Synthesis of three-dimensional nitrogen-doped reduced graphene oxide catalyst with high activity in acidic

Abstract

In this study, we synthesized a novel three-dimensional nitrogen-doped reduced graphene oxide (3D-NRGO) by integrating a sulfonated polystyrene (PSS) template method with nitrogen doping. The resulting 3D-NRGO was applied as an electrocatalyst for the oxygen reduction reaction (ORR) in acidic electrolyte. Owing to the synergistic effect arising from its three-dimensional structure and nitrogen doping, the catalyst demonstrates substantially augmented catalytic current density, a more positive ORR potential, excellent methanol tolerance, and prolonged operational stability. XPS and EDS characterization coupled with complementary analyses established that graphitic-N constitutes the paramount nitrogen species governing ORR activity in acidic media.

Supplementary Information

The online version contains supplementary material available at 10.1186/s11671-025-04348-x.

Introduction

The oxygen reduction reaction (ORR) constitutes a cornerstone process for fuel cell advancement [1]. While Pt/C-based catalysts historically represent the benchmark for ORR efficiency, intrinsic constraints like prohibitive cost, CO poisoning vulnerability, and inadequate long-term stability, their commercial scalability [2–4]. Consequently, pursuing cost-effective non-noble metal electrocatalysts has emerged as a pivotal research frontier. Previous studies have demonstrated promising alternatives to Pt-based catalysts. For example, Zhang et al. [5] engineered nitrogen-doped graphene manifesting superior ORR activity. Su et al. [6] fabricated nitrogen/sulfur-cofunctionalized three-dimensional (3D) graphene architectures as efficient metal-free electrocatalysts; analogously, Yan et al. [7] devised 3D nitrogen-doped graphene/MnO₂ nanoparticle hybrids exhibiting exceptional catalytic performance.

Notwithstanding the aforementioned catalysts’ efficacy in alkaline media, their acidic electrolyte performance persists as a constraint. Acidic systems hold greater practical utility for fuel cells, since alkaline electrolytes undergo rapid carbonation via CO₂ reaction, yielding performance-degrading precipitates. Conversely, acidic electrolytes exhibit CO₂ inertness. Thus, pioneering high-performance acidic ORR catalysts demands urgent attention.

Graphene has attracted significant attention due to its distinctive two-dimensional (2D) structure and exceptional chemical and physical properties [8–11]. As a novel nanocarbon material, it harbors versatile application potential [12]. Heteroatom-doped carbons have accrued considerable attention over preceding decades for their novel properties [13]. Doping studies reveal that nitrogen incorporation engenders unique electronic configurations, thereby modulating graphene’s reactivity and amplifying electrocatalysis [14, 15].

Three primary nitrogen configurations exist in the graphene framework: graphitic-N, pyridinic-N, and pyrrolic-N. Among these, pyridinic and graphitic configurations exert critical influence by rendering adjacent carbon atoms electrophilic. This electronic perturbation promotes oxygen adsorption, augmenting ORR kinetics [16–18].

The 3D structure not only retains the essential properties of 2D graphene materials and overcomes the stacking and reuniting problems of graphene, while concurrently delivering high specific surface area, robust mechanical strength, and optimized pathways for electrolytes and electron transfer [19–21]. Therefore, 3D graphene finds extensive application in diverse domains including energy storage [22, 23], catalysis [24, 25], sensing and separation [26–28]. Numerous studies indicate that template-assisted methodologies represent a prevalent approach for fabricating 3D graphene and doped variants. For example, Liu et al. [29] successfully use ethanol flame pyrolysis to remove the commercial polyurethane (PU) prepared 3D nitrogen-doped graphene. Sun et al. [30] fabricated 3D-NRGO exhibiting tunable pore diameters via a polystyrene template strategy.

In this paper, we developed a novel synthesis method for 3D nitrogen-doped reduced 3D-NRGO using PSS as a sacrificial template and polyaniline as a dual-function nitrogen source and structural stabilizer. Through a one-step high-temperature pyrolysis process, we simultaneously accomplished template removal, GO reduction, and in situ nitrogen doping. This methodology fosters the formation of a 3D porous framework while preferentially yielding graphitic-N configurations—critical determinants of electrocatalytic efficacy.

