A‐Site High‐Entropy Perovskite Enabling Sulfur‐Tolerant and Coking‐Resistant Anodes for Hydrocarbon‐Fueled Solid Oxide Fuel Cells

ABSTRACT

Solid oxide fuel cells (SOFCs) are capable of electrochemically converting fossil fuels such as natural gas and coal‐based syngas directly into electricity with high efficiency and minimal emissions, yet the state‐of‐the‐art nickel‐based anodes are susceptible to sulfur poisoning or coking when operated with sulfur‐containing or hydrocarbon fuels. Here, we report a high‐entropy strategy in which five equimolar cations are introduced at the A‐site to develop a highly active and robust perovskite anode, Pr_0.2Ba_0.2La_0.2Sr_0.2Ca_0.2FeO_3‐δ (PBLSCF). In situ reduction of PBLSCF leads to the exsolution of nanoscale Fe particles, enhancing its tolerance to sulfur poisoning and coking. Electrolyte‐supported single cells using PBLSCF anodes achieve a peak power density (PPD) of 1.22 W cm⁻² at 800°C in H₂, maintain stable operation for 1000 h and exhibit promising sulfur tolerance in 50 ppm H₂S‐H₂. Density functional theory (DFT) calculations reveal that the high‐entropy strategy reduces oxygen‐vacancy formation energy, contributing to improved sulfur tolerance and fuel oxidation performance. Furthermore, stable operation using a PBLSCF anode for 600 h is also achieved with propane as fuel. This work provides a synergistic strategy through A‐site high‐entropy engineering and in situ metal exsolution to achieve promising electrochemical performance and enhanced multi‐tolerance anodes for fuel‐flexible SOFCs.

A‐site high‐entropy engineering creates a robust Pr_0.2Ba_0.2La_0.2Sr_0.2Ca_0.2FeO_3‐δ (PBLSCF) perovskite‐based solid oxide fuel cell (SOFC) anode with in situ Fe nanoparticle exsolution to boost fuel oxidation and durability, achieving strong sulfur tolerance and coking resistance.

1. Introduction

Solid oxide fuel cells (SOFCs) are promising energy conversion devices capable of converting a wide range of fuels directly into electricity with high efficiency at elevated temperatures [1, 2, 3]. Among the different types of fuel cells, SOFCs offer several advantages, including high overall efficiency through combined heat and power, reversibility, and fuel flexibility. Various hydrocarbons, syngas, and ammonia can be used as fuels without the need for external reforming [4]. Conventional SOFC anodes are Ni‐based cermet materials (e.g., Ni‐YSZ) due to their excellent catalytic activity, high electronic conductivity, and mature processing [5, 6, 7]. However, when operating with hydrocarbon fuels, Ni readily catalyzes the cracking of C─H and C─C bonds, leading to coking on the anode surface [8]. The accumulated carbon blocks active sites, physically damages the Ni‐YSZ structure, and can react with Ni at high temperatures, causing anode expansion and micro‐cracking [9]. In addition to carbon deposition, Ni‐based anodes are highly sensitive to sulfur contaminants. Trace amounts of H₂S in the fuel can react with Ni to form NiS_x species, resulting in severe sulfur poisoning [10].

To overcome the limitations of Ni‐based anodes, various ceramic anode materials with inherent resistance to coking and sulfur poisoning have been developed. Among them, mixed ionic‐electronic conducting (MIEC) ceramic anodes have attracted considerable attention in SOFCs. Several synergistic strategies have been proposed to enhance coking resistance and sulfur tolerance; one of which is through the in situ exsolution of metal nanoparticles. By appropriately doping the B‐site with metals such as Ni, Co, Fe, or Cu, firmly anchored nano‐alloy particles can be exsolved from the oxide lattice under reducing atmospheres at elevated temperatures, serving as highly active reforming/catalytic sites [11, 12, 13, 14, 15]. On the other hand, A‐site engineering (e.g., tailoring the ratio of rare‐earth to alkaline‐earth cations or introducing A‐site deficiencies) can regulate the lattice parameters, thermal expansion coefficient (TEC), and oxygen vacancy concentration, thereby improving electrical conductivity and surface oxygen exchange kinetics [16, 17, 18, 19]. A‐site substitution of Sr with La or Pr in the Sr₂Fe_1.4Ni_0.1Mo_0.5O_6‐δ (SFNM) perovskite has been shown to strengthen the lattice stability and markedly enhance both electrochemical performance and resistance to H₂S poisoning, leading to improved sulfur tolerance in SFNM‐based ceramic anodes [20]. Another important approach is the infiltration strategy, in which precursor solutions are introduced into a porous ceramic scaffold to generate high surface area nanoparticles in situ (e.g., noble metals, CeO₂, or perovskites) [21]. Infiltrated electrodes typically exhibit high initial activity and improved coking resistance and sulfur tolerance, mainly due to their large active surface area and optimized distribution of triple‐phase boundaries [22]. However, solution infiltration often requires multiple infiltration and high‐temperature calcination steps, making the process time‐consuming and difficult to scale up for large‐area cell fabrication [23, 24].

Among the numerous perovskite anodes explored, Ba_0.5La_0.5FeO_3‐δ (BLF) has received increasing attention as a potential SOFC anode due to its excellent mixed ionic‐electronic conductivity, thermal expansion coefficient compatibility with common electrolytes, and superior stability in reducing atmospheres [25]. However, its catalytic activity toward fuel oxidation (especially hydrocarbon fuels) remains relatively low, and further compositional optimization is needed to enhance its electrochemical performance for practical SOFCs. Recently, A‐site high‐entropy engineering has emerged as a promising strategy for designing perovskite electrodes [26].

A‐site high‐entropy engineering is hypothesized to enhance electrochemical performance and multi‐foulant tolerance through cooperative lattice, defect, and surface‐chemistry effects while imparting a redox‐adaptive character to the perovskite [15, 27, 28]. Zhang et al. developed an A‐site high‐entropy engineered (LaSrPrBaCaGd)₂Fe_1.5Mo_0.5O_6‐δ (LSPBCGFM) fuel electrode, demonstrating a much higher current density (1.34 A cm⁻² at 800°C and 1.5 V) than Sr₂Fe_1.5Mo_0.5O_{6‐ δ} (SFM) (1.04 A cm⁻² at 800°C and 1.5 V) in CO₂ electrolysis with negligible performance degradation for over 400 h at 1.2 V, highlighting that A‐site high‐entropy design can simultaneously enhance performance and improve long‐term stability [29]. In our work, introducing multiple equimolar cations (Pr, Ba, La, Sr, Ca) increases configurational entropy and stabilizes the average perovskite framework, while the associated local lattice distortion lowers the oxygen‐vacancy formation energy and renders the lattice more readily reducible, thereby accelerating oxygen mobility, facilitating controlled surface reconstruction, improving fuel oxidation kinetics and decreasing polarization resistance. Simultaneously, the chemically heterogeneous A‐site suppresses deleterious surface segregation and modifies the electronic environment of Fe, mitigating stable metal‐sulfur bond formation and enabling dynamic sulfur removal pathways. Element‐specific roles are proposed as follows: Pr³⁺ enhances redox flexibility and suppresses Fe sulfidation via electronic modulation; La³⁺ stabilizes the perovskite backbone and maintains high electronic conductivity; Ba²⁺ increases surface basicity and promotes carbonate formation for carbon removal; Sr²⁺ tunes lattice strain and promotes oxygen‐vacancy generation and fast oxygen exchange; and Ca²⁺, with its smaller ionic radius, increases lattice distortion and defect density, further boosting oxygen mobility [20, 30, 31, 32]. Collectively, the high‐entropy A‐site creates a defect‐rich, chemically adaptive and redox‐responsive surface that promotes hydrocarbon activation while continuously oxidizing CH_x and sulfur intermediates, thereby delivering simultaneous performance enhancement, coking resistance, and sulfur tolerance without compromising operational stability.

