Heterostructured electrodes for Cr-tolerant solid oxide fuel cells

Abstract

Cr poisoning at the surface of Sr-doped perovskite oxides significantly reduces the durability of solid oxide fuel cells during stack operation under practical atmospheres. This poisoning is attributed to the interactions between vaporized Cr and segregated Sr at the surface, facilitated by the electrostatic interaction of oxygen vacancies and doped Sr. To address this issue, we designed a heterostructured electrode coated with a Cr-tolerant Sr-free material, which exhibits a low oxygen-vacancy concentration and contains reducible sites, preventing direct contact between vaporized Cr and Sr. By redistributing the charged defects near the heterointerface, excessive oxygen vacancies in the bulk electrode are reduced, leading to suppressed Sr segregation and Cr poisoning. Evaluation of thin-film model electrodes reveals significantly improved stability of the heterostructured electrode in both ambient air and a Cr atmosphere at 600 °C for 100 h as well as its enhanced oxygen reduction reaction kinetics. Furthermore, we demonstrate the feasibility of using this approach to fabricate a porous electrode, exhibiting high performance and high stability with nearly no degradation at 600 °C for 300 h. This paper presents a rational design strategy for Cr-tolerant hetero-structured electrodes, particularly based on Sr-doped perovskite oxides, considering physical, chemical, and electrochemical stabilities and performances.

---

1. Introduction

Solid oxide fuel cells (SOFCs) have emerged as highly promising energy-conversion devices for power generation owing to their high efficiency, power output, and scalability.¹ The primary focus of SOFC development has been to enhance electrochemical performance by improving the oxygen reduction reaction (ORR) activity of the air electrode with extensive material exploration efforts.^2,3 Consequently, significant advancements in the electrochemical performance of SOFCs have been achieved, particularly in the intermediate temperature regime (500–700 °C), through the development of state-of-the-art Sr-doped perovskite oxides such as La_0.6Sr_0.4CoO_3−δ (LSC),⁴ La_0.6Sr_0.4Co_0.2Fe_0.8O_3−δ (LSCF),² and Sm_0.5Sr_0.5CoO_3−δ (SSC).⁵ However, ensuring the high durability of perovskite oxides in practical atmospheres with metallic interconnectors in stacks has become a critical concern for further commercializing SOFCs.^1,6,7 During stack operation, metallic interconnectors, typically composed of chromia-based Cr–Fe alloys, are commonly utilized to achieve sufficient electronic conductivity for current collection, along with good mechanical properties and cost-effectiveness.⁷ However, challenges arise from the generation of vaporized CrO₃ when the metallic interconnector reacts with surrounding gases such as O₂ and H₂O, resulting in Cr poisoning of perovskite oxides. This rapid decrease in the ORR kinetics of air electrodes, especially Sr-doped perovskite oxides, is a significant obstacle.^8,9

According to the Cr poisoning mechanism of Sr-doped perovskite oxides (eqn (1)–(3)), Sr segregation plays a significant role in initiating the formation of Cr-related secondary phases. Consequently, Sr-doped perovskite oxides are highly susceptible to Cr poisoning when simultaneous Sr segregation occurs.^10–13 Vaporized Cr from the metallic interconnector reacts with SrO_x induced by Sr segregation at the electrode surface, resulting in the formation of Cr–Sr–O nuclei (eqn (1)) and facilitating the formation of insulating Cr-related phases, such as SrCrO₄ and Cr₂O₃ (eqn (2) and (3)).¹³ Furthermore, these insulating phases cause the segregated Sr clusters to grow larger, significantly blocking the electrochemical reaction sites with nonreactive clusters, thereby accelerating the electrochemical degradation of the electrode.¹⁰ Therefore, ensuring the long-term durability of Sr-doped perovskite oxides under a Cr atmosphere is crucial for the sustainability of SOFC systems.

CrO₃(g) + SrO(s) → Cr–Sr–O_x,nuclei(s) (1)

Cr–Sr–O_x,nuclei(s) + CrO₃(g) → Cr₂O₃(s) (2)

Cr–Sr–O_x,nuclei(s) + CrO₃(g) + SrO(s) → SrCrO₄(s) (3)

Various materials, including Sr-free^14,15 and high-entropy materials,¹⁶ have been developed to mitigate Cr poisoning by stabilizing the material. However, the use of single-phase materials as electrodes is challenging because the development of air electrodes requires a multifaceted approach that considers various conditions, including Cr tolerance, high ORR, chemical and structural compatibility, and practical manufacturing feasibility.^17–19 Therefore, due to their simplicity, effectiveness, and feasibility, heterostructured electrodes of Sr-doped perovskite oxides have been proposed for fabricating Cr-tolerant electrodes.^6,18,20,21 Among these approaches, coating discrete nanoparticles onto the electrode surface as a Cr-reactive catalyst has been extensively investigated to prevent Cr poisoning towards beneficial reactions.^18,21 Vaporized Cr preferentially reacts with the surface nanoparticles rather than with the bulk electrode, generating conductive and beneficial Cr-related phases at the surface, such as BaCrO₄ (ref. 21) and LaCrO₃,¹⁸ while preventing the formation of insulating Cr-related phases. However, these nanoparticles may not cover the entire electrode surface, and once Cr-reactive catalysts are fully consumed, Cr poisoning of the exposed electrode surface can eventually occur.¹⁹ Recently, conformal electrode surface coating with Cr-tolerant materials was reported, leading to significantly enhanced stability.^6,18,20 In particular, Sr-free materials have been introduced as coating layers for Cr-tolerant materials to inhibit the formation of Sr–Cr–O nuclei on the surface.^6,20 However, Cr poisoning can still occur unless Sr segregation from the perovskite oxide under the coating layer is prevented or at least substantially suppressed, necessitating a design that inhibits Sr segregation towards the surface.^19,22 Additionally, considering the impact of material properties such as the reducibility, conductivity, and catalytic activity of the coating layer in the heterostructured electrode on ORR kinetics, it is crucial to comprehensively investigate their effect on the electrochemical performance and stability of perovskite oxides.^23,24 However, a rational design for an effective heterostructured electrode with high performance and stability against Cr poisoning has not yet been thoroughly explored.

