyasu, M. Ohtaki and K. Eguchi, J. Electrochem. Soc., 1998, 145, 1004–1007.
133. J. B. Goodenough, Annu. Rev. Mater. Res., 2003, 33, 91–128.
134. N. P. Brandon, S. Skinner and B. C. H. Steele, Annu. Rev. Mater. Res., 2003, 331, 183–213.
135. H. Iwahara, H. Uchida, K. Ono and K. Ogaki, J. Electrochem. Soc., 1988, 135, 369–375.
136. H. Iwahara, T. Esaka, H. Uchida and N. Maeda, Solid State Ionics, 1981, 3, 359–363.
137. Z. Zhu, E. Guo, Z. Wei and H. Wang, J. Power Sources, 2018, 373, 132–138.
138. K. Xie, R. Yan, X. Chen, S. Wang, Y. Jiang, X. Liu and G. Meng, J. Alloys Compd., 2009, 473, 323–329.
139. E. Fabbri, D. Pergolesi and E. Traversa, Chem. Soc. Rev., 2010, 39, 4355.
140. S. Sun and Z. Cheng, J. Electrochem. Soc., 2017, 10, F3104–F3113.
141. C. Duan, J. Tong, M. Shang, S. Nikodemski, M. Sanders, S. Ricote, A. Almansoori and R. OHayre, Science, 2015, 349, 1321–1326.
142. L. Bi, S. Boulfrad and E. Travers, Chem. Soc. Rev., 2014, 43, 8255–8270.
143. J. Kim, A. Jun, O. Gwon, S. Yoo, M. Liu, J. Shin, T. Lim and G. Kim, Nano Energy, 2018, 44, 121–126.
144. V. Foglietti, N. Yang, A. Tebano, C. Aruta, E. Di Bartolomeo, S. Licoccia, C. Cantoni and G. Balestrino, Appl. Phys. Lett., 2014, 104, 81612.
145. Y. Shi, I. Garbayo, P. Muralt and J. L. M. Rupp, J. Mater. Chem. A, 2017, 5, 3900–3908.
146. A. Fluri, D. Pergolesi, V. Roddatis, A. Wokaun and T. Lippert, Nat. Commun., 2016, 7, 10692.
147. D. S. Aidhy, B. Liu, Y. Zhang and W. J. Weber, J. Phys. Chem. C, 2014, 118, 30139–30144.
148. L. Chen, C. L. Chen, X. Chen, W. Donner, S. W. Liu, Y. Lin, D. X. Huang and A. J. Jacobson, Appl. Phys. Lett., 2003, 83, 4737–4739.
149. W. Shen, J. Jiang and J. L. Hertz, J. Phys. Chem. C, 2014, 118, 22904–22912.
150. A. Peters, C. Korte, D. Hesse, N. Zakharov and J. Janek, Solid State Ionics, 2007, 178, 67–76.
151. W. L. Cheah and M. W. Finnis, J. Mater. Sci., 2012, 47, 1631–1640.
152. S. Lee, W. Zhang, F. Khatkhatay, H. Wang, Q. Jia and J. L. MacManus-Driscoll, Nano Lett., 2015, 15, 7362–7369.
153. S. M. Yang, S. Lee, J. Jian, W. Zhang, P. Lu, Q. Jia, H. Wang, T. W. Noh, S. V. Kalinin and J. L. MacManus-Driscll, Nat. Commun., 2015, 6, 8588.
154. X. B. Liao, Y. Ni, H. Yang and L. H. He, Appl. Phys. Lett., 2013, 103, 141903.
155. T. He, J. Feng, J. Ru, Y. Feng, R. Lian and J. Yang, ACS Nano, 2019, 13, 830–838.
156. C. Xia, Y. Li, Y. Tian, Q. Liu, Z. Wang, L. Jia, Y. Zhao and Y. Li, J. Power Sources, 2010, 195, 3149–3154.
157. R. A. De Souza, A. Ramadan and S. Hörner, Energy Environ. Sci., 2012, 5, 5445–5453.
158. P. P. Dholabhai, G. Pilania, J. A. Aguiar, A. Misra and B. P. Uberuaga, Nat. Commun., 2014, 5, 5043.
159. M. Sillassen, P. Eklund, N. Pryds, E. Johnson, U. Helmersson and J. Bøttiger, Adv. Funct. Mater., 2010, 20, 2071–2076.