In this work, a BLF‐based high‐entropy perovskite anode, Pr_0.2Ba_0.2La_0.2Sr_0.2Ca_0.2FeO_3‐δ (PBLSCF), is synthesized by introducing equimolar Pr, Ba, La, Sr, and Ca cations at the A‐site, aiming to achieve high‐entropy ceramic anodes with tolerance toward sulfur poisoning and carbon deposition. Under reducing atmospheres, PBLSCF induces the in situ exsolution of Fe nanoparticles from the perovskite lattice, forming a finely dispersed metal/oxide composite active interface that enhances fuel oxidation kinetics and multi‐foulant tolerance. To demonstrate the practical applicability of PBLSCF, electrolyte‐supported single cells are fabricated, and the electrochemical performance and long‐term stability of PBLSCF anodes are systematically evaluated under different fuels, including H₂, sulfur‐containing H₂, propane, and CH₄ (3% H₂O) mixtures. The results demonstrate that A‐site high‐entropy design combined with in situ metal exsolution is a promising synergistic strategy for achieving SOFC anodes with enhanced electrochemical performance, coking resistance, and sulfur tolerance.

2. Results and Discussion

To construct a fuel‐flexible SOFC anode with enhanced resistance to sulfur poisoning and coking, we rationally designed a high‐entropy A‐site engineered perovskite, as schematically illustrated in Figure 1a. The A‐site high‐entropy design increases lattice distortion and lowers oxygen‐vacancy formation energy, rendering PBLSCF more redox‐active and prone to controlled phase reconstruction under reducing atmospheres. This enhanced reducibility is expected to promote in situ Fe exsolution and formation of catalytically active interfaces without compromising long‐term structural integrity. As shown in Figure 1b, the high‐entropy perovskite PBLSCF exhibits a single perovskite structure in the as‐synthesized state. The XRD reflections match a cubic perovskite phase, with no observable impurity peaks. After treatment in H₂ at 800°C for 5 h, the XRD pattern shows a clear phase transformation. The new peaks can be assigned to Fe, Pr₂O₃, and Ruddlesden–Popper (R–P) layered SrPrFeO₄ phase, respectively. This behavior is consistent with recent reports on fuel‐electrode materials under reducing atmosphere at elevated temperatures, e.g., Pr_0.5Sr_0.5Cu_0.1Fe_0.8Mn_0.1O_3‐δ converts from perovskite to an R‐P phase with in situ formed FeCu₄ alloy nanoparticles under reducing conditions [33]. Thermodynamically, ABO₃ perovskites release oxygen to maintain charge neutrality under a reducing atmosphere at elevated temperatures, generating a high concentration of oxygen vacancies. When the oxygen content drops substantially, the system transforms to more stable structures. It has been reported that partial substitution of Ba²⁺ by Sr²⁺ at the A‐site in perovskites can lower the energy barrier for phase transformation and promote the formation of oxygen vacancies [34]. First, Sr‐ doping reduces the overall energy required for the perovskite to R–P phase reconstruction, and an increase in Sr content enhances the thermodynamic driving force for this structural transition. In addition, A‐site doping makes the oxygen vacancy formation energy more negative, indicating that oxygen vacancies are more easily generated under reducing atmospheres. The smaller ionic radius of Sr²⁺ (compared with Ba^{2 +}) leads to lattice contraction and changes in bond energy, which weaken the binding of oxygen ions within the lattice. Consequently, the lattice becomes more prone to oxygen loss and vacancy formation, thereby inducing structural instability and facilitating the reorganization into new phases. To further understand the lattice characteristics of the PBLSCF perovskite, its XRD diffraction peaks were compared with those of the conventional stable perovskite Ba_0.5La_0.5FeO₃ (BLF) (Figure S1). The main diffraction peaks of PBLSCF are slightly shifted to higher 2θ angles relative to BLF, indicating a smaller lattice parameter for PBLSCF. This difference can be attributed to the change in A‐site ionic radii. The average A‐site ionic radius in PBLSCF is smaller than that in BLF, due to the incorporation of smaller cations such as Ca²⁺ (1.34 Å) and Pr³⁺ (∼1.36 Å). According to the perovskite tolerance factor theory, a smaller average A‐site radius reduces the lattice parameter, resulting in peak shifts toward higher diffraction angles.

FIGURE 1.

(a) Schematic illustration of A‐site high‐entropy engineering strategy for fuel‐flexible SOFC anodes; (b) X‐ray diffraction patterns of as‐prepared PBLSCF and R‐PBLSCF reduced in H₂ at 800°C for 5 h; SEM images of (c) fresh PBLSCF electrode and (d) reduced PBLSCF electrode after electrochemical testing under H₂ conditions; (e) Enlarged HRTEM image of a reduced PBLSCF grain, with the inset showing the magnified lattice fringes of the Fe particle; (f) HAADF‐STEM image with corresponding EDS elemental mapping and line‐scan profiles of the nanoparticles; (g) In situ exsolved nanoparticles on R‐PBLSCF observed through SEM and EDS line‐scan profiles of the nanoparticles.

Figure 1c,d shows the surface morphologies of PBLSCF before and after electrochemical testing under H₂, respectively. The as‐synthesized sample shows a relatively smooth surface, whereas after reduction, numerous nanoparticles appear on the surface. These newly formed particles are uniformly distributed across the matrix, suggesting that Fe species exsolve from the perovskite lattice. To further reveal the exsolution evolution of Fe particles, the time‐dependent changes in the size and density of surface particles on PBLSCF were investigated after reduction at 800°C. As shown in Figures S2 and S3, with the reduction time increasing from 1 to 3 and 5 h, the exsolved Fe particles on the PBLSCF surface gradually grew, with average particle sizes of 48.58, 108.8, and 248.2 nm, respectively. This increase in particle size can be mainly attributed to the continuous outward diffusion of Fe species and the growth of exsolved particles under high‐temperature reducing conditions.