In this study, we designed a coating layer for a heterostructured electrode with a Sr-free material, a low oxygen vacancy concentration, and the presence of reducible sites to enhance the durability and performance of Sr-doped perovskite oxide against Sr segregation and Cr poisoning. Among the various candidates, LaCoO₃ (LCO) meets the requirements for a coating layer on the LSC electrode surface. We fabricated an LCO/LSC heterostructured thin-film model electrode using pulsed laser deposition (PLD). Initially, the Sr-free surface of LCO prevented Cr adsorption and preserved its surface in a Cr atmosphere. Furthermore, oxygen vacancies are redistributed from LSC to LCO at the heterointerface, suppressing Cr poisoning induced by electrostatically induced Sr segregation by oxygen vacancies at the LSC surface. Consequently, the heterostructured electrode exhibits excellent Cr tolerance and electrochemical stability at 600 °C for 100 h without the formation of secondary particles at the surface and in the bulk. Moreover, the comparable performance of the heterostructured electrode to that of the LSC electrode indicates that the increased oxygen vacancies in the LCO coating layer through charge redistribution significantly enhance the ORR kinetics. We further demonstrate the feasibility of the heterostructured electrode in an anode-supported single-cell configuration using infiltration as the coating technique. A single cell with the heterostructured electrode shows no degradation at 600 °C for 300 h without Cr poisoning. Our results suggest an effective and feasible design strategy for Cr-tolerant SOFCs.

2. Results and discussion

We designed a heterostructured electrode coated with Cr-tolerant and Sr-free materials to mitigate Cr poisoning in Sr-doped perovskite oxide electrodes, as shown in Fig. 1. In particular, a protective coating layer comprising an Sr-free material can prevent physical contact between vaporized Cr and Sr in the perovskite oxide, thus inhibiting their interaction to form detrimental SrCrO₄ and Cr₂O₃ phases. To ensure the continued absence of Sr in the coating layer during long-term operation, it is crucial to effectively inhibit Sr segregation toward the electrode surface, which could otherwise potentially segregate into the coating layer at elevated temperatures. Furthermore, to minimize the reduction in the electrochemical performance caused by the coating layer, the materials used in the coating layer need to exhibit sufficiently high oxygen ion conductivity, enabling their active participation in the ORR without significantly increasing the resistance. Successful integration of the heterostructured electrode necessitates that the materials used in the coating layer exhibit chemical and structural compatibility with the electrode materials, including insignificant undesired secondary reactions, identical crystal structures, and minimized mismatches in the lattice constants and thermal expansion coefficients to prevent stress-induced cracks or delamination during fabrication and operation at elevated temperatures.

Fig. 1 Schematic illustrating the design of a heterostructured electrode with enhanced Cr tolerance. (a) Mechanism of Cr poisoning in Sr-doped perovskite oxides. (b) Mechanism of high Cr tolerance in heterostructured electrodes.

LCO is one of the most promising candidates that satisfy the specified requirements for a Cr-tolerant coating layer for heterostructured electrodes, particularly when utilizing Sr-doped perovskite oxides as electrode materials for SOFCs, such as LSC and LSCF. LCO can be considered a Cr-tolerant material because of its Sr-free nature and the negligible reactivity of La and Co with vaporized Cr, thereby preventing the formation of Cr–Sr–O nuclei.²⁵ Furthermore, Sr segregation in the heterostructured electrode with LCO as a coating layer was mitigated by suppressing the electrostatic attraction by transferring oxygen vacancies from the electrode to the coating layer.^26,27 This transfer substantially reduces the oxygen vacancy concentration at the perovskite electrode surface, as shown in Fig. 1(b). LCO exhibits desired characteristics to function as a vacancy-accepting layer, including an intrinsically lower oxygen vacancy concentration ( = 0.02–0.09) than the electrode ( = 0.14–0.16 for LSC),²⁸ sufficiently high ionic conductivity facilitating the transfer of oxygen vacancies within the lattice (D = 8 × 10⁻¹⁰ cm² s⁻¹ at 873 K⁻¹ for LCO and D = 1 × 10⁻⁸ cm² s⁻¹ at 873 K⁻¹ for LSC),^28,29 the presence of a reducible host cation, Co, capable of adjusting its valence states (Co²⁺ and Co³⁺) to accommodate transferred oxygen vacancies,³⁰ and excellent chemical and structural compatibility with the LSC electrode in terms of an identical cubic structure, similar lattice constants (5.444 Å for LCO and 5.435 Å for LSC), and similar thermal expansion coefficients (22 × 10⁻⁶ K⁻¹ for LCO and 19 × 10⁻⁶ K⁻¹ for LSC).^31–33 Moreover, the redistribution of oxygen vacancies in the heterostructured electrode, particularly the transfer from the electrode to the coating layer, can activate the ORR activity of LCO (K = 7 × 10⁻¹¹ cm s⁻¹ at 873 K⁻¹), which is lower than that of LSC (K = 4 × 10⁻⁸ cm s⁻¹ at 873 K⁻¹).²⁹

To investigate the effect of the heterostructured electrode on mitigating Cr poisoning, we fabricated thin-film model electrodes to systematically analyze the surface characteristics and conducted an acceleration test of Cr poisoning-induced degradation of the perovskite oxide electrodes. Three model electrodes were fabricated: LSC, LCO (both with a thickness of ∼60 nm), and LCO/LSC electrodes (LSC electrode with a thickness of ∼60 nm and an LCO coating layer with a thickness of ∼5 nm), deposited on a GDC buffer layer (thickness of ∼20 nm) using PLD, as shown in Fig. 2 and S1.† In Fig. 2(a), the cross-sectional HR-TEM image of the LCO/LSC electrode shows that the LCO layer corresponds to the (200) plane with a lattice spacing of 0.265 nm, whereas the LSC layer corresponds to the (200) plane with a lattice spacing of 0.273 nm. Fig. 2(b) shows the XRD patterns of the LCO/LSC electrode, confirming the epitaxial growth of the LSC and LCO layers without secondary phase formation, which is consistent with the lattice distances of 0.268 and 0.274 nm for LCO and LSC, respectively, as observed in the HR-TEM analysis. Fig. 2(c) shows the STEM-EDX mapping, confirming the fabrication of a coating layer with a thickness of ∼5 nm without noticeable interdiffusion of Sr into the LCO layer during the PLD process.