160. H. Lee, N. Campbell, J. Lee, T. J. Asel, T. R. Paudel, H. Zhou, J. W. Lee, B. Noesges, J. Seo, B. Park, L. J. Brillson, S. H. Oh, E. Y. Tsymbal, M. S. Rzchowski and C. B. Eom, Nat. Mater., 2018, 17, 231–236.
161. D. S. Mebane and R. A. De Souza, Energy Environ. Sci., 2015, 8, 2935–2940.
162. R. A. De Souza and E. C. Dickey, Philos. Trans. R. Soc., A, 2019, 377, 20180430.
163. K. M. Kant, V. Esposito and N. Pryds, Appl. Phys. Lett., 2010, 97, 143110.
164. J. Jiang, X. Hu, W. Shen, C. Ni and J. L. Hertz, Appl. Phys. Lett., 2013, 102, 143901.
165. Y. Shi, A. H. Bork, S. Schweiger and J. L. M. Rupp, Nat. Mater., 2015, 14, 721–728.
166. G. F. Harrington, A. Cavallaro, D. W. Mccomb, S. J. Skinner and J. A. Kilner, Phys. Chem. Chem. Phys., 2017, 19, 14319–14336.
167. C. Korte, J. Keppner, A. Peters, N. Schichtel, H. Aydin and J. Janek, Phys. Chem. Chem. Phys., 2014, 16, 24575–24591.
168. W. Shen and J. L. Hertz, J. Mater. Chem. A, 2015, 3, 2378–2386.
169. S. Azad, O. A. Marina, C. M. Wang, L. Saraf, V. Shutthanandan, D. E. McCready, A. El-Azab, J. E. Jaffe, M. H. Engelhard, C. H. F. Peden and S. Thevuthasan, Appl. Phys. Lett., 2005, 86, 131906.
170. C. Korte, A. Peters, J. Janek, D. Hesse and N. Zakharov, Phys. Chem. Chem. Phys., 2008, 10, 4623.
171. X. Guo, Science, 2009, 324, 465.
172. A. Cavallaro, M. Burriel, J. Roqueta and E. Al, Solid State Ionics, 2010, 181, 592–601.
173. S. Sanna, V. Esposito, A. Tebano, S. Licoccia, E. Traversa and G. Balestrino, Small, 2010, 6, 1863–1867.
174. A. Cavallaro, M. Burriel, J. Roqueta, A. Apostolidis, A. Bernardi, A. Tarancón, R. Srinivasan, S. N. Cook, H. L. Fraser, J. A. Kilner, D. W. McComb and J. Santisoa, Solid State Ionics, 2010, 181, 592–601.
175. R. A. De Souza and A. H. H. Ramadan, Phys. Chem. Chem. Phys., 2013, 15, 4505–4509.
176. B. Li, J. Zhang, T. Kaspar, V. Shutthanandan, R. C. Ewing and J. Lian, Phys. Chem. Chem. Phys., 2013, 15, 1296–1301.
177. H. Aydin, C. Korte, M. Rohnke and J. Janek, Phys. Chem. Chem. Phys., 2013, 15, 1944–1955.
178. E. Ruiz-Trejo, K. Thyden, N. Bonanos and M. B. Mogensen, Solid State Ionics, 2016, 288, 82–87.
179. W. Lv, N. Feng, Y. Niu, F. Yang, K. Wen, M. Zou, Y. Han, J. Zhao and W. He, Solid State Ionics, 2016, 289, 168–172.
180. J. M. Polfus, T. Norby and R. Bredesen, Solid State Ionics, 2016, 297, 77–81.
181. J. Yang, M. Youssef and B. Yildiz, Phys. Chem. Chem. Phys., 2017, 19, 3869–3883.
182. N. Yang, C. Cantoni, V. Foglietti, A. Tebano, A. Belianinov, E. Strelcov, S. Jesse, D. Di Castro, E. Di Bartolomeo, S. Licoccia, S. V. Kalinin, G. Balestrino and C. Aruta, Nano Lett., 2015, 15, 2343–2349.
183. E. Fabbri, D. Pergolesi and T. Enrico, Sci. Technol. Adv. Mater., 2010, 11, 54503.
184. C. Korte, N. Schichtel, D. Hesse and J. Janek, Monatsh. Chem., 2009, 140, 1069–1080.
185. M. Zou, F. Yang, K. Wen, W. Lv, M. Waqas and W. He, Int. J. Hydrogen Energy, 2016, 41, 22254–22259.
186. J. L. M. Rupp, E. Fabbri, D. Marrocchelli, J. Han, D. Chen, E. Traversa, H. L. Tuller and B. Yildiz, Adv. Funct. Mater., 2014, 24, 1562–1574.
