Breaking the limits of Ruddlesden–Popper cathodes to achieve a game-changer for proton-conducting solid oxide fuel cells

Abstract

Ruddlesden–Popper (R–P) structured oxides are promising cathode materials for proton-conducting solid oxide fuel cells (H-SOFCs) due to their excellent thermal compatibility and chemical stability. However, the performance of R–P cathodes has not yet matched that of the widely studied perovskite cathodes, making the enhancement of R–P cathode performance critical for advancing H-SOFC technology. In this study, we introduce a high-entropy R–P oxide, La_0.4Pr_0.4Nd_0.4Ba_0.4Sr_0.4NiO_4+x (LPNBSN), synthesized using an entropy engineering strategy. Compared to conventional R–P oxides, LPNBSN demonstrates significant improvements in oxygen reduction reaction (ORR) activity, interstitial oxygen formation, and proton migration, thereby enhancing its performance as a cathode material for H-SOFCs. The LPNBSN-based fuel cell achieves a record-high peak power density of 2790 mW cm⁻² at 700 °C, surpassing previous R–P oxide cathode performances. Additionally, the high-entropy design induces favorable changes in the coordination environment and electronic state, which suppresses the formation of secondary phases during long-term high-temperature operation—an issue common in conventional R–P oxides—ensuring stable performance under operating conditions. The combination of exceptional power output and long-term stability makes LPNBSN a highly promising cathode material, revitalizing the potential of R–P oxides in H-SOFCs.

---

Broader context

Modern energy and environmental challenges necessitate sustainable solutions. Proton-conducting solid oxide fuel cells (H-SOFCs), similar to conventional SOFCs but operating at lower temperatures (600–700 °C), offer promise. However, this temperature reduction often increases cathode polarization resistance, hindering performance and making cathode development critical. Over the past decade, research on H-SOFC cathodes has primarily focused on perovskite materials, which have significantly enhanced cell performance. However, issues such as thermal expansion mismatches, chemical instability, and elemental segregation hinder their long-term reliability, prompting the search for alternative cathode materials. One promising alternative is Ruddlesden–Popper (R–P) structured oxides, which offer high catalytic activity, good stability, and suitability for intermediate-temperature operation. Despite their advantages, R–P cathodes currently underperform compared to perovskite counterparts. In this study, we introduce a novel high-entropy R–P oxide, La_0.4Pr_0.4Nd_0.4Ba_0.4Sr_0.4NiO_4+x (LPNBSN), as an H-SOFC cathode. This material not only achieves record-high performance among R–P cathodes for H-SOFCs but also rivals leading perovskite cathodes. Furthermore, LPNBSN effectively mitigates the Sr and Ni evolution issues commonly observed in conventional R–P oxides, significantly improving the fuel cell's operational stability. Given its high catalytic activity and enhanced stability, LPNBSN holds promise for broader applications in energy conversion technologies.

Introduction

The global energy and environmental challenges demand the development of sustainable materials and technologies.¹ Fuel cells, which directly convert fuel energy into electricity, have garnered significant attention in recent years.² Among the various types of fuel cells, solid oxide fuel cells (SOFCs) are among the most widely studied due to their all-solid-state architecture and flexibility in fuel selection.^3,4 Additionally, SOFCs operate at high temperatures, enabling the use of oxides as electrodes and eliminating the need for expensive noble metals as electrode catalysts.⁵ These advantages make SOFCs particularly attractive in the fuel cell community.⁶ However, the high operating temperature of traditional SOFCs—typically exceeding 800 °C—presents several challenges, including reduced cell lifespan, difficulties in sealing, and issues with interfacial diffusion.⁷ As a result, reducing the operating temperature of SOFCs has become a critical focus in the field of SOFC development.^8,9

Proton-conducting SOFCs (H-SOFCs), which use proton-conducting electrolytes, have emerged as a promising solution to reduce the operating temperature of SOFCs.^10,11 Compared to the oxygen-ion conducting electrolytes used in traditional SOFCs, proton-conducting electrolytes exhibit lower activation energies and higher conductivities at intermediate temperatures, making them well-suited for operation at lower temperatures.^12,13 Additionally, the by-product H₂O, generated during the fuel cell reaction, is produced at the cathode side, effectively avoiding the fuel dilution problem that arises in oxygen-ion conducting SOFCs (O-SOFCs).¹⁴ Despite the advantages of proton-conducting electrolytes, a significant challenge remains: cathode kinetics become sluggish as the operating temperature is reduced, significantly impeding fuel cell performance.¹⁵ Furthermore, the cathode reaction mechanism in H-SOFCs differs from that in O-SOFCs, involving proton participation, which necessitates the development of specialized cathodes.¹⁶ Therefore, designing cathodes optimized for H-SOFCs is crucial to overcoming these performance limitations.¹⁷

Many new cathode materials for H-SOFCs have been proposed, with perovskite oxides being the most widely investigated.^18,19 High-performing perovskite cathodes have significantly improved the performance of H-SOFCs. However, some perovskite cathodes face challenges such as thermal expansion mismatches, chemical instability, and elemental segregation, which hinder their long-term reliability.^20,21 This has driven the exploration of alternative cathode materials. One promising category of alternatives is Ruddlesden–Popper (R–P) structured oxides, which have gained significant attention due to their high catalytic activity, good stability, and suitability for intermediate temperature operation.²² Initially developed for O-SOFCs because of their superior oxygen-ion migration and electronic conductivity,²³ R–P cathodes have also been shown to facilitate proton migration, making them suitable for H-SOFCs.^24,25 While significant progress has been made in developing R–P cathodes for H-SOFCs, their performance is still much inferior to that of perovskite-based cathodes.²² To enhance the performance of R–P cathodes, compositional tailoring has emerged as the primary strategy.^24,26 Although modifying the metal cations in these oxides can lead to improvements, the performance of these modified R–P cathodes still lags behind that of the best-performing perovskite cathodes for H-SOFCs,²⁷ suggesting that further optimization is necessary.

One promising approach to further enhance cathode performance is entropy engineering. Initially applied to high-entropy alloys, this concept has been extended to oxide materials, where the incorporation of equal or near molar multiple metal cations into a single oxide phase can lead to enhanced catalytic activity due to the “cocktail effect” of multiple elements in a single phase.^28,29 Recent studies on high-entropy perovskite oxides have shown their potential to improve SOFC performance.^30,31 However, most of the existing high-entropy designs in H-SOFCs have focused on perovskite oxides,^32,33 with relatively few studies exploring high-entropy R–P structure cathodes.³⁴ Moreover, the rationale behind the performance improvements in high-entropy materials remains underexplored, with many studies focusing primarily on fuel cell output rather than the operational stability of the fuel cells. In this study, we introduce a high-entropy oxide based on the well-known Ln₂NiO_4+x (Ln = La, Nd, Pr) cathode oxides and investigate how entropy engineering affects the performance of H-SOFCs. Special attention is given to the impact of the high-entropy design on the operational stability of the cathode. Additionally, the underlying mechanisms behind the enhanced cathode performance and stability are explored.