Fig. 2 Structure of the heterostructured LCO/LSC model electrode. (a) Cross-sectional HR-TEM image, (b) XRD patterns, and (c) cross-sectional STEM-EDX elemental mapping images of La, Co, and Sr for the LCO/LSC electrode.

Fig. 3 shows the XPS analysis in the thickness direction, which confirms the redistribution of oxygen vacancies in the LCO/LSC heterostructured electrode. We deconvoluted the O 1s photoelectron spectra into three binding states: hydroxyl group (OH⁻), oxygen-related defects , and oxygen in the lattice .²⁶ The detailed fitting information is shown in Fig. S2.†Fig. 3(a) shows the stacked O 1s photoelectron spectra as a function of sputtering time in the thickness direction. The pronounced hydroxyl peaks at the outermost layer of all three model electrodes disappeared after the surface was etched by a few nanometers.^26,27 On the other hand, the oxygen-related defect peaks exhibited substantial changes in the thickness direction, and different trends were observed in the three model electrodes. Fig. 3(b) compares the depth profiles of the relative intensity ratio of oxygen-related defects to the oxygen in the lattice, , representing the oxygen vacancy concentration.^26,27 For all three electrodes, the intensity ratios of were highest at the surface due to the high reducibility of the B-site cation at the surface,³⁴ verifying the presence of excessive oxygen vacancies at the perovskite surface. These ratios decreased in the thickness direction, approaching values consistent with those reported in the literature.^27,34 For example, the LSC and LCO electrodes exhibited 1.7-fold higher ratios both at the surface (∼0.29 for LSC and ∼0.16 for LCO) compared to those in the bulk (∼0.17 for LSC and ∼0.09 for LCO). However, the LCO/LSC electrode showed relatively constant intensity ratios of 0.16–0.17 near the LCO–LSC interface, which were lower than those at the LSC surface (∼0.29) and close to those in the LSC bulk (∼0.17). Moreover, the LCO/LSC electrode exhibited higher intensity ratios near the LCO–LSC interface than the LCO electrode. The changes in the intensity ratios of near the LCO–LSC interface imply that the oxygen vacancies were transferred from the LSC electrode to the LCO coating layer.

Fig. 3 Redistribution of charged defects in the model electrodes. (a) Stacked O 1s photoelectron spectra. (b) Depth profiles of the relative intensity ratio of oxygen-related defects to the oxygen in the lattice. (c) Stacked Co 2p photoelectron spectra. (d) Depth profiles of the relative intensity ratio of the charge state of Co, [Co²⁺]/[Co³⁺].

The oxygen vacancy transfer was further evidenced by concurrent changes in the oxidation states of Co near the LCO–LSC interface. We deconvoluted the Co 2p photoelectron spectra into two oxidation states: Co²⁺ and Co³⁺.³⁵ The detailed fitting information is shown in Fig. S3.†Fig. 3(c) shows the stacked Co 2p photoelectron spectra as a function of the sputtering time in the thickness direction, comparing the relative intensity ratio of [Co²⁺]/[Co³⁺] in the thickness direction near the LCO–LSC interface. Interestingly, the depth profiles of the [Co²⁺]/[Co³⁺] ratio near the LCO–LSC interfaces, as shown in Fig. 3(d), exhibited a similar trend to that of the ratio, as shown in Fig. 3(b). The intensity ratios of [Co²⁺]/[Co³⁺] were the highest at the surface, ∼1.5 and ∼0.8, respectively, for LSC and LCO, whereas these ratios decreased in the thickness direction, approaching values consistent with those reported in the literature, ∼1.0 and ∼0.5, respectively.^26,36 However, the LCO/LSC electrode exhibited relatively constant intensity ratios of ∼1.0 near the LCO–LSC interface, which were lower than those at the LSC surface (∼1.5) and higher than those in the LCO electrode (∼0.8). These changes correspond to the more oxidized and reduced states of Co in the LSC electrode and LCO coating layers of the LCO/LSC electrode, respectively.^26,27 The concurrent changes in the and [Co²⁺]/[Co³⁺] ratios near the LCO–LSC interface confirm that the equilibrium defect concentration was altered near the LCO–LSC interface for charge neutrality in the system.