187. H. Jalili, J. W. Han, Y. Kuru, Z. Cai and B. Yildiz, J. Phys. Chem. Lett., 2011, 2, 801–807.
188. N. Schichtel, C. Korte, D. Hesse and J. Janek, Phys. Chem. Chem. Phys., 2009, 11, 3043–3048.
189. T. Nakamura, K. Yashiro, K. Sato and J. Mizusaki, Phys. Chem. Chem. Phys., 2009, 11, 3055.
190. T. Petrişor, A. Meledin, A. Boulle, R. B. Moş, M. S. Gabor, L. Ciontea and T. Petrişor, Appl. Surf. Sci., 2018, 433, 668–673.
191. S. Krzysztof, S. Wolfgang, B. Gustav and W. Rainer, Nat. Mater., 2006, 5, 312.
192. E. Navickas, Y. Chen, Q. Lu, W. Wallisch, T. M. Huber, J. Bernardi, M. Stöger-Pollach, G. Friedbacher, H. Hutter, B. Yildiz and J. Fleig, ACS Nano, 2018, 11, 11475–11487.
193. C. Chang, M. Chu, H. Jeng, S. Cheng, J. Lin, J. Yang and C. Chen, Nat. Commun., 2014, 5, 3522–3530.
194. M. J. Demkowicz, A. Misra and A. Caro, Curr. Opin. Solid State Mater. Sci., 2012, 16, 101–108.
195. L. Marc, D. Gerhard, A. Eduard and B. T. John, Science, 2008, 319, 1646–1649.
196. L. Sun, D. Marrocchelli and B. Yildiz, Nat. Commun., 2015, 6, 6294.
197. D. Marrocchelli, L. Sun and B. Yildiz, J. Am. Chem. Soc., 2015, 137, 4735–4748.
198. T. Yeh, R. Lin and J. Cherng, Thin Solid Films, 2015, 574, 66–70.
199. A. Ohtomo and H. Hwang, Nature, 2004, 427, 423–426.
200. E. Dagotto, Science, 2007, 318, 1076–1077.
201. J. Maier, Prog. Solid State Chem., 1995, 23, 171–263.
202. G. Gregori, R. Merkle and J. Maier, Prog. Mater. Sci., 2017, 89, 252–305.
203. M. Shirpour, B. Rahmati, W. Sigle, P. A. V. Aken, R. Merkle and J. Maier, J. Phys. Chem. C, 2012, 116, 2453–2461.
204. D. Lee, S. H. Baek, T. H. Kim, J. G. Yoon, C. M. Folkman, C. B. Eom and T. W. Noh, Phys. Rev. B, 2011, 84, 125305.
205. A. Karthikeyan, C. L. Chang and S. Ramanathan, Appl. Phys. Lett., 2006, 89, 1123–1127.
206. W. J. Bowman, J. Zhu, R. Sharma and P. A. Crozier, Solid State Ionics, 2015, 272, 9–17.
207. M. C. Göbel, G. Gregori and J. Maier, J. Phys. Chem. C, 2013, 117, 22560–22568.
208. H. Li, F. L. Souza and R. H. R. Castro, J. Eur. Ceram. Soc., 2018, 38, 1750–1759.
209. J. M. Polfus, T. Kazuaki, O. Fumiyasu, T. Isao and H. Reidar, Phys. Chem. Chem. Phys., 2012, 14, 12339–12346.
210. X. Guo, W. Sigle and J. Maier, J. Am. Ceram. Soc., 2010, 86, 77–87.
211. K. K. Adepalli, J. Yang, J. Maier, H. L. Tuller and B. Yildiz, Adv. Funct. Mater., 2017, 27, 1700243.
212. P. Lupetin, G. Gregori and J. Maier, Angew. Chem., Int. Ed., 2010, 49, 10123–10126.
213. A. Tschöpe, E. Sommer and R. Birringer, Solid State Ionics, 2001, 139, 267–280.
214. W. Tong, W. Zhang, B. Yu and J. Chen, Int. J. Hydrogen Energy, 2017, 42, 29900–29910.
215. S. B. Adler, Chem. Rev., 2004, 104, 4791–4844.