The resulting 3D-NRGO catalyst demonstrates exceptional ORR performance in acidic media, exhibiting both high activity and remarkable stability. Characterization data verify successful nitrogen incorporation with predominant graphitic-N speciation, whereas the 3D architecture confers enhanced mass transport and accessibility to active sites.

Experimental

Materials

Natural graphite powder was sourced from XF NANO, Inc. Potassium persulfate was purchased from Tianjin Chemical Reagent Co.,Ltd. Styrene, aniline and sodium dodecylsulfonatewas purchased from Shanghai Macklin Biochemical Technology Co., Ltd. All other reagents were purchased from XI LONG Chemical Reagent Co., Ltd. All reagents were of analytical reagent (AR). GO was synthesized from natural graphite via a modified Hummers method.

Preparation of graphite oxide

Pre-oxidation of graphite: A mixture of P₂O₅ (2.5 g), K₂S₂O₈ (2.5 g), and concentrated H₂SO₄ (98%, 12 mL) was introduced into a 500 mL beaker under vigorous stirring at 80 °C. Natural flake graphite (3 g) was subsequently added, followed by incubation in an 80 °C water bath for 4 h. Upon cooling to ambient temperature, the composite was diluted with 500 mL deionized water (DW) and equilibrated for 6 h. The resultant solid was exhaustively rinsed with DW until neutral pH (7.0) and dried at 60 °C for 12 h.

Hummer oxidation: The above sample was charged into a 500-mL three-neck flask. Under ice-bath cooling, concentrated H₂SO₄ (98%, 120 mL) was introduced, followed by gradual addition of KMnO₄ (15 g). The slurry was agitated magnetically for 2 h and subsequently at 35 °C for 45 min. DW (250 mL) was then dripped slowly into the mixture (maintained at 0–5 °C, followed by dilution with 500 mL DW and stirring for 5 min. Aqueous H₂O₂ (30%, 20 mL) was added dropwise, inducing vigorous gas evolution. The suspension was purified by sequential washing with 10% HCl (1 L) and DW (1.5 L) at 10,000 rpm until pH 7.0, dialyzed for 5–7 d, and lyophilized.

Synthesis of polystyrene

The polystyrene (PS) particles were prepared by emulsion polymerization as follows: 40 ml of Styrene (St, washed with a 0.05% aqueous solution of NaOH to remove the polymerization inhibitor in the monomer for use), 0.5 g of Sodium dodecyl sulfonate (SDS), and 200 ml of DW were charged into a 500-mL three-neck flask with a magnetic stirrer. The solution was degassed via N₂ bubbling (30 min, 25 °C). K₂S₂O₈ (0.2 g) was introduced under N₂ flow, initiating polymerization at 75 °C for 10 h. The latex was demulsified using 10% NaCl, filtered, washed, and dried.

Synthesis of sulfonated polystyrene template

PS (7.5 g) was sulfonated in concentrated H₂SO₄ (98%, 60 mL) within a 250 mL conical flask under magnetic stirring (40 °C, 12 h). The functionalized product was isolated by centrifugation, washed, and dried.

Synthesis of PSS/polyaniline/GO composites

PSS (0.3 g), DW (30 mL), and aniline (stoichiometry in Table 1) were mixed in a 250-ml single-neck flask. The composite was stirred in an ice bath (0–10 °C, ) and 4 ml of 2 mol/L HCl was added. A GO dispersion (4 ml 2 mol/L HCl, 40 ml 3.5 mg/ml oxidized graphite into a 100 ml beaker to stir well) was added dropwise, whereupon stirring continued for 1 h under ice-bath environment. An appropriate amount of 1.5 mol/L Na₂S₂O₈ solution was introduced, reacting for 6 h in the ice bath. The composite was then collected by filtration, washed, and lyophilized.