Fig. 4 and 5 show changes in the structure and chemistry at the surface of the three model electrodes upon annealing at 600 °C for 100 h in both ambient air and the Cr atmosphere. Given that Cr poisoning is initialized by the formation of Cr–Sr–O nuclei at the surface of the perovskite oxides, analyzing the changes in the three model electrodes in both the ambient air and the Cr atmosphere is crucial. Fig. 4(a) and (b) present the surface morphologies of the LSC and LCO/LSC electrodes at their initial state, after 100 h in ambient air, and after 100 h in the Cr atmosphere. Initially, the surface of LSC was clean and smooth. However, particle-like secondary phases with a size of 30–50 nm appeared after 100 h in ambient air. Moreover, larger secondary phases with a size of 100–150 nm appeared after 100 h in the Cr atmosphere. In addition to the changes in the surface structure, the LSC electrode exhibited considerable changes in surface chemistry, particularly in Sr and Cr. Fig. 4(c) and (d) compare the Sr 3d and Cr 2p photoelectron spectra of the LSC and LCO/LSC electrodes in their initial states after 100 h in ambient air and after 100 h in a Cr atmosphere. The Sr 3d spectra were deconvoluted into lattice (Sr_lattice) and non-lattice (Sr_non-lattice) components.^23,37 Detailed fitting information is provided in Fig. S4.† For the LSC electrode, the intensity ratios of [Sr_non-lattice]/[Sr_lattice] after 100 h in ambient air (∼0.85) and in a Cr atmosphere (∼1.24) were 1.4- and 2.0-fold higher than those in the initial state (∼0.61), respectively. These increases indicate the formation of secondary phases owing to Sr segregation towards the surface, surpassing the solubility limit of the perovskite oxide.^38,39 The concurrent increase in the intensity ratio of at the LSC surface, as shown in Fig. 3(a), supports Sr segregation induced by excessive oxygen vacancies at the surface, resulting in electrostatic attraction to negatively charged Sr in the lattice.^23,37,40 Furthermore, more significant structural and chemical changes at the surface in the Cr atmosphere confirmed accelerated surface deterioration by vaporized Cr.¹² Deconvolution of the Cr 2p photoelectron spectra after 100 h in a Cr atmosphere into three components, Cr⁶⁺, Cr³⁺, and hydroxyl, as shown in Fig. 4(c), provides clear evidence of Cr poisoning at the surface of the LSC electrode, representing the formation of SrCrO₄ and Cr₂O₃.⁹

Fig. 4 Changes in surface structure and chemistry in model electrodes. SEM images of the surfaces of the (a) LSC and (b) LCO/LSC electrodes before and after annealing at 600 °C. The Sr 3d and Cr 2p photoelectron spectra of the (c) LSC and (d) LSC/LCO electrodes after 100 h in ambient air and 100 h in a Cr atmosphere.

Fig. 5 Changes in the cross-sectional structure and chemistry after 100 h at 600 °C in a Cr atmosphere. Cross-sectional dark field STEM-EDX mapping images of (a) LSC and (b) LCO/LSC electrodes. Cross-sectional HR-TEM images of the (c) LSC and (d) LCO/LSC electrodes.

Cross-sectional TEM analysis was used to elucidate the mechanism of Cr poisoning by investigating localized changes after 100 h in a Cr atmosphere. Fig. 5(a) displays the STEM-EDX maps of the LSC electrode, showing a notably higher Sr intensity at the surface than in the bulk and the existence of Cr both at the surface and in the bulk. This indicates that Cr was incorporated into the bulk of the electrode, originating from SrCrO₄ at the surface, resulting in Cr poisoning both at the surface and in the bulk.⁴¹ The unusually high intensities of La and Cr in the GDC layer can be disregarded because they are artifacts caused by the overlapping of La L with Ce L and Cr K with Gd L in the EDX energy spectrum.^2,42 HR-TEM analysis consistently showed the pronounced formation of SrCrO₄ and Cr₂O₃ at the surface and in the bulk, verified by a lattice constant of 3.23 Å and 2.22, 3.62 Å, respectively, as shown in Fig. 5(c).^43,44

In contrast, the LCO/LSC electrode exhibited remarkable structural and chemical stability against Sr segregation and Cr poisoning, as shown in Fig. 4 and 5. The Sr 3d photoelectron spectra at the LSC surface of the LCO/LSC electrode remained unchanged after annealing in both the ambient air and Cr atmosphere, as shown in Fig. 4(b). No Cr-related photoelectron spectra were detected at the surface after annealing in the Cr atmosphere, as shown in Fig. 4(d). HR-TEM analysis in Fig. 5(b) and (d) confirmed that the coating layer and the bulk electrode maintained the LCO and LSC phases, respectively, verifying the absence of Sr segregation towards the LCO layer and Cr poisoning both at the surface and in the bulk. As expected, the LCO electrode showed no changes in the structure and chemistry at the surface after annealing both in the ambient air and Cr atmosphere, as shown in Fig. S5.† In addition, Raman spectra of the LSC, LCO/LSC, and LCO electrodes after annealing in the Cr atmosphere consistently revealed the formation of SrCrO₄ and Cr₂O₃ in the LSC electrode, while no such formation was observed in the LCO/LSC and LCO electrodes, as shown in Fig. S6.†

We evaluated the electrochemical performance and stability in both ambient air and a Cr atmosphere. Fig. 6(a) shows the changes in the polarization resistance (R_p) of the three model electrodes in a symmetrical cell configuration at 600 °C in the ambient air for 100 h. The R_p values of the three model electrodes evaluated at 450–600 °C are shown in Fig. S9.† At the initial state (0 h), although the LCO electrode showed ∼1.6-fold higher R_p of 23.8 Ω cm² compared to the LSC electrode (15.7 Ω cm²), the LCO/LSC electrode exhibited a comparable R_p of 15.8 Ω cm² compared to the LSC electrode. The improved performance of the LCO/LSC electrode is attributed to the transfer of oxygen vacancies from the LSC electrode to the LCO coating layer. This transfer near the LCO–LSC interface results in an increased oxygen vacancy concentration and the reduction of Co³⁺ to Co²⁺ within the LCO coating layer, enabling the LCO layer to actively participate in the ORR with increased ionic and electronic conductivities, as shown in Fig. 2.^45,46 The sheet resistances of the model electrodes were measured using a four-point probe, and the results revealed that the sheet resistance of the LCO/LSC electrode (0.067 Ω sq⁻¹) was ∼52% lower than that of the LCO electrode (0.102 Ω sq⁻¹), as shown in Fig. S7.† This improvement can be attributed to the release of electrons at the Co site to form a free electron and leave behind an electron hole, thereby facilitating electron migration, because adjacent electrons can fill the hole in an electrical field.⁴⁵ Thus, charge transfer at the heterostructured electrode is facilitated by increased electronic conductivity at the surface, leading to higher electrocatalytic activity.

Fig. 6 Electrochemical performance and stability of model electrodes at 600 °C. (a) Changes in R_p of model electrodes in ambient air for 100 h. (b) Nyquist plots and (c) Bode plots of model electrodes in ambient air for 100 h. (d) Changes in R_p of model electrodes in a Cr atmosphere for 100 h. (e) Nyquist plots and (f) Bode plots of model electrodes in a Cr atmosphere for 100 h.