216. T. Wu, B. Yu, W. Zhang, J. Chen and S. Zhao, RSC Adv., 2016, 6, 68379–68387.
217. S. Liu, W. Zhang, Y. Li and B. Yu, RSC Adv., 2017, 7, 16332–16340.
218. S. Liu, B. Yu, W. Zhang, Y. Zhai and J. Chen, Int. J. Hydrogen Energy, 2016, 41, 15952–15959.
219. S. J. Kim, A. M. Dayaghi, K. J. Kim and G. M. Choi, J. Power Sources, 2017, 344, 218–222.
220. L. Luo, Y. Lin, J. Shi, L. Cheng, Y. Wu and L. Sun, J. Inorg. Mater., 2016, 7, 756–760.
221. J. Miao, H. Y. Tian, X. Y. Zhou, K. H. Pang and Y. Wang, J. Appl. Phys., 2007, 101, 84101.
222. E. J. Crumlin, S. Ahn, D. Lee, E. Mutoro, M. D. Biegalski, H. M. Christen and Y. Shao-Horn, J. Electrochem. Soc., 2012, 159, F219–F225.
223. Y. Zheng, Y. Li, T. Wu, C. Zhao, W. Zhang, J. Zhu, Z. Li, J. Chen, J. Wang, B. Yu and J. Zhang, Nano Energy, 2019, 62, 521–529.
224. K. Świerczek, Solid State Ionics, 2011, 192, 486–490.
225. W. Zhang, H. Wang, K. Guan, Z. Wei, X. Zhang, J. Meng, X. Liu and J. Meng, ACS Appl. Mater. Interfaces, 2019, 11, 26830–26841.
226. E. Mutoro, E. J. Crumlin, M. D. Biegalski, H. M. Christen and Y. Shao-Horn, Energy Environ. Sci., 2011, 4, 3689–3696.
227. Y. Chen, W. Jung, Z. Cai, J. J. Kim, H. L. Tulle and B. Yildiz, Energy Environ. Sci., 2012, 5, 7979–7988.
228. M. Kubicek, Z. Cai, W. Ma, B. Yildiz, H. Hutter and J. Fleig, ACS Nano, 2013, 7, 3276–3286.
229. N. Tsvetkov, Y. Chen and B. Yildiz, J. Mater. Chem. A, 2014, 2, 14690–14695.
230. D. Lee, Y. L. Lee, W. T. Hong, M. D. Biegalski, D. Morgan and Y. Shao-Horn, J. Mater. Chem. A, 2015, 3, 2144–2157.
231. J. R. Petrie, C. Mitra, H. Jeen, W. S. Choi, T. L. Meyer, F. A. Reboredo, J. W. Freeland, G. Eres and H. N. Lee, Adv. Funct. Mater., 2016, 26, 1564–1570.
232. Y. Fujimaki, K. Mizuno, Y. Kimura, T. Nakamura, K. D. Bagarinao, K. Yamaji, K. Yashiro, T. Kawada, F. Iguchi, H. Yugami and K. Amezawa, ECS Trans., 2017, 78, 847–853.
233. C. Lenser, Q. Lu, E. Crumlin, H. Bluhm and B. Yildiz, J. Phys. Chem. C, 2018, 122, 4841–4848.
234. H. Chen, Z. Guo, L. A. Zhang, Y. Li, F. Li, Y. Zhang, Y. Chen, X. Wang, B. Yu, J. Shi, J. Liu, C. Yang, S. Cheng, Y. Chen and M. Liu, ACS Appl. Mater. Interfaces, 2018, 10, 39785–39793.
235. Y. Kuru, H. Jalili, Z. Cai, B. Yildiz and H. L. Tuller, Adv. Mater., 2011, 23, 4543–4548.
236. N. Tsvetkov, Q. Lu, L. Sun, E. J. Crumlin and B. Yildiz, Nat. Mater., 2016, 15, 1010–1016.
237. M. Sase, K. Sato and K. Yashiro, Electrochem. Solid-State Lett., 2009, 9, B135–B137.
238. C. Xia, Y. Cai, Y. Ma, B. Wang, W. Zhang, M. Karlsson, Y. Wu and B. Zhu, ACS Appl. Mater. Interfaces, 2016, 8, 20748–20755.
239. J. F. Shin, W. Xu, M. Zanella, K. Dawson, S. N. Savvin, J. B. Claridge and M. J. Rosseinsky, Nat. Energy, 2017, 2, 16214.