Table 1.

Ratio of the raw materials

Sample	PSS (g)	3.5 mg/ml GO (ml)	1.5 M Na2S2O8 (ml)	Ani (µl)
RGO	–	–	–	–
NRGO920	–	42	8	920
3D-NRGO230 3D-NRGO460	0.3 0.3	42 42	2 4	230 460
3D-NRGO690	0.3	42	6	690
3D-NRGO920	0.3	42	8	920
3D-NRGO920	0.3	42	10	1150

Preparation of three-dimensional nitrogen-doped graphene

Experimental sample: An appropriate amount of PSS/polyaniline/GO composites was placed in a quartz boat, and the composite was treated at 450 °C, for 30 min and then 800 °C, for 10 min in Ar atmosphere with a tube furnace.

Control sample: The preparation method of NRGO is the same as that of 3D-NRGO except that PSS was not added. RGO was prepared under the same high-temperature treatment conditions as 3D-NRGO.

The specific mixing ratio is shown in Table 1. The amount of added aniline is determined for each material x.

Characterization and instruments

The Scanning electron microscopy (SEM) images were gained at using JSM-6360 LA (JEOL Co., Japan). The Element mapping images and energy X-ray spectroscopy were obtained at using JSM-6360 LA. X-ray photoelectron spectroscopy (XPS) were recorded on an ESCALAB 250Xi X instrument (Thermo Fisher Scientific, USA), using Al Kα (hv = 1486.6 eV) as radiation source. Raman spectrum was recorded on a LABRAM-HR800 confocallaser micro-Raman spectrometer (Jobin Yvon, France) with a wavelength of 532 nm. X-ray diffraction (XRD) analyses were performed on a D8-Advance diffractometer (Bruker, German) operated at 40 kV and 40 mA, using a nickel-filtered Cu Kα(λ = 1.5406 × 10–10 m) as a radiation source. Brunauer–Emmett–Teller (BET) analysis studies were tested on Quantanchrome Instruments (Quadrasob SI).

Electrochemical measurement

Electrochemical measurement was performed on a CHI660E electrochemical workstation (Shang Hai Chenghua Instrument, China) in a standard three-electrode system. A platinum plate electrode and saturated calomel electrode (SCE, convert SCE to NHE E_NHE = E_SCE + 0.241 V) were acted as the counter electrode and reference electrode, respectively. The working electrode was a RGO, NRGOx(x = 920), 3D-NRGOx (x = 230, 460, 690, 920, 1150) modified glassy carbon electrode (GC, S = 0.07065 cm²) whose surface was coated with 1.0 μm of 0.3 μm and 0.05 μm aluminium oxide powder, followed by the addition of 5 µl of well-dispersed catalyst suspension [2 mg of nano-catalyst powder dispersed in 40 µl of Nafion (5 wt%) and 375 µl of ethanol solution, which was sonicated for 1 h], and then dried at room temperature.

The cyclic voltammetry (CV) data of ORR tests were collected at the room temperature in oxygen-saturated 0.5 M H₂SO₄ electrolyte and the potential range was from − 0.2 to 0.8 V (vs. SCE) at a scan rate of 100 mV s^− 1.

Results and discussion

Mechanism of the synthesis process

Scheme 1 The illustration of fabrication procedure of the 3D-NRGOx.

Scheme 1.

outlines the for the three-dimensional nitrogen-doped reduced graphene oxide (3D-NRGOx) (x = 230, 460, 690, 920, 1150) fabrication. PSS serves as a macroporous scaffold for electrostatic adsorption of aniline, which undergo oxidative polymerization on GO surfaces, affording PSS-PANI-GO hybrids. Subsequent pyrolysis effects PSS decomposition, GO reduction, and nitrogen doping, concurrently establishing the 3D-NRGOx framework. The specific mixing ratio is shown in Table 1. The amount of added aniline is determined for each material x

