High-purity pyrrole-type FeN₄ sites as a superior oxygen reduction electrocatalyst

Abstract

Atomically dispersed iron–nitrogen (FeN₄) catalysts have emerged as the most promising alternative to costly Pt-based counterparts in proton exchange membrane fuel cells (PEMFCs), but often they suffer from high overpotential and poor stability due to the diverse iron–nitrogen coordination structure. Herein, we demonstrate high-purity pyrrole-type FeN₄ sites for the first time, as a superior ORR electrocatalyst for PEMFCs. The high-purity pyrrole-type FeN₄ catalyst exhibited extremely outstanding ORR activity with an ultra-high active area current density of 6.89 mA m⁻² in acid medium, which exceeds that of most reported metal–nitrogen coordination catalysts. Experimental and theoretical analyses reveal that high-purity pyrrole-type coordination significantly modifies the atomic and electronic structures of FeN₄ sites, bringing with it high intrinsic catalytic activity, preferable O₂ adsorption energy and full four-electron reaction selectivity for ORR catalysis. Therefore, PEMFCs built with this high-purity FeN₄ catalyst achieve a high open-circuit voltage (1.01 V) and a large peak power density (over 700 mW cm⁻²). High-purity iron–nitrogen coordination would give new insights into highly efficient electrocatalysts for PEMFCs.

---

Broader context

Proton exchange membrane fuel cells (PEMFCs) have been targeted as promising clean power candidates with the merits of large power density, environmental compatibility and high efficiency. As the key reaction catalysts of PEMFCs, commercial Pt-based catalysts account for more than 26% of the total cell cost and the scarcity of Pt largely impedes their widespread commercialization. Recently, iron–nitrogen (FeN₄) coordination sites have emerged as the most promising non-noble catalysts for PEMFCs, but often they suffer from high overpotential and instability. At the heart of the performance-determining parameters, a high-purity FeN₄ structure has always been a researcher's dream since then. However, current methodologies for iron–nitrogen coordination sites usually require high-temperature pyrolysis, which inevitably leads to diverse structures of iron–nitrogen coordination. In this work, we report a high-purity pyrrole-type FeN₄ structure as a superior cathode material for PEMFCs and unveil its structure–activity relationship at the atomic scale. Benefiting from the regulated atomic and electronic structure, the high-purity pyrrole-type FeN₄ catalyst exhibited an extremely outstanding catalytic activity in a PEMFC system and exceeds most reported metal–nitrogen coordination catalysts. This work will open up a new avenue to design advanced electrocatalysts for PEMFCs.

Proton exchange membrane fuel cells (PEMFCs) are considered to be a clean power candidate due to their high energy conversion efficiency, large power density and environmentally friendly features.^1–3 As a key reaction of PEMFCs, the oxygen reduction reaction (ORR) at the cathode electrode suffers greatly from sluggish reaction kinetics and multiple proton-coupled electron transfer processes, which largely impedes the widespread commercialization of PEMFCs.^4–6 Recently, iron–nitrogen coordination sites (FeN₄) have emerged as the most promising non-noble catalysts for PEMFCs because of their encouraging ORR activity in acidic media.^7–9 However, the FeN₄ catalysts tend to be protonated or attacked by free radicals at the active sites, leading to high overpotential and insufficient stability of ORR catalysis.^10,11 Therefore, development of highly active FeN₄ sites with low overpotential and superior stability for PEMFCs is actively being pursued.

Over the past few years, much effort has been devoted to optimizing the FeN₄ configurations at the atomic scale, such as increasing the density of iron–nitrogen sites,^12,13 fabricating dual metal sites,^14,15 anion doping^16,17 and coordination number regulations.^18,19 Despite the excellent results, most research studies indicate that the properties are derived from the synergy of multiple coordination forms.^6,20 It is still a great challenge to figure out which coordination form is decisive for the performance.^21,22 In this regard, developing high-purity FeN₄ sites with defined structure is extremely important for understanding the structure–activity relationship at the atomic scale. In effect, theoretical calculations already gave us an inspiration that pyrrolic N can not only regulate the O₂ adsorption energy of bonded Fe atoms but also activate the neighbouring C atoms as active sites for the ORR.^23,24 However, the current synthesis of pyrrole nitrogen coordinated FeN₄ structures remains a challenge, which seriously hampers the commercial application of FeN₄ catalysts. Therefore, realizing high-purity pyrrole-type FeN₄ structures has become a vital issue for designing advanced electrocatalysts for PEMFCs.