240. Y. Gong, R. L. Patel, X. Liang, D. Palacio, X. Song, J. B. Goodenough and K. Huang, Chem. Mater., 2013, 25, 4224–4231.
241. Y. Gong, D. Palacio, X. Song, R. L. Patel, X. Liang, X. Zhao, J. B. Goodenough and K. Huang, Nano Lett., 2013, 13, 4340–4345.
242. A. Samreen, S. Saher, S. Ali, H. Seema and A. Qamar, Int. J. Hydrogen Energy, 2019, 44, 6223–6228.
243. E. Kim, H. Jung, K. An, J. Park, J. Lee, I. Hwang, J. Kim, M. Lee, Y. Kwon and J. Hwang, Ceram. Int., 2014, 40, 7817–7822.
244. T. Hong, M. Zhao and K. Brinkman, ACS Appl. Mater. Interfaces, 2017, 9, 8659–8668.
245. X. Zhu, H. Xia, Y. Li and Z. Lü, Mater. Lett., 2015, 161, 549–553.
246. J. Li, S. Wang, Z. Wang, R. Liu, T. Wen and Z. Wen, J. Power Sources, 2009, 194, 625–630.
247. D. Ding, M. Liu, Z. Liu, X. Li, K. Blinn, X. Zhu and M. Liu, Adv. Energy Mater., 2013, 3, 1149–1154.
248. M. E. Lynch, Y. Lei, W. T. Qin, J. J. Choi, M. F. Liu, K. Blinn and M. L. Liu, Energy Environ. Sci., 2011, 4, 2249–2258.
249. X. Zhang, Z. Hui and X. Liu, J. Power Sources, 2014, 269, 412–417.
250. W. Zhou, F. Liang, Z. Shao and Z. Zhu, Sci. Rep., 2012, 2, 327.
251. S. Stammler, R. Merkle, B. Stuhlhofer, G. Logvenov, K. Hahn, P. A. van Aken and J. Maier, Solid State Ionics, 2017, 303, 172–180.
252. J. Wang, T. Yang, Y. Wen, Y. Zhang, C. Sun and K. Huang, J. Power Sources, 2019, 414, 24–30.
253. G. M. Rupp, A. K. Opitz, A. Nenning, A. Limbeck and J. Fleig, Nat. Mater., 2017, 16, 640–645.
254. H. J. Choi, K. Bae, D. Y. Jang, J. W. Kim and J. H. Shim, J. Electrochem. Soc., 2015, 162, F622–F626.
255. A. Karimaghaloo, J. Koo, H. Kang, S. A. Song, J. H. Shim and M. H. Lee, Int. J. Pr. Eng. Man-GT, 2019, 6, 611–628.
256. Y. Chen, Y. Chen, D. Ding, Y. Ding, Y. Choi, L. Zhang, S. Yoo, D. Chen, B. DeGlee, H. Xu, Q. Lu, B. Zhao, G. Vardar, J. Wang, H. Bluhm, E. J. Crumlin, C. Yang, J. Liu, B. Yildiz and M. Liu, Energy Environ. Sci., 2017, 10, 964–971.
257. Y. Chen, S. Yoo, K. Pei, D. Chen, L. Zhang, B. DeGlee, R. Murphy, B. Zhao, Y. Zhang, Y. Chen and M. Liu, Adv. Funct. Mater., 2018, 28, 1704907.
258. Y. Chen, B. DeGlee, Y. Tang, Z. Wang, B. Zhao, Y. Wei, L. Zhang, S. Yoo, K. Pei, J. H. Kim, Y. Ding, P. Hu, F. F. Tao and M. Liu, Nat. Energy, 2018, 3, 1042–1050.
259. J. Chakhalian, J. W. Freeland, H. U. Habermeier, G. Cristiani, G. Khaliullin, M. V. Veenendaal and B. Keimer, Science, 2007, 318, 1114–1117.
260. B. Koo, K. Kim, J. K. Kim, H. Kwon, J. W. Han and W. Jung, Joule, 2018, 2, 1476–1499.
261. F. S. Baumann, J. Fleig, M. Konuma, U. Starke, H. U. Habermeier and J. Maier, J. Electrochem. Soc., 2005, 152, A2074–A2079.