Structure and composition analysis

The morphology and structure of the as-synthesized 3D-NRGOx subsequent to high-temperature thermal treatment were characterized by scanning electron-microscopy (SEM). Figure 1a–c under varying magnifications demonstrate that 3D-NRGOx (x = 920) constitutes crumpled graphene sheets (analogous to RGO Fig. 1d) and exhibits a three-dimensionally interconnected porous architecture with stochastically distributed voids (aligned with the 3D-NRGOx morphology evidenced in Fig. S1a–d). As can be seen from Fig. 1d, the NRGO₉₂₀ manifests dense stacking devoid of discernible porosity. This morphological divergence stems principally from the incorporation of the PSS template, which undergoes pyrolytic gasification under elevated temperatures, thereby engendering the expansive, cross-linked porous matrix.

Fig. 1.

SEM images of 3D-NRGO₉₂₀ (a–c), RGO (d) and NRGO₉₂₀ (e)

Figure 2 presents the nitrogen adsorption-desorption isotherms of 3D-NRGO₉₂₀, showing a BET surface area of 274 m² g⁻¹. Mesopore size distribution ascertained via the BJH methodology (inset) encompasses 2.0–35 nm (Mainly distributed around 3.7 nm). These data corroborate the PSS template’s proficiency in fabricating porous materials possessing integrated three-dimensional microporous networks and mesoporous conduits.

Fig. 2.

N₂ adsorption–desorption isotherms of 3D-NRGO₉₂₀ and the insert is the pore size distribution

Elemental mapping coupled with energy-dispersive X-ray spectroscopy (EDS) verified homogeneous dispersion of carbon (C), nitrogen (N), and oxygen (O) throughout the 3D-NRGO₉₂₀ sample (Fig. 3b–d). Quantitative analysis disclosed a nitrogen content of 12.75 at.%, with spatially uniform elemental distribution apparent in both the principal specimen (Fig. 3) and auxiliary 3D-NRGOx variants (Fig. S2, Supporting Information). These outcomes constitute unequivocal validation of nitrogen’s successful integration into the graphene lattice.

Fig. 3.

Element mapping images of 3D-NRGO₉₂₀ a typical scanning electron microscopy (SEM) image and corresponding element mapping images of b carbon, c Nitrogen, and d oxygen in the selected area, and e EDS images of 3D-NRGO₉₂₀

XPS was employed to quantify nitrogen content and elucidate chemical modifications ensuing from nitrogen doping. Figure 4 displays the XPS survey spectra of the synthesized materials. The results showed that the 3D-NRGOx are composed of the Carbon (C 1s = 285.3 eV), oxygen (O 1s = 533.7 eV) and Nitrogen (N 1s = 400.1 eV). Quantitative analysis of the peak areas affirmed pervasive nitrogen doping across the 3D-NRGO_X samples. Figure 5 presents the high-resolution N 1s XPS spectrum of the 3D-NRGO₉₂₀ catalyst. Spectral deconvolution employing a mixed Gaussian-Lorentzian line profile resolved three discrete nitrogen configurations: Peak 1 (~ 400.5 eV) corresponds to graphitic-N [31]. The peak 2 at ~ 399.7 eV and peak 3 at ~ 398.7 eV can be assigned to pyrrolic-N [29] and pyridinic-N [5], respectively.

Fig. 4.

XPS survey spectra of 3D-NRGO_X (_X = 230, 460, 690, 920, and 1150)

Fig. 5.

High-resolution XPS spectrum of 3D-NRGO₂₃₀(a), 3D-NRGO₄₆₀(b), 3D-NRGO₆₉₀(c), 3D-NRGO₉₂₀ (d) and 3D-NRGO₁₁₅₀ (e) showing N 1s