Materials and method

Synthesis of materials

Conventional Sr-doped Ln₂NiO_4+x (Ln = La, Nd, Pr) oxides, including La_1.2Sr_0.8NiO_4+x (LSN), Pr_1.2Sr_0.8NiO_4+x (PSN) and Nd_1.2Sr_0.8NiO_4+x (NSN), were synthesized using a wet chemical route with metal nitrates as precursors. Proper amounts of metal nitrates were mixed in a water solution with the addition of citric acid as the complexing agent. The molar ratio of citric acid to metal cations was set at 1.5. The pH value of the solution was adjusted to around 8, followed by the heating of the solution to evaporate water. The solution became viscous and finally ignited, producing powders. The powders were calcined at 1100 °C for 6 hours to obtain the target R–P phases. A high-entropy R–P oxide, La_0.4Pr_0.4Nd_0.4Ba_0.4Sr_0.4NiO_4+x (LPNBSN), was synthesized by co-doping La, Pr, Nd, Ba, and Sr at the Ln site in an equal molar ratio. The synthesis conditions were identical, with a calcination temperature of 1100 °C for 6 hours.

Characterization techniques

The phase composition of the as-prepared materials was analyzed by X-ray diffraction (XRD). The microstructure was characterized using transmission electron microscopy (TEM) coupled with energy dispersive X-ray spectroscopy (EDS) for elemental analysis. Thermal expansion behavior was assessed by preparing dense bars of LSN, PSN, NSN, and LPNBSN, which were tested using a Netzsch DIL 402E dilatometer in air, from room temperature to 1000 °C. Electrical conductivity relaxation (ECR) measurements were conducted on the dense bars by switching the atmosphere from dry to wet air (3% H₂O). Changes in conductivity were recorded, and the proton diffusion rate was inferred from the time required to reach a new equilibrium. The X-ray absorption spectroscopy (XAS) at the Ni K-edge was conducted at the BL14W1 beamline of the Shanghai synchrotron radiation facility (SSRF), with data acquisition in transmission mode. The XAS spectra were analyzed using the DEMETER software package. Energy calibration and normalization were carried out using the Athena software, while the k³-weighted Fourier transform of the extended X-ray absorption fine structure (EXAFS) in R-space was fitted using Artemis. Additionally, the χ(k) data from the EXAFS data analysis, exported from Athena, were processed through wavelet transform analysis using the Hama Fortran code.³⁵

Computational methods

First-principles calculations based on density functional theory (DFT) were performed using the Vienna ab initio simulation package (VASP).³⁶ The energy and force convergence criteria were set to 10⁻⁵ eV and 0.05 eV Å⁻¹, respectively, with a cutoff energy of 520 eV. The formation energy of interstitial oxygen was calculated according to , where E_f is the formation energy of the interstitial oxygen atom, the E_bulk–O and E_bulk are energies of the bulk with and without an interstitial oxygen atom, and the E_O2 is the energy of an oxygen molecule. To probe the oxygen reduction reaction (ORR) energies at the surface, the (001) surface was cleaved, and a 15 Å vacuum layer was added. The top two layers were fully relaxed, while the bottom four layers were fixed. The Gibbs free energies for ORR were determined by calculating total energies of the system and correcting for entropy changes (TΔS) at specific temperatures. For proton migration studies, the climbing-image nudged elastic band (CI-NEB) method was employed to calculate the energy barriers for proton movement through the materials.³⁷

Symmetrical cell measurements

The ORR activities of the cathodes were evaluated using symmetrical cell configurations. The BaCe_0.7Zr_0.1Y_0.2O_3−δ (BCZY) proton-conducting electrolyte was synthesized via wet chemical methods and calcined at 1000 °C for 6 h to obtain phase-pure powder. This powder was uniaxially pressed into pellets, with dense BCZY electrolytes subsequently prepared by sintering the green bodies at 1600 °C for 6 h. Symmetrical cells were fabricated by depositing identical cathode layers on both surfaces of the polished BCZY pellets, followed by co-firing at 900 °C for 2 h. Electrochemical impedance spectroscopy (EIS) measurements were performed across various temperatures and atmospheres. Impedance spectra were fitted using Relax IS software, with the cathode area-specific resistance (ASR) calculated as: ASR = R_p/2, where R_p is the total polarization resistance derived from the symmetrical cell measurement, accounting for the dual-electrode configuration.

Fabrication of fuel cells

The BCZY proton-conducting electrolyte was used in the fuel cell fabrication. The anode was prepared by mixing BCZY with NiO in a 2 : 3 weight ratio and pressing the mixture at ∼100 MPa. BCZY electrolyte powder was then dispersed on the top of the pressed anode substrate, followed by co-pressing at ∼150 MPa to form NiO + BCZY/BCZY bi-layers, which were sintered at 1350 °C for 6 h. The cathodes (LSN, PSN, NSN, or LPNBSN) were applied as slurries to the BCZY electrolyte surface and co-fired at 900 °C for 2 h, resulting in complete cells with the structure NiO + BCZY/BCZY/LPNBSN (or LSN, PSN and NSN). The effective area of the cathode was 0.28 cm².

Electrochemical performance testing

The electrochemical performance of the fuel cells was evaluated under typical H-SOFC conditions, using humidified H₂ as the fuel and static air as the oxidant. The impedance spectra were recorded over a frequency range of 1 MHz to 0.1 Hz using a Gamry Interface 5000E electrochemical workstation. The microstructure of the fuel cells was examined using scanning electron microscopy (SEM, Phenom XL, Thermo Scientific). Thermal cycling stability was evaluated by applying a constant voltage to the cell while recording the current during heating and cooling cycles. Long-term stability of the cell was assessed by monitoring the voltage over time under fuel cell operating conditions.

Results and discussion

Structure analysis and thermal expansion properties

Fig. 1(a) shows the XRD patterns of the synthesized LSN, PSN, NSN, and LPNBSN samples. The diffraction peaks of all samples correspond to the characteristic R–P perovskite structure, with no additional peaks indicating the presence of other phases. Slight shifts in the diffraction peaks are observed, which can be attributed to the different ionic radii of La, Pr, Nd, Ba, and Sr, leading to changes in the unit cell volume. To obtain more detailed crystallographic information, XRD Rietveld refinements were performed for LSN, PSN, NSN, and LPNBSN. As shown in Fig. 1(b), the refinements yielded reliable fitting parameters: R_p = 8%, 7%, 8%, and 9%; R_wp = 12%, 12%, 13%, and 11%; and χ² = 1.21, 1.15, 1.33, and 1.24 for LSN, PSN, NSN, and LPNBSN, respectively. Based on the space group I4/mmm and the tetragonal structure, the refined lattice parameters are: a = b = 3.82, 3.75, 3.70, and 3.79 Å, and c = 12.58, 12.49, 12.41, and 12.52 Å for LSN, PSN, NSN, and LPNBSN, respectively. High-resolution TEM (HR-TEM) images in Fig. 1(c) provide further insight into the crystal structure of these materials. The crystal planes are identified from the diffraction spots, with interplanar spacings observed at 2.05 Å, 2.82 Å, and 3.66 Å for the (1̄14̄), (013̄), and (1̄01̄) planes of LSN; 2.79 Å for the (103) plane of PSN; 2.76 Å, 3.55 Å, and 2.00 Å for the (103), (011), and (114) planes of NSN; and 2.81 Å, 3.63 Å, and 2.04 Å for the (103), (011), and (114) planes of LPNBSN. The calculated lattice parameters, based on the formula for spacing between crystal faces in a tetragonal system, are consistent with the XRD refinement results. TEM elemental mapping images in Fig. 1(d) further confirm the homogeneous elemental distribution across all the tested oxides, with no signs of segregation. Collectively, these results indicate the successful preparation of the high-entropy R–P oxide LPNBSN.

