The continued use of fossil fuel-based energy resources has led to price volatility, environmental pollution, and significant contributions to global warming [1–4]. These environmental challenges underscore the urgent need for a global transition toward alternative and sustainable energy sources. In this context, advancing energy storage and conversion systems for effective utilization of sustainable energy has become a pivotal task in addressing these pressing global crises [5–7]. Metal-air batteries are highly attractive electrochemical energy storage systems capable of storing energy from renewable sources, making them a key technology for energy storage and transportation [8–10]. However, the cathodic four-electron oxygen (O₂) reduction reaction (ORR) suffers from lower efficiency and performance than the anodic reaction due to sluggish kinetics, hindering further commercialization [11]. Therefore, the development of efficient electrocatalysts to accelerate ORR kinetics is crucial for advancing practical applications. Pt-group metal (PGM) catalysts have traditionally been considered the most effective due to their superior ORR activity. However, their widespread use is limited by high costs, low availability, and susceptibility to poisoning, which hinders broader commercial development and drives the search for alternative materials [12]. In response, non-noble transition metal oxides, single-atom catalysts, and carbon-based materials have emerged as promising candidates to overcome these limitations [13–18].

Among various alternatives, carbon-based materials have drawn significant attention due to their excellent electrical conductivity, low cost, and high selectivity for the four-electron ORR pathway [19–21]. In particular, biomass-derived carbon materials have recently garnered increasing interest for their potential to enhance sustainability in catalyst fabrication through the recycling of eco-friendly raw materials and the use of green production processes [22–24]. These materials typically exhibit high surface areas and inherent heteroatoms, such as B, F, S, P, O, and N, which contribute to remarkable ORR activity. As a result, they provide both cost advantages and environmental benefits by upcycling waste resources into high-value materials, presenting an economical and sustainable circular economy model. In the Republic of Korea alone, approximately 600,000 tons of mandarins are produced annually, with around 60,000 tons discarded as peel waste, contributing to resource inefficiency, environmental pollution, and methane emissions, which are ~25 times more potent than CO₂ in terms of global warming potential [25]. Upcycling mandarin peel waste (MPW) for catalyst production offers a practical solution to these issues by reducing costs and transforming agricultural waste into high-value-added materials. This strategy not only addresses environmental concerns but also supports the development of economically viable and sustainable energy technologies.

Despite recent advancements, carbon-based, metal-free catalysts often exhibit lower performance and efficiency than PGM catalysts. This gap highlights additional strategies, such as doping heteroatoms into carbon-based materials. Recent studies have explored introducing heteroatoms into mandarin peel-based carbon materials to modify electron density and structural defects, thereby enhancing their electrochemical activity. For example, Li et al. reported a manganese oxide-based material as an effective ORR catalyst using tangerine peel as a natural carbon support for metal oxides [26]. Qu et al. fabricated Fe₃C/Fe-encapsulated biochar derived from tangerine peel for the treatment of the air pollutant sulfur dioxide (SO₂)[27]. Karaman demonstrated both N- and S-co-doped carbon dots (N, S-CDs) derived from orange peel and anchored them on 3D graphene networks (N, S-CDs@3D-GNs) to enhance ORR electrocatalytic activity [28]. While various fruit peels have been explored as biomass precursors, studies specifically focused on MPW remain limited. In this context, our approach is distinctive in directly utilizing MPW as a metal-free catalyst with nitrogen doping to improve electrochemical performance, setting it apart from previous biomass-based studies.

Herein, we demonstrate a nitrogen-doped, metal-free, carbon-based catalyst, directly derived from MPW for application in zinc (Zn)-air batteries, with a focus on the ORR in alkaline electrolytes. MPW serves as a sustainable carbon precursor that can be readily converted into porous pristine carbon after a simple heat treatment under a nitrogen atmosphere. To further enhance catalytic performance toward ORR, nitrogen doping is introduced via heat treatment under NH₃ atmosphere, replacing carbon atoms in the structure with nitrogen atoms and resulting in remarkable electrocatalytic activity in the conversion of O₂ into H₂O. Although the exact nature of the active sites and the role of various nitrogen species in promoting ORR remain controversial, numerous previous studies have emphasized that nitrogen doping significantly enhances ORR performance of carbon-based electrocatalyst [29,30]. The optimized catalyst, nitrogen-doped MPW-derived catalyst-2 (NMPC-2) demonstrates excellent ORR activity, along with superior stability and durability in alkaline solutions, comparable to that of commercial Pt/C. Furthermore, NMPC-2-based zinc-air batteries exhibit high open-circuit voltage and exceptional discharge performance with notable stability. Overall, our work highlights a sustainable strategy for converting biomass waste into high-value electrocatalysts, offering both environmental and energy-related advantages.