Quantitative nitrogen analysis is tabulated in Table 2. XPS measurements revealed nitrogen doping concentrations of 4.58 wt%, 7.60 wt%, 6.94 wt%, 11.97 wt% and 6.96 wt% for 3D-NRGO₂₃₀, 3D-NRGO₄₆₀, 3D-NRGO₆₉₀, 3D-NRGO₉₂₀, and 3D-NRGO₁₁₅₀, respectively. These results implyed that during pyrolysis, PANI molecules underwent superficial decomposition on graphene while concurrently permeating the underlying matrix, thus homogeneous nitrogen incorporation was realized. Table 2 further delineates a volcano-type progression in nitrogen content relative to escalating aniline concentration during in situ polymerization: concentrations ascend to a zenith of 11.97 wt% (3D-NRGO₉₂₀) prior to diminishing at superior precursor loadings, the content is close to that analyzed by EDS (12.75%). This phenomenon originated predominantly from constrained pyrolysis kinetics—the thick polyaniline layers could not fully decompose within the fixed heating duration (30 min at 800 °C) at elevated aniline concentrations. Thus, while higher aniline loading provided more nitrogen source, the shortened residence time per polymer chain at high precursor loads ultimately decreased the doping yield. XPS quantitative analysis further established graphitic-N as the predominant configuration, surpassing both pyrrolic-N and pyridinic-N configurations (Table 2). This finding confirmed that in situ pyrolysis of polyaniline on GO at 800 °C favored graphitic-N formation, conferring significant catalytic advantages attributable to its augmented thermal resilience in acidic media, superior electron-donating capability, and optimal binding energetics toward oxygenated intermediates.

Table 2.

The nitrogen element type content of 3D-NRGO sample measured by XPS

Sample 3D-NRGOx	Total nitrogen content (%)	Graphite nitrogen content (%)	Pyrrole nitrogen content (%)	Pyridine nitrogen content (%)	Graphite nitrogen/Pyrrole nitrogen
230	4.58	1.83	1.69	1.06	1.08
460	7.60	4.03	1.59	1.98	2.52
690	6.94	3.33	1.73	1.88	1.92
920	11.97	7.18	2.75	2.04	2.60
1150	6.96	3.54	1.46	1.96	2.40

X-ray diffraction (XRD) was use to analyze the crystallographic structure of graphite, GO, reduced graphene oxide (RGO), NRGO₉₂₀, 3D-NRGO₉₂₀ (Fig. 6). Figure 6(a) exhibit the (002) characteristic diffraction peak of the raw material of graphite appeared at 26.4° (d₍₀₀₂₎ = 0.337 nm), and the peak is sharp. Compared with the graphite, GO displayed a broadened (002) diffraction peak shifted toward lower angles at 2θ = 10.4° (d (001) = 0.849 nm). This increased interlayer spacing signified successful intercalation of abundant oxygen-containing functional groups. The RGO diffraction peak appeared at 25.8° (d₍₀₀₂₎ = 0.345 nm), with reduced interlayer spacing relative to GO, indicating effective removal of most oxygen functionalities under thermal reduction. The (002) diffraction peak appeared at 25.4° (d ₍₀₀₂₎ = 0.350 nm) and 25.1° (d₍₀₀₂₎ = 0.369 nm), corresponded to NRGO₉₂₀ and 3D-NRGO₉₂₀, respectively. While that of 3D-NRGO₉₂₀ with three-dimensional structure was larger than that of 2D NRGO₉₂₀ layer, and the interlayer spacing of NRGO₉₂₀ was smaller than that of RGO. These results suggested that N doping led to the formation of more closely packed graphene nanostructures, primarily attributable to enhanced π-π interactions. Conversely, the three-dimensional architecture effectively mitigates π-π restacking. π-π re-stacking.

Fig. 6.

a X-ray diffraction patterns of graphite, GO, RGO, NRGO, and 3D-NRGO₉₂₀; b Raman spectra of GO, RGO, NRGO₉₂₀, and 3D-NRGO₉₂₀

The typical Raman spectra of GO, RGO, NRGO₉₂₀, and 3D-NRGO₉₂₀ are shown in Fig. 6b. They all showed two typical Raman peaks at about 1350 cm^− 1 and1580cm^− 1, corresponding to the typical D band and G band of graphene, respectively. Despite minimal peak position variations, their I_D/I_G ratios diverge (0.98, 1.01, 1.08, 1.09). The I_D/I_G values of RGO was higher than that of GO, which might be ascribed to defect generation during thermal reduction of oxygen groups [32]. Higher ratios in NRGO₉₂₀ and 3D-NRGO₉₂₀ relative to RGO reflected increased defect density from nitrogen doping [33]. Comparable I_D/I_G values for 3D-NRGO₉₂₀ and NRGO₉₂₀ demonstrated that three-dimensional structuring introduces no additional defects.

