English

• The pervasive discharge of antibiotics like oxytetracycline (OTC) into aquatic systems poses escalating ecological and public health risks due to incomplete metabolic assimilation and environmental persistence [1–3]. Conventional wastewater treatments exhibit limited removal efficiencies and risks of secondary pollution, necessitating advanced oxidation processes (AOPs) for contaminant mineralization [4]. PMS-based AOPs, distinguished by robust oxidation capacity and broad pH adaptability, outperform ozone- or H₂O₂-driven systems [5,6] However, traditional PMS activation methods relying on UV irradiation, thermal energy, or transition metals face challenges including energy inefficiency and metal leaching, underscoring the imperative for sustainable metal-free catalysts [7,8].

  Carbonaceous materials, particularly biochar derived from agro-industrial waste, offer promise due to their tunable porosity, surface functionality, and environmental compatibility [9]. Nitrogen (N) doping enhances PMS activation by modulating electron density and creating defect-mediated active sites, improving adsorption and catalytic efficiency. In contrast, single-atom doping with oxygen (O), sulfur (S), or boron (B) yields limited improvements, primarily altering surface polarity without establishing stable catalytic centers [10]. To enhance the performance of N-doped biochar catalysts, metal-N co-doped catalysts (e.g., Fe/N, Co/N, Mn/N) have been developed [11,12]. However, this approach introduces risks of metal leaching and increases production and disposal costs, despite improving catalytic activity. Critically, oxygen's high electronegativity and lone-pair electrons theoretically synergize with nitrogen's electron-donating properties, optimizing sp² carbon lattice polarity and generating dual-active centers [13]. Such N, O co-doping could mitigate the limitations of single dopants by enhancing interfacial electron transfer kinetics and stabilizing metastable radical intermediates, which is a hypothesis that requires rigorous validation through both experimental and computational methods.

  Importantly, while heteroatom-doped carbons typically activate PMS via radical pathways (SO₄^•−, ^•OH, O₂^•−), non-radical mechanisms (¹O₂ generation, electron transfer) demonstrate superior selectivity and resistance to environmental interference, making them ideal for complex matrices [14]. Tailoring catalysts to favor non-radical pathways demands precise control over dopant configurations (e.g., type, coordination) and defect engineering. However, the interplay between multi-heteroatom doping, surface defects, and PMS activation mechanisms remains poorly resolved, particularly regarding ¹O₂ generation pathways [15]. Addressing these gaps is critical for designing catalysts that balance high reactivity with environmental adaptability, guiding practical applications in real-world water remediation [16]. Furthermore, current research predominantly evaluates catalytic performance through short-term (< 1 h) static batch experiments, emphasizing initial kinetics but neglecting hydrodynamic effects on mass transfer and long-term stability [17]. These idealized conditions poorly replicate natural aquatic environments, where turbulent flows govern pollutant-catalyst interactions.

  To address these research gaps, we developed a nonmetallic N, O-doped biochar catalyst (NOBC_BM) using wheat straw as a precursor to facilitate green synthesis and in-situ heteroatom doping. This study aims to: (1) Demonstrate that N, O co-doping overcomes the limitations of single N-doped biomass catalysts to achieve efficient OTC degradation; (2) elucidate non-radical degradation mechanisms, reactive sites, and pollutant degradation pathways, demonstrating system resilience to environmental perturbations; (3) simulate the real water flow conditions and consider the influence of time factors on the degradation process; (4) study the impact of purified water on plant growth and evaluate the ecological safety and environmental friendliness of the catalyst. By exploring the synergistic effects of N, O co-doping and employing holistic evaluation methods, this work not only enhances understanding of heteroatom-engineered biochar catalysis but also offers a scalable, sustainable strategy for mitigating antibiotic pollution in aquatic ecosystems, thereby advancing the field toward practical, environmentally responsible solutions.

  Following ball milling, NSBC_BM displayed fragmented flake-like graphite structures (Fig. 1a), indicative of mechanical shear forces disrupting the original biomass architecture. The synergistic combination of ball milling and N/O co-doping in NOBC_BM resulted in a highly porous, irregular microstructure (Fig. 1b). This structural evolution arises from two key processes: (1) thermal decomposition of dicyandiamide (DCD) during pyrolysis generated gaseous byproducts that fracture the carbon matrix, and (2) oxygen atom incorporation induced lattice strain, creating interconnected pores and surface defects [18]. Elemental mapping confirmed homogeneous distribution of carbon, nitrogen (22.8%), and oxygen (25.3%) throughout NOBC_BM (Fig. 1c, Fig. S1a in Supporting information), ensuring uniform active site accessibility. Fig. S1b (Supporting information) displayed N₂ adsorption-desorption isotherms for NSBC_BM and NOBC_BM, revealing type Ⅳ curves with H3 hysteresis loops characteristic of mesoporous materials. Structural parameters in Table S1 (Supporting information) showed that NOBC_BM possessed enhanced textural properties relative to NSBC_BM, including the larger BET surface area (105.55 vs. 99.26 m²/g) and the higher mesopore volume (0.59 vs. 0.48 cm³/g). These improvements correlated with superior dispersity, accelerated mass transfer, and elevated OTC adsorption capacity.