Herein, we highlight a high-purity pyrrole-type FeN₄ structure for the first time, as the cathode material for PEMFCs. By removing additional carbon atoms with the assistance of ammonia gas, high-purity pyrrole-type FeN₄ sites could be successfully achieved via chemical configuration transformation from pyridine nitrogen to pyrrole nitrogen. Benefiting from the regulated atomic and electronic structures, high-purity pyrrole-type FeN₄ sites exhibit high intrinsic catalytic activity, preferable O₂ adsorption energy and full four-electron reaction selectivity. As expected, the exclusive high-purity FeN₄ structure as a reaction site boosts the ORR catalytic reaction, achieving a high open-circuit voltage and a large peak power density in PEMFCs. High-purity FeN₄ sites will open up a new avenue for the design of highly-efficient electrocatalyts for PEMFC systems.

In this work, high-purity pyrrole-type FeN₄ sites (denoted as HP-FeN₄) were derived from the pyrolysis of a polyaniline precursor. By removing additional carbon atoms with the assistance of ammonia, the chemical configuration of pyridine nitrogen was transformed to pyrrole nitrogen and a pyrrole-type FeN₄ structure was obtained (Fig. 1a). High-resolution transmission electron microscopy (Fig. 1b) shows highly disordered carbon structures with randomly oriented graphitic domains of HP-FeN₄ and no obvious nanoparticles were found. Meanwhile, only two broad carbon diffraction peaks of (002) and (100) at 26° and 44° were observed in the X-ray diffraction (XRD) pattern of HP-FeN₄ (Fig. S3, ESI†),²⁵ which was similar to the NC material without Fe doping. These results suggest that no Fe-based compound existed in the HP-FeN₄ catalyst. The Fe species were highly dispersed as tiny clusters or single sites on the graphene framework. The uniform dispersion of Fe atoms was confirmed by aberration-corrected high-angle annular dark field scanning transmission electron microscopy (HAADF-STEM) as shown in Fig. 1c. Bright spots corresponding to the atomically dispersed Fe sites were homogeneously distributed across the carbon framework. The associated energy-dispersive spectroscopy (EDS) mapping shows the uniform distribution of Fe, N and C atoms (Fig. 1d). For comparison, the characterization of traditional FeN₄ catalysts synthesized under an Ar atmosphere, which are denoted as FeN₄, is presented in Fig. S1–S5 of the ESI.† The traditional FeN₄ catalyst exhibited similar features to HP-FeN₄ in both carbon structure and element distribution. Therefore, this ammonia-assisted strategy keeps the morphology and elemental dispersion of the traditional FeN₄ material and Fe atoms are uniformly dispersed as single sites or clusters in the HP-FeN₄ material.

Fig. 1 High-purity FeN₄ sites. (a) Preparation process of high-purity pyrrole-type FeN₄ structure. The balls in grey, blue and orange represent C, N and Fe atoms, respectively. (b) HRTEM and (c) HAADF-STEM image of HP-FeN₄ material. (d) EDS elemental mapping profiles of HP-FeN₄. (e) Fourier transformations of the EXAFS spectra for Fe foil, Fe₂O₃, HP-FeN₄ and FeN₄. (f) First-shell fitting of Fourier transformations of EXAFS spectra for HP-FeN₄. Top and bottom spectra are magnitude and imaginary part, respectively. Inset: The structure of the iron site in the HP-FeN₄ material.

Synchrotron X-ray absorption spectroscopy (XAS) measurements were carried out to determine the electronic structure and local geometry around implanted Fe atoms. Fig. S6 (ESI†) shows Fe K-edge X-ray absorption near edge structure (XANES) spectra of HP-FeN₄ catalyst and standard compounds. HP-FeN₄ shows Fe K-edge absorption energy between Fe foil and Fe₂O₃, which demonstrates that the Fe atoms exhibited a positive valence state below +3. Besides, both HP-FeN₄ and FeN₄ show clear pre-edge peaks at 7113.5 eV, which are similar to the characteristics of Hemin, suggesting the typical FeN₄ configurations.²⁶Fig. 1e displays the Fourier transform of extended X-ray absorption fine structure (FT-EXAFS) spectra. The Fe–Fe interactions of Fe foil were not observed in the FT-EXAFS of HP-FeN₄, which confirms the HAADF-STEM results and verifies the existence of atomically dispersed Fe sites on the graphene framework. The main peaks at approximately 1.55 Å can be ascribed to the single scattering of photoelectrons excited by an X-ray from Fe atoms to the nearest neighbored N atoms i.e., the Fe–N shell. Fig. 1f shows the best fit of the FT-EXAFS of HP-FeN₄ compared with the experimental one. The fitting result shows that Fe was bonded by four nitrogen atoms with Fe–N bonds, suggesting the Fe–N₄ coordination structure of this catalyst. Moreover, traditional FeN₄ exhibits a similar local geometric structure as that of HP-FeN₄. And the fitting parameters of both the samples are summarized in Fig. S7 and Table S1 (ESI†). In this regard, the ammonia-assisted strategy has little effect on the bond length and coordination number of iron–nitrogen sites. And Fe atoms of the HP-FeN₄ material are basically in a coordination form with the Fe–N₄ structure.

