Dual‐Functional g‐C₃N₄ Templated Synthesis of N‐Doped Hierarchical Porous Carbon for High‐Performance Zinc‐Ion Hybrid Capacitors

ABSTRACT

The pursuit of high‐performance cathode materials is essential for advancing aqueous zinc‐ion hybrid capacitors (ZIHCs). However, conventional porous carbons often suffer from limited capacity and unsatisfactory rate capability due to insufficient active sites, mismatched pores, and low nitrogen‐doping levels. Herein, we propose a novel strategy for synthesizing nitrogen‐doped porous carbon (NPC) with a hierarchical pore structure using graphitic carbon nitride (g‐C₃N₄) as a dual‐function soft template and nitrogen source and potassium citrate as a combined carbon precursor and activating agent. The optimal material, NPC‐0.5 (with a g‐C₃N₄/potassium citrate mass ratio of 1:3), exhibits a high specific surface area (805 m² g⁻¹), a well‐defined hierarchical pore network, and a nitrogen content of 7.94 at% dominated by graphitic‐N species, which collectively enhance Zn²⁺ storage and facilitate rapid ion transport. When employed as a cathode for ZIHCs, the NPC‐0.5 delivers a high specific capacity of 172 mAh g⁻¹ at 0.1 A g⁻¹, excellent rate capability (41% capacity retention at 20 A g⁻¹), and outstanding long‐term cycling stability (86% capacity retention after 65,000 cycles at 10 A g⁻¹). This work provides an efficient and scalable approach for fabricating high‐performance nitrogen‐doped carbon cathodes for advanced ZIHCs.

A novel nitrogen‐doped hierarchically porous carbon is synthesized via a dual‐functional g‐C₃N₄ templating strategy. Synergistic K⁺ etching and co‐pyrolysis with potassium citrate yield a high surface area and robust graphitic‐N doping. As a zinc‐ion hybrid capacitor cathode, it delivers a high specific capacity and ultra‐long stability over 65,000 cycles.

1. Introduction

The extensive reliance on fossil fuels has led to severe environmental pollution, spurring urgent demand for advanced energy storage devices (ESDs). In the context of global carbon peak and carbon neutrality initiatives, the development of eco‐friendly systems capable of efficient energy storage and rapid release has become a paramount scientific priority [1]. Among various ESDs, batteries are widely deployed in electric vehicles and portable electronics due to their high energy density and environmental compatibility [2, 3]. However, their further application is limited by relatively low power density and poor cycling stability. Supercapacitors (SCs) [4, 5, 6], another important class of ESDs, are regarded promising owing to their high power density and long cycle life, yet they suffer from lower energy density compared to batteries [7]. Therefore, it is imperative to develop hybrid energy storage systems that synergistically combine the high energy density of batteries with the high power density and longevity of SCs. Aqueous zinc‐ion hybrid capacitors (ZIHCs) have emerged as a representative example of such hybrid systems. Zinc metal is considered as an ideal anode material due to its small ionic radius (0.74 Å), high theoretical capacity (823 mAh g⁻¹ or 5855 mAh cm⁻¹), low redox potential (−0.76 V vs. SHE), and minimal polarization. As a result, ZIHCs are recognized for their high energy density, high power density, cost‐effectiveness, and inherent safety. The performance of ZIHCs is heavily dependent on electrode materials, particularly the capacitive cathode, whose specific capacity must match that of the zinc anode to avoid limiting the overall energy density [8]. Consequently, enhancing the zinc storage capacity of cathode materials represents a critical research direction.

Porous carbon materials, characterized by high specific surface area (SSA) and tunable pore structures, have demonstrated excellent capacitive performance in ZIHCs [9]. However, due to insufficient active sites and mismatched pore‐ion sizes, the capacity of porous carbon materials remains far below the theoretical value, which limits their energy density. Template approaches provide an effective route for synthesizing porous carbon materials with controlled pore structures and abundant active sites [10]. These methods are generally categorized into hard and soft templating. Hard templates (e.g., SiO₂, MgO, Al₂O₃ nanoparticles) are often associated with high cost and complex removal processes, limiting their scalability [11]. In contrast, soft templating, based on the self‐assembly of precursors and templating agents followed by carbonization, allows for more intimate contact and easier template removal, demonstrating great potential for practical application [12]. Recent advances have demonstrated the effectiveness of sustainable template strategies, such as the use of recyclable NaCl templates for preparing N/O co‐doped porous carbon with enhanced zinc‐ion storage performance [13]. Additionally, heteroatom doping, particularly nitrogen doping, can effectively enhance the electrochemical properties of porous carbon materials by introducing additional active sites and modifying their electronic structure [14, 15]. However, achieving high nitrogen doping levels through conventional soft templating remains challenging due to the limited nitrogen content in most precursors [16]. Graphitic carbon nitride (g‐C₃N₄), composed solely of carbon and nitrogen, presents a unique opportunity as it can serve simultaneously as a soft template and a rich nitrogen source, enabling simultaneous pore formation and nitrogen doping during thermal decomposition [17, 18, 19]. Despite these advantages, reports on the application of g‐C₃N₄‐templated, highly nitrogen‐doped porous carbons in ZIHCs remain scarce. Moreover, the specific influence of such a template on the SSA and pore size distribution of the resulting carbons requires further investigation.