  Figure 1

  

  Figure 1.  SEM images of (a) NSBC_BM, (b) NOBC_BM. (c) EDS mapping images of NOBC_BM. (d) FTIR patterns of Straw, NSBC_BM and NOBC_BM. (e) Raman patterns for NSBC_BM and NOBC_BM. (f) C 1s, (g) N 1s and (h) O 1s XPS spectrum of NSBC_BM, NOBC_BM.

  DownLoad: Full-Size Img PowerPoint

  Surface functionalization was systematically characterized through X-ray photoelectron spectroscopy (XPS) and FTIR spectroscopy. FTIR spectra (Fig. 1d) revealed the enhanced oxygen-containing functionalities in NOBC_BM compared to NSBC_BM, particularly intensified C–O (1220 cm⁻¹) and C═O (1560 cm⁻¹) stretching vibrations [19]. These observations aligned with XPS quantification: C 1s deconvolution showed that C–O and C═O contributions increased from 17% and 7% in NSBC_BM to 28% and 17% in NOBCBM, respectively (Fig. 1f). Concurrently, N 1s spectra demonstrated a doubling of oxidized nitrogen species (6% vs. 3% in N 1s XPS, Fig. 1g), confirming oxygen doping's role in modifying chemical environments. These oxygen-rich functional groups enhanced PMS adsorption and electron transfer kinetics [20].

  Raman spectroscopy data from Fig. 1e showed two characteristic bands: The D band at 1360 cm⁻¹ and the G band at 1571 cm⁻¹. The intensity ratio of the D and G bands (S_D1/S_G) is a well-established indicator for evaluating defect concentration in biochar materials [21]. Notably, the S_D1/S_G value of NOBC_BM (5.49) was higher than that of NSBC_BM (2.32), indicating that O doping increased the defect density in the composite. The unique structure of NOBC_BM significantly enhances its targeting of PMS and OTC through a series of synergistic mechanisms. In terms of electronic modulation of N, O doping, the pyridine nitrogen atom in the NOBC_BM structure significantly enhances the adsorption capacity of PMS, allowing PMS molecules to be adsorbed onto the catalyst surface more efficiently (The proof is presented in DFT calculations) [21]. Meanwhile, the presence of graphitic nitrogen atoms enhanced the electron conduction efficiency, ensuring smooth and unobstructed electron transfer during the catalytic reaction. In addition, the carbonyl oxygen atom in the structure further promoted the directional activation of PMS through surface dipole polarization, which enables the PMS molecules to be more efficiently converted into active oxide species on the catalyst surface (Figs. 1d, f, h).

  The catalytic performance of catalysts with varying N/O doping ratios was evaluated for OTC degradation (20 mg/L) (Fig. 2a). Single oxygen-doped catalysts (OSBC_BM) exhibited markedly inferior activity compared to N-doped counterparts (NSBC_BM), confirming nitrogen doping sites as primary active centers. Notably, dual N/O doping achieved a 2.3-fold enhancement in k_obs (0.1523 min⁻¹) over single N-doped catalysts (0.0664 min⁻¹), demonstrating the critical role of synergistic non-metal interactions in amplifying catalytic efficiency (Fig. 2b). While low oxygen content optimized activity by fine-tuning electronic structures and defect density, excessive oxygen loading suppressed performance, likely due to pore blockage at active sites. The catalyst synthesized at a 1:1:10 mass ratio (straw: MCHP: DCD) demonstrated peak efficiency (94% degradation; k_obs = 0.1523 min⁻¹), outperforming other formulations (Fig. 2b). This breakthrough highlights the transformative potential of dual heteroatom engineering in overcoming the limitations of single-dopant systems, providing a scalable strategy for advanced water purification technologies.

  Figure 2

  

  Figure 2.  Degradation of OTC (a) in different O doping ratios and (b) k_obs values; Degradation of OTC (c) in different PMS dosage and (d) k_obs values. (e) Adsorption of OTC by NOBC_BM. (f) Correlations between different PMS dosage and k_obs. (g) Normalized k_obs comparison of OTC degradation by NOBC_BM/PMS at different PMS concentrations. (h) Normalized k_obs comparison of OTC degradation by NOBC_BM/PMS and other nonmetallic catalyst/PMS systems.