In order to eliminate the influence of carbon structure on material properties, Brunauer–Emmett–Teller (BET) characterization was performed to analyze the surface area and pore size distribution of the as-prepared catalysts. As shown in Fig. S10 (ESI†), HP-FeN₄ and FeN₄ exhibit BET specific surface areas of 609.9 and 527.7 m² g⁻¹, respectively. N₂ sorption isotherms of both the HP-FeN₄ and FeN₄ catalysts possess a pronounced hysteresis loop, and can be identified as type-IV isotherms, suggesting the existence of a mesoporous structure in the carbon framework.²⁷ Moreover, Barrett–Joyner–Halenda (BJH) pore size distribution curves derived from the adsorption branch confirm that these two catalysts were rich in mesopores with a diameter of 3–7 nm (Fig. S11, ESI†). Raman spectroscopy was also used to characterize the carbon structure of the two catalysts. As shown in Fig. S12 (ESI†), both HP-FeN₄ and FeN₄ show broad D and G peaks, revealing the disordered carbon structures. The I_D/I_G ratios of HP-FeN₄ and FeN₄ are 1.025 and 1.027, suggesting the similar structural defects of the two catalysts. Specifically, HP-FeN₄ exhibits similar C 1s spectra compared to FeN₄, which further reveals that HP-FeN₄ and FeN₄ exhibit similar surface oxidation degree (Fig. S13, ESI†). The graphitization degree and hierarchical pore structure were insensitive to the ammonia-assisted strategy. And the adjacent structure of FeN₄ sites would play a key role in affecting the ORR performance. Moreover, a large number of mesopores in HP-FeN₄ will promote the electrolyte wetting of the physical surface to increase the portion of electrochemically available active sites.^28,29 Thus, the kinetic accessibility to active sites could be largely improved for the ORR process, which increased the utilization of active sites.

Iron–nitrogen sites of HP-FeN₄ were confirmed to be high-purity as pyrrole-type coordination by virtue of Soft X-ray absorption spectrocopy (sXAS) based on the synchrotron radation facility, which is sensitive to the local chemical configuration around the probed element and can be employed to clarify at the electronic level the chemical physics of the coordinated low-z atoms, such as N in the FeN₄ site. As shown in Fig. 2a, C K-edge XANES spectra present two typical spectroscopic features for HP-FeN₄ and FeN₄ materials: the 1s core electron of carbon into the π*_C=C (283.8 eV) and σ*_C–C (290.5 eV).^30,31 The HP-FeN₄ material shows similar features to highly graphitized carbon. This result demonstrates that the introduction of Fe and N atoms maintains the electronic structure and high electron conductivity of carbon supports.³² Moreover, the N K-edge spectra of HP-FeN₄ materials are dominated by two typical spectroscopic features: 1s → π* transition at the region of 393.0–401.0 eV and 1s → σ* transition at 401.0–410.0 eV, respectively.³³ In the region of 1s → π* transition, two obvious resonances labeled as peak a and peak b can be observed. The resonance of peak a can be ascribed to the π* transition in the C–N–C portion of pyridinic N sites.³⁴ And peak b is the π* transition of pyrrolic/graphitic type N-groups, consistent with previous reports.³⁵ Notably, peak b split into double peaks for HP-FeN₄ while FeN₄ kept unchanged, indicating that the pyrrolic N was bonded to Fe atoms under NH₃ pyrolysis. The formation of Fe–pyrrolic bonds will induce a charge transfer between the metal atom and the N-decorated graphene, thereby promoting the electron transfer capacity between active FeN₄ sites and the carbon supports during the ORR process.³⁶

Fig. 2 Pyrrole-type FeN₄ coordination. (a) sXAS spectra of the C K-edge and N K-edge for HP-FeN₄ and FeN₄. Deconvoluted features of (b) peak a and (c) peak b of the N K-edge spectra. (d) High-resolution N 1s XPS spectra of HP-FeN₄ and FeN₄. (e) Comparison of the pyrrolic N content (from XPS spectra) of the HP-FeN₄ precursor pyrolysed for different time under NH₃ atmosphere.