262. W. Lee, J. W. Han, Y. Chen, Z. Cai and B. Yildiz, J. Am. Chem. Soc., 2013, 135, 7909–7925.
263. S. N. Stephen, K. Rainer, V. John and A. B. Dawn, ACS Nano, 2013, 7, 6330–6336.
264. W. C. Chueh, Y. Hao, W. Jung and S. M. Haile, Nat. Mater., 2012, 11, 155–161.
265. J. W. Han and B. Yildiz, J. Mater. Chem., 2011, 21, 11899–18983.
266. B. Koo, H. Kwon, Y. Kim, H. G. Seo, J. W. Han and W. Jung, Energy Environ. Sci., 2018, 11, 71–77.
267. T. Mayeshiba and D. Morgan, Solid State Ionics, 2017, 311, 105–117.
268. X. Liu, L. Zhang, Y. Zheng, Z. Guo, Y. Zhu, H. Chen, F. Li, P. Liu, B. Yu, X. Wang, J. Liu, Y. Chen and M. Liu, Adv. Sci., 2019, 1801898,  DOI:10.1002/advs.201801898.
269. F. Li, Y. Li, H. Chen, H. Li, Y. Zheng, Y. Zhang, B. Yu, X. Wang, J. Liu, C. Yang, Y. Chen and M. Liu, ACS Appl. Mater. Interfaces, 2018, 10, 36926–36932.
270. Y. A. Mastrikov, R. Merkle, E. Heifets, E. A. Kotomin and J. Maier, J. Phys. Chem. C, 2010, 114, 3017–3027.
271. J. Hwang, Z. Feng, N. Charles, X. R. Wang, D. Lee, K. A. Stoerzinger, S. Muy, R. R. Rao, D. Lee, R. Jacobs, D. Morgan and Y. Shao-Horn, Mater. Today, 2019 10.1039/C4RA11973H.
272. D. Pesquera, G. Herranz, A. Barla, E. Pellegrin, F. Bondino, E. Magnano, F. Sánchez and J. Fontcuberta, Nat. Commun., 2012, 3, 1189.
273. J. M. Rondinelli and N. A. Spaldin, Adv. Mater., 2011, 23, 3363–3381.
274. W. C. Jung and H. L. Tuller, Energy Environ. Sci., 2012, 5, 5370–5378.
275. Y. Chen, H. Téllez, M. Burriel, F. Yang, N. Tsvetkov, Z. Cai, D. W. McComb, J. A. Kilner and B. Yildiz, Chem. Mater., 2015, 27, 5436–5450.
276. S. He, K. F. Chen, M. Saunders, Z. Quadir, S. W. Tao, J. T. S. Irvine., C. Q. Cui and S. P. Jiang, Solid State Ionics, 2018, 325, 176–188.
277. Y. Li, W. Zhang, T. Wu, Y. Zheng, J. Chen, B. Yu, J. Zhu and M. Liu, Adv. Energy Mater., 2018, 8, 1801893.
278. Y. L. Lee, J. Kleis, J. Rossmeisl, S. H. Yang and D. Morgan, Energy Environ. Sci., 2011, 4, 3966–3970.
279. M. B. Mogensen, C. Chatzichristodoulou, C. Graves, K. V. Hansen, K. K. Hansen, A. Hauch, T. Jacobsen and K. Norrman, ECS Trans., 2016, 72, 93–103.
280. G. M. Rupp, H. Téllez, J. Druce, A. Limbeck, T. Ishihara, J. Kilner and J. Fleig, J. Mater. Chem. A, 2015, 3, 22759–22769.
281. Y. Wen, T. Yang, D. Lee, H. N. Lee, E. J. Crumlin and K. Huang, J. Mater. Chem. A, 2018, 6, 24378–24388.
282. H. Ding, A. Virkar, M. Liu and F. Liu, Phys. Chem. Chem. Phys., 2013, 15, 489–496.
283. D. Ding, M. Liu and M. Liu, ECS Trans., 2013, 57, 1801–1810.
284. J. Suntivich, K. J. May, H. A. Gasteiger, J. B. Goodenough and Y. Shao-Horn, Science, 2011, 9, 1383–1385.
285. P. Bo, C. Gang, W. Tao, J. Zhou, J. Guo, Y. Cheng and W. Kai, J. Power Sources, 2012, 201, 174–178.
286. A. Yamada, Y. Suzuki, K. Saka, M. Uehara, D. Mori, R. Kanno, T. Kiguchi, F. Mauvy and J. C. Grenier, Adv. Mater., 2010, 20, 4124–4128.