In this study, we prepared nitrogen‐doped porous carbon (NPC) with a hierarchical pore structure via one‐step high‐temperature pyrolysis and chemical activation, using g‐C₃N₄ as both a nitrogen source and a template, and potassium citrate as a carbon source and an activating agent. This novel approach integrates self‐activation, templating, and heteroatom doping into a single efficient process, where carbonization, activation, and template removal occur simultaneously. We systematically investigated the effect of the g‐C₃N₄‐to‐potassium citrate ratio on the SSA, nitrogen doping, and electrochemical performance of the resulting materials, and revealed the core mechanism of this dual‐functional system. During pyrolysis, potassium citrate decomposes before g‐C₃N₄, releasing K⁺ ions that intercalate into the g‐C₃N₄ interlayers. These intercalated ions exert a catalytic etching effect, lowering the thermal decomposition temperature of g‐C₃N₄ and precisely replicating its layered porous structure into the emerging carbon matrix. Meanwhile, nitrogen‐rich gases released from g‐C₃N₄ decomposition further expand the pore structure, forming an interconnected hierarchical pore network. This addresses the pore‐ion size mismatch typical of conventional porous carbons while eliminating the need for complex template removal. Concurrently, nitrogen‐containing active species from g‐C₃N₄ are in‐situ incorporated into the carbon skeleton, achieving high‐level nitrogen doping dominated by highly conductive graphitic‐N. This provides abundant active sites for Zn²⁺ storage and enhances the electronic conductivity of the material. The optimal sample, prepared with a mass ratio of 1:3 (denoted as NPC‐0.5), exhibited a high SSA of 805 m² g⁻¹ and an effective nitrogen‐doping level of 7.94 at%. When assembled as a cathode in ZIHCs, the NPC‐0.5 delivered a high specific capacity of 172 mAh g⁻¹ at 0.1 A g⁻¹ and demonstrated exceptional long‐term stability, retaining 85% of its initial capacity after 65,000 cycles at 10 A g⁻¹, highlighting its excellent electrochemical performance and durability.

2. Results and Discussion

2.1. Morphology and Structural Analysis

Figure 1a illustrates the preparation process for NPC. Initially, g‐C₃N₄ was synthesized via the thermal polycondensation of urea [20]. The precursor mixture was then obtained by thoroughly grinding g‐C₃N₄ with potassium citrate, followed by carbonization in a tube furnace at 600 °C under a N₂ atmosphere. Based on the mass ratios of g‐C₃N₄ to potassium citrate in the precursors (0.25:1.5, 0.5:1.5, and 0.75:1.5), the resulting samples were labeled as NPC‐0.25, NPC‐0.5, and NPC‐0.75, respectively. The sample prepared without g‐C₃N₄ was denoted as PC. Furthermore, to clarify the specific role of K⁺ ions in the activation process, potassium citrate was replaced with an equimolar amount of citric acid, and the product obtained under this condition was designated as NC‐0.5. To elucidate the material transformation and activation mechanism during the pyrolysis of the precursor, thermogravimetric analysis (TGA) was firstly conducted (Figure S1). The TGA curves reveal distinct thermal behaviors for the individual components. Potassium citrate exhibits early initial decomposition between 200°C and 300°C [21], yet it shows a high residual mass of 69% at 600°C due to the formation of thermally stable inorganic salts and carbon. In contrast, g‐C₃N₄ displays excellent low‐temperature thermal stability with negligible mass loss below 350°C, but undergoes rapid gasification at higher temperatures, retaining about 56% of its mass at 600°C. For the NPC‐0.5 precursor, the theoretical residual mass at 600°C, assuming a simple physical mixture, is 66%. However, the actual residual mass drops significantly to 57%, providing direct evidence that the early decomposition of potassium citrate generates abundant K⁺ ions that intercalate into g‐C₃N₄ and exert a catalytic etching effect. This effect lowers the activation energy for g‐C₃N₄ decomposition, accelerating its conversion into nitrogen‐rich gases at reduced temperatures [22, 23, 24, 25]. These in‐situ released gases further contribute to pore expansion, while the nitrogen species derived from the decomposed g‐C₃N₄ are effectively incorporated into the nascent carbon matrix, achieving nitrogen doping. The synergistic decomposition not only increases the SSA but also introduces hierarchical pores and nitrogen doping, which are expected to increase Zn²⁺ adsorption sites and improve contact between the electrode and the electrolyte.

FIGURE 1.

(a) Schematic of the synthesis process for NPC‐x; (b‐e) SEM images of PC, NPC‐0.25, NPC‐0.5, and NPC‐0.75, respectively; (f) TEM images of NPC‐0.5; (g) HR‐TEM image of NPC‐0.5; (h) TEM images of PC‐0.5; (i‐l) HAADF‐STEM image and corresponding elemental mapping images of NPC‐0.5.

Figure 1b–e show the scanning electron microscopy (SEM) images of PC and the NPC‐x samples. The PC (Figure 1b) exhibits a regular polygonal sheet‐like structure with uniform interlayer gaps, indicating that carbon produced by the pyrolysis of potassium citrate grows and stacks freely in the absence of a template. Upon introducing g‐C₃N₄, the morphology of NPC‐x undergoes a great change. The NPC‐0.25 (Figure 1c) appears as aggregates of spherical particles, suggesting that a small amount of g‐C₃N₄ disrupts carbon nucleation and growth. However, its templating effect is not fully realized, likely because the g‐C₃N₄ is completely etched by potassium citrate during pyrolysis. Its layered template structure fails to be preserved, primarily contributing to pore creation and promoting carbon spheroidization. When g‐C₃N₄ and potassium citrate are combined at a mass ratio of 1:3, a synergistic effect is achieved in NPC‐0.5. Potassium citrate fully catalyzes the decomposition of the g‐C₃N₄ template and precisely replicates its ordered porous structure into the final carbon material. The SEM image of NPC‐0.5 (Figure 1d) closely resembles the morphology of the g‐C₃N₄ precursor (Figure S2a), which exhibits a loose, porous structure composed of irregularly stacked 2D sheets with abundant interlayer voids and surface pores, indicating the successful replication of an open, interconnected carbon framework with a hierarchical pore structure. Similarly, the NC‐0.5 (Figure S2b) displays a comparable morphology, confirming that citric acid effectively catalyzes the decomposition of g‐C₃N₄ and enables the resulting carbon to inherit its structural features. In contrast, the NPC‐0.75 (Figure 1e) displays an irregular morphology, which is attributed to excess g‐C₃N₄ coupled with an insufficient amount of activating agent, preventing the complete decomposition of g‐C₃N₄.