To further confirm the coordination N structure of FeN₄ sites, the region of 1s → π* transition for N K-edge sXAS was fitted and is shown in Fig. 2b and c. As shown in Fig. 2b, the high-resolution spectrum of peak a for the FeN₄ material was deconvoluted into two peaks which can be ascribed to the pyridinic N (395.7 eV) and Fe–pyridinic N sites (396.4 eV). In particular, HP-FeN₄ exhibits only one peak at 395.7 eV, which demonstrates that there is no Fe–pyridinic N coordination in the HP-FeN₄ catalyst. In contrast, peak b of HP-FeN₄ can be divided into three characteristic peaks at 397.5, 398.0 and 398.7 eV, which are assigned to pyrrolic N, Fe–pyrrolic N and graphitic N sites, respectively (Fig. 2c). Notably, only one peak of graphitic N for the FeN₄ catalyst was observed, suggesting that NH₃ pyrolysis results in the change of neighbouring N structure around the FeN₄ sites and promotes the formation of Fe–pyrrolic N coordination structure. HP-FeN₄ manifests an obvious Fe–pyrrolic N characteristic peak, and no Fe–pyridinic N site was detected in the sXAS spectra. The above results confirm that HP-FeN₄ possesses a high-purity FeN₄ structure and almost all of the FeN₄ sites are pyrrole-type.

The coordination N structure of the HP-FeN₄ catalyst can also be confirmed by the Fourier transform infrared (FT-IR) spectra and X-ray photoelectron spectroscopic (XPS) analysis. In Fig. S14 (ESI†), the FT-IR spectra of FeN₄ and HP-FeN₄ show three characteristic peaks at 1404, 1637 and 3459 cm⁻¹, respectively. The wider band at ∼3459 cm⁻¹ corresponds to C–OH and N–H stretching vibrations, whereas the peaks at around 1637 cm⁻¹ and 1404 cm⁻¹ could be assigned to the typical CN heterocycle stretching modes.^37,38 Moreover, both HP-FeN₄ and FeN₄ show a peak at low wavenumbers of 833 cm⁻¹, which can be related to the Fe–N bonds. Specifically, the peak at 873 cm⁻¹ belongs to the typical Fe–N stretch (pyrrole), which is consistent with FePc.³⁹ Therefore, the pyrrole-type FeN₄ structure was successfully achieved via the ammonia assisted pyrolysis strategy.

Fig. S15 (ESI†) shows the high resolution spectra of N 1s for different catalysts. The N 1s spectra of NC can be deconvoluted into three peaks, which correspond to pyridinic N (398.5 eV), pyrrolic N (399.7 eV) and graphitic N (401.1 eV), respectively.^40,41 Specifically, the pyridinic N of the FeN₄ catalyst shifted to a higher binding energy compared with NC materials (Fig. S15(a), ESI†), which can be ascribed to the interaction between Fe atoms and pyridinic N species, suggesting the formation of Fe–pyridinic N bonds.⁴² In contrast, the binding energy of pyrrolic N increased from 399.7 eV to 399.9 eV for the HP-FeN₄ catalyst (Fig. S15(b), ESI†), which demonstrates that Fe–pyrrolic N moieties have been formed during the NH₃ pyrolysis. The different N species content can be obtained by calculating the area of the N species peak in XPS spectra. As shown in Fig. 2d and Table S2 (ESI†), HP-FeN₄ exhibits significantly increased pyrrolic N content (22.1 at%) and decreased pyridinic N content (41.7 at%) compared with FeN₄ synthesized under Ar (5.9 at% and 54.2 at%, respectively). This result reveals that the FeN₄ structure was successfully converted from pyridine-type to pyrrole-type via the ammonia-assisted strategy and maintains its high purity.

The content of pyrrole-type FeN₄ sites can also be tuned by controlling the annealing atmosphere and chemical reaction time. Fig. S15(c) (ESI†) displays the N 1s XPS spectra of the HP-FeN₄ precursor annealed under NH₃ for different lengths of time. When the NH₃ annealing time was 0 min, few pyrrolic N species were observed and almost all Fe atoms were coordinated by pyridinic N. More pyrrole-type FeN₄ sites were formed by increasing the NH₃ annealing time. A high-purity pyrrole-type FeN₄ structure was obtained with an annealing time of 60 min. As shown in Fig. 2e, it was obviously observed that the content of pyrrolic N increased with the pyrolysis time increasing, which further demonstrates that NH₃ atmosphere plays an important role in forming the pyrrole-type FeN₄ structure. Based on the above results, high-purity pyrrole-type FeN₄ sites were successfully achieved by controlling the ammonia-assisted reaction conditions and it exhibits perfect atom utilization and efficient mass transfer capability, which makes it a promising electrocatalyst for the ORR process.

