Eliminating the reliance on platinum group metals-based electrocatalysts to fulfil high activity and stability is crucial for the large-scale application of proton-exchange membrane fuel cells (PEMFCs) [1]. Currently, transition metal/nitrogen co-doped carbon (M/N-C) based materials, especially Fe/N-C electrocatalysts, have demonstrated exceptional potential owing to their high intrinsic activity and good durability toward oxygen reduction reaction (ORR) [2]. Nevertheless, their practical application has been limited by three fundamental limitations: (i) insufficient active-site density [3]; (ii) the scaling relationship between *O, *OH, and *OOH adsorption energies increasing ORR overpotential [4]; (iii) Fenton reaction-induced demetallization under acidic conditions [5].

In a recent study, Zhao et al. [6] reported a novel curvature-engineering strategy to synthesize curved-surface Fe/N−C (CS Fe/N–C) catalyst, which features Fe single atoms anchored within inner shells of nanoscaled hollow multi-shelled structure (HoMS) (Figure 1a). High-angle annular dark field scanning transmission electron microscopy (STEM) measurement corroborates that up to 97.6% Fe atoms are embedded within the inner curved surface (Figure 1b), and the loading of Fe atoms in the curved region is as high as ~1.6 nm⁻². X-ray absorption spectroscopy results confirm the FeN₄C₁₀ coordination geometry of CS Fe/N–C, and Mössbauer spectroscopy analysis further reveals that highly active low-spin state is dominant, accounting for 57.9%. In addition, the theoretical calculations verify that only the increase of the curvature for Fe-N₄ will unexpectedly strengthen the binding strength toward reaction intermediates, leading to a decrease in ORR activity. The single-atom Fe sites are primarily embedded within the inner shells curve-surface of the HoMS, which effectively weakens the binding strength and breaks the linear scaling relationship between ΔG_*OH, ΔG_*O and ΔG_*OOH, thereby boosting the electrocatalytic activity (Figure 1c). Simultaneously, the graphitized outer shell serves to protect the Fe atoms via electrostatic repulsion, suppressing the Fenton reaction or metal leaching during acidic ORR, thus enhancing the electrocatalytic stability.

Figure 1 (a) Schematic of the CS Fe/N–C structure. (b) Distribution of Fe atomic fractions in each layer. Scale bar, 5 nm. (c) The relationship between ΔG*OH, ΔG*OH–ΔG*O and the theoretical overpotential. (d) Plots for H2–air fuel cell polarization (solid symbols) and power density (hollow symbols) under 1.0 bar H2–air. (e) Long-term durability testing of CS Fe/N–C and 2D Fe/N–C at a constant 0.6  V. Reprinted with permission from Ref. [6]. Copyright©2025, Springer Nature.

Owing to its unique morphological and structural characteristics, the CS Fe/N–C electrocatalyst has exhibited outstanding activity toward ORR in 0.1 mol L⁻¹ HClO₄ (a half-wave potential of 0.82 V vs. reversible hydrogen electrode (RHE) with a current density of 5.8  mA cm⁻²). More importantly, when the CS Fe/N–C is further applied to a 5 cm² fuel cell and tested under the practical H₂-air conditions, it delivers a high power density (P_max) of 0.75 W cm⁻² at 1.0 bar (Figure 1d), which is superior to the 2D Fe/N–C counterpart and on par with the state-of-the-art performance achieved under the comparable operational conditions. Besides, the CS Fe/N--C based fuel cell also achieves a P_max of 1.25 W cm⁻² at 1.0 bar under H₂–O₂ conditions. Moreover, the CS Fe/N–C-based fuel cell shows minimal current density loss at 0.8  V after 30,000 accelerated stress test (AST) cycles under H₂-air conditions. Additionally, it can also operate stably for 303 h at a constant voltage of 0.6 V and retain 86% of its initial current density (Figure 1e). These results collectively highlight the outstanding durability of CS Fe/N–C.

In summary, the work by Zhao et al. [6] developed a novel strategy to construct the Fe/N−C catalyst with Fe single atoms primarily anchored within the inner curved surface of the nano-HoMS, which realizes remarkable ORR activity and best combination performance in PEMFCs, outperforming most of the reported platinum group metal-free PEMFC performances. We believe that the proposed curvature-engineering strategy has great promise for promoting high-performance electrocatalysts toward broader catalytic applications, and may even provide a viable pathway to overcome the limitations of conventional volcano charts.

Conflict of interest

The authors declare no conflict of interest.

References

© The Author(s) 2025. Published by Science Press and EDP Sciences.

All Figures

Figure 1 (a) Schematic of the CS Fe/N–C structure. (b) Distribution of Fe atomic fractions in each layer. Scale bar, 5 nm. (c) The relationship between ΔG*OH, ΔG*OH–ΔG*O and the theoretical overpotential. (d) Plots for H2–air fuel cell polarization (solid symbols) and power density (hollow symbols) under 1.0 bar H2–air. (e) Long-term durability testing of CS Fe/N–C and 2D Fe/N–C at a constant 0.6  V. Reprinted with permission from Ref. [6]. Copyright©2025, Springer Nature.

In the text