The transmission electron microscopy (TEM) image (Figure 1f) reveals that NPC‐0.5 is composed of aggregated fine nanoparticles, forming a loose and porous sponge‐like network structure. The HRTEM image (Figure 1g) displays a highly disordered carbon arrangement consisting of randomly stacked, curved nanocrystals, which generate numerous open and closed pores as well as edge defects. The selected area electron diffraction (SAED) pattern (inset in Figure 1g) shows diffuse diffraction rings, confirming a highly amorphous carbon structure with long‐range disorder and defect‐rich nanocrystalline domains. In contrast, the PC presents a dense, smooth‐edged sheet‐like or blocky structure with minimal porosity and structural defects (Figure 1h). Elemental distribution in NPC‐0.5 was further examined by high‐angle annular dark‐field scanning TEM (HAADF‐STEM) and energy‐dispersive X‑ray (EDX) elemental mapping (Figure 1i–l). The results demonstrate a uniform distribution of C, O, and N elements throughout the material, which clearly confirms the successful incorporation of nitrogen into the carbon framework. The oxygen signal primarily originates from the carbonized residue of potassium citrate. Such a disordered carbon structure rich in defects and nitrogen dopants is anticipated to enhance the Zn²⁺ adsorption and storage capacity, while the open and interconnected porous network promotes rapid electrolyte ion transport, thereby significantly reducing diffusion resistance and thereby improving the rate capability of the material.

To elucidate the structural and pore characteristics of these materials, X‐ray diffraction (XRD), Raman spectroscopy, Fourier transform infrared spectroscopy (FTIR), X‐ray photoelectron spectroscopy (XPS), and nitrogen adsorption‐desorption measurements were performed. The XRD patterns (Figure 2a) reveal that PC, NPC‐0.25, and NPC‐0.5 exhibit two broad diffraction peaks at approximately 22.2° and 43.1°, corresponding to the (002) and (100) planes of graphitic carbon (JCPDS No. 41–1487), respectively [26]. Notably, the (002) peak of NPC‐0.5 is significantly broadened, indicating its amorphous structure, consistent with the SAED results. In contrast, the NPC‐0.75 shows a distinct diffraction peak at 27°, which can be attributed to residual incompletely decomposed g‐C₃N₄ (Figure S3a). This observation perfectly aligns with the aforementioned TGA results, further confirming that thermal effects alone at 600°C are insufficient to fully decompose g‐C₃N₄, and the catalytic etching by K⁺ ions is crucial. Furthermore, the mass ratio of g‐C₃N₄ to potassium citrate significantly affects the completeness of the reaction: a ratio of 1:3 (NPC‐0.5) allows complete conversion, whereas an excess of g‐C₃N₄ (NPC‐0.75) leads to an incomplete reaction and residual g‐C₃N₄. It is noteworthy that the (002) peak of NC‐0.5 is broader than that of NPC‐0.5 (Figure S3b), indicating that even if g‐C₃N₄ can be completely decomposed, the absence of potassium ions will result in a carbon skeleton with higher disorder and richer defects, as well as smaller crystal size. This comparison strongly demonstrates that potassium citrate is not only indispensable in promoting the decomposition of g‐C₃N₄, but also play a key catalytic graphitization role in the subsequent rearrangement of carbon atoms and the growth of graphite microcrystals.

FIGURE 2.

Structural and chemical characterization of PC and NPC‐x samples. (a) XRD patterns; (b) Raman spectra; (c) FTIR spectra; (d) XPS survey spectra; (e‐g) High‐resolution XPS spectra of C 1s, O 1s, and N 1s regions; (h) N₂ adsorption‐desorption isotherms, and (i) the corresponding pore size distribution.

The Raman spectra of all samples (Figure 2b and Figure S3c) display characteristic D and G bands at 1350 cm⁻¹ and 1590 cm⁻¹, corresponding to structural disordered sp³ carbon and the ordered sp² graphitic carbon, respectively [27]. The graphitization degree of the materials was evaluated by the intensity ratio of the D to G bands (ID/IG) [28]. Except for NPC‐0.75, which was affected by fluorescence interference from residual g‐C₃N₄, all potassium‐containing samples exhibit a clear decrease in the ID/IG ratio, indicating that K⁺ ions promote pore formation and catalyze the transformation of amorphous carbon into ordered graphitic microcrystals at elevated temperatures, a phenomenon often termed “catalytic graphitization” [29]. This process may locally enhance the degree of order while preserving the porous structure by repairing dangling bonds at the edges of sp² carbon domains. In contrast, NC‐0.5, prepared without potassium, shows no such decrease in ID/IG (Figure S3c), confirming that proton catalysis alone cannot induce similar graphitic ordering. Taken together with its high nitrogen content and poorly developed pore structure, this result highlights the unique role of K⁺ ions in optimizing the structural ordering of the carbon skeleton, which helps improve conductivity and structural stability. The chemical structures of the materials were further characterized by FTIR. As shown in Figure 2c, NPC‐x series exhibit strong characteristic absorption peaks at around 3400, 1630, and 1400 cm⁻¹, corresponding to the N‐H/O‐H stretching vibrations, C = N bending vibrations, and C‐N stretching vibrations [30, 31], respectively. In contrast, these signals are very weak in the PC sample. This marked difference strongly confirms the successful incorporation of nitrogen into the carbon framework of the NPC‐x materials.