To theoretically evaluate the ORR activity and selectivity on the pyrrole-type FeN₄ structure, density functional theory (DFT) calculations were carried out to investigate the free energetics of the ORR reaction mechanism. As shown in Fig. 3a and b, pyrrole-type FeN₄ shows stronger electron depletion (blue area) around the Fe atom than pyridine-type FeN₄, but weaker electron depletion is observed around the neighboring pyrrolic N. This result reveals that iron has a distinctly different charge distribution on both structures, and the valence state of iron in the pyrrole-type structure is more positive than that in the pyridine-type structure. In the free energy diagram of the ORR (Fig. 3c), lower Gibbs free energy differences between O₂ and OOH* of pyrrole-type FeN₄ suggest its preferable oxygen adsorption for the ORR.⁴³ Moreover, pyrrole-type FeN₄ exhibits lower thermodynamic overpotential (0.35 eV) from the initial state (O₂) to the final state (H₂O) than that of pyridine-type (0.67 eV) in a 4e⁻ pathway, giving theoretical evidence for a highly efficient ORR catalytic activity. In particular, the thermodynamic onset potential of pyrrole-type FeN₄ (0.88 V) is also higher than that of pyridine-type FeN₄, which further confirms that pyrrole-type coordination is more conducive to the ORR reaction (Fig. S24, ESI†). At a potential of 1.23 V, the reduction of OOH* to H₂O₂ is endothermic by 1.77 eV on the pyrrole-type FeN₄ structure, higher than that of pyridine-type (1.57 eV) (Fig. 3c), revealing that the 2e⁻ reduction pathway is largely suppressed and pyrrole-type FeN₄ would have high selectivity for the 4e⁻ reduction process. Fig. 3d shows the atomic configuration details of the ORR reaction, which involves four electron transfer steps. Initially, an O₂ molecule adsorbs on the top of Fe atom and then transfers into OOH*. Next, H₂O is desorbed and O* is formed on the central Fe atom. Finally, OH* is formed and further converts to a H₂O molecule. Overall, the DFT results reveal that pyrrole-type FeN₄ could lower the limiting potential and is more conducive to 4e⁻ reaction, which would be an ideal active site for ORR catalysis.

Fig. 3 Theoretical calculations. Calculated charge density difference of (a) pyrrole-type FeN₄ and (b) pyridine-type FeN₄. Yellow and blue regions represent electron accumulation and electron depletion, respectively. Iso-surface 0.005 e⁻ Å⁻³. (c) Free energy diagram of the oxygen reduction reaction on pyrrole-type FeN₄ and pyridine-type FeN₄. (d) Adsorption configurations of the intermediates during the ORR process on pyrrole-type FeN₄ (the balls in grey, blue, orange, red and pink represent C, N, Fe, O and H atoms, respectively).

To investigate the catalytic activity of high-purity pyrrole-type FeN₄ sites, electrochemical performance of ORR is evaluated by employing a rotating disk electrode (RDE) method. As shown in Fig. 4a, HP-FeN₄ exhibits higher onset potential (0.95 V vs. RHE) and more positive half-wave potential (0.80 V vs. RHE) than that of FeN₄ (0.86 V and 0.71 V vs. RHE, respectively), which is comparable to the commercial Pt/C catalyst. Compared to both FeN₄ (80 mV dec⁻¹) and Pt/C (88 mV dec⁻¹), a small Tafel slope of 72 mV dec⁻¹ can also be observed for HP-FeN₄, indicating its fast kinetics in the oxygen reduction process (Fig. 4b). The current density relative to the BET specific surface area at 0.8 V vs. RHE has also been compared. In Fig. 4c, HP-FeN₄ exhibits the largest current density of 6.89 mA m⁻², which is 6.9 and 54.3 times that of traditional FeN₄ and NC, and it surpasses most reported metal–nitrogen coordination catalysts (Table S3, ESI†). The electrochemical activity has been greatly improved for HP-FeN₄ while the specific surface area has not changed significantly, which demonstrates that high-purity pyrrole-type iron–nitrogen sites are the key factor for performance improvement. Moreover, the ORR activity of HP-FeN₄ annealing under NH₃ for different times was also evaluated. As shown in Fig. 4d, HP-FeN₄ catalysts exhibit more positive half-wave potential and smaller Tafel slope with increasing NH₃ annealing time. This result gives direct evidence for the high intrinsic catalytic activity of pyrrole-type FeN₄ sites.

Fig. 4 Electrocatalytic performance. (a) ORR polarization curves and (b) corresponding Tafel plots of HP-FeN₄, FeN₄ and NC catalysts in O₂-saturated 0.5 M H₂SO₄ and 20% Pt/C in 0.1 M HClO₄ under a rotating rate of 1600 rpm. (c) Comparison of current density relative to BET surface area at 0.8 V and half-wave potential for HP-FeN₄, FeN₄ and NC catalysts. (d) ORR catalytic activity of HP-FeN₄ precursor pyrolysed for different times under NH₃ atmosphere. (e) Peroxide yield and electron transfer number of HP-FeN₄, FeN₄, NC and Pt/C during the ORR process. (f) Polarization and power density curves of HP-FeN₄, FeN₄ and NC based membrane electrode assemblies in PEMFCs. Cell temperature: 80 °C; RH: 100%, H₂/O₂: 200 kPa.