  DownLoad: Full-Size Img PowerPoint

  The NOBC_BM/PMS system's OTC removal efficiency was systematically evaluated under varying PMS doses (0–0.5 mmol/L) (Fig. 2c). OTC removal efficiency exhibited a dose-dependent increase, rising from 32% at 0 mmol/L PMS (adsorption-only phase) to 94% at 0.2 mmol/L PMS, with k_obs reaching 0.1833 min⁻¹ (Fig. 2d). Notably, NOBC_BM achieved higher removal efficiency than NSBC_BM at all tested PMS concentrations, even surpassing NSBC_BM's performance at 0.5 mmol/L PMS when using only 0.2 mmol/L PMS. This stark contrast underscores dual N/O doping's ability to amplify catalytic performance while minimizing oxidant consumption, justifying the economic selection of 0.2 mmol/L PMS for subsequent studies. Adsorption equilibrium occurred rapidly within 40 min (Figs. 2c and e). NOBC_BM demonstrated a 32 mg/g adsorption capacity, exceeding NSBC_BM's 30 mg/g, directly attributable to oxygen doping-induced surface area expansion and adsorption site proliferation. In catalytic systems, k_obs values showed a linear correlation with PMS concentration (Fig. 2f), driven by enhanced electron transfer and reactive oxygen species (ROS) generation at elevated oxidant levels. Standardized analysis (normalized to unit PMS/catalyst dosage) revealed NOBC_BM's k_N values (up to 91.45 min⁻¹ g⁻² L²) were 2.2-fold higher than NSBC_BM's (up to 41.29 min⁻¹ g⁻² L²) (Fig. 2g), highlighting dual doping's synergistic role in optimizing active site accessibility and PMS utilization efficiency. Compared with the k_N values of the other non-metallic doped catalysts reported in the literature, the superior performance of NOBC_BM was further verified, achieving a standardized degradation rate increase of 2 to 12 times under the same conditions (Fig. 2h, Tables S2 and S3 in Supporting information). These findings collectively demonstrate the transformative potential of N/O co-doping in advancing metal-free catalytic systems for sustainable water remediation.

  The catalytic performance of NOBC_BM was dose-dependent (Figs. S2a and b in Supporting information). At 0.05 g/L, OTC degradation reached equilibrium within 30 min (92.3% removal). Increasing the dosage to 0.1 g/L enhanced removal efficiency to 98.7% (k_obs = 0.1917 min⁻¹), demonstrating dual N/O doping's ability to amplify active site availability and PMS utilization. Compared to NSBC_BM, NOBC_BM's superior performance stems from synergistic effects: oxygen doping introduces additional defect sites, while nitrogen optimizes electron transfer, collectively increasing ROS generation per unit catalyst. The OTC concentration significantly influenced system performance (Figs. S2c and d in Supporting information). In NSBC_BM/PMS systems, increasing OTC from 10 mg/L to 30 mg/L reduced removal efficiency from 91% to 62% (k_obs = 0.1520–0.0479 min⁻¹). In contrast, NOBC_BM/PMS maintained higher stability under identical conditions (86%–97%), attributable to enhanced active site density and PMS activation efficiency from N/O co-doping. This dual doping strategy minimizes catalyst deactivation at elevated pollutant concentrations by preventing active site saturation.

  To elucidate the degradation pathway of OTC in the NOBC_BM/PMS system, researchers employed free radical quenchers to characterize the contribution of ROS to OTC oxidation [22]. From Figs. 3a-c, at PMS = 0.05 mmol/L, the inhibitory effect of adding EtOH and TBA on SO₄^•‒ and ^•OH was ~9%. This inhibitory effect weakened as the PMS concentration increased. As free radical trapping agents, EtOH and TBA have limited radical capturing ability. When the radical generation rate exceeds the trapping agent's capacity, some radicals escape and participate in the target reaction, weakening the inhibition. The minimal inhibition observed from quencher addition indicated that SO₄^•‒ and ^•OH did not serve as dominant oxidants in this system. To evaluate the role of O₂^•−, SOD was introduced as a specific scavenger [23]. Experimental results revealed negligible effects of O₂^•− on OTC degradation, confirming its minor role as a reactive intermediate during the reaction.

  Figure 3

  

  Figure 3.  OTC quenching experiment and k_obs at PMS concentrations of (a) 0.05, (b) 0.2 and (c) 0.5. (d) EPR spectrums of NOBC_BM/PMS systems. (e) Different concentrations of TEMP quench ¹O₂. (f) Ultraviolet spectrum of DPA-¹O₂.

  DownLoad: Full-Size Img PowerPoint

  The electron paramagnetic resonance (EPR) technique with DMPO spin trapping reagent showed weak signals of SO₄^•‒, ^•OH, and O₂^•− in the NOBC_BM/PMS system (Fig. 3d) [24]. This indicated the catalytic system generates free radicals by oxidizing DMPO to DMPOX. However, free radicals have short lifetimes (usually microseconds or nanoseconds). They may rapidly decompose after generation. In EPR assays, radicals can be detected with a capture agent, but in the actual reaction, they may be quickly consumed and unable to effectively participate in the target reaction. The ¹O₂ is commonly generated in PMS-based AOPs. TEMP was selected as the ¹O₂ spin trapping reagent [25]. Fig. 3d clearly shows a large amount of ¹O₂ generated in the NOBC_BM/PMS system. TEMP also verified the role of ¹O₂. In the NOBC_BM/PMS system, OTC removal decreased from 94.1% to 41.8% (Fig. 3e). This means 0.03 mol/L TEMP inhibited OTC degradation by up to 52.3%, confirming ¹O₂ as an effective reactive species for OTC removal [26]. The oxidation of 9,10-diphenylanthracene (DPA) by ¹O₂ produces a unique peak near 378 nm, which can detect ¹O₂ [27]. In the NOBC_BM/PMS system, the absorption peak intensity near 378 nm decreased over 10 min, furtherly indicating continuous ¹O₂ consumption (Fig. 3f). Furthermore, the accelerated OTC degradation observed upon replacing H₂O with D₂O (Fig. S3 in Supporting information), resulted in complete removal and an increase in k_obs from 0.206 min⁻¹ to 0.277 min⁻¹. This rate enhancement confirms the significant contribution of ¹O₂.