Electrochemical investigation

The electrocatalytic performance of 3D-NRGO₉₂₀ for the ORR was systematically evaluated in 0.5 M H₂SO₄ electrolyte using a standard three-electrode configuration. Cyclic voltammetry (CV) measurements were performed using a 3D-NRGO₉₂₀-modified glassy carbon (GC) working electrode in both N₂- and O₂-saturated 0.5 M H₂SO₄ electrolyte to characterize the ORR activity. Apparently, ORR peak for 3D-NRGO₉₂₀ is observed at 0.151 V (vs. NHE) in O₂ saturated solution, whereas the significant ORR peak does not emerge in N₂ saturated solution (Fig. 7a). The expanded CV area under N₂ arose from two factors: First, the mesoporous structure of 3D-NRG₉₂₀ (BET: 274 m²/g) provides substantial charge storage sites, enabling full electric double-layer capacitance expression. Second, pyridinic-N (XPS: 389.7 eV) and pyrrolic-N (XPS: 400.5 eV) configurations introduced significant pseudo-capacitance, augmenting charge storage. Under O₂ saturation, cathodic ORR currents partially offset capacitive currents, reducing the net CV area. As show in Fig. 7b, comparative studies were performed for the 3D-NRGO₉₂₀, NRGO₉₂₀ and RGO, the current density peaks of both 3D-NRGO₉₂₀ and NRGO₉₂₀ were higher than that of RGO, corresponding to 8.14 mA/cm², 6.87 mA/cm² and 6.22 mA/cm², respectively. The CV analysis data demonstrated that N doping could provide more active sites for electrochemical catalysis. The superior current density of 3D-NRGO₉₂₀ over NRGO₉₂₀ originated from multidimensional electron transfer pathways within its 3D architecture. Methanol tolerance tests (Fig. 7d) exhibited unaltered ORR peak potential and current density upon adding 1 M methanol, underscoring exceptional catalytic stability for direct methanol fuel cell applications.

Fig. 7.

a CV curves of 3D-NRGO₉₂₀ in O₂-Saturated 0.5 M H₂SO₄ solution. Scan rate:100 mV s^− 1, Scan direction: negative to positive, b CV curves of RGO, NRGO₉₂₀, and 3D-NRGO₉₂₀ in O2-Saturated 0.5 M H₂SO₄ solution. Scan rate:100 mV s^− 1, Scan direction: negative to positive, c LSV curves of RGO, NRGO₉₂₀ and 3D-NRGO₉₂₀ in a O₂ saturated 0.5M H₂SO₄ solution at a rotation rate of 1600 rpm. Scan rate: 10 mV s^− 1, Scan direction: positive to negative, d CV curves of 3D-NRGO₉₂₀ in O₂-Saturated 0.5 M H₂SO₄ solution before and after the addition of 1 M methanol Scan rate:100 mV s^− 1, Scan direction: negative to positive

Chronopotentiometric analyses (Fig. 8) compared ORR stability of RGO, NRGO₉₂₀, and 3D-NRGO₉₂₀ on glassy carbon electrodes at identical loadings. NRGO₉₂₀ demonstrated superior stability to RGO, while 3D-NRGO₉₂₀ exhibited enhanced oxygen redox kinetics with a 318 mV positive potential shift (− 116 vs. NHE) and higher ORR potential. This further demonstrated that the 3D structure could improve the stability and catalytic activity than 2D N-doped graphene.

Fig. 8.