Figure 1e presents the high‐resolution transmission electron microscopy (HRTEM) image of the reduced sample, where a lattice spacing of 0.202 nm is clearly observed (inset), consistent with the (110) plane of metallic Fe. Figure 1f illustrates the STEM‐EDS elemental mapping of the reduced sample, indicating that Pr, Ba, La, Sr, and Ca are mainly distributed in the perovskite matrix region, while O is also concentrated in the oxide matrix. In contrast, Fe is strongly enriched at the particle surface, forming local Fe‐rich regions. The line‐scan analysis at the lower right further confirms that Fe intensity increases significantly at the surface, while Sr and Ca signals decrease, suggesting that the surface region primarily consists of exsolved metallic Fe, distinct from the A‐site element distribution in the perovskite bulk. Figure 1g presents the SEM image and corresponding EDS line‐scan analysis of the reduced PBLSCF (R‐PBLSCF) sample, further confirming the exsolution of metallic Fe on the particle surfaces.

X‐ray photoelectron spectroscopy (XPS) was carried out to investigate the surface chemical states of the R‐PBLSCF and oxidized PBLSCF (O‐PBLSCF) samples. As shown in Figure 2a, five fitted peaks are observed in the Fe 2p spectrum of the O‐PBLSCF sample, corresponding to Fe³⁺, Fe²⁺, and their associated satellite features. In contrast, a new peak at 706.70 eV appears clearly in the Fe 2p spectrum of the R‐PBLSCF sample (Figure 2b), indicating the presence of metallic Fe⁰ [35], which is consistent with the TEM and EDS observations. As shown in Table S1, the surface Fe species in O‐PBLSCF are present only as Fe^{3 +} and Fe^{2 +}, with atomic fractions of 47.5% and 52.5%, respectively, and no Fe⁰ is detected. In contrast, for R‐PBLSCF, the relative proportions of Fe^{3 +}, Fe^{2 +}, and Fe⁰ are 42.9%, 51.9%, and 5.2%, respectively. The decrease in oxidized Fe species, together with the appearance of Fe⁰ in R‐PBLSCF clearly indicates a reduction‐induced Fe exsolution process. This enhances the redox adaptability of the PBLSCF surface under reducing conditions and makes a small fraction of the Fe more prone to selective reduction into metallic Fe⁰. Meanwhile, the relatively low Fe⁰ content suggests that although the high‐entropy effect promotes Fe exsolution, the majority of Fe in the parent oxide remains in the oxidized state.

FIGURE 2.

(a) Fe 2p XPS spectrum of O‐PBLSCF; (b) Fe 2p XPS spectrum of R‐PBLSCF; (c) O 1s XPS spectrum of O‐PBLSCF; (d) H₂‐TPR profiles of PBLSCF and BLF; (e) TGA curves of PBLSCF and BLF samples, measured under 4% H₂/N₂ from RT to 900°C; (f) N₂ adsorption–desorption isotherms and BET surface areas of O‐PBLSCF and R‐PBLSCF; (g) Electrochemical impedance spectra of symmetric cells with PBLSCF electrodes under humidified H₂ at 700–800°C; (h) TEC of PBLSCF and BLF from RT to 1000°C; (i) Electrical conductivity of BLF and PBLSCF as a function of temperature in 10% H₂/N₂.

O 1s spectra of O‐PBLSCF and oxidized BLF (O‐BLF) are presented in Figure 2c and Figure S4. Each O 1s spectrum can be deconvoluted into four characteristic components, assigned to lattice oxygen (O²⁻, 528.50 eV), adsorbed oxygen species (O⁻/O₂ ²⁻, 530.30 eV), hydroxyls or carbonates (OH⁻/CO₃ ²⁻, 532.30 eV), and physisorbed water (H₂O, 534.53 eV) [36, 37, 38]. According to the fitted results summarized in Table S2, the proportion of adsorbed oxygen in O‐PBLSCF reaches 17.67%, notably higher than the 9.53% observed for O‐BLF. This indicates that O‐PBLSCF possesses a higher concentration of oxygen vacancies and more surface active sites available for oxygen adsorption [39]. This enhancement is consistent with the A‐site high‐entropy design, where multi‐cation size and charge disorder introduce significant local lattice distortion and broaden the oxygen binding‐energy landscape, thereby lowering the oxygen‐vacancy formation energy. Moreover, the chemically heterogeneous A‐site environment suppresses cation ordering and promotes defect stabilization near the surface, leading to a higher population of reactive adsorbed oxygen species. The differences in oxygen‐vacancy concentration and surface reactive oxygen species further lead to distinct catalytic behaviors between the two materials.

Figure 2d presents the H₂ temperature‐programmed reduction (H₂‐TPR) profiles of two perovskite materials measured in a 10% H₂/Ar from room temperature to 900°C. In the temperature range of 200°C–500°C, two distinct reduction peaks are detected at 340.79°C and 436.03°C for BLF and PBLSCF, respectively, corresponding to the reduction of Fe³⁺ to Fe²⁺ [40]. It is noteworthy that a subsequent reduction peak appears at approximately 750°C, which can be ascribed to the reduction of Fe²⁺ to metallic Fe⁰, representing the temperature range for the exsolution of Fe nanoparticles [39]. Compared to BLF, PBLSCF exhibits more pronounced reduction peaks and quantitative peak fitting and integration further support this interpretation. The integrated H₂ consumption of PBLSCF is higher than that of BLF, indicating a greater overall reducible oxygen inventory and enhanced redox activity of the high‐entropy material. Notably, the reduction onset temperature of BLF is slightly lower than that of PBLSCF, suggesting that the high‐entropy A‐site configuration improves the intrinsic phase stability of the perovskite lattice against premature reduction. Therefore, PBLSCF exhibits a desirable combination of structural robustness (higher onset temperature) and enhanced redox capacity (larger peak area). This behavior is fully consistent with our hypothesis that A‐site high‐entropy engineering creates a redox‐adaptive but structurally stable lattice, which facilitates controlled Fe exsolution and oxygen‐vacancy generation while maintaining bulk integrity. The TPR results also correlate well with the XPS, O₂‐TPD, and electrochemical measurements discussed below, collectively confirming the synergistic role of the high‐entropy design in promoting both activity and durability.