The XPS survey spectra (Figure 2d) confirm the presence of C, N, and O elements in all samples. The nitrogen content determined by XPS (Table S1) varies significantly: PC (1.42 at%), NC‐0.5 (24.57 at%), NPC‐0.25 (4.30 at%), NPC‐0.5 (7.94 at%), and NPC‐0.75 (23.42 at%). The nitrogen content in these NPC‐x series increases with the amount of g‐C₃N₄ added, while the trace nitrogen in PC originates from annealing in a nitrogen atmosphere. Notably, the NC‐0.5 exhibits an extremely high nitrogen content, indicating that H⁺ ions only partially disrupt the g‐C₃N₄ framework, leaving behind residual C─N bonds and thus enriched nitrogen. In contrast, K⁺ ions in NPC‐0.5 break C─N bonds thoroughly, achieving both effective template decomposition and controlled nitrogen doping. The C 1s spectra (Figure 2e) can be deconvoluted into groups such as C─C/C = C (284.8 eV), C─O/C─N (286.2 eV), and C = O (288.5 eV). As the amount of g‐C₃N₄ increases, the proportion of C─N bonds in the NPC‐x series rises significantly, further confirming the successful incorporation of nitrogen into the carbon skeleton. The O 1s spectra (Figure 2f) show that the oxygen‐containing functional groups (C = O, C─O) are present in all samples, primarily derived from the carbonization residue of citrate. The N 1s spectra (Figure 2g) can be deconvoluted into pyrrolic nitrogen (N‐5, 399.1 eV) and graphitic nitrogen (N‐Q, 400.1 eV). N‐5 can introduce defect sites to enhance ion adsorption, while N‐Q improves charge transport mobility [32]. From NPC‐0.25 to NPC‐0.5, the proportion of N‐Q increases significantly, indicating that K⁺ ions promote the doping of nitrogen in a more conductive form at the optimal ratio. The N 1s spectrum of NPC‐0.75 highly matches that of g‐C₃N₄ (Figure S4a), confirming the presence of abundant undecomposed template. In contrast, NC‐0.5 contains more N‐5 (Figure S4b), leading to high defect density but poor conductivity. These comparisons further highlight the key role of K⁺ ions in optimizing nitrogen doping configuration—acting not only as a pore‐forming agent but also as a “doping morphology optimizer.”

The SSA and pore size distribution of these samples were further investigated by low‐temperature nitrogen adsorption‐desorption measurements (Figures 2h,i and S5). NPC‐0.5 exhibits a Type‐IV isotherm with an H4 hysteresis loop, characteristic of a hierarchical pore system containing both micropores and mesopores. Its BET SSA is as high as 803 m² g⁻¹, with a micropore volume of 0.244 cm³ g⁻¹ and a micropore area of 445 m² g⁻¹ (Table S2), significantly outperforming these of the other samples. This well‐developed porosity originates from the synergistic effect of the structural guidance provided by the g‐C₃N₄ template and the deep activation/etching effect of K⁺. Although PC possesses a moderate SSA of 432 m² g⁻¹, its pore system is primarily microporous with poorly developed mesopores, reflecting the randomness of the K⁺ activation process in the absence of a structural template. In contrast, NPC‐0.25 and NPC‐0.75 exhibit very low SSAs (17 and 21 m² g⁻¹, respectively). The former results from the collapse of the pore structure due to an insufficient amount of g‐C₃N₄ template, while the latter is due to pore blockage by excessive residual g‐C₃N₄. Notably, the distinct pore size distributions of NC‐0.5 and NPC‐0.5 (Figure S5) reflect different pore‐forming mechanisms for citric acid and potassium ions, as detailed in the Supporting Information.

Structural and compositional analysis of the NPC‐x series reveals a clear correlation between the g‐C₃N₄‐to‐potassium citrate mass ratio and the resulting material properties. At a low ratio (NPC‐0.25), the limited amount of g‐C₃N₄ is insufficient to maintain a stable template structure during pyrolysis, leading to pore collapse and a dramatically reduced SSA of 17 m² g⁻¹. Conversely, at an excessive ratio (NPC‐0.75), incomplete decomposition of g‐C₃N₄ leaves residual template residues that block pores and limit nitrogen accessibility, resulting in a low SSA of 21 m² g⁻¹ and a high nitrogen content dominated by inactive species. Only at the optimal ratio (NPC‐0.5) does a balanced synergy occur: potassium citrate effectively catalyzes the decomposition of g‐C₃N₄, enabling the formation of a well‐interconnected hierarchical pore structure with an SSA of 805 m² g⁻¹, while simultaneously facilitating nitrogen incorporation in a favorable graphitic configuration. This optimal configuration provides abundant active sites, enhanced conductivity, and rapid ion transport pathways, which are expected to boost the electrochemical performance of NPC‐0.5.