The schematic of the catalyst preparation process is illustrated in Fig. A1. Mandarin peel waste (MPW) was first washed and freeze-dried, then used as a carbon precursor. The pre-treated MPW was carbonized at 1000°C for 2 h under Ar atmosphere, yielding a black powder form of MPW-based carbon (MPC). Subsequently, following an effective N-doping strategy reported in previous studies [31–33], the pristine carbon was treated under an NH₃ atmosphere at 1000°C for designated durations to activate the electrocatalytic surface for enhanced ORR activity. This process resulted in nitrogen-doped MPW-based carbon (NMPC) catalysts, named NMPC-1, NMPC-2, and NMPC-3, corresponding to NH₃ activation times of 10, 30, and 120 minutes, respectively.

The surface morphology and porous surface of MPC and NMPC catalysts were examined using scanning electron microscope (SEM) and transmission electron microscope (TEM) images. These analyses highlight structural changes that occur before and after NH₃ activation. Fig. 1B shows the SEM image of the MPC catalyst, revealing a powdered chunk-like morphology. After NH₃ activation, the NMPC-2 catalyst exhibits an etched carbon structure with increased surface roughness due to the formation of small wrinkles, as shown in Fig. 1C, indicating improved porosity and surface area. Consistent with the SEM observations, the TEM images clearly confirm the roughened surface of NMPC-2 compared to MPC with amorphous carbon structures (Fig. 1D and E). For comparison, the STEM images of NMPC-1 and NMPC-3 are presented in Fig. S1. Moreover, scanning TEM (STEM)-energy dispersive X-ray spectroscopy (EDS) map images demonstrate a uniform elemental distribution of C, O, and N in both MPC and NMPC-2 (Fig. 1F and G). Although C, O, and N were present in both MPC and NMPC-2 before and after activation, SEM-EDS analysis revealed a substantial increase in N content, from 2.5 to 6.3 at.% in NMPC-2, indicating that N doping was effectively achieved during the NH₃ activation process. The effective incorporation of N-doped sites in NMPC-2 is further supported by the presence of C-N bonding observed in the X-ray photoelectron spectroscopy (XPS) C 1s spectrum (Fig. S2).

The specific surface area and porous structure of the catalysts were analyzed using Brunauer-Emmett-Teller (BET) measurements of N₂ adsorption-desorption isotherms and Barrett-Joyner-Halenda (BJH) pore size distribution calculations. The specific surface area of the electrocatalysts increases with longer thermal exposure under an NH₃ atmosphere (Fig. 2A), accompanied by an increase in total pore volume from 0.24, 0.53, 0.79, and 1.09 cm³ g⁻¹ for MPC, NMPC-1, NMPC-2, and NMPC-3, respectively (Table S1). This increase in surface area is attributed to the etching effect of NH₃ radicals generated during the NH₃ activation process, as corroborated by the previous SEM and TEM images. In addition to the overall trend of increasing specific surface area with longer activation durations, NMPC-2 appears to contain some amount of mesoporous structure with pore diameters in the range of 2–3 nm (Fig. 2B). This is also supported by the narrow hysteresis loop observed in the N₂ adsorption-desorption isotherm of NMPC-2 (Fig. 2A), which is indicative of a hierarchical pore structure composed of both micropores and mesopores [34]. Although NMPC-3 exhibits a higher total specific surface area, the mesoporous feature becomes less pronounced with activation durations exceeding 30 min. This reduction in mesoporosity is likely due to the collapse of surface mesopores during extended NH₃ treatment, leaving a structure dominated by micropores within the NMPC-3 electrocatalyst.