Fig. 1 (a) XRD patterns, (b) Rietveld refinements, (c) HR-TEM and (d) elemental mapping for LSN, PSN, NSN and LPNBSN.

The entropy design does not significantly alter the thermal expansion behavior of the R–P oxides. The thermal expansion coefficient (TEC) of LPNBSN is 14.24 × 10⁻⁶ K⁻¹, which is similar to that of LSN, PSN, and NSN, as shown in Fig. S1 (ESI†). This value is also comparable to that of the BCZY proton-conducting electrolyte (11.09 × 10⁻⁶ K⁻¹),³⁸ suggesting that the high-entropy design does not disrupt the compatibility of thermal expansion properties between the traditional R–P cathode oxides and the electrolyte.

High-entropy design of LPNBSN cathodes for enhanced oxygen reduction reaction and proton migration in H-SOFCs

Unlike perovskite oxides, which conduct oxygen-ions and protons via oxygen vacancies, R–P oxides rely on interstitial oxygens for oxygen and proton transport.^39,40 Therefore, the formation of interstitial oxygen is crucial for R–P cathodes, and this process is facilitated by the high-entropy design. Fig. S2 (ESI†) shows the formation energies of interstitial oxygen in LSN, PSN, NSN, and LPNBSN. The formation energies are 0.53, 0.74, 0.46, and 0.27 eV for LSN, PSN, NSN, and LPNBSN, respectively. Compared to conventional LSN, PSN, and NSN oxides, the LPNBSN shows a smaller interstitial oxygen formation energy, indicating a lower energy barrier for the formation of interstitial oxygen due to the high-entropy design. This reduction in the energy barrier potentially enhances oxygen diffusion, proton migration, and overall cathode performance.

The LPNBSN high-entropy oxide also exhibits superior oxygen reduction reaction (ORR) activity compared to the conventional LSN, PSN, and NSN oxides. The overall chemical reaction for fuel cells can be written as: 2H₂ + O₂ → 2H₂O. In contrast to ORR in O-SOFC cathodes, which involves only O₂ adsorption and dissociation at the cathode, H-SOFCs also involve proton combination and H₂O release at the cathode, with intermediates *OOH, *O, and *OH formed,⁴¹ as shown in Fig. 2(a). The Gibbs free energies (ΔG) for each step on the surface of LSN, PSN, NSN, and LPNBSN are calculated, revealing that the formation of *OOH is the potential rate-limiting step for all these cathodes, as it is the only uphill step requiring energy input. The energy barriers for this step are 1.68, 1.79, 1.52 and 0.85 eV for LSN, PSN, NSN, and LPNBSN. Notably, the high-entropy design results in a significantly reduced ORR energy barrier. In contrast, the energy steps for the formation of *O and *OH are downhill for all cathodes, suggesting these steps are thermodynamically favourable. To further explore the improved ORR activity, XAS was used to probe the electronic states of LSN and LPNBSN. Fig. 2(b) presents X-ray absorption near-edge spectroscopy (XANES) data at the K-edge of Ni, revealing that the white line peak for LPNBSN appears at a higher energy than that for LSN. This indicates a higher oxidation state of Ni in LPNBSN. With a higher oxidation state, Ni tends to adopt a low-spin state (t⁶_2ge¹_g), which is favorable for ORR activity according to Shao-Horn theory.⁴² Furthermore, as the oxidation state increases and the number of d-electrons decreases, the electronegativity of the transition metal ions improves.⁴³ This effect shifts the energy of the Ni d-states closer to the O 2p band, raising the O 2p band center relative to the Fermi level. The calculated O-p band centers for LPNBSN and LSN are −1.44 eV and −1.91 eV, respectively, supporting the XAS findings. The O-p band centers for PSN and NSN are calculated to be −2.12 eV and −1.59 eV, respectively, as shown in Fig. S3 (ESI†), with LPNBSN having the highest O-p band center, suggesting the best ORR activity. This trend of calculated ORR activity aligns with the O-p band centers, as shown in Fig. 2(c), in agreement with literature reports.⁴⁴

Fig. 2 (a) Free energies of the ORR procedure. (b) XANES for LSN and LPNBSN. (c) ORR barrier and O p-band center values for LSN, PSN, NSN and LPNBSN. (d) Schemed proton migration routes in R–P oxides. (e) Energy of each step and (f) energy barrier of the neighbouring steps for proton migration in LSN, PSN, NSN and LPNBSN.

One motivation for using R–P cathodes in H-SOFCs is their potential for proton migration, which facilitates the formation of triple-phase boundaries (TPBs) at the cathode.⁴⁵ It is therefore crucial to compare proton migration in high-entropy oxide LPNBSN with that in LSN, PSN, and NSN. Proton migration occurs through a combination of rotational and hopping mechanisms.⁴⁶ The proton rotates to find a favorable orientation for hopping and then jumps to a neighboring oxygen site. This process enables proton movement through the oxide lattice, contributing to proton conductivity. Once the proton defect (OH) is formed, the proton (H) rotates around an oxygen atom, aligning in the most favorable orientation for further migration. The hopping step, which involves the proton moving from one oxygen atom to another, is usually the rate-limiting step for proton migration.⁴⁷Fig. 2(d) illustrates the proton migration pathways, which involve both rotation within the oxygen atoms and hopping between adjacent oxygens. Fig. 2(e) shows the calculated energy for each step in LSN, PSN, NSN, and LPNBSN, revealing significant differences in energies for each step among the oxides. Fig. 2(f) presents the energy barriers between adjacent steps, highlighting that the energy barrier for proton rotation is generally smaller than that for proton hopping, in agreement with the literature. Notably, LPNBSN exhibits smaller energy barriers for proton migration in nearly all steps compared to LSN, PSN, and NSN, suggesting that the high-entropy design greatly facilitates proton transport in LPNBSN. Furthermore, the highest energy barrier, identified as the rate-determining step for LSN, PSN, and NSN, occurs between states 4 and 5, with values of 0.76, 1.01 and 0.65 eV for LSN, PSN, and NSN, respectively. In contrast, the highest energy barrier in LPNBSN is significantly smaller (0.53 eV), indicating superior proton transport performance. Experimental studies further validate these findings. ECR measurements were conducted to assess proton diffusion and surface exchange in LSN, PSN, NSN, and LPNBSN. Upon switching from dry to wet air, the formation of proton defects alters material conductivities, and the time to reach equilibrium reflects proton diffusion and surface exchange rates.⁴⁸ As shown in Fig. S4 (ESI†), LPNBSN exhibits a shorter relaxation time than LSN, PSN, and NSN, suggesting faster proton diffusion and surface exchange. Fitting the ECR curves yields proton diffusion coefficients (D_h) and proton surface exchange coefficients (k_h) that are several times larger for LPNBSN than for the other oxides, as indicated in Table S1 (ESI†), confirming enhanced proton diffusion kinetics in LPNBSN.