  A premix experiment was designed to investigate the electron transfer mechanism driving OTC degradation in the NOBC_BM/PMS system. NOBC_BM and PMS were premixed for 5, 10, and 15 min before OTC addition [28]. As shown in Fig. S4a (Supporting information), OTC degradation efficiency decreased by only ~10% after 15 min of pre-mixing. This minimal reduction confirms that non-radical pathways dominate the reaction. The open circuit potential (OCP) measures electrode potential without current flow, reflecting material stability and reaction tendency under specific conditions. A higher OCP allows catalysts to reach the required reaction potential more easily, reducing overpotential and enhancing efficiency [29]. Fig. S4b (Supporting information) showed NOBC_BM had an initial OCP of 0.2389 V, significantly higher than NSBC_BM (0.1338 V), confirming superior catalytic performance. Introducing PMS and OTC alters the catalyst's surface electronic state, affecting active sites and OCP values. This provided insights into electron transfer processes during pollutant oxidation [30]. It can be seen that the OCP value decreases when OTC is added, indicating that electrons play a role in the degradation of OTC. In electrochemical impedance spectroscopy (EIS), the diameter of the semicircular arc reflects electron transfer resistance, with smaller radii indicating higher conductivity [31]. As depicted in Fig. S4c (Supporting information), the radius followed the order NSBC_BM > NOBC_BM, demonstrating superior electrical conductivity in NOBC_BM. Tafel polarization curves (Fig. S4d in Supporting information) showed a corrosion potential order of NOBC_BM > NSBC_BM, suggesting from a thermodynamic perspective that NOBC_BM electrodes exhibit greater resistance to corrosion compared to NSBC_BM [32].

  XPS analysis identified catalytic active sites in NOBC_BM through chemical state evolution of carbon, nitrogen, and oxygen. Deconvolution of the C 1s spectrum revealed three characteristic peaks (Fig. 1f): C–C/C═C (284.6 eV), C–O (286.0 eV), and C═O/C–N (288.7 eV). A marked decrease in C–/C═C (from 55% to 51%) and C═O/C–N (from 17% to 7%) contents suggested graphitic structure disruption, implicating these sites in non-radical mechanisms such as electron transport and singlet oxygen generation. The N 1s spectrum (Fig. 2g) resolved four coordination states: Pyridinic N (397.9 eV), pyrrolic N (399.4 eV), graphitic N (400.5 eV), and oxidized N (403.6 eV). Pyridinic and graphitic N, due to nitrogen's higher electronegativity, modified nearby carbon charge distributions. Post-reaction, pyridinic and pyrrolic N contents decreased from 46% to 44% and 31% to 29%, likely from PMS/contaminant conversion via adjacent activated carbon. Concurrent O 1s analysis (Fig. 1h) demonstrated surface oxidation through intensified oxygen functionality signals, consistent with reactive species formation during catalysis. Complementary FTIR analysis (Fig. S5 in Supporting information) tracked functional group evolution across reaction cycles. Pre-catalysis spectra exhibited prominent C═O stretching (1553 cm⁻¹) and C–O vibrations (1235 cm⁻¹). After reaction, the intensity of oxygen-containing functional group stretching vibrations in NOBC_BM decreased, particularly for C═O groups, which aligned with the XPS analysis results (Figs. 1f and h) and suggested these groups serve as active sites in the reaction.

  Additionally, in the NOBC_BM system, PMS activation began with chemisorption. N/O co-doping induced electronic modulation, forming localized electron-deficient regions. Pyridine nitrogen's lone pairs coordinated with PMS's terminal O atom, weakening the O–O bond (bond length increased from 1.471 Å to 1.472 Å via DFT calculations) (Fig. 4a). The high electronegativity of carbonyl oxygen (C═O) polarizes adjacent carbon atoms, creating localized electron-deficient regions that asymmetrically redistribute charge. This dipole polarization weakens the PMS O–O bond, elongating it from 1.472 Å to 1.476 Å and reducing bond strength. Consequently, the polarized carbon framework facilitates heterolytic O–O cleavage through electron transfer, preferentially generating ¹O₂-dominated non-radical species rather than radical pathways, as conclusively validated through EPR and quenching experiments. Defect structures from ball milling (e.g., five-membered rings, edge carbons) acted as electron transfer channels, enhancing electron flow from NOBC_BM to PMS. EIS showed NOBC_BM had a lower transfer resistance compared with that of NSBC_BM, confirming superior electron transfer. DFT calculations were performed to explore the activation of PMS by N/O co-doped biochar and material defects. As shown in Fig. 4a, PMS adsorption on NOBC_BM exhibited an energy (E_ad) of -3.154 eV, which was more negative than that on BC and NSBC_BM. This stronger adsorption facilitates the formation of NOBC_BM/PMS* complexes, promotes PMS activation, and induces the generation of ¹O₂, which were the key steps in the non-radical degradation pathway. Additionally, this non-metal doping was also confirmed to result in a longer O–O bond in the catalyst, which is favorable for the production of reactive species from PMS. The Fig. 4b displayed the molecular electrostatic potential (ESP) of the materials, where positive and negative potential regions are represented by blue and red, respectively. Nitrogen atoms, with higher electronegativity, created localized potential wells by lowering the electrostatic potential around them, which enhanced the adsorption of PMS and contaminant. Oxygen atoms, with even stronger electronegativity, further reduced the potential around nitrogen atoms, improving adsorption capacity for PMS and contaminants and boosting catalytic activity. Consequently, NOBC_BM exhibited higher electron density than NSBC_BM. This finding suggests that OTC may be more effectively adsorbed onto the NOBC_BM material.