Consistent with this, the O₂‐TPD results (Figure S5) show that PBLSCF reveals a pronounced oxygen desorption peak in the range of approximately 300°C–500°C, whereas BLF releases oxygen only at higher temperatures. Oxygen desorption in the medium‐temperature range is typically attributed to the migration of active lattice oxygen species located on or near the surface [41]. The strong low‐temperature oxygen desorption of PBLSCF indicates the presence of abundant oxygen vacancies and highly mobile lattice oxygen. In other words, PBLSCF possesses higher oxygen mobility and a greater amount of accessible surface oxygen compared with BLF, which is closely related to its superior ability to rapidly form and refill oxygen vacancies. This behavior is consistent with the A‐site high‐entropy design, where multi‐cation size and charge mismatch introduce pronounced local lattice distortion and a broadened oxygen binding‐energy distribution, effectively lowering the average oxygen‐vacancy formation energy. In addition, the chemically heterogeneous A‐site environment disrupts long‐range cation ordering and promotes defect percolation pathways, facilitating faster bulk‐surface oxygen exchange. This enhanced oxygen mobility and surface oxygen activity are particularly beneficial under hydrocarbon‐fuel environments such as propane. When hydrocarbon intermediates or carbon deposits form on the anode surface, the active lattice oxygen in PBLSCF can oxidize them, thereby preventing carbon accumulation. Indeed, perovskite oxides with fast oxygen migration capabilities have been demonstrated to effectively suppress carbon deposition by supplying lattice oxygen to oxidize residual carbon species [34].

Figure 2e shows the thermogravimetric analysis (TGA) curves of the samples measured from room temperature to 900°C under a 4% H₂/N₂ atmosphere, exhibiting different degrees of weight loss. The slight weight loss of PBLSCF below approximately 300°C can be attributed to the desorption of adsorbed water. As the temperature further increases, the weight loss of BLF powder reaches approximately 1.76%, while that of PBLSCF is 1.83%. This may be due to the change in the oxidation state of Fe at high temperatures, resulting in a greater increase in oxygen vacancies in PBLSCF and thus leading to weight loss. This result is consistent with the H₂‐TPR findings, confirming that more Fe species are reduced under a reducing atmosphere. In the temperature range of 500–900°C, the weight loss of BLF is 0.51%, whereas that of PBLSCF is 1.14%. The sharp drop in the curve is associated with the reduction of Fe species to metallic Fe, which leads to a decrease in oxygen content.

According to the Brunauer–Emmett–Teller (BET) analysis (Figure 2f), after thermal treatment in a hydrogen atmosphere, the specific surface area of the PBLSCF sample increased from 0.95 to 2.30 m² g⁻¹. This increase in surface area originated from the formation of Fe nanoparticles, which provide abundant active sites for the propane oxidation reaction, thereby enhancing the catalytic activity of the SOFC anode. Figure 2g shows the electrochemical impedance spectroscopy (EIS) results, demonstrating that PBLSCF exhibits superior electrochemical performance compared with BLF under humidified H₂ condition. At 800°C, 750°C, and 700°C, the polarization resistances (R_p) of PBLSCF are 0.143, 0.250, and 0.495 Ω cm², respectively, which are significantly lower than those of BLF (0.176, 0.274, and 0.536 Ω cm², Figure S6). Furthermore, under air atmosphere, PBLSCF exhibits even lower resistances of 0.038, 0.063, and 0.129 Ω cm² at 800°C, 750°C, and 700°C (Figure S7), respectively, further confirming its excellent oxygen reduction reaction (ORR) activity and good mixed ionic‐electronic conductivity.

In the temperature range of 200°C–500°C, PBLSCF shows a clearly lower expansion rate than BLF in Figure 2h. The average linear TEC of PBLSCF is approximately 10.1 × 10⁻⁶ K⁻¹, compared to 14.1 × 10⁻⁶ K⁻¹ for BLF. In the higher temperature range of 500°C–1000°C, the average TEC of PBLSCF is about 16.2 × 10⁻⁶ K⁻¹, lower than of BLF (18.8 × 10⁻⁶ K⁻¹). This reduction can be attributed to the multi‐cation size and mass disorder at the A‐site, which introduces local lattice strain and strong lattice distortion that stiffen the average metal‐oxygen framework and suppress cooperative lattice expansion. In addition, the high configurational entropy tends to flatten the lattice anharmonicity and broaden phonon scattering, effectively lowering the thermal expansion response. The presence of smaller‐radius cations (e.g., Ca²⁺ and Pr³⁺) also reduces the average A‐site ionic radius and tolerance factor, further constraining lattice dilatation at elevated temperatures.

Under a 10% H₂/N₂ reducing atmosphere (Figure 2i), the electrical conductivities of both materials increase markedly with increasing temperature, exhibiting a trend opposite to that observed in air. In air (Figure S8), both samples display typical p‐type conduction behavior, where conductivity gradually decreases with rising temperature due to thermally induced lattice oxygen release and the consequent reduction in hole concentration. Although a strongly reducing environment decreases the p‐type carrier concentration by generating oxygen vacancies and lowering the valence state of transition‐metal cations, thermally activated small‐polaron hopping becomes dominant at elevated temperatures, leading to an overall increase in conductivity with temperature. Notably, PBLSCF maintains higher conductivity than BLF under both atmospheres. Under reducing conditions at 800°C, PBLSCF achieves a conductivity of 1.04 S cm⁻¹, which is significantly higher than that of BLF (0.43 S cm⁻¹). The higher conductivity of PBLSCF could be attributed to the A‐site high‐entropy effect, which introduces lattice distortion and defect‐mediated electronic pathways that facilitate small‐polaron hopping between Fe cations.

To further substantiate the role of A‐site high‐entropy engineering in regulating lattice distortion and oxygen defect energetics, spin‐polarized DFT+U calculations were performed using relaxed periodic bulk models of BLF and PBLSCF. Both compositions were represented using special quasi‐random structure (SQS) [42] supercells to approximate the A‐site cation distributions, with the dashed boxes in Figure 3a,b indicating the periodic simulation cell boundaries. Detailed computational parameters, including the Hubbard U values applied to Fe 3d and Pr 4f states, are provided in the Supporting Information. The projected density of states (PDOS) of BLF and PBLSCF are shown in Figure 3c,d. In both systems, the electronic states near the Fermi level are mainly associated with Fe 3d and O 2p orbitals, indicating that the Fe–O framework dominates the redox‐active electronic structure. In PBLSCF, Pr 4f states are also present, reflecting the contribution of Pr‐containing A‐site chemistry to the overall electronic structure.

FIGURE 3.

DFT analysis of lattice distortion and oxygen‐vacancy formation in BLF and PBLSCF. Relaxed bulk structures of (a) BLF and (b) PBLSCF. The dashed boxes indicate the periodic cell boundaries; spin‐polarized projected density of states (PDOS) of (c) BLF and (d) PBLSCF, showing Fe 3d, O 2p, and Pr 4f contributions; The Fermi level is set to 0 eV. Distributions of oxygen‐vacancy formation energies for (e) BLF and (f) PBLSCF, with mean values of 1.09 and 0.86 eV, respectively. Distributions of Fe─O─Fe bond angles for (g) BLF and (h) PBLSCF, with mean values of 172.1° and 167.1°, respectively.