2.2. Electrochemical Performance

The electrochemical performance of the PC and NPC‐x series materials was evaluated using a three‐electrode system in a 6 M KOH aqueous electrolyte. Cyclic voltammetry (CV) measurements were firstly conducted in a potential window of ‐1.0 to 0.0 V (vs. HgO/Hg). As shown in Figure 3a, at a scan rate of 20 mV s⁻¹, the CV curves of PC and NPC‐x electrodes exhibit quasi‐rectangular profiles, indicative of fundamental electric double‐layer capacitive (EDLC) behavior. Notably, distinct and reversible redox peaks are observed in the CV curves of the NPC‐x electrodes, originating from Faradaic pseudocapacitance contributed by nitrogen‐containing functional groups introduced via g‐C₃N₄ templating [28]. In the alkaline electrolyte, these functional groups undergo highly reversible electrochemical processes; in particular, pyrrolic nitrogen (N‐5) species at defect sites provide additional pseudocapacitance via highly reversible proton‐coupled electron transfer reactions. Consequently, the NPC‐0.5 electrode exhibits the largest integrated CV area, corresponding to the highest specific capacitance. In contrast, the PC electrode exhibits the lowest current density with a featureless, nearly rectangular CV curve, reflecting its limited active sites and EDLC‐based charge storage mechanism. The superior performance of NPC‐0.5 is attributed to the synergistic integration of abundant electroactive sites and a well‐developed hierarchical pore structure, which ensures full exposure of active sites and rapid diffusion of electrolyte ions.

FIGURE 3.

Electrochemical performance of PC and NPC‐x samples tested in a three‐electrode system. (a) CV curves at 20 mV s⁻¹; (b) CV curves of NPC‐0.5 at scan rates from 10 to 100 mV s⁻¹; (c) GCD curves of NPC‐0.5 at current densities from 0.5 to 20 A g⁻¹; (d) GCD curves at 0.5 A g⁻¹; (e) Specific capacitances at various current densities; (f) Nyquist plots from EIS measurements.

Figure 3b presents the CV curves of the optimal NPC‐0.5 electrode at various scan rates from 10 to 100 mV s⁻¹. As the scan rate increases, the quasi‐rectangular shape is well‐maintained without significant distortion, and the characteristic redox peaks remain clearly discernible even at a high rate of 100 mV s⁻¹. This indicates exceptional rate capability and rapid charge/ion transport kinetics. Furthermore, no obvious polarization shift of the redox peaks is observed at higher scan rates, demonstrating the low internal resistance of the NPC‐0.5 electrode. In contrast, the CV curves of PC enclose smaller areas and weaker current responses (Figure S6a), reflecting its limited active sites, which is consistent with its relatively low SSA (432 m² g⁻¹). The NPC‐0.25 exhibits a lower scan rate dependence (Figure S6b). Its low SSA (17 m² g⁻¹) is caused by pore collapse, which limits ion reachability and thus results in a lower variation in the current response as the scan rate increases. While the NPC‐0.75 (Figure S6c) has numerous active sites derived from g‐C₃N₄, its extremely low SSA (21 m² g⁻¹), together with pore blockage by excessive undecomposed g‐C₃N₄ (forming dense “dead volume”) and poor intrinsic conductivity, collectively hinders effective ion diffusion to the internal active sites. This significantly reduces the EDLC contribution and renders the pseudocapacitive sites largely inaccessible. The CV curves of NC‐0.5 (Figure S6d) show slightly deteriorated symmetry. Its low SSA (123 m² g⁻¹) and N‐5‐dominated defects (with poor conductivity) restrict high‐rate ion/charge transport, resulting in moderate polarization.

Galvanostatic charge‐discharge (GCD) measurements (Figure 3c,d) further indicate the capacitive characteristics of these materials. Among them, the NPC‐0.5 exhibits nearly symmetrical triangular shapes at various current densities and the longest discharge time (Figure 3d), indicating highly reversible EDLC behavior and the highest specific capacitance. Specifically, the NPC‑0.5 delivers specific capacitances of 237.5, 197.0, 179.2, 161.0, 155.4, 152.0, and 136.0 F g⁻¹ at 0.5, 1, 2, 5, 7, 10, and 20 A g⁻¹, respectively (Figure 3e). Electrochemical impedance spectroscopy (EIS) was used to analyze charge transport and ion diffusion kinetics of the electrodes (Figure 3f). A magnified view of the high‐to‐medium frequency region (inset in Figure 3f) reveals that all Nyquist plots begin with a minimally discernible, depressed semicircular arc. This subtle feature indicates the presence of a finite but extremely small charge‐transfer resistance (R_ct) at the electrode‐electrolyte interface for all samples, confirming generally efficient charge transfer kinetics. The most diagnostic differences emerge in the low‐frequency region, which governs capacitive behavior. The low‐frequency line for PC exhibits a markedly shallower slope. This Warburg‐like impedance feature suggests greater diffusive resistance for ions within its predominantly microporous structure, consistent with its rate capability limitations. The profiles for NPC‐0.25 and NPC‐0.75 lie between these two extremes, aligning with their respective structural compromises. In contrast, the plot for NPC‐0.5 shows a nearly vertical line relative to the real axis, characteristic of ideal capacitor‐like behavior, reflecting rapid and unhindered ion diffusion through its interconnected hierarchical pore network. Through the synergistic effect of the g‐C₃N₄ template and K⁺ activation, the NPC‐0.5 successfully constructs a highly conductive 3D network, thereby exhibiting excellent capacitive performance and rapid charge transport kinetics.