Enhanced performance of LPNBSN cathodes in H-SOFCs: a comparative analysis with conventional R–P cathodes

Due to the enhanced ORR activity and proton transport properties, it is reasonable to predict that LPNBSN will outperform conventional R–P cathodes. The ORR activity of the LPNBSN cathode compared with the conventional R–P cathodes is experimentally studied by using the symmetrical cell measurements. Fig. S5 (ESI†) shows the area specific resistance (ASR) of LSN, PSN, NSN, and LPNBSN cathodes obtained from the symmetrical cell measurements. The ASR of LPNBSN is much smaller than that for LSN, PSN and NSN, suggesting improved ORR activity by using the high-entropy design. In addition, we have conducted EIS analysis on symmetrical cells under varying oxygen partial pressures (pO₂), providing mechanistic insights into the ORR kinetics. The LPNBSN symmetrical cell is tested under different oxygen partial pressures. Fig. S6(a) (ESI†) shows that the overall polarization resistance increases with the decreased in the oxygen partial pressure. The EIS plot is fitted with the equivalent circuit, as shown in Fig. S6(b) (ESI†). EIS deconvolution reveals three distinct processes governing ORR kinetics in the cathode. The high-frequency process (R_H), which is independent of pO₂, is regarded as the process for charge transfer and water formation.^49,50 In contrast, both middle-frequency process (R_M) and low-frequency process (R_L) decrease with the increasing pO₂ with a linear correlation of and , as shown in Fig. S6(c) (ESI†), which corresponds to the oxygen reduction and surface diffusion of oxygen species, respectively.⁴⁹ LPNBSN demonstrates reduced resistance for each sub-process (R_H, R_M, R_L) compared to conventional R–P cathodes (Fig. S7, ESI†), proving that the high-entropy effect enhances every critical step in the ORR pathway. The DFT calculations reveal lower proton migration energy barriers in LPNBSN, facilitating rapid proton delivery to the triple-phase boundary (TPB) and efficient water formation, thus reducing R_H. In addition, LPNBSN has a lower ORR energy barrier, smaller formation energy for interstitial oxygen formation and a closer O p-band center to the Fermi level, making the reaction at the middle- and low-frequency ranges faster than the conventional R–P oxides, thus decreasing the R_M and R_L. The high-entropy design in LPNBSN enhances all critical ORR sub-processes—proton transfer, oxygen reduction, and oxygen species diffusion, showing great promise in fuel cell applications. Fig. 3(a) presents the performance of an H-SOFC using LPNBSN. The peak power densities (PPD) of the cell are 1872, 2221, and 2790 mW cm⁻² at 600, 650, and 700 °C, respectively. In comparison, the performance of H-SOFCs utilizing conventional LSN, PSN, and NSN cathodes is also measured, with results shown in Fig. S8 (ESI†). The PPD values for the LSN cell are 332, 605, and 1014 mW cm⁻²; for the PSN cell, they are 284, 487, and 882 mW cm⁻²; and for the NSN cell, 492, 791, and 1234 mW cm⁻² at 600, 650, and 700 °C, respectively. The cell using LPNBSN consistently demonstrates superior performance compared to those with conventional LSN, PSN, and NSN cathodes across the entire temperature range (Fig. 3(b)), indicating that the high-entropy strategy significantly enhances R–P cathode performance for H-SOFCs. To eliminate the potential influence of cathode/electrolyte interfacial reactions on performance, we examined the chemical compatibility between the R–P cathodes and the BCZY electrolyte. LPNBSN (LSN, PSN, NSN) powders were mixed with BCZY electrolyte powder and co-fired at 900 °C for 2 h, replicating the thermal treatment conditions for cathode fabrication. XRD was used to analyze the phase composition of the mixtures before and after the treatments, as shown in Fig. S9 (ESI†). No new phases were observed after thermal treatment, suggesting good chemical compatibility between LPNBSN (LSN, PSN, NSN) and BCZY. This rules out undesirable interfacial reactions as a source of the observed differences in fuel cell performance. Furthermore, cross-sectional images of the tested LSN, PSN, NSN, and LPNBSN cells (Fig. S10, ESI†) reveal similar microstructures across all these cells, indicating that microstructural differences are unlikely to account for performance disparities. Notably, the performance improvements are more pronounced at lower temperatures. For instance, the PPD of the LPNBSN cell at 700 °C is 175%, 216%, and 126% higher than that of LSN, PSN, and NSN cells, respectively. In contrast, at 600 °C, the PPD of the LPNBSN cell is 463%, 558%, and 280% higher than that of the LSN, PSN, and NSN cells (Fig. S11, ESI†). These results underscore the significant benefit of LPNBSN in enhancing low-temperature H-SOFC operation, aligning with the motivation behind developing high-performance H-SOFCs.

Fig. 3 (a) Performance of the fuel cell using LPNBSN cathode. (b) PPD comparison for fuel cells using LSN, PSN, NSN and LPNBSN cathodes. (c) EIS plot of the LPNBSN cell measured at 700 °C. (d) Comparison of R_p, R_o, R_tot and (e) the ratio between R_p and R_o for fuel cells using LSN, PSN, NSN and LPNBSN cathodes measured at 700 °C. (f) Performance comparison between the current LPNBSN cathode with other R–P cathodes reported in the literature. (g) Performance comparison between the fuel cell using the current LPNBSN cathode with most representative perovskite cathodes reported in the past decade. The PPD is measured at 600 °C. BCFZY: BaCo_0.4Fe_0.4Zr_0.1Y_0.1O₃; PBSCF: PrBa_0.5Sr_0.5Co_1.5Fe_0.5O₅, PLD: pulsed laser deposition; BCCY: BaCo_0.7(Ce_0.8Y_0.2)_0.3O₃; PNC: PrNi_0.5Co_0.5O₃; LCCN: La_0.7Ca_0.3Co_0.8Ni_0.2O₃; Ba_0.875FZ: Ba_0.875Fe_0.875Zr_0.125O₃; HE-PBSLCC: Pr_0.2Ba_0.2Sr_0.2La_0.2Ca_0.2CoO₃; LCN91: LaCo_0.9Ni_0.1O₃; BSCFW: (Ba/Sr)(Co/Fe/W)O₃, PBSCF: PrBa_0.5Sr_0.5Co_1.5Fe_0.5O₅; C-BSCF: Ba_0.5Sr_0.5Co_0.8Fe_0.2O₃, H-BSCF: Ba₄Sr₄(Co_0.8Fe_0.2)₄O₁₆. (h) Cycling stability test for the LPNBSN cell. (i) Operational stability for fuel cells using LSN, PSN, NSN and LPNBSN cathodes measured at 600 °C under the fuel cell working condition.