287. K. Chen, N. Ai and S. P. Jiang, Int. J. Hydrogen Energy, 2014, 39, 10349–10358.
288. Y. Wan, B. Hu and C. Xia, Electrochim. Acta, 2017, 252, 171–179.
289. Y. Song, W. Wang, L. Ge, X. Xu, Z. Zhang, P. S. B. Julião, W. Zhou and Z. Shao, Adv. Sci., 2017, 4, 1700337.
290. S. Cui, J. Li, X. Zhou, G. Wang, J. Luo, K. T. Chuang, Y. Bai and L. Qiao, J. Mater. Chem. A, 2013, 1, 9689–9696.
291. B. Hua, N. Yan, M. Li, Y. Q. Zhang, Y. F. Sun, J. Li, T. Etsell, P. Sarkar, K. Chuang and J. L. Luo, Energy Environ. Sci., 2016, 9, 207–215.
292. W. Wang, C. Su, Y. Wu, R. Ran and Z. Shao, Chem. Rev., 2013, 113, 8104–8151.
293. L. Yang, Y. Choi, W. Qin, H. Chen, K. Blinn, M. Liu, P. Liu, J. Bai, T. A. Tyson and M. Liu, Nat. Commun., 2011, 2, 357.
294. J. Qu, W. Wang, Y. Chen, X. Deng and Z. Shao, Appl. Energy, 2016, 164, 563–571.
295. A. L. Vincent, A. R. Hanifi, J. Luo, K. T. Chuang, A. R. Sanger, T. H. Etsell and P. Sarkar, J. Power Sources, 2012, 215, 301–306.
296. J. Xu, X. Zhou, X. Dong, L. Pan and K. Sun, Int. J. Hydrogen Energy, 2017, 42, 15632–15640.
297. J. Zhou, T. H. Shin, C. Ni, G. Chen, K. Wu, Y. Cheng and J. T. S. Irvine, Chem. Mater., 2016, 28, 2981–2993.
298. L. Ye, M. Zhang, P. Huang, G. Guo, M. Hong, C. Li, J. T. S. Irvine and K. Xie, Nat. Commun., 2017, 8, 14785.
299. B. Hua, W. Zhang, L. Meng, W. Xin, C. Bo, P. Jian and L. Jian, J. Power Sources, 2014, 247, 170–177.
300. E. Lay, C. Metcalfe and O. Kesler, J. Power Sources, 2012, 218, 237–243.
301. M. Asamoto, S. Miyake, K. Sugihara and H. Yahiro, Electrochem. Commun., 2009, 11, 1508–1511.
302. D. L. Rosa, A. Sin, M. L. Faro, G. Monforte, V. Antonucci and A. S. Aricò, J. Power Sources, 2009, 193, 160–164.
303. S. Islam and J. M. Hill, J. Mater. Chem. A, 2014, 2, 1922–1929.
304. M. D. McIntyre, J. D. Kirtley, A. Singh, S. Islam, J. M. Hill and R. A. Walker, J. Phys. Chem. C, 2015, 119, 7637–7647.
305. A. Singh and J. M. Hill, J. Power Sources, 2012, 214, 185–194.
306. W. Yue, Y. Li, Y. Zheng, T. Wu, C. Zhao, J. Zhao, G. Geng, W. Zhang, J. Chen, J. Zhu and B. Yu, Nano Energy, 2019, 62, 64–78.
307. B. Hua, M. Li, Y. Sun, Y. Zhang, N. Yan, J. Li, T. Etsell, P. Sarkar and J. Luo, Appl. Catal., B, 2017, 200, 174–181.
308. W. Wang, L. Gan, J. P. Lemmon, F. Chen, J. T. S. Irvine and K. Xie, Nat. Commun., 2019, 10, 1550.
309. Y. Li, K. Xie, S. Chen, H. Li, Y. Zhang and Y. Wu, Electrochim. Acta, 2015, 153, 325–333.
310. D. Neagu, G. Tsekouras, D. N. Miller, H. Ménard and J. T. S. Irvine, Nat. Chem., 2013, 5, 916–923.
311. G. Yang, W. Zhou, M. Liu and Z. Shao, ACS Appl. Mater. Interfaces, 2016, 8, 35308–35314.
312. G. Tsekouras, D. Neagu and J. T. S. Irvine, Energy Environ. Sci., 2012, 6, 256–266.
313. D. Neagu, T. Oh, D. N. Miller, H. Ménard, S. M. Bukhari, S. R. Gamble, R. J. Gorte, J. M. Vohs and J. T. S. Irvine, Nat. Commun., 2015, 6, 8120.