Chronopotentiograms of ORR at 1 mA/cm² in O₂-Saturated 0.5 M H₂SO₄ solution at room temperature at same loadings on GC of three material a 3D-NRGO₉₂₀, b NRGO₉₂₀, and c RGO

To study the kinetics of the 3D-NRGOx catalyst, we conducted a rotating disk electrode (RDE), acquiring linear sweep voltammetry curves for 3D-NRGO₉₂₀ at rotation rates spanning 400–3600 rpm (Fig. 9a). The Koutecky–Levich plots (j^− 1 vs. ω^−1/2 ) were obtained from ORR curves at different potential. The slopes of the linear regression lines were utilized to calculate the electron transfer number (n) per the Koutecky–Levich equation [34–36].

Fig. 9.

a RDE curves of 3D-NRGO₉₂₀ in O₂-saturated 0.5 M H₂SO₄ solution at various rotation rates. Sweep rate:10 mv s^− 1, Scan direction: positive to negative, b corresponding Koutecky–Levich plots at different potentials

where J denotes the measured current density; Jk represents the kinetic current densities; ω is the electrode rotating rate (ω = 2 πN, N is the linear rotation speed); n signifies the number of electrons transferred in the oxygen reduction; F is the Faraday constant (F = 96485 C mol^− 1); C₀ is the bulk concentration of O₂ (C₀ = 1.03 × 10^– 6 mol cm^− 3); D₀ stands for the diffusion coefficient of O₂ in the 0.5 M H₂SO₄ electrolyte (D₀ = 2.1 × 10^− 5 cm² s^− 1); v is the kinematic viscosity of the electrolyte (v = 0.01 cm² s^− 1). The average value of n approximates 4, indicating a predominant four-electron transfer pathway.

The ORR mechanism

Figure 10a demonstrates that 3D-NRGO₉₂₀ exhibits the highest catalytic current density, followed by 3D-NRGO₁₁₅₀ and 3D-NRGO₄₆₀. This performance trend correlates directly with the graphitic-N to pyrrolic-N ratio, as shown in Fig. 10b and quantified in Table 3. The enhanced activity could be attributed to the key factors: First, pyrrolic-N readily converted to graphitic-N at high temperatures (800 °C), as confirmed by XPS analysis (graphitic-N content: 7.18 wt% in 3D-NRGO₉₂₀ and Peak intensity ratio (graphitic-N/pyrrolic-N = 2.60). Second, graphitic-N possessed superior stability and catalytic activity in acidic electrolytes (Fig. 10a).

Fig. 10.

a CV curves of 3D-NRGO₂₃₀, 3D-NRGO₄₆₀, 3D-NRGO₆₉₀, 3D-NRGO₉₂₀ and 3D-NRGO₁₁₅₀ in O₂-Saturated 0.5 M H₂SO₄ solution. Scan rate: 100 mV s^− 1, Scan direction: negative to positive, b Graphite nitrogen and pyrrole nitrogen ratio with the catalytic current density diagram

Table 3.

The data of current density and ratio of Graphitic-N: Pyrrolic-N

Sample	Current density (mA/cm2)	Graphitic-N: Pyrrolic-N
3D-NRGO230	2.10	1.08
3D-NRGO460	6.60	2.52
3D-NRGO690	3.77	1.92
3D-NRGO920	8.14	2.60
3D-NRGO1150	6.72	2.40

The reaction mechanism of 3D-NRGOx (x = 230, 460, 690, 920, 1150) is depicted in Fig. 11a. Pyrrolic-N, bonding to two carbon atoms with a lone electron pair, is unstable and readily forms pyrrolic-N–H, precluding its role in ORR. Pyridinic-N is also a type of nitrogen that bonds to two carbon atoms in the graphene framework with a basic lone pair of electrons. Since the lone pair of electrons is not delocalized into the aromatic π-system, the pyridinic-N can be protonated to pyridinic-N–H [37]. Liu et al. proved that the more Pyridinic-N content in the acidic electrolyte, the faster the performance decreased [38]. Thus, although catalytically active, pyridinic-N is prone to protonation-induced deactivation. Conversely, graphitic-N bonds to three carbon atoms within the graphene plane without accessible lone pairs, rendering it resistant to protonation and explaining the high stability of 3D-NRGOx. Ikeda et al. and Niwa et al. propose graphitic-N as the primary ORR active site and stability source in N-doped carbon materials under acidic conditions [39, 40].