To verify the ORR catalytic pathways of these electrocatalysts, rotating ring-disk electrode (RRDE) measurements were performed to monitor the formation of peroxide species (H₂O₂) during the ORR process. As shown in Fig. 4e, the measured H₂O₂ yields of HP-FeN₄, FeN₄ and NC were below 10% over the potential range of 0.2–0.8 V, suggesting an electron transfer number over 3.8. Specifically, HP-FeN₄ exhibited a Pt-like property with H₂O₂ yields below 2%, much lower than the FeN₄ catalyst, which demonstrated higher product selectivity of pyrrole-type FeN₄ sites. This result reveals that a pyrrole-type FeN₄ structure could restrain the generation of H₂O₂, thereby reducing Fenton reactions during oxygen reduction and enhancing the stability of FeN₄ sites. Electrochemical impedance spectroscopy (EIS) was also employed to characterize the interface reactions and electrode kinetics in the ORR. As shown in Fig. S21 (ESI†), HP-FeN₄ exhibits a smaller charge transfer resistance than that of FeN₄ and NC, suggesting that the high-purity pyrrole-type FeN₄ structure possessed a lower charge transfer resistance and allowed much faster shuttling of electrons during the ORR process. The stability test of HP-FeN₄ was also performed. As shown in Fig. S17 (ESI†), only 26 mV loss of half-wave potential after 10 000 potential cycles in O₂-saturated 0.5 M H₂SO₄ was observed, suggesting its excellent stability. Therefore, purity and coordination N structure essentially determine the activity and stability of the FeN₄ site. Pyrrolic N coordination and high-purity FeN₄ sites achieve significantly enhanced catalytic activity and stability and would favor improved performance of PEMFCs.

To corroborate the ORR activity revealed by the half-cell characterization, HP-FeN₄, FeN₄ and NC were assembled into the membrane electrode assembly (MEA) for evaluating the actual PEMFC performance. Fig. S29 (ESI†) shows the working principle scheme of the PEMFC and digital photographs of a single cell assembling process. As shown in Fig. S19 (ESI†), the open circuit voltage (OCV) of HP-FeN₄ was as high as 1.01 V, better than 0.92 V of FeN₄ and 0.84 V of NC, suggesting the good intrinsic catalytic performance of high-purity pyrrole-type FeN₄ sites. Fig. 4f presents polarization and power density curves for the three catalysts. The HP-FeN₄ catalyst was capable of generating current densities of 0.55 and 2.66 A cm⁻² at 0.6 and 0.2 V with H₂/O₂ total pressure of 200 kPa, which outperformed the FeN₄ and NC catalyst. The corresponding peak power density of HP-FeN₄ was up to 700 mW cm⁻², suggesting its excellent application prospects. Based on the above results, the greatly improved PEMFC performance of the HP-FeN₄ catalyst can be ascribed to the following features: (1) high-purity FeN₄ sites enable significantly enhanced intrinsic activity for the ORR. (2) The pyrrole-type FeN₄ configuration possesses preferred O₂ adsorption energy and ultra-high four-electron reaction selectivity. (3) The presence of a large number of mesopores greatly increased the utilization of active sites. Due to the combination of enhanced catalytic activity and high exposure of active sites, the ORR processes of HP-FeN₄ are largely accelerated, and a high power density is achieved when integrated into a PEMFC system.

In conclusion, we have successfully developed a high-purity pyrrole-type FeN₄ structure as a high-efficiency PEMFC electrocatalyst. The as-prepared pyrrole-type FeN₄ sites exhibit significantly enhanced intrinsic activity, preferred O₂ adsorption energy and complete four-electron reaction selectivity, leading to an excellent ORR activity and stability. As expected, the high-purity FeN₄ catalyst exhibits an ultra-high active area current density of 6.89 mA m⁻² in acid medium, which exceeds most reported metal–nitrogen coordination catalysts. Moreover, the high-purity FeN₄ site based PEMFCs show excellent performance with a large power density and a high open circuit voltage, suggesting a wider perspective for practical applications. A high-purity iron–nitrogen configuration will open up new avenues to design advanced non-noble metal catalysts for reversible energy conversion systems.

Conflicts of interest

The authors declare no competing interests.

Acknowledgements

This work was financially supported by the National Basic Research Program of China (2015CB932302), Natural Science Foundation of China (No. 91745113, 11621063), National Program for Support of Top-notch Young Professionals, and the Fundamental Research Funds for the Central Universities (No. WK 2060190084). We also appreciate the support from the Major/Innovative Program of Development Foundation of Hefei Center for Physical Science and Technology. This work was partially carried out at the USTC Center for Micro and Nanoscale Research and Fabrication. Minglong Chen and Xiaojun Wu thank the Supercomputing Center of USTC for the computational resources and the support of the Natural Science Foundation of China (No. 21573204, 21421063) and the Ministry of Science and Technology of the People's Republic of China (2018YFA0208603).