  Figure 4

  

  Figure 4.  (a) Binding of PMS onto different catalyst configurations with different adsorption energies (E_ed) and l_o-o binding lengths. (b) Electron density distribution of different structures.

  DownLoad: Full-Size Img PowerPoint

  The Fukui function f ⁻ is associated with a molecule's ability to lose electrons. In chemical reactions, a higher f ⁻ value indicates the molecule is more prone to electron loss, favoring electrophilic reactions. f ⁺ relates to the molecule's capacity to accept electrons. A higher f ⁺ value signifies greater readiness for nucleophilic reactions. f ⁰ reflects changes in the molecule's electron cloud, indicating the influence of free radicals on reactivity without significant electron gain or loss. Analysis of f ⁻, f ⁺, and f ⁰ revealed that O2, N8, and N32 were favorable for electrophilic reactions; C11, C17, and C20 were prone to nucleophilic reactions; and N8 was susceptible to radical reactions (Figs. S6 and S7, Table S4 in Supporting information). These atomic sites were thus likely the most reactive during the degradation process.

  By integrating identified intermediates, Fukui function results, and HPLC-MS data, the potential degradation pathways of OTC in the NOBC_BM/PMS system were discussed. Three distinct OTC removal pathways were proposed in Fig. 5a. Pathways 1 and 2 involved hydroxylation and demethylation, leading to the formation of P1 (m/z 476) and P2 (m/z 433), respectively, which was consistent with result of Fukui function. Subsequently, P1 and P2 were transformed into P4 and then P5. In the third pathway, the OTC molecule lost H₂O to form P3 (m/z 443), which then underwent decarbonylation to produce P6 (m/z 415). These intermediates further transformed into P7 and P8. Finally, intermediates P9, P10, P11, P12, and P13 were formed and further oxidized to smaller molecules such as CO₂ and H₂O. The toxicological profiles of OTC and its degradation intermediates, including oral LD₅₀ (rats) and LC₅₀ (blackhead fish), were computationally modeled using T.E.S.T. software (Table S5 in Supporting information). Initial toxicity assessments revealed significant acute risks. OTC exhibited an oral LD50 of 1165.80 mg/kg in rats (Fig. 5b) and an LC₅₀ of 1.12 mg/kg in blackhead fish (Fig. 5c). Post-degradation intermediates demonstrated reduced toxicity, with LD₅₀ and LC₅₀ values exceeding those of parent OTC for most species. Notably, stable intermediates P2, P5, P9-P12 showed markedly diminished toxicity. While developmental toxicity increased for select intermediates (P3, P6, P8), the majority displayed nontoxic characteristics. These findings suggest extended reaction cycles could enhance detoxification, as terminal intermediates exhibit lower toxic potential than initial products.

  Figure 5

  

  Figure 5.  (a) LC-MS analysis of OTC degradation products and the possible pathways for OTC degradation in NOBC_BM/PMS system. (b, c) Toxicity analysis of OTC and its degradation intermediates. (d–g) Changes of three-dimensional fluorescence during OTC degradation.

  DownLoad: Full-Size Img PowerPoint

  Three-dimensional fluorescence techniques were utilized to characterize the degradation of OTC molecules. Fluorescence spectra were classified into five regions: aromatic protein (Ⅰ and Ⅱ), xanthate (Ⅲ), soluble microbial by-product (Ⅳ), and humic acid (Ⅴ) fluorescence. As shown in Figs. 5d–g and Table S6 (Supporting information), the pre-reaction OTC solution displayed dominant fluorescence intensity in the V region, corresponding to humic acid-like substances with molecular structures containing aromatic rings, polycyclic compounds, and functional groups (amide, phenolic hydroxyl, carboxyl) at molecular edges. However, after 50 min of reaction, most fluorescent substances vanished, indicating decomposition of OTC molecules during catalysis. This process involved complete mineralization of benzene rings and carbonyl groups into H₂O and CO₂.

  The NOBC_BM/PMS system maintained robust OTC degradation efficiency (96%–98%) across pH 3–11 (Fig. 6a). Enhanced activity under acidic conditions (pH 5 vs. 11) correlated with PMS's preferential generation of SO₄^•− (oxidation potential: 2.5–3.1 V) over ^•OH in low-pH environments. Alkaline conditions promoted surface-bound ^•OH formation but reduced overall reactivity. The system's pH resilience stems from its non-radical-dominated mechanism, which minimizes pH sensitivity compared to radical-dependent pathways. In aquatic ecosystems, inorganic anions often react with free radicals, potentially compromising the efficiency of pollutant degradation. To assess their influence, experimental trials were conducted to investigate how coexisting anions affect OTC degradation within the NOBC_BM/PMS system. As shown in Fig. 6b, only HCO₃⁻ caused minimal inhibition within 30 min, possibly due to its hydrolysis altering solution pH. Other coexisting anions had no significant inhibitory effect, indicating they did not substantially affect efficiency. These results validated the non-radical degradation pathway, aligning with earlier findings and underscoring the system's resilience to anion interference.