Furthermore, the LCO/LSC electrode exhibited significantly improved stability compared with the LSC electrode, both in ambient air and in a Cr atmosphere. After 100 h in ambient air, the LSC electrode showed rapid degradation with a ∼1.9-fold increase in R_p compared to its initial state. The corresponding Nyquist and Bode plots show a noticeable increase in the low-frequency range (10⁰–10² Hz), as shown in Fig. 6(b) and (c). After 100 h in a Cr atmosphere, the degradation of the LSC electrode was more pronounced than that in ambient air, exhibiting a ∼3.5-fold increase in R_p from its initial state, as shown in Fig. 6(d). The increase in the low-frequency range in the Bode plots was similar to that in ambient air, as shown in Fig. 6(c) and (f). However, a substantial increase in the high-frequency range (>10² Hz) was observed in the Cr atmosphere, which was absent in ambient air. In contrast, the LCO/LSC and LCO electrodes maintained their initial R_p for 100 h both in ambient air and in the Cr atmosphere, as shown in Fig. 6(a) and (d), respectively, without any noticeable changes in the Nyquist plots and Bode plots in both the ambient air and Cr atmosphere.

To precisely determine the role of a heterostructured electrode coated with Cr-tolerant materials in terms of stability in both ambient air and the Cr atmosphere, we conducted distribution of relaxation time (DRT) analysis using impedance spectra. By deconvoluting the two distinct peaks in the Bode plot of the LSC electrode, we identified a low-frequency component (R_LF, 10⁰–10² Hz), representing the oxygen diffusion and adsorption reaction at the electrode surface, and a high-frequency component (R_HF, >10² Hz), corresponding to the charge transfer reactions at the interface.⁵ The detailed DRT analysis and peak deconvolution are shown in Fig. S8 and Table S1,† respectively.

Fig. 7(a) shows the changes in the R_LF and R_HF of the LSC electrode in ambient air for 100 h. From the initial state (0 h), the R_LF continuously increased to ∼2.0-fold its value at 100 h, dominating the total R_p, whereas the R_HF remained negligible. The increase in R_LF can be attributed to hindered gas diffusion and adsorption reactions due to the formation of secondary phases such as SrO_x and Sr(OH)_x.¹¹ Moreover, Fig. 7(b) shows the accelerated degradation of the LSC electrode in a Cr atmosphere with a ∼2.4-fold increase in R_LF at 100 h compared to its initial state. This rapid increase in the R_LF in the Cr atmosphere can be attributed to the formation of Cr-induced secondary phases such as SrCrO₄ and Cr₂O₃ with increased surface coverage compared to those in ambient air, further hindering gas diffusion and adsorption, as shown in Fig. 4. Moreover, R_HF exhibited a substantial ∼100-fold increase at 100 h compared with its initial state, constituting a significant portion of the total R_p. This rapid increase in the R_HF, stemming from hindered charge-transfer reactions in the bulk electrode, can be ascribed to the accelerated degradation in the Cr atmosphere compared to that in ambient air. As shown in Fig. 5, Cr poisoning is evident both at the surface and in the bulk of the LSC electrode owing to Sr instability, impeding both ionic and electronic conduction, which has detrimental effects on the charge transfer reaction.⁴⁷ The concurrent changes in the activation energies of R_LF and R_HF after the stability test are evident in Fig. S9 and S10.† This indicates that the significant deterioration in the ORR activity for both gas diffusion and adsorption and the charge-transfer reaction in the Cr atmosphere can be attributed to the formation of additional Cr-induced secondary phases at the LSC surface and in bulk, respectively.

Fig. 7 Electrochemical impedance spectroscopy analysis of the LSC and LCO/LSC model electrodes. Deconvoluted R_p values of the LSC electrode as a function of time (a) in ambient air and (b) in a Cr atmosphere. Deconvoluted R_p values of the LCO/LSC electrode as a function of time (c) in ambient air and (d) in a Cr atmosphere.

In contrast, the LCO/LSC electrode exhibits significantly improved stability in ambient air and Cr atmospheres. Both the R_LF and R_HF of the LCO/LSC electrode remained unchanged compared with their initial states, as shown in Fig. 7(c) and (d), with no noticeable secondary phases at the surface or in the bulk, as shown in Fig. 4 and 5, respectively. Moreover, the activation energy of the LCO/LSC electrode (1.32 eV) was comparable to that of the LSC electrode (1.31 eV) but higher than that of the LCO electrode (1.45 eV) in the initial state, as shown in Fig. S9.† This demonstrates the effectiveness of the LCO/LSC electrode in preventing Cr poisoning, owing to the high Cr tolerance of the LCO coating layer and the suppression of Sr segregation towards the surface to inhibit the interaction with vaporized Cr during long-term operation while maintaining ORR activities similar to those of the LSC electrode in its initial state.

In consideration of manufacturing feasibility, the heterostructured electrode should be produced using cost-effective and scalable techniques that are compatible with existing fabrication methods. Moreover, uniform and conformal coverage of the coating layer with Cr-tolerant materials on the porous electrodes should be ensured. Wet chemical-based infiltration is one of the most commonly employed techniques for fabricating nanostructured electrodes with high uniformity, ease, and repeatability.^48–50Fig. 8(a) shows the morphology of the LCO-infiltrated LSC electrode (LCO/LSC cell) with an LCO coating layer thickness of less than 10 nm. We evaluated the electrochemical performance, stability, and structural and chemical changes of the heterostructured electrode in a Ni-GDC anode-supported single-cell configuration in a Cr atmosphere, as shown in Fig. S11.†Fig. 8(b) and (c) show that the peak power densities of the LCO/LSC cell were compatible with those of the LSC cell in the temperature range of 600–450 °C in the Cr atmosphere, which was consistent with the electrochemical performance evaluation of model electrodes in a symmetric cell configuration, as shown in Fig. 6 and 7. Fig. 8(d)–(h) demonstrate the significantly improved stability of the LCO/LSC cell compared with that of the LSC cell in a Cr atmosphere. During the galvanostatic stability test at a constant current density of 0.5 A cm⁻² at 600 °C for 300 h, the LSC cell exhibited a degradation of ∼12% with ∼1.7-fold increased R_p compared to its initial state. Fig. 8(e) shows noticeable increases in both the low frequency (10–10²) and high frequency (10²–10³), corresponding to gas diffusion and adsorption reactions and charge transfer reactions at the cathode, respectively.⁵Fig. 8(g), S12–S14, and Table S2† reveal considerable Sr segregation and the formation of Cr-related secondary phases, such as SrCrO₄ and Cr₂O₃, with notable structural changes in the electrode morphologies.