Fig. 3(c) displays the EIS plots for the LPNBSN cell tested at 700 °C, revealing an exceptionally low polarization resistance (R_p) of 0.009 Ω cm²—one order of magnitude lower than that for conventional LSN, PSN, and NSN cathodes, as shown in Fig. S12 (ESI†). To the best of our knowledge, this is the smallest R_p ever reported for R–P cathodes in H-SOFCs. The distribution of relaxation times (DRT) method was employed to separate the anode and cathode contributions in R_p for the LPNBSN cell, which enabled the identification and quantification of specific electrochemical processes. Analysis of the EIS data for the LPNBSN cell at 700 °C, via DRT, as shown in Fig. S13(a) (ESI†), reveals five distinct peaks, each corresponding to a dominant electrochemical process, with the peak area providing a quantitative assessment of the associated reaction process. Specifically, P1 is related to gas diffusion at the anode,⁵¹ P2 and P3 are related to the processes involving oxygen adsorption, dissociation, diffusion, and oxygen reduction at the cathode, P4 is related to hydrogen adsorption and dissociation at the anode, and P5 is related to proton incorporation and diffusion at the cathode/electrolyte interface.^52,53 Crucially, the combined polarization resistance attributed to the anode-related processes (P1 and P4, totally about 31.9%) is significantly less than that associated with the cathode processes, as shown in Fig. S13(b) (ESI†). This indicates that the anode reaction kinetics are not the primary contribution to the overall polarization resistance and are not the rate-limiting steps within the overall cell reaction. Fig. 3(d) shows the values of R_p, ohmic resistance (R_o), and total resistance (R_tot) for the LSN, PSN, NSN, and LPNBSN cells at 700 °C. The R_o values are comparable for all four cells, but the R_p values differ significantly. The LPNBSN cell exhibits the lowest R_p, leading to a substantially lower R_tot than the other cells, which accounts for its superior performance. Fig. 3(e) illustrates the ratio of R_p to R_o. For the LPNBSN cell, the low R_p/R_o ratio indicates that the electrode polarization contributes minimally to the total cell resistance relative to ohmic losses, enabling exceptional power density. In contrast, the R_p/R_o ratios for the LSN, PSN, and NSN cells are 1.02, 1.21, and 0.77, respectively, suggesting that R_p is comparable to R_o in these cells. The similar R_o values across all cells reflect the similar electrolyte and interfacial contact resistances. The significantly reduced R_p/R_o ratio in the LPNBSN cell indicates an accelerated cathode reaction, which minimizes R_p relative to R_o, thereby improving the overall fuel cell performance. Not only does the LPNBSN cathode outperform the LSN, PSN, and NSN cathodes, as demonstrated in this study, but it also surpasses all reported R–P structure cathodes for H-SOFCs, as shown in Fig. 3(f).^26,54–60 Although significant progress has been made in developing R–P structure cathodes for H-SOFCs, their performance still lags behind that of perovskite cathodes. The advent of LPNBSN represents a transformative step in R–P cathode development, significantly boosting H-SOFC performance. Even compared to many recently reported high-performing perovskite cathodes, the LPNBSN cathode demonstrates superior performance. Fig. 3(g) shows the performance of most representative cathodes developed for H-SOFCs over the past decade.^{18,19,30,61–67} Different new perovskite cathode materials have been proposed for H-SOFCs in the past 10 years, and indeed the performance of the H-SOFCs have been greatly boosted. Even numerous works have been carried out and substantial advancements have been achieved in perovskite oxide cathodes, the LPNBSN cathode offers competitive performance, providing renewed potential for R–P cathodes in H-SOFC applications.

Enhanced stability and thermal cycling performance of LPNBSN cathode in H-SOFCs

As H₂O is formed at the cathode side in H-SOFCs, the chemical stability of the LPNBSN material under the steam condition is examined. The LPNBSN powder was treated in an air atmosphere containing 30% H₂O at 600 °C for 10 h, and XRD results shown in Fig. S14 (ESI†) indicate that the phase of LPNBSN remains stable even after the treatment, suggesting good chemical stability of LPNBSN against steam. In addition to the excellent fuel cell performance and good chemical stability, the H-SOFC using the LPNBSN cathode also exhibits outstanding stability under operational conditions. Fig. 3(h) shows the current stability of the cell under a constant voltage of 0.7 V at temperatures between 600 and 700 °C. After 10 cycles, the current remains stable, demonstrating the fuel cell's good thermal cycling ability. The microstructure observation shown in Fig. S15 (ESI†) indicates that a good cathode/electrolyte interfacial condition is still maintained even after the cycling test. Furthermore, the LPNBSN cathode displays excellent operational stability, as shown in Fig. 3(i). No degradation in cell voltage is observed after 150 hours of operation, indicating the good operational stability of the LPNBSN cathode for H-SOFCs. In contrast, fuel cells with conventional LSN, PSN, and NSN cathodes show much smaller voltages, attributed to the inferior performance of these traditional R–P cathodes compared to the high-entropy LPNBSN cathode. While the LPNBSN cell operates stably, cells with LSN, PSN, and NSN cathodes exhibit performance degradation under the same testing conditions, demonstrating the comparative stability advantage afforded by the high-entropy design principle over conventional R–P cathodes. The 150-hour's test is a short-term stability test, which may not be sufficient to demonstrate the potential of LPNBSN as a cathode for practical applications. Therefore, a stability test exceeding 500 hours is further performed for the LPNBSN cell under the fuel cell working condition, and the result is shown in Fig. S16 (ESI†). One can see that the fuel cell utilizing the LPNBSN cathode maintains remarkably stable performance (>500 h) under continuous operating conditions, further demonstrating the good operational stability of the cell.

Mechanisms behind superior long-term performance

The performance degradation for fuel cells with conventional R–P cathodes may stem from phase changes during long-term operation, as it is known that some R–P oxides undergo phase decomposition at high temperatures, negatively impacting their performance.⁶⁸ To explore this, phase decomposition in LSN, PSN, NSN, and LPNBSN oxides was examined. The powders of these materials were heated at 600 °C for 200 hours in air, and XRD was employed to analyze the phase changes before and after the treatment. Fig. 4(a) presents the XRD patterns of these materials before and after long-term thermal treatment. Notably, peaks corresponding to SrCO₃ and NiO appear in the LSN, PSN, and NSN samples after treatment, indicating partial phase decomposition at high temperatures. In contrast, no new peaks are observed for LPNBSN, even after the same thermal treatment, suggesting that LPNBSN exhibits better thermal stability than the conventional R–P oxides. Elemental mapping (Fig. 4(b)) further supports this finding. In the LSN, PSN, and NSN samples, significant accumulation of Sr and Ni elements is observed after the thermal treatment, whereas the elements in LPNBSN remain uniformly distributed, even after the thermal treatment under the same condition. The X-ray photoelectron spectroscopy (XPS) analysis further confirms segregation behavior in the materials after thermal treatment. Fig. S17(a) (ESI†) presents deconvoluted Sr 3d spectra for LSN, PSN, NSN, and LPNBSN. Due to spin–orbit coupling, the Sr 3d level splits into distinct 3d_3/2 and 3d_5/2 components. Spectral fitting assigns lower binding energy peaks to lattice Sr species, while higher binding energy features correspond to surface Sr species.⁶⁹ The surface-to-lattice Sr ratio (quantified by fitted peak areas) serves as a direct indicator of Sr segregation extent.^70,71 Post-treatment analysis reveals substantial increases in surface-to-lattice Sr ratios for LSN, PSN, and NSN (Fig. S17(b), ESI†), indicating significant Sr surface enrichment due to segregation and strontium carbonate formation. Conversely, LPNBSN maintains a nearly constant ratio, demonstrating superior structural stability and segregation resistance. Ni speciation was analyzed via Ni 3p XPS to avoid La 3d_3/2 satellite interference.^72,73 Spectral deconvolution confirms the coexistence of Ni²⁺ and Ni³⁺ states, with Ni²⁺ signals originating from both lattice-incorporated Ni and segregated NiO.⁷⁴ The Ni²⁺/Ni³⁺ peak area ratio, therefore, serves as a semi-quantitative indicator of segregation. As shown in Fig. S18 (ESI†), LSN, PSN, and NSN exhibit significantly increased Ni²⁺/Ni³⁺ ratios after treatment, confirming NiO exsolution. LPNBSN maintains a constant ratio, indicating negligible Ni segregation and preserved structural integrity. These results confirm that SrO and NiO exsolution occur in conventional R–P oxides, while the high-entropy design of LPNBSN effectively suppresses such exsolution. Sr segregation, often reported for SOFC cathodes, including both perovskite and R–P structure cathodes, can impair cathode performance by blocking active sites.⁷⁵ Similarly, NiO, commonly used in the anode, does not contribute to the cathodic reaction and may hinder performance by obstructing the active cathode area. Moreover, the formation of new phases could alter the stoichiometry and catalytic properties of the original R–P cathode, further reducing its performance. The absence of SrCO₃ and NiO in LPNBSN is likely a key factor in its superior long-term operational stability.