To avoid bias from selecting a specific local oxygen environment, the oxygen‐vacancy formation energies were calculated for all oxygen sites in the SQS supercells, with the site‐resolved values and nearest‐neighbor A‐site classifications summarized in Tables S3 and S4. As shown in Figure 3e,f, PBLSCF exhibits a lower average oxygen‐vacancy formation energy than BLF (0.86 vs 1.09 eV) and, importantly, a broader distribution of vacancy formation energies. This indicates that the chemically diverse A‐site environment not only makes oxygen vacancies thermodynamically more accessible on average, but also broadens the oxygen‐binding‐energy landscape and creates locally favorable sites for oxygen‐vacancy formation. To further quantify the associated structural distortion, the Fe─O─Fe bond‐angle distributions were analyzed. The average Fe─O─Fe bond angle decreases from 172.1° in BLF to 167.1° in PBLSCF (Figure 3g,h), confirming a more distorted Fe─O─Fe network in PBLSCF. Such distortion, induced by the size and charge mismatch among the five A‐site cations, weakens the uniformity of Fe–O bonding and facilitates oxygen removal from selected local environments. These DFT results provide direct computational support that A‐site high‐entropy engineering facilitates oxygen‐vacancy formation in PBLSCF, leading to a higher oxygen‐vacancy concentration and more redox‐active lattice behavior, consistent with the XPS and O₂‐TPD results.

The electrochemical performance of the electrolyte‐supported single cell PBLSCF|La_0.8Sr_0.2Ga_0.83Mg_0.17O_3‐δ (LSGM)|La_0.6Sr_0.4Co_0.2Fe_0.8O_3‐δ (LSCF) with PBLSCF as the anode was evaluated. Figure S10 shows the cross‐sectional SEM image of the cell, where the porous PBLSCF anode is tightly bonded to the LSGM electrolyte without any delamination, indicating good interfacial compatibility. As shown in Figure 4a,b, the PBLSCF cell achieved excellent PPDs of 1.22, 0.838, 0.508, and 0.288 W cm⁻² at 800°C, 750°C, 700°C, and 650°C under humidified H₂, respectively. In comparison, the BLF cell delivered PPD values of 0.963, 0.646, 0.394, and 0.204 W cm⁻² under the same conditions. The superior performance of the PBLSCF cell can be attributed to its overall low resistance. As shown in Figure S11, the R_p values of the PBLSCF cell were 0.165, 0.215, 0.372, and 0.716 Ω cm² at 800°C, 750°C, 700°C, and 650°C, respectively. In contrast, the BLF single cell exhibited a R_p of 0.302 Ω cm² at 750°C. To further elucidate the electrode reaction mechanisms, a distribution of relaxation time (DRT) analysis was conducted based on the EIS results (Figure S12). According to the frequency domains, the responses can be categorized into low‐frequency (LF), intermediate‐frequency (IF), and high‐frequency (HF) regions, which are generally attributed to gas diffusion processes (adsorption or dissociation), surface exchange reactions, and oxygen ion transport across the electrode/electrolyte interface, respectively [43, 44]. The results demonstrate that in the LF region, PBLSCF exhibits a reduced resistance, indicating enhanced gas transport capability and improved adsorption/dissociation kinetics. In the IF region, PBLSCF also shows a lower resistance, suggesting accelerated surface reaction kinetics.

FIGURE 4.

I‐V‐P curves of (a) BLF and (b) PBLSCF cells measured under humidified H₂ conditions; I‐V‐P curves of (c) BLF and (d) PBLSCF cells measured under humidified 50 ppm H₂S‐H₂ condition; (e) SEM image of the tested R‐PBLSCF anode/electrolyte interface with exsolved Fe nanoparticles; (f) Electrochemical performance comparison at 800°C with the reported high‐performance anodes; (g) H₂O‐TPD profile of PBLSCF corresponding to different water‐related species; (h) Schematic illustration of the hydroxyl‐assisted sulfur removal mechanism on the PBLSCF anode; (i) Long‐term stability test of the PBLSCF single cell operated at a current density of 0.2 A cm⁻² in humidified H₂; (j) Stability test of the PBLSCF single cell operated at a current density of 0.15 A cm⁻² in humidified 50 ppm H₂S‐H₂.

The performance of single cells with different anodes was evaluated using a fuel containing 50 ppm H₂S‐H₂. As shown in Figure 4c, the BLF‐based cell exhibited lower PPD values of 0.801, 0.487, 0.294, and 0.135 W cm⁻² at 800, 750°C, 700°C, and 650°C, respectively. In contrast, the PBLSCF‐based cell (Figure 4d) delivered PPDs of 1.07, 0.762, 0.406, and 0.213 W cm⁻² under the same conditions. Compared with the results in pure H₂, both anodes showed a slight decrease in power density and an increase in R_p (Figures S13 and S14), indicating a certain degree of sulfur poisoning. As shown in Figure 4e, nanoparticles are observed after performance test and are attributed to exsolved metallic Fe.

To further demonstrate the superior electrochemical performance of the PBLSCF anode, a conventional pure‐phase SFM single cell was introduced for comparison (Figure S15). As shown in Figure S16, the cells using the SFM‐fuel electrode, fabricated and evaluated under the same conditions as the cells using the PBLSCF‐fuel electrode, delivered PPDs of 0.771, 0.523, 0.281, and 0.143 W cm⁻² at 800°C–650°C, with corresponding R_p values of 0.257, 0.358, 0.579, and 1.15 Ω cm², respectively. When operated in 50 ppm H₂S‐H₂, the PPD at 800 °C was 0.654 W cm⁻². The PPD was also compared with those of various anode materials reported in recent literature (Figure 4f and Table S5). Among the systems evaluated, the PBLSCF anode exhibits a relatively high PPD, highlighting the excellent electrochemical performance of PBLSCF as an anode material.

For some perovskite‐based anode materials, certain B‐site transition metal oxides tend to be reduced and exsolved as metallic particles such as Fe, Co, or Ni under a reducing atmosphere. The exsolved Fe⁰ particles can readily react with H₂S to form iron sulfides (e.g., FeS), which deposit on the active sites of the anode surface and consequently lead to performance degradation [20]. The main reaction mechanism can be described as follows:

$${\mathrm{H}}_2\mathrm{S}+\mathrm{Fe}\to \mathrm{FeS}+{\mathrm{H}}_2$$ (1)

On the one hand, the formation of FeS through the exsolution of Fe from the perovskite lattice indicates that metallic Fe is no longer available to catalyze H₂ oxidation. On the other hand, the poor conductivity of FeS and the sulfur‐covered layer hinders the reaction between the fuel gas and the anode surface. The alkaline‐earth metals at the A‐site may also undergo side reactions with sulfur under reducing conditions, further aggravating anode poisoning. For example, Sr, Ca, or Ba may exsolve from the surface as their oxides (such as SrO, CaO, or BaO) and react with H₂S to form the corresponding sulfides [45]. Taking Sr as an example:

$$\mathrm{SrO}+{\mathrm{H}}_2\mathrm{S}\to \mathrm{SrS}+{\mathrm{H}}_2\mathrm{O}$$ (2)

These alkaline‐earth metal sulfides are generally electrical insulators, and their deposition on the anode surface further hinders electrochemical reactions, deteriorating the electrical conductivity and catalytic activity of the anode.