  Figure 6

  

  Figure 6.  (a) Degradation of OTC at different pH by NOBCBM/PMS system. (b) Degradation of OTC in different anions by NOBCBM/PMS system. (c) Degradation of OTC in different water bodies by NOBCBM/PMS system. (d) Stability tests of NOBCBM under four replicate experiments and removal efficiency. (e) Membrane reactor setup and (f) the degradation efficiency of OTC at different times. (g) Column reactor setup and (h) the degradation efficiency of OTC at different times. (i) Floater reactor setup and (j) the degradation efficiency of OTC at different times.

  DownLoad: Full-Size Img PowerPoint

  The system's performance was further evaluated using real water samples (tap water, Jiazi Lake water on campus, and Daming Lake water in the city) to assess interfering substances. As shown in Fig. 6c, OTC degradation efficiency remained high in all samples, confirming the non-radical pathway was dominant and less affected by natural water constituents, ensuring stable removal efficiency. As shown in Fig. 6d, NOBC_BM stability was tested over four recycling cycles. The NOBC_BM/PMS system maintained 88.4% OTC removal efficiency after four cycles, indicating significant in situ recoverability despite slight performance decline. Catalysts were loaded onto different reactors to systematically investigate the degradation of pollutants by the NOBC_BM/PMS system under dynamic flow conditions, simulating real water flow and assessing the impact of reaction time. Detailed schematics of the experimental setups were shown in Figs. 6e, g and i. As shown in Fig. 6f, the NOBC_BM/PTFE membrane achieved approximately 90.9% OTC removal, maintaining 86.8% efficiency after 8 h of continuous operation, indicating high stability. The slight efficiency decline may result from uneven thickness and pore size distribution during catalyst loading, causing heterogeneous degradation across different membrane regions. NOBC_BM was embedded in absorbent cotton for column experiments. Fig. 6h showed that OTC degradation initially increased to 96.1% as the catalyst-solution contact area expanded. The system retained 91.7% removal efficiency after 8 hours, demonstrating NOBC_BM stability and the column's practical potential. Two NOBC_BM sponges were placed in a beaker, with PMS-OTC solution pumped in and treated solution pumped out via peristaltic pumps; samples were taken every 120 min. As shown in Fig. 6j, OTC degradation stabilized above 80% over three days, though slightly lower than other reactors, likely due to catalyst detachment from floater movement.

  Wheat and mung bean growth experiments evaluated purified water's effects and catalyst eco-safety. Long-term tests assessed NOBC_BM/PMS system stability for engineering applications. OTC-contaminated water caused high germination but unhealthy seedlings of wheat (short/yellow/bent stems, death/abnormalities; Fig. 7a, Fig. S8 in Supporting information). Purified water enhanced growth. Rhizome lengths differed significantly between control and contaminated groups, but not control vs. purified (Fig. 7b). Contaminated-group wet weight (2.63 g) was significantly lower; dry weight differences were minor, suggesting impaired water uptake (Fig. 7c). Contaminated water induced stunted growth, weak stems, and deformities of mung bean growth (Fig. 7d, Fig. S9 in Supporting information). Purified water improved plant height/leaf expansion. Rhizome development differed significantly between control and contaminated groups, but not control vs. purified (Fig. 7e). Contaminated-group wet weight (5.26 g) was lower than control (6.31 g); its lowest dry weight indicated biomass disruption, while purified water matched control levels (Fig. 7f). Microbial toxicity assessment revealed higher toxicity and reduced E. coli density in solutions treated with OTC-contaminated water. In contrast, E. coli growth in the NOBC_BM/PMS-treated solution matched the control group (Fig. S10 in Supporting information), indicating minimal toxicity from its degradation products. Biochar catalyst/PMS systems had lower environmental impact than PMS alone due to better efficiency, reduced energy, and renewable biomass (Fig. 7g). N/O doping further reduced impacts by enhancing active sites/PMS activation, but increased marine eutrophication risk from nitrogen compounds (e.g., C₂H₄N₄) during pyrolysis.

  Figure 7

  

  Figure 7.  The growth status of (a) wheat seeds and (d) mung bean seeds on the 7^th day (Column 1: distilled water; Column 2: OTC wastewater; Column 3: purified water); The t-test in the significance analysis of the rhizome lengths of (b) wheat and (e) mung bean grown under different water bodies. The asterisk indicates a significant difference compared with the background group at the 95% confidence level. The rhizome weight of (c) wheat and (f) mung bean plants grown in different water bodies after 7 days. (g) Impact scores of various descriptors in LCA for different systems.