Fig. 8 Evaluation of electrochemical performance and stability of the heterostructured electrode in a single cell configuration. (a) HR-TEM image of the LCO/LSC electrode. I–V–P curves of (b) the LSC and (c) LCO/LSC cells in the temperature range of 600–450 °C. (d) Stability test in a Cr atmosphere at 600 °C at 500 mA cm⁻² for 300 h with wet hydrogen and dry air at the fuel and air electrodes, respectively. Nyquist and Bode plots of (e) LSC and (f) LCO/LSC cells before and after the stability test. Sr 3d and Cr 2p photoelectron spectra at the surface of the (g) LSC and (h) LCO/LSC electrodes before and after the stability test.

These results indicate that the degradation in the electrochemical performance of the LSC cell was largely attributed to Sr segregation and Cr poisoning in the electrode, which is consistent with the structural and chemical changes in the model electrodes, as shown in Fig. 4 and 5. In contrast, the LCO/LSC cell exhibited no discernible degradation in the Cr atmosphere and maintained its initial electrochemical performance, microstructure, and chemical composition, as shown in Fig. 8(f), (h) and S15.† These results demonstrate that a heterostructured electrode with a Cr-tolerant coating layer can be directly applied to powder-based porous electrodes using conventional fabrication processes to enhance the stability against Sr segregation and Cr poisoning. Furthermore, the similar degradation characteristics of the model electrode and single cell in a Cr atmosphere verified that the structural, chemical, and electrochemical changes observed in the model electrode could serve as key indicators for analyzing the Cr poisoning phenomenon.

3. Conclusions

We demonstrated that a heterostructured electrode coated with Sr-free perovskite oxides can effectively mitigate the Sr segregation and Cr poisoning of Sr-doped perovskite oxides under practical conditions while maintaining the performance of the bulk electrode. Coating with Sr-free and Cr-tolerant materials prevents direct contact between vaporized Cr and segregated Sr at the surface. Furthermore, the transfer of oxygen vacancies from the bulk electrode to the coating layer enabled the continuous suppression of Sr segregation and Cr poisoning, as well as the activation of the coating layer for ORR kinetics. Consequently, the heterostructured electrode exhibited excellent electrochemical performance and stability in a Cr atmosphere in a single-cell configuration at 600 °C for 300 h without noticeable degradation. This paper presents rational design strategies for developing high-Cr tolerance and high-performance electrodes for SOFCs.

4. Experimental section

4.1. Model electrode fabrication

To prepare the PLD target, powders of LSC (Kceracell Co., Ltd), LCO (Kceracell Co., Ltd), and Gd_0.1Ce_0.9O_1.95 (GDC, Rhodia, ULSA grade) were pressed in the form of discs and sintered at 1200 °C. The model electrodes were symmetrically deposited on the YSZ (100) single crystal (MTI Korea) surface by PLD under the corresponding process conditions: 520 mJ of laser power per pulse at 5 Hz, a substrate temperature of 700 °C, and an oxygen partial pressure of 30 mTorr. A GDC layer was deposited with a thickness of ∼20 nm to prevent the formation of SrZrO₃.⁵¹ For the LSC and LCO model electrodes, the LSC and LCO layers were subsequently deposited as bulk electrodes with thicknesses of ∼60 nm. For the LCO/LSC model electrode, an LCO layer with a thickness of approximately 5 nm was additionally deposited on the LSC bulk electrode.

4.2. Characterization of model electrodes

X-ray diffraction (XRD, X'Pert Pro) with Cu Kα (λ = 1.5406 Å) was conducted at room temperature to characterize the crystal structure of model electrodes. X-ray photoelectron spectroscopy (XPS, ESCA Lab 250 XPS spectrometer) with Al Kα was used to identify the chemical state of model electrodes with the thickness direction using Ar⁺ sputtering. Secondary electron microscopy (SEM, JSM-IT800) was used to characterize changes in the morphologies of the electrodes. High-resolution transmission electron microscopy (HR-TEM, NEO ARM/JEOL) and scanning transmission electron microscopy-energy-dispersive X-ray spectroscopy (STEM-EDX, NEO ARM/JEOL) were used to characterize the structural and chemical changes, respectively, in cross-sectional views of the electrodes. Electrochemical impedance spectroscopy (EIS) analysis was conducted at 600 °C in ambient air and the Cr atmosphere to evaluate the stability of model electrodes with a frequency range of 10⁻² to 10⁵ Hz using a potentiostat (GAMRY, Gamry Reference 600). A setup enabling concurrent assessments was employed to evaluate the model electrodes in identical atmospheres. The sheet resistance was assessed using a 4-point probe (NEXTRON MPS-CHH) in ambient air. A sample deposited as a thin film on a Si wafer was placed on the instrument and measured at 650 °C. The measurement results were calculated using the van der Pauw method using the following equation:where R_12,34 and R_23,41 are the resistance values along the measurement direction and R_S is the sheet resistance value.