The PBLSCF anode incorporates five equimolar A‐site cations (Pr, Ba, La, Sr, and Ca), forming a high‐entropy multi‐component composition. It has been reported that such multi‐cation doping can effectively suppress surface segregation and secondary phase formation of single components (such as Sr or Ba) through the entropy stabilization effect [46]. Moreover, studies have shown that partial substitution of La with Pr at the A‐site prevents the formation of FeS and SrS phases in 50 ppm H₂S‐H₂, whereas these sulfides were detected in undoped or La‐only doped samples [20]. This indicates that doping can modify the chemical environment of Fe (valence state and bonding energy), significantly enhancing the resistance to sulfidation.

In addition, some studies have pointed out that the hydrophilicity of perovskite anodes plays a key role in their sulfur‐tolerant mechanism [10, 47]. A highly hydrophilic anode surface facilitates the adsorption and dissociation of water molecules at oxygen vacancies, generating surface hydroxyl species that serve as active oxygen sources for H₂S oxidation. To evaluate the water adsorption and dissociative hydration capability of PBLSCF, H₂O‐TPD measurement was carried out. As shown in Figure 4g, the H₂O‐TPD profile of PBLSCF shows three desorption regions centered at approximately 219.3°C, 328.1/428.6°C, and 669.3°C, which can be assigned to molecularly adsorbed water (Type I), dehydration of structural hydroxyl clusters (Type II), and strongly bonded hydroxyl species (Type III), respectively [48]. In particular, the pronounced high‐temperature desorption peak at 669.3°C indicates that H₂O on the PBLSCF surface does not merely exist as weakly physisorbed species, but can undergo dissociative adsorption to form thermally stable surface hydroxyl species. Meanwhile, as shown in Figure 5e, the in situ diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS) display a distinct hydroxyl‐group stretching vibration band in the 3200–3700 cm⁻¹ region, which further confirms the formation of surface O‐H species under hydration conditions [49].

FIGURE 5.

I‐V‐P curves of (a) BLF and (b) PBLSCF cells measured under humidified C₃H₈ conditions; (c) Long‐term stability test of the PBLSCF single cell operated at a current density of 0.15 A cm⁻² in humidified C₃H₈; (d) DRT plots of BLF and PBLSCF electrodes in humidified C₃H₈ at 750°C; in situ DRIFTS spectra of PBLSCF under humidified propane conditions at (e) RT to 400°C and (f) 400°C to 700°C; (g) Schematic illustration of the propane conversion and coking resistance mechanism over the PBLSCF electrode; (h) In situ Raman spectra of PBLSCF at 700°C under humidified C₃H₈ for various exposure times.

These water‐derived surface hydroxyl species are closely associated with the proposed sulfur‐tolerance mechanism of PBLSCF. DFT calculations indicate that PBLSCF has a low oxygen vacancy formation energy, suggesting that oxygen vacancies can be readily generated on its surface. This feature facilitates H₂O adsorption and dissociation, thereby promoting the formation of surface hydroxyl species. In a sulfur‐containing atmosphere, H₂S may first interact with metal active sites on the anode surface, forming adsorbed sulfur species. These hydroxyl species, together with lattice oxygen, may further promote the oxidation of adsorbed sulfur intermediates and alleviate their accumulation on Fe‐related active sites. Therefore, the water activation capability, oxygen‐vacancy‐regulated surface chemistry, and lattice oxygen mobility of PBLSCF may synergistically participate in the transformation of sulfur species, thereby mitigating the electrochemical performance degradation induced by sulfur poisoning. As shown in Figure 4h, the detailed processes are as follows [50]:

$${\mathrm{H}}_2\mathrm{O}+{\mathrm{O}}_{\mathrm{O}}^{\mathrm{x}}+{\mathrm{V}}_{\mathrm{O}}^{··}\to 2{(\mathrm{OH})}_{\mathrm{O}}^{·}$$ (3)

$${\mathrm{H}}_2\mathrm{S}+\mathrm{alloy}\to {\mathrm{S}}_{\mathrm{ads}}^{\ast }+{\mathrm{H}}_2$$ (4)

$${\mathrm{S}}_{\mathrm{ads}}^{\ast }+2{(\mathrm{OH})}_{\mathrm{O}}^{·}\to \mathrm{S}{\mathrm{O}}_2+{\mathrm{H}}_2\mathrm{O}+2{\mathrm{e}}^{-}$$ (5)

$$\mathrm{S}{\mathrm{O}}_2+{\mathrm{O}}^{2-}+2{(\mathrm{OH})}_{\mathrm{O}}^{·}\to {\mathrm{SO}}_4^{2-}+{\mathrm{H}}_2\mathrm{O}$$ (6)

As shown in Figure 4i, the PBLSCF cell operated continuously for 1000 h at a constant current density of 0.20 A cm⁻² in pure H₂, with almost no voltage degradation, demonstrating excellent electrochemical and structural stability. In 50 ppm H₂S‐H₂ (Figure 4j), the cell maintained stable operation for 97 h at 700 °C with an average voltage degradation rate of 0.78 mV h⁻¹, demonstrating promising sulfur‐tolerant stability of the PBLSCF anode. Figure S17 shows that the PBLSCF anode maintained a porous structure after stability testing, with surface nanoparticles attributed to exsolved Fe species.

Figure 5 evaluates the electrochemical performance of two different anodes using propane as the fuel. As shown in Figure 5a,b, under humidified propane, the BLF cell shows PPDs of 0.209 and 0.075 W cm⁻² at 750°C and 700°C, respectively. In contrast, the PBLSCF cell delivers superior performance, achieving PPDs of 0.535 and 0.221 W cm⁻² under the same conditions. As illustrated in Figure S18, the PBLSCF single cell shows a lower R_p than BLF at all tested temperatures. The comparison of the PBLSCF anode operated at 750°C under three different fuel atmospheres is shown in Figure S19. As shown in Figure 5c, a long‐term stability test was conducted at 700°C under humidified propane at 0.15 A cm⁻². After continuous operation for 600 h, the cell voltage remains stable throughout the test. Its stability was compared with that of electrolyte‐supported SOFCs fueled with propane or ethane reported in the literature (Table S6). The electrochemical stability of the PBLSCF anode is superior to that of most electrolyte‐supported SOFCs operated on propane or ethane. In Figure 5d, under C₃H₈ fuel conditions, PBLSCF exhibits reduced peak intensities in both the LF and IF regions of the DRT compared with BLF. The decrease in the LF response indicates improved gas diffusion as well as enhanced adsorption and dissociation kinetics of propane molecules, while the reduced IF peak reveals accelerated surface exchange and interfacial reaction kinetics. This synergistic enhancement is mainly attributed to the lattice disorder and strain effects induced by A‐site high‐entropy doping, which promotes oxygen vacancy formation and enhances oxygen ion mobility, thereby improving surface activity and oxygen supply capability.