  DownLoad: Full-Size Img PowerPoint

  In summary, N/O co-doped biochar (NOBC_BM) was synthesized via mechanical ball milling and pyrolysis. SEM, FTIR, and Raman analyses showed NOBC_BM had a porous structure with high defect density. Degradation tests revealed NOBC_BM removed 94% of OTC (20 mg/L) in 30 min, with a kinetic constant (k_obs = 0.1523 min⁻¹) over twice that of NSBC_BM (0.0664 min⁻¹). Its high PMS adsorption energy (−3.154 eV) promoted preferential O-site binding, accelerating electron transfer to PMS and generating ¹O₂. EPR, quenching, and electrochemical experiments confirmed ¹O₂ and e⁻ as the main reactive species for OTC degradation. DFT calculations showed O-doping increased electrostatic potential, redistributed electron density at doping sites, and enhanced N-doped catalytic activity. LCA indicated favorable environmental performance. Dynamic flow tests on NOBC_BM/PTFE membranes and packed columns showed > 90% OTC removal after 8 h of continuous operation. Seven-day plant growth experiments with wheat and mung bean demonstrated significantly higher morphological parameters (plant height, rhizome) in purified water vs. OTC-contaminated water. Statistical analysis showed no differences between purified water and control groups, confirming effective toxicity removal. These results highlight NOBC_BM's long-term application potential, ecological safety, and environmental friendliness.

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  CRediT authorship contribution statement

  Enyu Zhao: Validation, Methodology, Investigation. Xin Tan: Funding acquisition, Conceptualization. Ran Liu: Investigation. Zihan Yu: Validation, Methodology, Investigation, Conceptualization. Runbin Zeng: Investigation. Wei Hong: Funding acquisition. Haiqiang Qi: Methodology, Investigation. Xuguang Li: Supervision, Conceptualization. Liangguo Yan: Supervision, Conceptualization. Xing Xu: Supervision, Conceptualization. Wen Song: Writing – review & editing, Validation, Supervision, Funding acquisition, Conceptualization.

  Acknowledgments

  This work was funded by the Department of Science and Technology of Shandong Province (No. 2024TSGC0020), National College Students' Innovation and Entrepreneurship Training Program (202510427012), the National Natural Science Foundation of China (No. 52000087), and Shandong Provincial Natural Science Foundation, China (No. ZR2020QE229).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2025.111935.

  
  
• 1. [1]

     M. Jangra, D.Y. Travin, E.V. Aleksandrova, et al., Nature 640 (2025) 1022–1030. doi: 10.1038/s41586-025-08723-7

  2. [2]

     J. Du, B. Zhang, J. Li, et al., Chin. Chem. Lett. 31 (2020) 2575–2582. doi: 10.1016/j.cclet.2020.07.050

  3. [3]

     H. Liu, M. Cheng, Y. Liu, et al., Sep. Purif. Technol. 355 (2025) 129659. doi: 10.1016/j.seppur.2024.129659

  4. [4]

     Q. Zhang, D. Zheng, B. Bai, et al., Chem. Eng. J. 500 (2024) 157134. doi: 10.1016/j.cej.2024.157134

  5. [5]

     Y. Zhang, S. Liu, D. Chen, et al., Chin. Chem. Lett. 35 (2024) 108666. doi: 10.1016/j.cclet.2023.108666

  6. [6]

     Q. Tian, B. Yu, Z. Li, et al., Chin. Chem. Lett. 36 (2025) 110322. doi: 10.1016/j.cclet.2024.110322

  7. [7]

     Y. Ren, Y. Chu, N. Li, et al., Chem. Eng. J. 455 (2023) 140859. doi: 10.1016/j.cej.2022.140859

  8. [8]

     X. Zhou, X. Li, Y. Xiang, et al., Chin. Chem. Lett. 36 (2025) 110664. doi: 10.1016/j.cclet.2024.110664

  9. [9]

     J. Yang, J. Pignatello, C. Yang, et al., Appl. Catal. B 47 (2024) 123793.

  10. [10]

      V. Chesnyak, D. Perilli, M. Panighel, et al., Sci. Adv. 10 (2024) eado8956. doi: 10.1126/sciadv.ado8956

  11. [11]

      M. Ao, J. Wei, C.S. He, et al., Chin. Chem. Lett. 36 (2025) 109882. doi: 10.1016/j.cclet.2024.109882

  12. [12]

      X. Dai, T. Liu, Y. Du, et al., Chin. Chem. Lett. 36 (2025) 110548. doi: 10.1016/j.cclet.2024.110548

  13. [13]

      H. Chen, H. Liao, Q. Zhou, et al., Chin. Chem. Lett. 35 (2024) 11071.

  14. [14]

      R. Wang, Q. Shi, W. Wang, et al., Environ. Sci. Nano 12 (2025) 2407–2420. doi: 10.1039/d5en00004a

  15. [15]

      Y. Ren, J. Li, C. Liu, et al., J. Hazard. Mater. 485 (2025) 136799. doi: 10.1016/j.jhazmat.2024.136799

  16. [16]

      C.H. Gu, S. Wang, A.Y. Zhang, et al., Nat. Commun. 15 (2024) 577. doi: 10.1038/s41467-024-44857-4

  17. [17]

      Y. Sun, Y. Zang, B. He, et al., Sci. Adv. 11 (2025) eadq2948. doi: 10.1126/sciadv.adq2948

  18. [18]

      J. Wang, B. Wang, Y. Li, et al., J. Hazard. Mater. 480 (2024) 136442. doi: 10.1016/j.jhazmat.2024.136442

  19. [19]