Fig. 4 (a) XRD patterns for LSN, PSN, NSN and LPNBSN before and after the thermal treatment. (b) Elemental mapping for LSN, PSN, NSN and LPNBSN after the thermal treatment. (c) Sr surface segregation energy for LSN, PSN, NSN and LPNBSN. (d) Calculated Ni–O bond length in LSN and LPNBSN. (e) EXAFS and (f) WT simulation for Ni–O in LSN and LPNBSN.

The mechanism behind the enhanced thermal stability of LPNBSN is further investigated. To address the Sr segregation issue, the thermodynamic tendency for Sr surface segregation was assessed by calculating the segregation energy according to the equation: , where E_seg is Sr surface segregation energy, E_{(La/Pr/Nd/Ba)–O} is the total energy for the La/Pr/Nd/Ba–O terminated surface, E_Sr–O is the total energy for the Sr–O terminated surface formed by exchanging the Sr–O slab to the surface of the original La/Pr/Nd/Ba–O terminated surface. As shown in Fig. 4(c), the high entropy of LPNBSN shows a positive E_seg value of 11.85 meV, and negative values of −6.57, −12.92 and −9.35 meV are obtained for LSN, PSN, and NSN, respectively. The negative energy value means the Sr segregation reaction tends to happen for LSN, PSN, and NSN from the thermodynamically point of view. The segregated SrO can then react with CO₂ to form SrCO₃, explaining the detection of SrCO₃ in LSN, PSN, and NSN. Comparatively, the positive E_seg value for LPNBSN is relatively high, which means a high energy barrier needs to be overcome when the Sr segregation reaction occurs, preventing the formation of SrCO₃ after thermal treatment. Regarding NiO exsolution, we compared the Ni coordination environment in LPNBSN with the typical R–P oxide LSN through DFT calculations. Fig. 4(d) shows that the Ni–O bond length in LPNBSN is between 1.87 and 1.91 Å, whereas in LSN, it ranges from 1.92 to 1.96 Å. The shorter Ni–O bond length in LPNBSN suggests a stronger Ni–O bond, which is further supported by the electronic structure changes in Ni due to the high-entropy design. A higher oxidation state of Ni in LPNBSN results in stronger electrostatic attraction to the oxygen ligands, enhancing bond strength. Fourier transform EXAFS and wavelet transform (WT) simulations (Fig. 4(e) and (f)) provide additional evidence of the stronger Ni–O coordination in LPNBSN, with a scattering signal at ∼1.88 Å, compared to ∼1.92 Å for LSN. This shorter bond distance implies stronger metal–oxygen bonds in LPNBSN, making it more difficult for the bonds to break and form NiO during thermal treatment. As a result, the LPNBSN not only shows excellent fuel cell output and also exhibits good long-term stability under the working condition, serving as a promising cathode for H-SOFCs.

Conclusions

In this study, we employed an entropy engineering strategy to design a novel high-entropy R–P oxide, LPNBSN, as a cathode material for H-SOFCs. Compared to conventional R–P oxides, LPNBSN demonstrated superior ORR activity and enhanced interstitial oxygen formation. Notably, proton migration in LPNBSN was more favorable than in traditional LSN, PSN, and NSN oxides, significantly enhancing its performance in H-SOFCs. Consequently, the LPNBSN-based H-SOFC achieved an impressive fuel cell output of 2790 mW cm⁻² at 700 °C with a remarkably low polarization resistance (0.009 Ω cm²), outperforming conventional R–P oxide cathodes. This performance represents the highest ever reported for H-SOFCs utilizing R–P cathodes. Furthermore, when compared to widely studied perovskite cathodes, LPNBSN exhibited sufficient competence, thus rejuvenating the potential of R–P oxides in H-SOFC applications. The high-entropy strategy also enhanced the thermal stability of LPNBSN. Unlike conventional LSN, PSN, and NSN R–P oxides, which experience Sr and Ni exsolution during the high-temperature operation, LPNBSN maintained phase stability. DFT and experimental studies revealed that the high Sr-segregation energy in LPNBSN effectively suppresses Sr segregation, a common issue in conventional R–P oxides. Additionally, the high oxidation state of Ni in LPNBSN, facilitated by the entropy design, strengthens the electrostatic interaction with surrounding oxygen ligands, thereby preventing Ni exsolution and contributing to the material's superior stability. As a result, the LPNBSN cathode not only addresses the performance degradation typically observed in conventional R–P oxides but also offers a promising solution for high-performance, long-term stable H-SOFCs.

Author contributions

Yanru Yin: investigation, formal analysis, software, writing – original draft. Hongfang Huang: investigation, validation. Samir Boulfrad: formal analysis. Hailu Dai: investigation. Yueyuan Gu: investigation. Shoufu Yu: validation. Lei Bi: project administration, supervision, writing – review & editing.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting this article have been included as part of the ESI.†

Acknowledgements

This work was supported by the National Natural Science Foundation of China (Grant Number: 52272216), the High-level Talents Program of Hunan and the Startup Funding for Talents at University of South China.