As mentioned earlier, PBLSCF possesses a higher concentration of oxygen vacancies. These oxygen vacancies enhance oxygen‐ion transport and promote the reactivity of surface active oxygen species. This vacancy enrichment is consistent with the A‐site high‐entropy effect, where multi‐cation size and charge disorder lower the oxygen‐vacancy formation energy and creates a broader distribution of oxygen binding strengths, as supported by the DFT, XPS and O₂‐TPD results discussed above. BET analysis shows that the partial exsolution of Fe after reduction increases the specific surface area of PBLSCF, exposing more active sites and facilitating the adsorption and oxidation of propane. The facilitated Fe exsolution behavior also agrees with the redox‐adaptive nature revealed by H₂‐TPR, confirming that the high‐entropy lattice promotes controlled surface reconstruction rather than bulk instability.

As shown in Figure 5e,f, in situ DRIFTS measurements were carried out using humidified propane as the fuel from RT to 700°C. Several intermediate species were identified in the infrared spectra, including CHO* species, monodentate and bidentate carbonates, as well as absorption bands assigned to olefins produced from propane dehydrogenation. The rich surface chemistry observed here is consistent with the defect‐rich and chemically heterogeneous surface created by A‐site high‐entropy engineering. Propane is first adsorbed and activated on the anode surface. As shown in Figure 5e, from room temperature to approximately 150–200°C, the spectra are dominated by the C–H stretching vibrations of propane located at 3000–2800 cm⁻¹. Meanwhile, the absorption feature observed in the 3200–3700 cm⁻¹ region confirms the presence of hydroxyl species under humidified conditions. With increasing temperature in Figure 5f, the intensity of the C–H stretching band at 2967 cm⁻¹ gradually decreases, indicating enhanced consumption and conversion of propane at elevated temperatures. Meanwhile, olefinic peaks appear at 922 and 974 cm⁻¹. As the temperature increases, the degree of dehydrogenation becomes stronger, leading to higher signals.

During the dehydrogenation process, smaller hydrocarbon fragments (denoted as CH_x) may remain on the surface. These are potential precursors for carbon deposition. The high‐entropy perovskite anode provides active oxygen species that promote the oxidation of CH_x fragments. This behavior directly correlates with the enhanced lattice‐oxygen mobility identified by O₂‐TPD and the lower oxygen‐vacancy formation energy revealed by DFT calculations. On one hand, H₂O in the fuel gas may dissociate on the surface to generate hydroxyl groups as shown in Figure 5e; on the other hand, lattice oxygen from the perovskite can also react with CH_x fragments, partially oxidizing them to form oxygen‐containing CHO* species at 1720 cm⁻¹. The detection of CHO* in the DRIFTS spectra confirms that this step occurs [37, 51, 52]. The presence of CHO* indicates that carbon fragments produced from propane cracking do not polymerize into solid carbon, but are instead captured by oxygen species supplied by the anode, forming CH‐O bonded intermediates. This reflects the oxidative removal function of the PBLSCF anode.

The generated CO₂ is further stabilized on the surface as carbonate species as shown in Figure 5f. Because the anode contains multiple alkaline and rare‐earth A‐site cations, CO₂ is readily adsorbed to form carbonate species [51, 53]. The strengthened carbonate signals are consistent with the increased surface basicity introduced by Ba²⁺/Sr²⁺/Ca²⁺ in the high‐entropy A‐site, which provides diverse adsorption environments for CO₂ capture. The observed monodentate and bidentate carbonates show that CO₂ is anchored to different surface coordination environments. The formation of these carbonates serves as a temporary storage form of CO₂, preventing its further reduction into solid carbon. In summary, the ability of the high‐entropy perovskite anode to sustain this reaction cycle arises from its intrinsic material characteristics. The B‐site transition metal Fe provides redox‐active sites for C–H bond cleavage and oxidation reactions, while the A‐site high‐entropy design offers alkaline sites that stabilize carbonate intermediates and enhances thermal stability and oxygen transport properties. Together with the TPR, XPS, O₂‐TPD, and conductivity results, these findings consistently verify that A‐site high‐entropy engineering creates a redox‐adaptive, defect‐rich surface that enables continuous carbon removal. Through the synergistic contribution of these multiple elements, the anode exhibits strong resistance to carbon deposition and maintains stable performance. Figure 5g schematically illustrates the propane conversion and coking resistance mechanism of the PBLSCF electrode.

Figure 5h shows the in situ Raman spectra of the PBLSCF electrode recorded at 700°C under a humidified propane atmosphere with different exposure times. The bands in the range of 500–1000 cm⁻¹ are assigned to the lattice vibrations of PBLSCF and its reduced R‐P derivative phase. With increasing exposure time to humidified propane, the positions and intensities of the main peaks remain almost unchanged, indicating that the bulk structure of PBLSCF is highly stable in the reducing hydrocarbon atmosphere. This structural robustness is consistent with the entropy‐stabilized A‐site framework, which maintains the average perovskite integrity even while enabling local redox adaptation observed in the H₂‐TPR results. In the range of 1200–1700 cm⁻¹, where the D band (∼1350 cm⁻¹) and the graphitic G band (∼1580 cm⁻¹) of carbon species are expected, no discernible peaks are observed even after prolonged propane exposure. This clearly demonstrates the excellent coking resistance of the PBLSCF electrode in humidified propane at 700°C. To further verify the coking resistance of the PBLSCF anode under propane fuel conditions, TG analysis was performed in air on the PBLSCF powder after propane treatment. As shown in Figure S20, the sample mass remains stable in the low‐temperature region, followed by a noticeable mass gain in the range of approximately 250°C–370°C. This is mainly attributed to the re‐oxidation of exsolved Fe particles in the air atmosphere. As the temperature increases, the sample mass shows a gradual decreasing trend, which may be related to the increase in oxygen vacancies at high temperatures [17]. Notably, the TG curve shows no distinct weight loss from carbon oxidation, suggesting that carbon accumulation on PBLSCF is negligible. The absence of carbon signatures correlates well with the enhanced lattice‐oxygen activity and surface basicity identified from O₂‐TPD and DRIFTS analyses, both of which originate from the A‐site high‐entropy design. Such behavior is likely related to the surface reconstruction of PBLSCF under reducing conditions. The reconstructed surface, consisting of an R‐P phase together with exsolved Fe nanoparticles, provides a high concentration of mobile lattice oxygen and fast oxygen‐transport pathways, enabling rapid oxidation of adsorbed CH_x intermediates and thus suppressing their further polymerization into solid carbon. Importantly, the high‐entropy lattice promotes this controlled reconstruction by lowering the oxygen‐vacancy formation energy while suppressing detrimental phase segregation, thereby creating a redox‐adaptive yet structurally stable catalytic interface.