314. S. Liu, Q. Liu and J. Luo, ACS Catal., 2016, 6, 6219–6228.
315. H. Wei, K. Xie, J. Zhang, Y. Zhang, Y. Wang, Y. Qin, J. Cui, J. Yan and Y. Wu, Sci. Rep., 2015, 4, 5156.
316. K. Xie, J. Zhang, S. Xu, B. Ding, G. Wu, T. Xie and Y. Wu, Fuel Cells, 2014, 14, 1036–1045.
317. W. Qi, C. Ruan, G. Wu, Y. Zhang, Y. Wang, K. Xie and Y. Wu, Int. J. Hydrogen Energy, 2014, 39, 5485–5496.
318. S. Liu, Q. Liu and J. Luo, J. Mater. Chem. A, 2016, 4, 17521–17528.
319. Y. Song, Y. Yin, L. Li, Z. Chen and Z. Ma, Chem. Commun., 2018, 54, 12341–12344.
320. Y. Sun, J. Li, Y. Zeng, B. S. Amirkhiz, M. Wang, Y. Behnamian and J. Luo, J. Mater. Chem. A, 2015, 3, 11048–11056.
321. Y. Sun, Y. Zhang, J. Chen, J. Li, Y. Zhu, Y. Zeng, B. S. Amirkhiz, J. Li, B. Hua and J. Luo, Nano Lett., 2016, 16, 5303–5309.
322. T. Wei, L. Jia, H. Zheng, B. Chi, J. Pu and J. Li, Appl. Catal., A, 2018, 564, 199–207.
323. N. Zhou, Y. Yin, Z. Chen, Y. Song, J. Yin, D. Zhou and Z. Ma, J. Electrochem. Soc., 2018, 165, F629–F634.
324. C. Yang, J. Li, Y. Lin, J. Liu, F. Chen and M. Liu, Nano Energy, 2015, 11, 704–710.
325. C. Yang, Z. Yang, C. Jin, G. Xiao, F. Chen and M. Han, Adv. Mater., 2012, 24, 1439–1443.
326. N. Hou, T. Yao, P. Li, X. Yao, T. Gan, L. Fan, J. Wang, X. Zhi, Y. Zhao and Y. Li, ACS Appl. Mater. Interfaces, 2019, 11, 6995–7005.
327. K. Lai and A. Manthiram, Chem. Mater., 2018, 30, 2515–2525.
328. S. Liu, K. T. Chuang and J. Luo, ACS Catal., 2016, 6, 760–768.
329. Z. Du, H. Zhao, S. Yi, Q. Xia, Y. Gong, Y. Zhang, X. Cheng, Y. Li, L. Gu and K. Świerczek, ACS Nano, 2016, 10, 8660–8669.
330. Y. Sun, J. Li, L. Cui, B. Hua, S. Cui, J. Li and J. Luo, Nanoscale, 2015, 7, 11173–11181.
331. Z. Wang, Y. Yin, Y. Yu, Y. Song, Z. Ma and J. Yin, Int. J. Hydrogen Energy, 2018, 43, 10440–10447.
332. J. Li, B. Wei, Z. Cao, X. Yue, Y. Zhang and Z. Lü, ChemSusChem, 2018, 11, 254–263.
333. D. Zubenko, S. Singh and B. A. Rosen, Appl. Catal., B, 2017, 209, 711–719.
334. W. Wang, C. Zhu, K. Xie and L. Gan, J. Power Sources, 2018, 406, 1–6.
335. J. Lu, C. Zhu, C. Pan, W. Lin, J. P. Lemmon, F. Chen, C. Li and K. Xie, Sci. Adv., 2018, 4, 5100.
336. H. Han, J. Park, S. Y. Nam, K. J. Kim, G. M. Choi, S. S. P. Parkin, H. M. Jang and J. T. S. Irvine, Nat. Commun., 2019, 10, 1471.
337. O. Kwon, K. Kim, S. Joo, Y. J. Hu and G. Kim, J. Mater. Chem. A, 2018, 6, 15947–15953.
338. O. Kwon, S. Sengodan, K. Kim, G. Kim, H. Y. Jeong, J. Shin, Y. W. Ju, J. W. Han and G. Kim, Nat. Commun., 2017, 8, 15967.
339. B. H. Park and G. M. Choi, Solid State Ionics, 2014, 262, 345–348.

---

Footnote

† These two authors contributed equally to this article.

---