To evaluate the Zn²⁺ storage performance, ZIHCs were assembled using zinc foil as the anode, NPC‐x/PC materials as the cathode, and a 2 M ZnSO₄ solution as the electrolyte. As shown in Figure 4a, at a scan rate of 20 mV s⁻¹, only the NPC‐0.5 electrode exhibits a quasi‐rectangular CV curve, indicating an ideal EDLC behavior and highly reversible charge storage process. In contrast, the CV curves of PC, NPC‐0.25, and NPC‐0.75 show significant distortion and tilting, reflecting significant polarization and high ion transport resistance. In addition, the NPC‐0.5 also shows the largest integrated CV area, indicating the highest specific capacity among all samples. Figure 4b shows the GCD curves measured at 0.1 A g⁻¹. The NPC‐0.5 demonstrates the highest specific capacity, and its slightly nonlinear curves implies a hybrid storage mechanism involving both EDLC and pseudocapacitance. The GCD curves of NPC‐0.5 tested at different current densities (0.1‐20 A g⁻¹) all exhibit high symmetry (Figure 4c), indicating excellent reversibility. The specific capacities calculated from the GCD data are shown in Figure 4d. The capacities of NPC‐0.5 at 0.1, 0.3, 0.5, 1, 2, 5, 10, and 20 A g⁻¹ are 172, 133.8, 124.9, 117.7, 107.5, 94.9, 83.5, and 70.1 mAh g⁻¹, respectively, all significantly higher than those of the other cathodes. From 0.1 to 20 A g⁻¹, the NPC‐0.5 exhibits a capacity retention of 41%, far exceeding the values for PC (25.2%), NPC‐0.25 (24.5%), and NPC‐0.75 (7.3%), indicating excellent rate capability (Figure 4e). The superior electrochemical performance of NPC‐0.5 is attributed to its unique hierarchical pore structure: the high SSA provides abundant adsorption sites, while the rich mesopores serve as fast ion diffusion channels, allowing the electrolyte to fully infiltrate the pores. Furthermore, the high content of graphitic nitrogen significantly enhances the electronic conductivity of material.

FIGURE 4.

Electrochemical performance of PC and NPC‐x samples tested in ZIHCs. (a) CV curves at 20 mV s⁻¹; (b) GCD profiles at 0.1 A g⁻¹; (c‐e) GCD profiles, specific capacities from the GCD data, and rate capability at current densities from 0.1 to 20 A g⁻¹; (f) Nyquist plots with the equivalent circuit model shown in the inset; (g) Linear fits of Z′ versus ω^−1/2; (h) Long‐term cycling stability and the corresponding Coulombic efficiency at 10 A g⁻¹.

To further investigate the charge transport behavior, the electrodes were analyzed using EIS patterns (Figure 4f). The Nyquist plots of all electrodes show a semicircle in the high‐frequency region and a linear feature in the low‐frequency region. Fitting with an equivalent circuit yielded series parameters (Table S3). The solution resistances (R_s) of PC, NPC‐0.25, NPC‐0.5, and NPC‐0.75 are 1.59, 1.47, 1.82, and 2.17 Ω, respectively, indicating comparable intrinsic electronic conductivity. The slightly higher R_s for NPC‐0.75 may originate from the poor conductivity of residual g‐C₃N₄. Notably, the R_ct of NPC‐0.5 (73 Ω) is significantly lower than that of PC (105 Ω) and NPC‐0.25 (166 Ω), indicating a lower reaction energy barrier and faster charge transfer kinetics at the electrode‐electrolyte interface. Furthermore, the Warburg segment in the low‐frequency region of the Nyquist plot shows that NPC‐0.5 presents the steepest slope, indicating the fastest Zn²⁺ diffusion rate among all cathode materials. The diffusion behavior of Zn²⁺ can be quantitatively analyzed using the following relations [33]:

$$\omega =2\pi f$$ (1)

$$Z^{\prime }=R_s+R_{ct}+\sigma {\omega }^{-1/2}$$ (2)

where ω is the angular frequency, f is the test frequency, Z′ is the real part of the impedance, and σ is the Warburg coefficient that is obtained from the linear slope of Z′ versus ω ^−1/2 and is inversely proportional to the ion diffusion capability. As shown in Figure 4g, the σ value for NPC‐0.5 is 21, significantly lower than the other samples (PC: 242, NPC‐0.25: 113, and NPC‐0.75: 54). Given that the Zn²⁺ diffusion coefficient (D_Zn ²⁺) is inversely proportional to σ, the NPC‐0.5 cathode possesses the most favorable Zn²⁺ diffusion kinetics among the materials studied. These EIS results fully demonstrate that the NPC‐0.5 facilitates rapid Zn²⁺ diffusion and efficient charge transfer, providing a solid kinetic foundation for its outstanding electrochemical performance. To evaluate the cycling performance of the NPC‐0.5 cathode, a long‐term stability test was carried out at a high current density of 10 A g⁻¹ (Figure 4h). The result indicates that NPC‐0.5 exhibits excellent cycling stability, retains 86% of its initial capacity after 65,000 cycles, with Coulombic efficiency consistently close to 100%. The SEM images of NPC‐0.5 after 65,000 cycles (Figure S7) shows that its morphology remains largely unchanged compared to the pre‐cycled state, indicating superior structural stability of the material during prolonged cycling and further confirming its excellent durability.