Notes and references

1. M. K. Debe, Nature, 2012, 486, 43.
2. L. Chong, J. Wen, J. Kubal, F. G. Sen, J. Zou, J. Greeley, M. Chan, H. Barkholtz, W. Ding and D.-J. Liu, Science, 2018, 362, 1276.
3. J. Greeley, I. E. L. Stephens, A. S. Bondarenko, T. P. Johansson, H. A. Hansen, T. F. Jaramillo, J. Rossmeisl, I. Chorkendorff and J. K. Nørskov, Nat. Chem., 2009, 1, 552.
4. M. Shao, Q. Chang, J.-P. Dodelet and R. Chenitz, Chem. Rev., 2016, 116, 3594–3657.
5. X. Huang, Z. Zhao, L. Cao, Y. Chen, E. Zhu, Z. Lin, M. Li, A. Yan, A. Zettl, Y. M. Wang, X. Duan, T. Mueller and Y. Huang, Science, 2015, 348, 1230.
6. X. Wang, Z. Li, Y. Qu, T. Yuan, W. Wang, Y. Wu and Y. Li, Chem, 2019, 5, 1–26.
7. U. Martinez, S. Komini Babu, E. F. Holby, H. T. Chung, X. Yin and P. Zelenay, Adv. Mater., 2019, 1806545.
8. Y. Shao, J.-P. Dodelet, G. Wu and P. Zelenay, Adv. Mater., 2019, 1807615.
9. G. Wu, Front. Energy, 2017, 11, 286–298.
10. J. Herranz, F. Jaouen, M. Lefèvre, U. I. Kramm, E. Proietti, J.-P. Dodelet, P. Bogdanoff, S. Fiechter, I. Abs-Wurmbach, P. Bertrand, T. M. Arruda and S. Mukerjee, J. Phys. Chem. C, 2011, 115, 16087–16097.
11. Y. He, S. Hwang, D. A. Cullen, M. A. Uddin, L. Langhorst, B. Li, S. Karakalos, A. J. Kropf, E. C. Wegener, J. Sokolowski, M. Chen, D. Myers, D. Su, K. L. More, G. Wang, S. Litster and G. Wu, Energy Environ. Sci., 2019, 12, 250–260.
12. X. Wan, X. Liu, Y. Li, R. Yu, L. Zheng, W. Yan, H. Wang, M. Xu and J. Shui, Nat. Catal., 2019, 2, 259–268.
13. V. Yarlagadda, M. K. Carpenter, T. E. Moylan, R. S. Kukreja, R. Koestner, W. Gu, L. Thompson and A. Kongkanand, ACS Energy Lett., 2018, 3, 618–621.
14. J. Wang, Z. Huang, W. Liu, C. Chang, H. Tang, Z. Li, W. Chen, C. Jia, T. Yao, S. Wei, Y. Wu and Y. Li, J. Am. Chem. Soc., 2017, 139, 17281–17284.
15. L. Lin, Z. K. Yang, Y.-F. Jiang and A.-W. Xu, ACS Catal., 2016, 6, 4449–4454.
16. Y. Chen, S. Ji, S. Zhao, W. Chen, J. Dong, W.-C. Cheong, R. Shen, X. Wen, L. Zheng, A. I. Rykov, S. Cai, H. Tang, Z. Zhuang, C. Chen, Q. Peng, D. Wang and Y. Li, Nat. Commun., 2018, 9, 5422.
17. H. Shen, E. Gracia-Espino, J. Ma, K. Zang, J. Luo, L. Wang, S. Gao, X. Mamat, G. Hu, T. Wagberg and S. Guo, Angew. Chem., Int. Ed., 2017, 56, 13800–13804.
18. H. Shen, E. Gracia-Espino, J. Ma, H. Tang, X. Mamat, T. Wagberg, G. Hu and S. Guo, Nano Energy, 2017, 35, 9–16.
19. M. Xiao, H. Zhang, Y. Chen, J. Zhu, L. Gao, Z. Jin, J. Ge, Z. Jiang, S. Chen, C. Liu and W. Xing, Nano Energy, 2018, 46, 396–403.
20. Q. Jia, E. Liu, L. Jiao, S. Pann and S. Mukerjee, Adv. Mater., 2018, 1805157.
21. Q. Jia, N. Ramaswamy, H. Hafiz, U. Tylus, K. Strickland, G. Wu, B. Barbiellini, A. Bansil, E. F. Holby, P. Zelenay and S. Mukerjee, ACS Nano, 2015, 9, 12496–12505.
22. A. Zitolo, V. Goellner, V. Armel, M.-T. Sougrati, T. Mineva, L. Stievano, E. Fonda and F. Jaouen, Nat. Mater., 2015, 14, 937.