XPS analysis was performed to investigate the surface chemical states of the NMPC-based electrocatalysts. As shown in the survey spectra (Fig. S3), no significant peaks corresponding to metal elements were observed, confirming the metal-free nature of the MPW-derived electrocatalysts. As shown in Fig. 2C-E, the N 1s peaks in all electrocatalysts can be deconvoluted into four chemical states: pyridinic, pyrrolic, graphitic, and oxidized N. Although the exact relationship between the electrocatalytic activity of ORR and the active sites in N-doped carbon-based catalysts is not fully understood, pyridinic N is known to facilitate O₂ binding at neighboring C sites by reducing the adsorption energy, thereby accelerating the electrochemical conversion of O₂ to H₂O. In addition, the formation of graphitic N sites contributes to improved electrical conductivity of the carbon backbone, leading to higher current density [35]. Therefore, the proportions of pyridinic and graphitic N are considered the most critical configurations for achieving excellent ORR performance. The combined proportion of graphitic and pyridinic N increases with longer activation durations and reaches a maximum in NMPC-2, which is expected to exhibit the highest electrocatalytic activity toward ORR among all the tested electrocatalysts (Fig. 2F). Indeed, NMPC-2 shows the highest ORR activity, as discussed in a later section. These results indicate that NH₃ activation effectively induces the catalytically active N-doped sites and improves ORR performance, consistent with previous studies, although further optimization is still necessary [36,37].

The ORR performance of NMPC-based electrocatalysts was evaluated by electrochemical measurements using rotating ring disk electrodes (RRDE) in a half-cell configuration, as shown in Fig. 3. All NMPC-based electrocatalysts exhibit superior electrocatalytic activity toward ORR compared to MPC in O₂-saturated 0.1 M KOH, demonstrating the effectiveness of the activation process (Fig. 3A). Furthermore, NMPC-2 electrocatalyst exhibits the highest electrocatalytic activity toward ORR, characterized by the most positive onset potential, defined as the potential (vs. the reversible hydrogen electrode, RHE) at a current density of −0.1 mA cm⁻² (E_onset = 0.953 V), and the highest current density at 0.8 V (vs. RHE) (J_{@0.8 V} = 3.49 mA cm⁻²). These values are somewhat comparable to those of commercial Pt/C (E_onset = 0.952 V, J_{@0.8 V} = 4.0 mA cm⁻²). Notably, the ORR activity of NMPC-2 also surpasses that of other reported metal-free electrocatalysts in the past five years (Table S2). The highest electrocatalytic activity of NMPC-2 aligns well with the findings from BET and XPS analyses, which reveal the presence of a mesoporous structure and the highest combined proportions of pyridinic and graphitic N. Not only the micropores but also the well-developed mesopores in NMPC-2 are expected to enhance mass transport and facilitate access to active sites [32,38]. In addition, the presence of pyridinic and graphitic N contributes to the formation of active sites for ORR, thereby rationalizing the superior activity of NMPC-2. To further investigate the effect of surface area, the double-layer capacitance (C_dl) was estimated from cyclic voltammograms (CVs) acquired at various scan rates in the non-Faradic region using Ar-saturated 0.1 M KOH (Fig. S4). As shown in Fig. 3B and Table S1, NMPC-2 exhibits the highest C_dl value of 73.6 mF cm⁻², implying the largest electrochemically active surface area (ECSA). However, NMPC-3 exhibited the largest specific surface area due to the longest activation durations under the NH₃ atmosphere (Fig. 2A). This result clearly indicates that the highest specific surface area measured by BET does not necessarily correspond to the largest ECSA, emphasizing the importance of a well-balanced porous structure comprising both mesopores and micropores for efficient ion accessibility in aqueous electrolytes. In NMPC-3, the excessive formation of micropores in the deeper electrocatalytic regions, without accompanying mesoporous structures, restricts ion adsorption and desorption during potential scans, resulting in a slightly lower ECSA compared to NMPC-2, despite its higher specific surface area. The pore formation mechanism during NH₃ activation is schematically suggested in Fig. S5. The observed correlation between N content, pore structure, and ORR activity further underlines the importance of optimizing NH₃ activation conditions [39,40]. To further elucidate the ORR pathway, the number of electrons transferred (n) and H₂O₂ yield were determined by measuring the currents at a constant potential of 1.5 V (vs. RHE) on the Pt ring electrode (Fig. 3C). NMPC-2 exhibits the highest n value, ranging from 3.5 to 3.9, along with the lowest H₂O₂ yield (<20%), indicating that the ORR predominantly follows a four-electron pathway to produce H₂O.