4.3. Single cell fabrication

A single cell was fabricated in an anode-supported configuration to demonstrate the model electrodes in a 3D cathode. An anode support was prepared by ball-milling NiO (Kojundo Chemical), GDC (Rhodia, LSA), and PMMA (Alfa Aesar) with a dispersant and sintered at 900 °C for 5 h. The gradient anode functional layer and GDC electrolyte were deposited by spin coating and co-sintered at 1400 °C for 5 h.⁵² For the LSC cathode, the LSC powder was screen-printed with a thickness of ∼10 μm and sintered at 900 °C for 3 h. The LCO/LSC cathode was fabricated using additional infiltration of the LCO precursor solution prepared by dissolving La(NO₃)₃·6H₂O (Alfa Aesar), Co(NO₃)₂·6H₂O (Sigma-Aldrich), ethylene glycol (PEG, Alfa Aesar), and urea (Alfa Aesar) in a solution with deionized water and ethanol. After infiltration, the cells with LCO/LSC electrodes were sintered at 800 °C for 3 h to synthesize the perovskite oxide LCO phase.

4.4. Electrochemical evaluation of the single cell

Current–voltage (I–V) curves were obtained and EIS measurements were conducted to evaluate the electrochemical performance and stability. For the Cr-poisoning atmosphere, pieces of SUS303 were placed near the cathode to allow the vaporized CrO₃ to diffuse into the air.

Data availability

The data supporting this article have been included as part of the ESI.† No software or code has been included.

Author contributions

Sehee Bang: conceptualization, methodology, writing – original draft, writing – review & editing. Jongseo Lee: conceptualization, methodology, writing – original draft. Joon Gyu Kim: data curation, investigation. Jinwoo Kim: data curation, investigation. Mingi Choi: conceptualization, methodology. Yan Chen: conceptualization, methodology. Wonyoung Lee: conceptualization, supervision, writing – review & editing.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea Government (MSIT) (2022R1A2C3012372, 2022R1A4A1031182, and 2021R1C1C2006657), the Korea Institute of Energy Technology Evaluation and Planning (KETEP) (20223030040080), the Competency Development Program for Industry Specialists of the Korean Ministry of Trade, Industry and Energy (MOTIE) operated by the Korea Institute for Advancement of Technology (KIAT) (P0017120, HRD program for Foster R&D specialist of parts for ecofriendly vehicle (xEV)), H2KOREA funded by the ministry of Education (2022 Hydrogen fuel cell-004, innovative Human Resources Development Project for Hydrogen Fuel Cells), and the Guangdong Provincial Science and Technology Program Project (2023A0505050096).