To gain deeper insight into the underlying mechanism of its superior electrochemical performance, electrochemical kinetic analysis was performed. The CV curves collected at various scan rates (2‐100 mV s⁻¹) maintain well‑preserved shapes without noticeable distortion (Figure 5a), indicating a highly reversible reaction process with fast kinetics. The peak current (i) and the scan rate (v) in the CV curves satisfy the following relationship [34]:

$$i=a{v}^b$$ (3)

where a and b are adjustable parameters. A b‐value of 0.5 indicates a diffusion‐controlled process, while b‐value of 1.0 indicates dominant capacitive behavior. The calculated b‐value of NPC‐0.5 is 0.905 ± 0.001 (Figure 5b), implying that the energy storage is predominantly capacitive process, yet retains a limited diffusion‑controlled contribution. To further quantify the proportion of capacitive and diffusion‑controlled contributions, the current response at a given potential can be deconvoluted using the equation [35]:

$$i\left(V\right)=k_{1}v+k_2{v}^{1/2}$$ (4)

FIGURE 5.

(a) CV curves of NPC‐0.5 at various scan rates; (b) The linear relationships between log(i) and log(v) of NPC‐0.5 curves; (c) The capacitive contribution of the NPC‐0.5 at 20 mV s⁻¹. (d) The contribution ratio of capacitive capacities at various scan rates.

The k₁ and k₂ were determined by plotting i/v ^1/2 against v ^1/2, and the contribution ratios of capacitive and diffusion processes at different scan rates were calculated accordingly. As shown in Figure 5c, the capacitive contribution accounts for 72% at 20 mV s⁻¹. When the scan rate increases to 100 mV s⁻¹, the capacitive contribution increases to 89% (Figure 5d). These results confirm that the capacity is primarily capacitive in nature and exhibits fast kinetics even at high scan rates. Notably, both capacitive and diffusion contributions coexist across all scan rates, indicating that the overall energy storage arises from the combined effects of EDLC and faradaic reactions. Therefore, Zn²⁺ storage in the material involves both physical/chemical adsorption and bulk‑phase Zn²⁺ diffusion, with diffusion behavior being more significant particularly at low scan rates or small current densities.

To further understand the effect of structure on electrochemical kinetics, we extended the kinetic analysis to PC, NPC‐0.25, and NPC‐0.75 (Figure S8–S10). From the CV curves, the b‐values were determined to be 0.768 ± 0.009, 0.721 ± 0.023, and 0.865 ± 0.01 for PC, NPC‐0.25, and NPC‐0.75, respectively. The relatively low b‐values of PC and NPC‐0.25 indicate stronger diffusion‐controlled behavior, consistent with their limited SSA and underdeveloped pore networks that restrict ion access. The NPC‐0.75 exhibits a moderately high b‐value, reflecting a more capacitive response at higher scan rates; however, its overall capacity remains low due to its low SSA and residual g‐C₃N₄ that blocks active sites. Capacitive contributions quantified from the CV curves further illustrate these trends. At 20 mV s⁻¹, the capacitive contributions of PC, NPC‐0.25, and NPC‐0.75 are 40%, 13%, and 54%, respectively. As the scan rate increases to 100 mV s⁻¹, these values become 72%, 29%, and 83%. The consistently low capacitive contribution of NPC‐0.25 across all scan rates aligns with its collapsed pore structure and minimal ion‐accessible surface. The NPC‐0.75, despite showing relatively high capacitive contribution at high scan rates, still suffers from limited absolute capacity due to its low SSA and pore blockage. Together, these comparisons demonstrate that NPC‐0.5, with its hierarchical pore architecture and high graphitic nitrogen content, enables efficient surface‐controlled charge storage and rapid ion transport, leading to its superior capacity and rate capability.

To elucidate the critical role of K⁺ activation, we systematically compared NC‐0.5 with NPC‐0.5. Although both materials are derived from g‐C₃N₄, replacing potassium citrate with citric acid in the synthesis of NC‐0.5 eliminates the K⁺‐mediated activation effect, resulting in markedly different structural characteristics in terms of SSA, porosity, and nitrogen configuration. The absence of strong K⁺ activation during synthesis results in a dense carbon structure with low SSA, poorly developed porosity, and limited electronic conductivity. Consequently, NC‐0.5 exhibits significantly lower CV response and rate capability in a three‐electrode system (Figure S11a, b), along with higher impedance and slower ion diffusion as revealed by EIS (Figure S11c). Similarly, in full‐device ZIHC testing, the NC‐0.5 shows consistently inferior performance in CV response, EIS behavior, specific capacity, and rate capability (Figure S11d–h). Altogether, these results indicate that such a structure cannot provide the necessary conditions for efficient ion storage and rapid electrochemical kinetics.

3. Conclusions

In conclusion, we have successfully developed a facile and scalable strategy to fabricate N‐doped hierarchical porous carbon (NPC) for high‐performance ZIHCs by employing g‐C₃N₄ as a dual‐functional soft template and nitrogen source, together with potassium citrate as an integrated carbon precursor and activating agent. The optimized NPC‐0.5 material, synthesized at a g‐C₃N₄/potassium citrate mass ratio of 1:3, possesses a high SSA (805 m² g⁻¹), a well‐defined hierarchical pore network, and a substantial N‐doping level (7.94 at%) with a dominant graphitic‐N configuration. These structural features synergistically provide abundant active sites, facilitate rapid ion transport, and enhance Zn²⁺ storage capability. Control experiments further reveal the critical role of K⁺ ions in pore creation, catalytic graphitization, and nitrogen configuration optimization. When evaluated as a ZIHC cathode, NPC‐0.5 delivers a remarkable specific capacity of 172 mAh g⁻¹ at 0.1 A g⁻¹, outstanding rate performance (41% capacity retention at 20 A g⁻¹), and exceptional long‐term cycling stability with 86% capacity retention after 65,000 cycles at 10 A g⁻¹. This work not only demonstrates an efficient template‐activation‐doping integrated approach for designing advanced carbon cathodes but also offers valuable insights into the development of high‐energy, high‐power, and durable hybrid energy storage systems.