The durability of NMPC-2 was evaluated by chronoamperometric measurements at 0.8 V (vs. RHE) in O₂-saturated 0.1 M KOH solution (Fig. 3D). Although the initial current density of NMPC-2 is lower than that of commercial Pt/C, NMPC-2 retains 84% of its initial current density after 10,000 s. In contrast, Pt/C shows a continuous decline in current density, with approximately a 50% decrease over the same period. Notably, NMPC-2 surpasses Pt/C in current density beyond 4,400 s, indicating superior long-term durability. These results, in conjunction with its high ORR activity, demonstrate that NMPC-2 possesses both excellent durability and electrocatalytic activity, underscoring its potential as a promising alternative to noble metal electrocatalysts in alkaline media. As shown in Fig. 3E, the stability of NMPC-2 was further validated through cyclic voltammograms acquired during an accelerated stability test (AST) involving 40,000 potential cycles between 0.6 and 1.0 V (vs. RHE). The changes in the potential at −2.25 mA cm⁻² and the current density at 0.8 V (vs. RHE) before and after 10,000, 20,000, and 40,000 potential cycles are presented in Fig. 3F. The results indicated a gradual yet marginal decline in ORR activity with increased potential cycles in O₂-saturated 0.1 M KOH solution. A#er 40,000 cycles of AST, the potential at −2.25 mA cm⁻² and the current density at 0.8 V (vs. RHE) decreased by only 10.2 mV and 0.18 mA cm⁻², respectively, demonstrating the exceptional long-term stability of NMPC-2 electrocatalysts.

To further demonstrate the practical applicability of the NMPC-2 electrocatalyst derived from upcycled bio-waste, it was employed as the air cathode in a single-cell Zn-air battery (ZAB), paired with a Zn foil anode (Fig. 4A). The NMPC-2 electrocatalyst was deposited onto carbon paper via electrospray, with a loading of 1 mg cm⁻², to fabricate the electrode (Fig. S6). A 6.0 M KOH solution served as the electrolyte and was continuously circulated during operation using a peristaltic pump. The battery achieved an open circuit voltage of 1.48 V and a maximum power density of 68 mW cm⁻² (Fig. 4B). During galvanostatic discharge across current densities ranging from 2 to 20 mA cm⁻² and back to 2 mA cm⁻², the battery exhibits stable voltage responses, indicating reliable and steady operation

(Fig. 4C). When the discharge current returned to the initial 2 mA cm⁻², the cell voltage recovered to 1.29 V, showing only a minimal 0.01 V drop from the initial 1.30 V. Additionally, during an 18 h discharge test at a constant current density of 5 mA cm⁻², the voltage remained stable over the first 14 h, followed by only a 7% decline after 18 h, confirming the excellent durability of NMPC-2-based electrode (Fig. S7). After normalization to the mass of consumed Zn, the ZAB delivered a specific capacity of 819.9 mA h g⁻¹ and an energy density of 1,040.4 W h kg⁻¹, approaching the theoretical maximum of 1,218 W h kg⁻¹ for ZABs. The durability of the NMPC-2-based Zn-air battery under chronopotentiometric charge–discharge cycling at 2 mA cm⁻² was evaluated, as shown in Fig. 4D. The initial discharge voltage of 1.30 V rapidly declined to 1.00 V within the first 6 h, possibly due to electrochemical oxidative degradation of the carbon-based framework during the charging process, which generally induces oxygen evolution reaction. Nevertheless, the discharge performance remained stable over the remaining 70 h of testing, while the charging voltage gradually decreased, indicating improved charge efficiency over time. These results highlight the strong potential of NMPC-2 for practical deployment in energy storage devices.

In this study, we demonstrate a simple and sustainable approach to synthesizing NMPC-based electrocatalysts by upcycling agricultural mandarin peel waste through thermal carbonization followed by NH₃ activation. This strategy provides a practical route for converting biomass-derived waste into high-value electrocatalytic materials, contributing to the development of environmentally responsible energy technologies. Among the synthesized catalysts, NMPC-2 exhibits outstanding ORR activity and stability in alkaline media, outperforming other biomass-derived and metal-free electrocatalysts reported over the past 5 years. Notably, its performance was comparable to that of commercial Pt/C, underscoring the potential of NMPC-2 as a viable, cost-effective alternative to noble-metal-based electrocatalysts. The enhanced catalytic activity of NMPC-2 is attributed to its unique hierarchical porosity, which promotes efficient O₂ diffusion and increases the density of accessible N-doped active sites. Furthermore, NH₃ activation during thermal treatment effectively incorporated pyridinic and graphitic N, thereby improving surface conductivity and accelerating ORR kinetics. When applied as the air cathode in a ZAB, NMPC-2 delivered excellent electrochemical performance and durability, confirming its strong promise for practical energy storage applications. Owing to the scalability of the synthesis process, further efforts toward scale-up and cell-level development are currently underway to address remaining challenges and enable real-world implementation.