References

1. S. Z. Golkhatmi, M. I. Asghar and P. D. Lund, Renewable Sustainable Energy Rev., 2022, 161, 112339.
2. K. Develos-Bagarinao, T. Ishiyama, H. Kishimoto, H. Shimada and K. Yamaji, Nat. Commun., 2021, 12, 3979.
3. J. G. Lee, J. H. Park and Y. G. Shul, Nat. Commun., 2014, 5, 4045.
4. H. S. Yoo, S. J. Kim, Y. T. Megra, J. Lee, J. W. Suk and W. Lee, Appl. Surf. Sci., 2023, 639, 158188.
5. J. Lee, S. Hwang, M. Ahn, M. Choi, S. Han, D. Byun and W. Lee, J. Mater. Chem. A, 2019, 7, 21120–21127.
6. J. Huang, F. Liang, S. Zhao, L. Zhao, N. Ai, S. P. Jiang, X. Wang, H. Fang, Y. Luo and K. Chen, Adv. Funct. Mater., 2024, 34, 2310402.
7. T. Horita, Ceram. Int., 2021, 47, 7293–7306.
8. J. Hong, S. J. Heo and P. Singh, Appl. Surf. Sci., 2020, 530, 147253.
9. K. Chen, J. Hyodo, A. Dodd, N. Ai, T. Ishihara, L. Jian and S. P. Jiang, Faraday Discuss., 2015, 182, 457–476.
10. H. Zhang, K. Xu, F. He, Y. Zhou, K. Sasaki, B. Zhao, Y. M. Choi, M. Liu and Y. Chen, Adv. Energy Mater., 2022, 12, 2200761.
11. N. Ni, C. C. Wang, S. P. Jiang and S. J. Skinner, J. Mater. Chem. A, 2019, 7, 9253–9262.
12. L. Zhao, J. Drennan, C. Kong, S. Amarasinghe and S. P. Jiang, J. Mater. Chem. A, 2014, 2, 11114–11123.
13. T. Horita, Ceram. Interfaces, 2021, 47, 7293–7306.
14. A. Shaur, S. U. Rehman, H.-S. Kim, R.-H. Song, T.-H. Lim, J.-E. Hong, S.-J. Park and S.-B. Lee, ACS Appl. Mater. Interfaces, 2020, 12, 5730–5738.
15. M. Yang, E. Bucher and W. Sitte, J. Power Sources, 2011, 196, 7313–7317.
16. X. Han, Y. Ling, Y. Yang, Y. Wu, Y. Gao, B. Wei and Z. Lv, Adv. Funct. Mater., 2023, 33, 2304728.
17. J. Huang, Z. Xie, N. Ai, C. C. Wang, S. P. Jiang, X. Wang, Y. Shao and K. Chen, Chem. Eng. J., 2022, 431, 134281.
18. Y. Niu, Y. Zhou, W. Lv, Y. Chen, Y. Zhang, W. Zhang, Z. Luo, N. Kane, Y. Ding, L. Soule, Y. Liu, W. He and M. Liu, Adv. Funct. Mater., 2021, 31, 2100034.
19. K. Pei, Y. Zhou, K. Xu, Z. He, Y. Chen, W. Zhang, S. Yoo, B. Zhao, W. Yuan, M. Liu and Y. Chen, Nano Energy, 2020, 72, 104704.
20. Y. Chen, S. Yoo, X. Li, D. Ding, K. Pei, D. Chen, Y. Ding, B. Zhao, R. Murphy, B. deGlee, J. Liu and M. Liu, Nano Energy, 2018, 47, 474–480.
21. J. Huang, Q. Liu, S. P. Jiang, L. Zhao, N. Ai, X. Wang, Y. Shao, C. Guan, H. Fang, Y. Luo and K. Chen, Appl. Catal., B, 2023, 321, 122080.
22. H. A. Ishfaq, M. Z. Khan, Y. M. Shirke, S. Qamar, A. Hussain, M. T. Mehran, R.-H. Song and M. Saleem, Appl. Catal., B, 2023, 323, 122178.
23. N. Tsvetkov, Q. Lu, L. Sun, E. J. Crumlin and B. Yildiz, Nat. Mater., 2016, 15, 1010–1016.
24. M. Choi, S. Kim, J. Paik and W. Lee, Korean J. Chem. Eng., 2020, 37, 1346–1351.
25. P. Qiu, J. Lin, L. Lei, Z. Yuan, L. Jia, J. Li and F. Chen, ACS Appl. Energy Mater., 2019, 2, 7619–7627.
26. M. Choi and W. Lee, Chem. Eng. J., 2022, 431, 134345.
27. M. Choi, I. A. M. Ibrahim, K. Kim, J. Y. Koo, S. J. Kim, J.-W. Son, J. W. Han and W. Lee, ACS Appl. Mater. Interfaces, 2020, 12, 21494–21504.
28. M. V. Ananyev, N. M. Porotnikova and E. Kh. Kurumchin, Solid State Ionics, 2019, 341, 115052.
29. A. V. Berenov, A. Atkinson, J. A. Kilner, E. Bucher and W. Sitte, Solid State Ionics, 2010, 181, 819–826.
30. L. Zhuang, L. Ge, Y. Yang, M. Li, Y. Jia, X. Yao and Z. Zhu, Adv. Mater., 2017, 29, 1606793.
31. T. Itoh, M. Inukai, N. Kitamura, N. Ishida, Y. Idemoto and T. Yamamoto, J. Mater. Chem. A, 2015, 3, 6943–6953.
32. M. Matsuda, K. Ihara and M. Miyake, Solid State Ionics, 2004, 172, 57–61.
33. V. V. Kharton, E. N. Naumovich, A. V. Kovalevsky, A. P. Viskup, F. M. Figueiredo, I. A. Bashmakov and F. M. B. Marques, Solid State Ionics, 2000, 138, 135–148.
34. C. Zhao, Y. Li, W. Zhang, Y. Zheng, X. Lou, B. Yu, J. Chen, Y. Chen, M. Liu and J. Wang, Energy Environ. Sci., 2020, 13, 53–85.
35. Y. Zhu, L. Zhang, B. Zhao, H. Chen, X. Liu, R. Zhao, X. Wang, J. Liu, Y. Chen and M. Liu, Adv. Funct. Mater., 2019, 29, 1901783.
36. C. Feng, Q. Gao, G. Xiong, Y. Chen, Y. Pan, Z. Fei, Y. Li, Y. Lu, C. Liu and Y. Liu, Appl. Catal., B, 2022, 304, 121005.
37. S. Koohfar, M. Ghasemi, T. Hafen, G. Dimitrakopoulos, D. Kim, J. Pike, S. Elangovan, E. D. Gomez and B. Yildiz, Nat. Commun., 2023, 14, 7203.
38. B. Koo, K. Kim, J. K. Kim, H. Kwon, J. W. Han and W. Jung, Joule, 2018, 2, 1476–1499.
39. J. Y. Koo, H. Kwon, M. Ahn, M. Choi, J.-W. Son, J. W. Han and W. Lee, ACS Appl. Mater. Interfaces, 2018, 10, 8057–8065.
40. W. Lee, J. W. Han, Y. Chen, Z. Cai and B. Yildiz, J. Am. Chem. Soc., 2013, 135, 7909–7925.
41. A. M. Mehdi, A. Hussain, R. H. Song, T.-H. Lim, W. W. Kazmi, H. A. Ishfaq, M. Z. Khan, S. Qamar, M. W. Syed and M. T. Mehran, RSC Adv., 2023, 13, 25029–25053.
42. R. Wang, Z. Sun, Y. Lu, S. Gopalan, S. N. Basu and U. B. Pal, J. Power Sources, 2020, 476, 228743.
43. K.-Y. Lai and A. Manthiram, Chem. Mater., 2018, 30, 2838–2847.
44. Z. Cao and C. Zuo, RSC Adv., 2017, 7, 40243–40248.
45. Q. Ji, L. Bi, J. Zhang, H. Cao and X. S. Zhao, Energy Environ. Sci., 2020, 13, 1408–1428.
46. M. S. D. Read, M. S. Islam, G. W. Watson, F. King and F. E. Hancock, J. Mater. Chem., 2000, 10, 2298–2305.
47. N. Ni, S. J. Cooper, R. Williams, N. Kemen, D. W. McComb and S. J. Skinner, ACS Appl. Mater. Interfaces, 2016, 8, 17360–17370.
48. S. Pirou, B. Talic, K. Brodersen, A. Hauch, H. L. Frandsen, T. L. Skafte, A. H. Persson, J. V. T. Hogh, H. Henriksen, M. Navasa, X.-Y. Miao, X. Georgolamprou, S. P. V. Foghmoes, P. V. Hendriksen, E. R. Nielsen, J. Nielsen, A. C. Wulff, S. H. Jensen, P. Zielke and A. Hagen, Nat. Commun., 2022, 13, 1263.
49. S. J. Kim, D. Woo, D. Kim, T. K. Lee, J. Lee and W. Lee, Int. J. Extreme Manuf., 2023, 5, 015506.
50. M. Choi, J. Lee and W. Lee, Int. J. Precis. Eng. Manuf., Green Technol., 2019, 6, 53–61.
51. S. Hwang, J. Lee, G. Kang, M. Choi, S. J. Kim, W. Lee and D. Byun, J. Mater. Chem. A, 2021, 9, 11683–11690.
52. S. Bang, J. Lee and W. Lee, J. Power Sources, 2023, 553, 232290.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4ta04215h

‡ These authors contributed equally to this work.

---