23. L. Yang, D. Cheng, H. Xu, X. Zeng, X. Wan, J. Shui, Z. Xiang and D. Cao, Proc. Natl. Acad. Sci. U. S. A., 2018, 115, 6626.
24. K. Liu, G. Wu and G. Wang, J. Phys. Chem. C, 2017, 121, 11319–11324.
25. H. Yu, L. Shang, T. Bian, R. Shi, G. I. N. Waterhouse, Y. Zhao, C. Zhou, L.-Z. Wu, C.-H. Tung and T. Zhang, Adv. Mater., 2016, 28, 5080–5086.
26. Q. Liu, X. Liu, L. Zheng and J. Shui, Angew. Chem., Int. Ed., 2018, 57, 1204–1208.
27. M. Graglia, J. Pampel, T. Hantke, T.-P. Fellinger and D. Esposito, ACS Nano, 2016, 10, 4364–4371.
28. S. H. Lee, J. Kim, D. Y. Chung, J. M. Yoo, H. S. Lee, M. J. Kim, B. S. Mun, S. G. Kwon, Y.-E. Sung and T. Hyeon, J. Am. Chem. Soc., 2019, 141, 2035–2045.
29. H. T. Chung, D. A. Cullen, D. Higgins, B. T. Sneed, E. F. Holby, K. L. More and P. Zelenay, Science, 2017, 357, 479.
30. I. Y. Kim, S. Kim, X. Jin, S. Premkumar, G. Chandra, N.-S. Lee, G. P. Mane, S.-J. Hwang, S. Umapathy and A. Vinu, Angew. Chem., 2018, 130, 17381–17386.
31. P. Chen, N. Zhang, S. Wang, T. Zhou, Y. Tong, C. Ao, W. Yan, L. Zhang, W. Chu, C. Wu and Y. Xie, Proc. Natl. Acad. Sci. U. S. A., 2019, 116, 6635.
32. D. Deng, X. Chen, L. Yu, X. Wu, Q. Liu, Y. Liu, H. Yang, H. Tian, Y. Hu, P. Du, R. Si, J. Wang, X. Cui, H. Li, J. Xiao, T. Xu, J. Deng, F. Yang, P. N. Duchesne, P. Zhang, J. Zhou, L. Sun, J. Li, X. Pan and X. Bao, Sci. Adv., 2015, 1, e1500462.
33. Y. Liang, H. Wang, J. Zhou, Y. Li, J. Wang, T. Regier and H. Dai, J. Am. Chem. Soc., 2012, 134, 3517–3523.
34. Y. Zheng, Y. Jiao, Y. Zhu, L. H. Li, Y. Han, Y. Chen, A. Du, M. Jaroniec and S. Z. Qiao, Nat. Commun., 2014, 5, 3783.
35. N. Meng, J. Ren, Y. Liu, Y. Huang, T. Petit and B. Zhang, Energy Environ. Sci., 2018, 11, 566–571.
36. Y. Tong, P. Chen, T. Zhou, K. Xu, W. Chu, C. Wu and Y. Xie, Angew. Chem., Int. Ed., 2017, 56, 7121–7125.
37. F. Arcudi, L. Đorđević and M. Prato, Angew. Chem., 2016, 128, 2147–2152.
38. F. Wang, P. Chen, Y. Feng, Z. Xie, Y. Liu, Y. Su, Q. Zhang, Y. Wang, K. Yao, W. Lv and G. Liu, Appl. Catal., B, 2017, 207, 103–113.
39. Z. Zhang, M. Dou, J. Jing and W. Feng, Nano Energy, 2017, 34, 338–343.
40. J. Zhang, Z. Zhao, Z. Xia and L. Dai, Nat. Nanotechnol., 2015, 10, 444.
41. P. Chen, T. Zhou, L. Xing, K. Xu, Y. Tong, H. Xie, L. Zhang, W. Yan, W. Chu, C. Wu and Y. Xie, Angew. Chem., Int. Ed., 2017, 56, 610–614.
42. X. Li, W. Bi, M. Chen, Y. Sun, H. Ju, W. Yan, J. Zhu, X. Wu, W. Chu, C. Wu and Y. Xie, J. Am. Chem. Soc., 2017, 139, 14889–14892.
43. A. Kulkarni, S. Siahrostami, A. Patel and J. K. Nørskov, Chem. Rev., 2018, 118, 2302–2312.

---

Footnotes

† Electronic supplementary information (ESI) available: Experimental details, SEM, XPS, TEM and additional electrochemical data. See DOI: 10.1039/c9ee03027a

‡ These authors contributed equally to this work.

---