In conventional oxide‐ion‐conducting fuel cells such as PBLSCF|LSGM|LSCF, PBLSCF mainly serves as the anode for electrochemical oxidation of the fuel. In contrast, in a proton‐conducting cell configuration (PBLSCF|Ni+BaZr_0.4Ce_0.4Y_0.1Yb_0.1O_3‐δ(BZCYYb)| BZCYYb|BaCo_0.4Fe_0.4Zr_0.1Y_0.1O_3‐δ (BCFZY)) using CH₄ (3% H₂O) as the fuel, PBLSCF is employed as a separate catalytic layer, emphasizing its catalytic activity beyond electrochemical function. As shown in Figure 6a, the cross section of the single cell shows a porous electrode and a dense electrolyte layer about 10 µm thick. The BCFZY sample can form a pure perovskite phase (Figure S21). Figure 6b illustrates that the R‐PBLSCF catalytic layer is well bonded to the Ni‐BZCYYb anode without visible delamination. Figure 6c shows the surface morphology of R‐PBLSCF after hydrogen reduction, where in situ formed Fe particles are observed. These Fe enhance humidified CH₄ reforming activity and effectively suppress coking [44].

FIGURE 6.

(a) Cross‐sectional microstructure of a single cell with BCFZY air electrode; (b) The microstructure of the anode‐support/catalytic‐layer interface; (c) The microstructure of the R‐PBLSCF electrode after hydrogen reduction; I‐V‐P curves of the single cells (d) without the anode catalytic layer and (e) with PBLSCF as the anode catalytic layer measured under humidified H₂; (f) EIS spectra of the single cells without and with PBLSCF anode catalytic layer; (g) I‐V‐P curves of the single cells with PBLSCF as the anode catalytic layer measured under CH₄ (3% H₂O); (h) Durability test of single cells without and with the anode catalytic layer at 600°C and 0.15 A cm⁻²; (i) Raman spectrum of the PBLSCF anode catalytic layer after stability testing.

Figure 6d,e shows the performance of proton‐conducting fuel cells (PCFCs) with and without a PBLSCF catalytic layer under H₂ (3% H₂O). The cell with the PBLSCF layer exhibited higher power density. At 750–600°C, the PPDs of the PBLSCF‐modified cell reaches 1.339, 1.129, 0.910, and 0.681 W cm⁻², respectively. In contrast, the corresponding values for the cell without the catalytic layer are 1.044, 0.892, 0.719, and 0.531 W cm⁻². The Fe‐containing nanoparticles offer a high surface area and provide abundant active sites for H₂ adsorption and dissociation. The EIS results (Figure 6f) further confirm that the PBLSCF‐modified PCFC shows a notably lower polarization resistance. For example, at 650°C, the PBLSCF‐modified PCFC exhibits an R_p of 0.086  Ω cm², which is only 64% of the value for the unmodified cell (0.134 Ω cm²). The increased number of active sites reduces the R_p by enabling more reactants to participate in the electrochemical reaction.

At high temperatures, methane can undergo thermal decomposition to form carbon and hydrogen, as well as carbon deposition through the disproportionation of CO. These reactions can lead to carbon accumulation on the electrode surface. One effective strategy to suppress carbon deposition on the anode is to add a small amount of steam to the methane fuel. Introducing a low concentration of water vapor (e.g., 3% H₂O) provides the oxidant needed for internal reforming, partially converting CH₄ into H₂ and CO via steam reforming (CH₄ + H₂O → CO + 3H₂). While adding steam helps prevent carbon buildup, the steam content should not be too high. Excessive water (high‐humidity fuel) can dilute the fuel, lower its heating value, and increase system complexity. A water content of around 3% H₂O has been shown in practice to effectively suppress carbon deposition on Ni‐based anodes without significantly reducing system efficiency [22]. In addition, at high temperatures, steam can react with solid carbon through the gasification reaction (C + H₂O → CO + H₂), which is considered the main mechanism by which water helps remove or prevent carbon deposition.

The electrochemical performance of the single cell with a catalytic layer was evaluated under CH₄ (3% H₂O) fuel conditions, as shown in Figure 6g. The PPDs reached 0.677, 0.461, 0.293, and 0.159 W cm⁻² at 750°C, 700°C, 650°C, and 600°C, respectively, indicating good catalytic activity for methane utilization at intermediate temperatures. Figure 6h presents a long‐term stability test at 600°C under 0.15 A cm⁻² (CH₄‐3% H₂O). Without the catalyst, the cell exhibited an initial voltage of ∼0.629 V, which rapidly declined within a short period, suggesting severe performance degradation. This instability is likely due to insufficient methane reforming activity at 600°C, leading to increased anode polarization and possibly carbon deposition. In contrast, with the catalyst layer, the cell maintained a stable voltage over 250 h of operation, with only minor fluctuations. Figure 6i shows the ex situ Raman spectrum of the anode surface after the stability test (1200–1800 cm⁻¹). No noticeable D‐band (∼1350 cm⁻¹) or G‐band (∼1580 cm⁻¹) signals were observed, indicating the absence of detectable disordered or graphitized carbon species. The absence of carbon accumulation supports the conclusion that the catalyst effectively prevents carbon formation, contributing to the long‐term stability observed in the electrochemical performance.

3. Conclusion

In this study, a high‐entropy perovskite anode material, PBLSCF, was developed by introducing A‐site high‐entropy doping into the BLF framework. Under reducing conditions, PBLSCF partially transformed into a R‐P phase and formed in situ Fe nanoparticles. Compared to BLF, PBLSCF exhibited higher oxygen vacancy concentration, faster oxygen ion mobility, and improved electrical conductivity, confirming its superior catalytic activity. Electrolyte‐supported cells using PBLSCF anode showed PPDs of 0.838, 0.762, and 0.535 W cm⁻² at 750°C under humid H₂, 50 ppm H₂S‐H₂, and C₃H₈ atmospheres, respectively, and demonstrated excellent stability at 700°C. Furthermore, in situ Raman measurements under humidified propane revealed no detectable D or G bands even after 30 h of operation, providing excellent coking resistance of PBLSCF. The performance gains originate from the unique role of A‐site high‐entropy engineering. The multi‐cation size and charge disorder introduce pronounced lattice distortion and a broadened oxygen binding‐energy distribution, which lowers oxygen‐vacancy formation energy and promotes rapid oxygen transport. At the same time, the entropy‐stabilized but redox‐adaptive lattice suppresses detrimental cation segregation while enabling controlled Fe exsolution and dynamic surface reconstruction under reducing conditions. This cooperative effect creates a defect‐rich, chemically heterogeneous surface capable of efficiently activating hydrocarbons while continuously oxidizing CH_x and sulfur intermediates. As a result, PBLSCF demonstrates outstanding resistance to both sulfur poisoning and carbon deposition across multiple fuel environments. This work establishes A‐site high‐entropy engineering as a powerful and generalizable strategy to simultaneously achieve high activity, structural robustness, and multi‐foulant tolerance in next‐generation fuel‐flexible SOFC anodes.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File: adma74091‐sup‐0001‐SuppMat.docx.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.