4. Experimental Section

4.1. Synthesis of g‐C₃N₄

The g‐C₃N₄ samples were prepared through thermal polycondensation of urea. The procedure is briefly described as follows: 20 g of urea was placed in a lidded ceramic crucible, and the crucible was entirely wrapped with aluminum foil to prevent the loss of reactants due to volatilization. The crucible was then transferred to a muffle furnace. Under an air atmosphere, the temperature was raised from room temperature to 550°C at a heating rate of 10°C min⁻¹, followed by isothermal calcination at 550°C for 2 h. After the reaction, the sample was allowed to cool naturally inside the furnace to room temperature, yielding a pale‐yellow solid product. The amount of urea precursor can be appropriately adjusted based on the actual capacity of the selected ceramic crucible.

4.2. Synthesis of PC, NPC‑x, and NC‑0.5

Porous carbon materials, including PC, the NPC‐x series (NPC‐0.25, NPC‐0.5, NPC‐0.75), and the control sample NC‐0.5, were synthesized via a common pyrolysis route under a nitrogen atmosphere, using potassium citrate or its mixture with g‐C₃N₄ as the carbon precursors. In each case, the precursors were finely ground and placed in a nickel boat, which was then transferred to a tube furnace. The pyrolysis was carried out by heating to 600°C at a rate of 5°C min⁻¹ under N₂, holding for 2 h, and subsequently allowing the furnace to cool naturally to room temperature. The resulting pyrolyzed product was immersed in 1 M HCl for 30 min to remove soluble impurities, repeatedly washed with deionized water until the filtrate reached neutrality (pH ≈ 7), and finally dried under vacuum at 80°C for 12 h. PC was obtained using 1.5 g of potassium citrate alone, whereas the NPC‐x samples were prepared by mixing 1.5 g of potassium citrate with varying amounts of g‐C₃N₄ (0.25 g, 0.5 g, and 0.75 g for NPC‐0.25, NPC‐0.5, and NPC‐0.75, respectively) prior to pyrolysis. For the control sample NC‐0.5, potassium citrate was replaced with an equimolar amount of citric acid (0.94 g), while all other synthesis and post‐treatment steps remained identical to those employed for NPC‐0.5.

4.3. Materials Characterizations

The crystallographic and morphological features of samples were characterized using XRD (Smart Lab), SEM (FEI NovaNano450), and TEM (FEI Tecnai G2 F20). The structural and surface properties of samples were evaluated using Raman spectroscopy (Thermo Fisher DXR instrument equipped with a 532 nm laser wavelength), FTIR spectroscopy (Nicolet iS10), and XPS spectroscopy (Thermo Fisher Nexsa system with an Al Kα source and C1s for calibration). The specific surface area and pore parameters of samples were collected through a nitrogen adsorption‐desorption analyzer (ASAP‐2460). TGA was performed using a thermal analyzer (Netzsch STA 449 F5) in a N₂ atmosphere, with the temperature ranging from room temperature to 800°C.

4.4. Electrochemical Measurements

The active materials (PC, NPC‐x, and NC‐0.5) were homogeneously mixed with carbon black and a polytetrafluoroethylene (PTFE) binder in a mass ratio of 8:1:1 in anhydrous ethanol. The mixture was coated onto a titanium mesh current collector and then dried overnight in an oven at 80°C. The ZIHCs was assembled as a CR2032‐type coin cell. The titanium mesh coated with the active material served as the positive electrode, a zinc foil as the negative electrode, and a 2 M ZnSO₄ aqueous solution as the electrolyte. The total mass loading of the active material was controlled between 1 and 2 mg cm⁻². For the three‐electrode system, the resulting slurry was pasted onto a 1 cm × 1 cm piece of nickel foam and dried at 60°C for 12 h. The loading density of the porous carbon was approximately 2 mg cm⁻². The electrochemical performance of the synthesized materials in 6 M KOH electrolyte was investigated using a three‐electrode system with a platinum plate as the counter electrode and a Hg/HgO electrode as the reference electrode. The electrochemical performance of the electrodes was evaluated using a Neware battery testing system. Additionally, CV and EIS experiments were conducted on a CHI760E electrochemical workstation. All tests were performed using electrodes with similar mass loadings and within a controlled voltage window of 0.2 to 1.8 V (vs. Zn²⁺/Zn) to ensure consistency in the experimental conditions. In the three‐electrode system, the specific capacitance of the electrode is calculated from the GCD curves using the following formula:

$$C_{sp}=\frac{I\times \mathrm{\Delta }t}{\mathrm{\Delta }E\times m}$$ (5)

where C_sp is the specific capacitance (F g⁻¹), I (A) is the current, Δt (s) is the discharge time, ΔE (V) is the potential difference in discharge excluding the IR drop, m (g) is the mass of active material. In two electrode system, the specific capacitance of a single electrode can be calculated as the following equation:

$$C_{sp}=\frac{4I\mathrm{\Delta }t}{mV}$$ (6)

where C_sp (F g⁻¹) is the specific capacitance of a single electrode, I (A) is the current, Δt (s) is the discharge time, m (g) is the total mass of the active material in the SCs, and V is the potential window (V).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File 1: chem71050‐sup‐0001‐SuppMat.docx.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.