References

1. F. Zhang, T. Liao, D. C. Qi, T. Y. Wang, Y. N. Xu, W. Luo, C. Yan, L. Jiang and Z. Q. Sun, Natl. Sci. Rev., 2024, 11, nwae199.
2. Z. Gao, L. V. Mogni, E. C. Miller, J. G. Railsback and S. A. Barnett, Energy Environ. Sci., 2016, 9, 1602–1644.
3. H. Zhang, Y. C. Zhou, K. Pei, Y. X. Pan, K. Xu, Y. Ding, B. T. Zhao, K. Sasaki, Y. M. Choi, Y. Chen and M. L. Liu, Energy Environ. Sci., 2022, 15, 287–295.
4. C. R. Jiang, J. J. Ma, G. Corre, S. L. Jain and J. T. S. Irvine, Chem. Soc. Rev., 2017, 46, 2889–2912.
5. Y. Zhang, B. Chen, D. Q. Guan, M. G. Xu, R. Ran, M. Ni, W. Zhou, R. O’Hayre and Z. P. Shao, Nature, 2021, 591, 246–251.
6. C. C. Duan, R. J. Kee, H. Y. Zhu, C. Karakaya, Y. C. Chen, S. Ricote, A. Jarry, E. J. Crumlin, D. Hook, R. Braun, N. P. Sullivan and R. O’Hayre, Nature, 2018, 557, 217–222.
7. J. F. Shin, W. Xu, M. Zanella, K. Dawson, S. N. Savvin, J. B. Claridge and M. J. Rosseinsky, Nat. Energy, 2017, 2, 16214.
8. J. A. Kilner and M. Burriel, Annu. Rev. Mater. Res., 2014, 44, 365–393.
9. H. Yu, I. Jeong, S. Jang, D. Kim, H. Im, C. W. Lee, E. D. Wachsman and K. T. Lee, Adv. Mater., 2024, 36, 202306205.
10. M. D. Zou, J. Conrad, B. Sheridan, J. W. Zhang, H. Huang, S. L. Mu, T. Y. Zhou, Z. Y. Zhao, K. S. Brinkman, H. Xiao, F. Peng and J. H. Tong, ACS Energy Lett., 2023, 8, 3545–3551.
11. I. Zvonareva, X. Z. Fu, D. Medvedev and Z. P. Shao, Energy Environ. Sci., 2022, 15, 439–465.
12. Y. Yamazaki, F. Blanc, Y. Okuyama, L. Buannic, J. C. Lucio-Vega, C. P. Grey and S. M. Haile, Nat. Mater., 2013, 12, 647–651.
13. D. Pergolesi, E. Fabbri, A. D’Epifanio, E. Di Bartolomeo, A. Tebano, S. Sanna, S. Licoccia, G. Balestrino and E. Traversa, Nat. Mater., 2010, 9, 846–852.
14. Z. Wang, Y. H. Wang, J. Wang, Y. F. Song, M. J. Robson, A. Seong, M. T. Yang, Z. Q. Zhang, A. Belotti, J. P. Liu, G. Kim, J. Lim, Z. P. Shao and F. Ciucci, Nat. Catal., 2022, 5, 777–787.
15. N. Wang, C. M. Tang, L. Du, R. J. Zhu, L. X. Xing, Z. Q. Song, B. Y. Yuan, L. Zhao, Y. Aoki and S. Y. Ye, Adv. Energy Mater., 2022, 12, 2201882.
16. R. R. Peng, T. Z. Wu, W. Liu, X. Q. Liu and G. Y. Meng, J. Mater. Chem., 2010, 20, 6218–6225.
17. B. Y. Yuan, N. Wang, C. M. Tang, L. Meng, L. Du, Q. W. Su, Y. Aoki and S. Y. Ye, Nano Energy, 2024, 122, 109306.
18. Y. S. Song, Y. B. Chen, W. Wang, C. Zhou, Y. J. Zhong, G. M. Yang, W. Zhou, M. L. Liu and Z. P. Shao, Joule, 2019, 3, 2842–2853.
19. S. Choi, C. J. Kucharczyk, Y. G. Liang, X. H. Zhang, I. Takeuchi, H. I. Ji and S. M. Haile, Nat. Energy, 2018, 3, 202–210.
20. T. Hu, Y. S. Xu, K. Xu, F. Zhu and Y. Chen, SusMat, 2023, 3, 91–101.
21. Y. F. Bu, S. Joo, Y. X. Zhang, Y. F. Wang, D. D. Meng, X. L. Ge and G. Kim, J. Power Sources, 2020, 451, 227812.
22. A. P. Tarutin, J. G. Lyagaeva, D. A. Medvedev, L. Bi and A. A. Yaremchenko, J. Mater. Chem. A, 2021, 9, 154–195.
23. C. N. Munnings, S. J. Skinner, G. Amow, P. S. Whitfield and L. J. Davidson, Solid State Ionics, 2006, 177, 1849–1853.
24. L. F. Zhang, F. Yao, J. L. Meng, W. W. Zhang, H. C. Wang, X. J. Liu, J. Meng and H. J. Zhang, J. Mater. Chem. A, 2019, 7, 18558–18567.
25. P. Zhong, K. Toyoura, L. L. Jiang, L. B. Chen, S. A. Ismail, N. Hatada, T. Norby and D. L. Han, Adv. Energy Mater., 2022, 12, 2200392.
26. X. Y. Li, Z. M. Chen, Y. Yang, D. M. Huan, H. Su, K. Zhu, N. Shi, Z. M. Qi, X. S. Zheng, H. B. Pan, Z. L. Zhan, C. R. Xia, R. R. Peng, S. Q. Wei and Y. L. Lu, Appl. Catal., B, 2022, 316, 121627.
27. F. Liu and C. Duan, Nat. Energy, 2023, 8, 1071–1072.
28. W. H. Shi, H. W. Liu, Z. Z. Li, C. H. Li, J. H. Zhou, Y. F. Yuan, F. Jiang, K. Fu and Y. G. Yao, SusMat, 2022, 2, 186–196.
29. C. Oses, C. Toher and S. Curtarolo, Nat. Rev. Mater., 2020, 5, 295–309.
30. F. He, Y. C. Zhou, T. Hu, Y. S. Xu, M. Y. Hou, F. Zhu, D. L. Liu, H. Zhang, K. Xu, M. L. Liu and Y. Chen, Adv. Mater., 2023, 35, 202209469.
31. Z. Wang, J. W. Li, M. K. Yuan, J. T. Gao, H. R. Hao, A. M. Abdalla, L. L. Xu, Z. Lv and B. Wei, Sustainable Mater. Technol., 2024, 40, e00969.
32. Y. H. Wang, M. J. Robson, A. Manzotti and F. Ciucci, Joule, 2023, 7, 848–854.
33. X. Han, Y. H. Ling, Y. Yang, Y. J. Wu, Y. Gao, B. Wei and Z. Lv, Adv. Funct. Mater., 2023, 33, 202304728.
34. X. L. Li, T. Chen, C. X. Wang, N. Sun, G. J. Zhang, Y. C. Zhou, M. Wang, J. Zhu, L. Xu and S. R. Wang, Adv. Funct. Mater., 2024, 34, 202411216.
35. Z. M. Xia, H. Zhang, K. C. Shen, Y. Q. Qu and Z. Jiang, Phys. B, 2018, 542, 12–19.
36. G. Kresse and J. Furthmuller, Phys. Rev. B: Condens. Matter Mater. Phys., 1996, 54, 11169–11186.
37. G. Henkelman, B. P. Uberuaga and H. Jónsson, J. Chem. Phys., 2000, 113, 9901–9904.
38. Y. R. Yin, D. D. Xiao, S. Wu, E. H. Da'as, Y. Y. Gu and L. Bi, SusMat, 2023, 3, 697–708.