Fig. 11.

a Schematic of the protonation reaction on 3D-NRGO catalyst in acidic electrolyte [35], b The proposed ORR catalytic cycle for the N structure of 3D-NRGO [41]

The reaction mechanism of graphitic-N for ORR is show in Fig. 11b. Graphitic-N confers electrophilic character, inducing electropositivity on adjacent carbon atoms. O₂ molecules adsorb non-dissociatively on the carbon atoms adjacent to graphitic-N, followed by weakening of the O=O double bond and formation of a single-bonded –O–O– intermediate (State II). Upon reaching this stable configuration and receiving electrons in acidic electrolyte, a selective H⁺ attack occurs at the terminal oxygen atom, facilitated by hydrophobic –CH moieties, yielding a hydroperoxyl intermediate (–O–OH, State III). This intermediate readily dissociates into a terminal oxo group (–O–) and a hydroxyl group (–OH). Subsequently, the first water molecule (H₂O) forms through proton (H⁺) transfer to the hydroxyl group (–OH), completing the initial 2e⁻ reduction step. The remaining oxygen atom forms a carbonyl group (C=O), which subsequently cleaves the adjacent C–N bond in Step (3), yielding a stable formyl group (CHO) and a nitrogen site that converts to pyridinic-N (State IV). In the other model structures considered in this paper, the ring-opening at the graphene edge was not observed. When H⁺ carried and attached to –O– in step (4), the broken C–N bond reconnects, and N transformed back into graphene nitrogen (state (V)), leading to the formation of a second H₂O molecule in step (5) and returning to state (I) [41]. This cyclic reaction mechanism underscores the exceptional stability of graphitic-N in acidic ORR. Collectively, constructing a 3D porous structure with high graphitic-N content confers excellent ORR performance in acidic electrolyte.

Conclusions

In conclusion, the 3D-NRGO catalysts can be prepared by incorporating template method and nitrogen doping. The 3D-NRGO₉₂₀ show the highest ORR activity in acidic electrolyte, surpassing NRGO₉₂₀ and RGO as an electrocatalyst. Moreover, the 3D-NRGO demonstrates exceptional stability and methanol tolerance for acidic ORR. XPS analysis revealed a maximum nitrogen doping content of 11.97 wt%. Graphitic-N appears to play the paramount role in acidic ORR activity. These results suggest its potential as an ideal cathode catalyst for acidic methanol fuel cells. The unique porous structure holds promise for applications in the catalyst, fuel cells, energy storage and biosensor systems.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Concept and Design: Professor zhang xin conceived the initial concept of the study, designed the overall research framework, and determined the research direction and key issues.Data Collection: wu min cong was responsible for the experimental operations and the collection of raw data, including [specific experimental operations], ensuring the accuracy and integrity of the data.Data Analysis: yao yue utilized [specific analysis methods or software] to conduct an in-depth analysis of the collected data, interpreted the results, and provided support for the discussion.Writing and Revision: wu min cong wrote the first draft of the paper, and yao yue revised the paper multiple times to optimize the language expression and logical structure.Funding Support: Professor zhang xin obtained the funds required for the study, providing material and human resources guarantees for the research.

Funding

All the expenses related to this research, including but not limited to personnel costs, material procurement, and equipment usage, were Supported by Zhang Xin’s research group, Shantou university, the Natural Science Research Project of Higher Education in Anhui Province (No. 2024AH050325), the Science and Technology Planning Project of Bengbu city (No. 2022gx10) and the Foundation of Anhui Science and Technology University (No. HCWD202001).

Data availability

Sequence data that support the findings of this study have been deposited in the European Nucleotide Archive with the primary accession code PRJWB13140, or Data is provided within the manuscript or supplementary information files.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.