      W. Song, Z. Yu, H. Li, et al., Chem. Eng. J. 498 (2024) 155601. doi: 10.1016/j.cej.2024.155601

  20. [20]

      H. Shi, Y. He, Y. Li, et al., J. Hazard. Mater. 443 (2023) 130345. doi: 10.1016/j.jhazmat.2022.130345

  21. [21]

      W. Song, Y. Ji, Z. Yu, et al., Int. J. Biol. Macromol. 281 (2024) 136525. doi: 10.1016/j.ijbiomac.2024.136525

  22. [22]

      Y. Lei, Y. Yu, X. Lei, et al., Environ. Sci. Technol. 57 (2023) 5433–5444. doi: 10.1021/acs.est.2c09338

  23. [23]

      Y. Sun, W. Han, F. Zhang, et al., Appl. Catal. B 345 (2024) 123689.

  24. [24]

      L. Wang, Q. Li, C. Xu, et al., Water Res. 255 (2024) 121519.

  25. [25]

      K. Yin, J. Yang, Y. Li, et al., Chin. Chem. Lett. 35 (2024) 109847. doi: 10.1016/j.cclet.2024.109847

  26. [26]

      L. Tang, S. Zhou, F. Li, et al., Environ. Sci. Technol. 57 (2023) 18550–18562. doi: 10.1021/acs.est.2c06008

  27. [27]

      S. Xu, P. Wang, X. Mi, et al., J. Hazard. Mater. 466 (2024) 133321.

  28. [28]

      J. Mao, K. Yin, Y. Zhang, et al., Appl. Catal. B: Environ. 342 (2024) 123428.

  29. [29]

      X. Qi, K. Obata, C. -K. Hung, et al., Nat. Commun. 16 (2025) 3646.

  30. [30]

      Q. Zeng, Y. Wen, X. Duan, et al., Appl. Catal. B 346 (2024) 123752.

  31. [31]

      M. Cai, Q. Zhu, X. Hou, et al., ACS Energy. Lett. 9 (2024) 5587–5595. doi: 10.1021/acsenergylett.4c02650

  32. [32]

      R. Wu, R. Gao, Z. Li, et al., J. Mater. Sci. Technol. 238 (2025) 266–275.

• 

  Figure 1  SEM images of (a) NSBC_BM, (b) NOBC_BM. (c) EDS mapping images of NOBC_BM. (d) FTIR patterns of Straw, NSBC_BM and NOBC_BM. (e) Raman patterns for NSBC_BM and NOBC_BM. (f) C 1s, (g) N 1s and (h) O 1s XPS spectrum of NSBC_BM, NOBC_BM.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  Degradation of OTC (a) in different O doping ratios and (b) k_obs values; Degradation of OTC (c) in different PMS dosage and (d) k_obs values. (e) Adsorption of OTC by NOBC_BM. (f) Correlations between different PMS dosage and k_obs. (g) Normalized k_obs comparison of OTC degradation by NOBC_BM/PMS at different PMS concentrations. (h) Normalized k_obs comparison of OTC degradation by NOBC_BM/PMS and other nonmetallic catalyst/PMS systems.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  OTC quenching experiment and k_obs at PMS concentrations of (a) 0.05, (b) 0.2 and (c) 0.5. (d) EPR spectrums of NOBC_BM/PMS systems. (e) Different concentrations of TEMP quench ¹O₂. (f) Ultraviolet spectrum of DPA-¹O₂.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  (a) Binding of PMS onto different catalyst configurations with different adsorption energies (E_ed) and l_o-o binding lengths. (b) Electron density distribution of different structures.

  下载: 全尺寸图片 幻灯片
  

  Figure 5  (a) LC-MS analysis of OTC degradation products and the possible pathways for OTC degradation in NOBC_BM/PMS system. (b, c) Toxicity analysis of OTC and its degradation intermediates. (d–g) Changes of three-dimensional fluorescence during OTC degradation.

  下载: 全尺寸图片 幻灯片
  

  Figure 6  (a) Degradation of OTC at different pH by NOBCBM/PMS system. (b) Degradation of OTC in different anions by NOBCBM/PMS system. (c) Degradation of OTC in different water bodies by NOBCBM/PMS system. (d) Stability tests of NOBCBM under four replicate experiments and removal efficiency. (e) Membrane reactor setup and (f) the degradation efficiency of OTC at different times. (g) Column reactor setup and (h) the degradation efficiency of OTC at different times. (i) Floater reactor setup and (j) the degradation efficiency of OTC at different times.

  下载: 全尺寸图片 幻灯片
  

  Figure 7  The growth status of (a) wheat seeds and (d) mung bean seeds on the 7^th day (Column 1: distilled water; Column 2: OTC wastewater; Column 3: purified water); The t-test in the significance analysis of the rhizome lengths of (b) wheat and (e) mung bean grown under different water bodies. The asterisk indicates a significant difference compared with the background group at the 95% confidence level. The rhizome weight of (c) wheat and (f) mung bean plants grown in different water bodies after 7 days. (g) Impact scores of various descriptors in LCA for different systems.

  下载: 全尺寸图片 幻灯片
•