39. A. Grimaud, F. Mauvy, J. M. Bassat, S. Fourcade, L. Rocheron, M. Marrony and J. C. Grenier, J. Electrochem. Soc., 2012, 159, B683–B694.
40. H. C. Tian, W. Y. Li, L. Ma, T. Yang, B. Guan, W. Y. Shi, T. L. Kalapos and X. B. Liu, ACS Appl. Mater. Interfaces, 2020, 12, 49574–49585.
41. A. B. Munoz-Garcia, M. Tuccillo and M. Pavone, J. Mater. Chem. A, 2017, 5, 11825–11833.
42. J. Hwang, R. R. Rao, L. Giordano, Y. Katayama, Y. Yu and Y. Shao-Horn, Science, 2017, 358, 751–756.
43. W. T. Hong, M. Risch, K. A. Stoerzinger, A. Grimaud, J. Suntivich and Y. Shao-Horn, Energy Environ. Sci., 2015, 8, 1404–1427.
44. Y. L. Lee, J. Kleis, J. Rossmeisl, Y. Shao-Horn and D. Morgan, Energy Environ. Sci., 2011, 4, 3966–3970.
45. M. Papac, V. Stevanovi, A. Zakutayev and R. O’Hayre, Nat. Mater., 2021, 20, 301–313.
46. E. Fabbri, D. Pergolesi and E. Traversa, Chem. Soc. Rev., 2010, 39, 4355–4369.
47. K. D. Kreuer, Annu. Rev. Mater. Res., 2003, 33, 333–359.
48. R. Z. Ren, Z. H. Wang, X. G. Meng, X. H. Wang, C. M. Xu, J. S. Qiao, W. Sun and K. N. Sun, ACS Appl. Energy Mater., 2020, 3, 4914–4922.
49. F. He, T. Z. Wu, R. R. Peng and C. R. Xia, J. Power Sources, 2009, 194, 263–268.
50. E. Fabbri, L. Bi, D. Pergolesi and E. Traversa, Energy Environ. Sci., 2011, 4, 4984–4993.
51. J. Chen, J. Li, L. Jia, I. Moussa, B. Chi, J. Pu and J. Li, J. Power Sources, 2019, 428, 13–19.
52. J. Wang, Z. Li, H. Zang, Y. Sun, Y. Zhao, Z. Wang, Z. Zhu, Z. Wei and Q. Zheng, Int. J. Hydrogen Energy, 2022, 47, 9395–9407.
53. N. Shi, F. Su, D. Huan, Y. Xie, J. Lin, W. Tan, R. Peng, C. Xia, C. Chen and Y. Lu, J. Mater. Chem. A, 2017, 5, 19664–19671.
54. K. Dong, J. Hou, L. N. Miao, Z. Z. Jin, D. Wang, Y. Teng and W. Liu, Ceram. Int., 2020, 46, 19335–19342.
55. L. Ma, J. Y. Gong, C. J. Jin, D. D. Yang and J. Hou, J. Alloys Compd., 2023, 945, 169359.
56. Y. H. Ling, T. M. Guo, Y. Y. Guo, Y. Yang, Y. F. Tian, X. X. Wang, X. M. Ou and P. Z. Feng, J. Adv. Ceram., 2021, 10, 1052–1060.
57. Q. Wang, J. Hou, Y. Fan, X. A. Xi, J. Li, Y. Lu, G. Huo, L. Shao, X. Z. Fu and J. L. Luo, J. Mater. Chem. A, 2020, 8, 7704–7712.
58. Z. Z. Chen, J. L. Wang, D. M. Huan, S. J. Sun, G. P. Wang, Z. P. Fu, W. H. Zhang, X. S. Zheng, H. B. Pan, R. R. Peng and Y. L. Lu, J. Power Sources, 2017, 371, 41–47.
59. X. Y. Li, Z. M. Chen, D. M. Huan, B. B. Qiu, K. Zhu, Z. M. Qi, H. J. Liu, C. R. Xia, R. R. Peng and Y. L. Lu, ACS Mater. Lett., 2023, 5, 2896–2905.
60. J. Y. Gong and J. Hou, J. Mater. Sci. Technol., 2024, 186, 158–163.
61. C. C. Duan, J. H. Tong, M. Shang, S. Nikodemski, M. Sanders, S. Ricote, A. Almansoori and R. O'Hayre, Science, 2015, 349, 1321–1326.
62. H. P. Ding, W. Wu, C. Jiang, Y. Ding, W. J. Bian, B. X. Hu, P. Singh, C. J. Orme, L. C. Wang, Y. Y. Zhang and D. Ding, Nat. Commun., 2020, 11, 1907.
63. W. J. Bian, W. Wu, B. M. Wang, W. Tang, M. Zhou, C. R. Jin, H. P. Ding, W. W. Fan, Y. H. Dong, J. Li and D. Ding, Nature, 2022, 604, 479–485.
64. N. Wang, B. Y. Yuan, C. M. Tang, L. Du, R. J. Zhu, Y. Aoki, W. B. Wang, L. X. Xing and S. Y. Ye, Adv. Mater., 2022, 34, 202203446.
65. N. Wang, B. Y. Yuan, F. Y. Zheng, S. Y. Mo, X. H. Zhang, L. Du, L. X. Xing, L. Meng, L. Zhao, Y. Aoki, C. M. Tang and S. Y. Ye, Adv. Funct. Mater., 2024, 34, 202309855.
66. S. C. Zhao, W. J. Ma, W. W. Wang, Y. L. Huang, J. Wang, S. J. Wang, Z. Shu, B. B. He and L. Zhao, Adv. Mater., 2024, 36, 2405052.
67. Z. Q. Liu, Y. S. Bai, H. N. Sun, D. Q. Guan, W. H. Li, W. H. Huang, C. W. Pao, Z. W. Hu, G. M. Yang, Y. L. Zhu, R. Ran, W. Zhou and Z. P. Shao, Nat. Commun., 2024, 15, 472.
68. X. M. Xu, Y. L. Pan, Y. J. Zhong, R. Ran and Z. P. Shao, Mater. Horiz., 2020, 7, 2519–2565.
69. Y. Li, W. Zhang, T. Wu, Y. Zheng, J. Chen, B. Yu, J. Zhu and M. Liu, Adv. Energy Mater., 2018, 8, 201801893.
70. E. Ostrovskiy, Y.-L. Huang and E. D. Wachsman, J. Mater. Chem. A, 2021, 9, 1593–1602.
71. W. Feng, G. Zou, T. Liu, R. Li, J. Yu, Y. Guo, Q. Liu, X. Zhang, J. Wang, N. Ta, M. Li, P. Zhang, X. Cao, R. Yu, Y. Song, M. Liu, G. Wang and X. Bao, Energy Environ. Sci., 2025, 18, 2273–2284.
72. P. Chandrasekharan Meenu, P. K. Samanta, T. Yoshida, N. J. English, S. P. Datta, S. A. Singh, S. Dinda, C. Chakraborty and S. Roy, ACS Appl. Energy Mater., 2022, 5, 503–515.
73. H. Li, Y. Song, M. Xu, W. Wang, R. Ran, W. Zhou and Z. Shao, Energy Fuels, 2020, 34, 11449–11457.
74. N. Han, S. Feng, W. Guo, O. M. Mora, X. Zhao, W. Zhang, S. Xie, Z. Zhou, Z. Liu, Q. Liu, K. Wan, X. Zhang and J. Fransaer, SusMat, 2022, 2, 456–465.
75. K. F. Chen and S. P. Jiang, Electrochem. Energy Rev., 2020, 3, 730–765.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5ee00993f

---