English

• Nitrogen-contained refractory organic wastewater is mainly derived from industrial activities including chemical manufacturing, pharmaceutical production, and textile dyeing. These wastewaters frequently contain inorganic nitrogen (e.g., nitrate (NO₃⁻-N) and ammonium (NH₄⁺-N)) and refractory organic contaminants (e.g., phenolic compounds, heterocyclic compounds, and polycyclic aromatic hydrocarbons), conferring high organic load, complexity, toxicity, and resistance to degradation. Conventional biological processes, which rely on microbial nitrogen conversion, are largely ineffective against such refractory contaminants [1]. Synergistic bottlenecks between nitrogen transformation and organic contaminant degradation (e.g., electron donor competition and inhibition by intermediates) further amplify treatment challenges [2]. Meanwhile, ammonia (NH₃) can be recovered as high-value products (e.g., struvite and ammonium sulfate) via adsorption or ion exchange for agricultural and industrial applications [3]. Existing technologies are unable to synergistically achieve contaminant mineralization and targeted nitrogen transformation (from NO₃⁻-N to NH₄⁺-N) for NH₃ recovery [4]. Therefore, developing integrated technology for simultaneous contaminant degradation and targeted nitrogen transformation is an urgent task for treating nitrogen-contained refractory organic wastewater.

  Zero-valent iron (ZVI)-based wastewater treatment processes are effective in treating nitrogen-contained refractory organic wastewaters. As a reactive metal (E⁰ = −0.44 V), microscale ZVI (mZVI) removes refractory organic contaminants synergistically via direct reduction and indirect oxidation. In the direct reduction process, mZVI as the electron donor can release electrons via the oxidation of Fe⁰ to Fe(II), thereby directly reducing electron-deficient contaminants, including heavy metals [5], NO₃⁻-N [6], nitroaromatics [7], and chlorinated organic contaminants [8]. Concurrently, hydrogen atoms (H^•) generated during mZVI corrosion can also indirectly reduce above organic contaminants [9,10]. In the indirect oxidation process, mZVI corrosion releases electrons for reducing O₂ via two electrons transfer to generate H₂O₂, initiating Fenton reactions to generate hydroxyl radicals (^•OH) for efficient organic contaminant degradation (Text S1 in Supporting information) [11,12]. Efficiency enhancements involve coupling with oxidants (e.g., H₂O₂ [13] and O₃ [14]), constructing Fe/Cu systems [15], and employing iron-carbon micro-electrolysis systems which utilize the micro-galvanic cells formed by iron and carbon to enhance electron transfer [16]. Currently, mZVI-based processes demonstrate engineering-scale implementation in environmental remediation, enabled by synergistic direct reduction and indirect oxidation pathways.

  First employed for NO₃⁻-N removal in the 1970s [17], ZVI preferentially reduces NO₃⁻-N into NH₄⁺-N (or NH₃) and N₂ via direct electron transfer through its surface corrosion layer (Text S2 in Supporting information), releasing Fe(II) [2]. Complete NO₃⁻-N reduction occurs within hours under acidic conditions [18], dominated by mZVI-surface direct electron transfer [19]. Furthermore, the mZVI-based processes demonstrate efficiently degrade refractory organic contaminants. For instance, the mZVI/O₃ process attains an 89.5% COD removal rate for p-nitrophenol (PNP)-containing wastewater within 60 min [14]. Through synergistic approaches of direct reduction and indirect oxidation, the mZVI-based processes enable simultaneous conversion from NO₃⁻-N to NH₄⁺-N and degradation of refractory organic contaminants, preliminarily validating its feasibility for co-treatment of nitrogen and organic contaminants in nitrogen-contained refractory organic wastewater.

  In early studies, researchers focused only on the reaction kinetics of single solute systems, without considering the potential effects of coexisting electron acceptors [2]. As research progressed, it was found that when NO₃⁻-N coexists with organic contaminants, its high oxidation leads to high-reactivity with ZVI, thereby inhibiting organic contaminants reduction efficiency (e.g., halogenated hydrocarbon dehalogenation) through competitive occupation of reactive sites and electron consumption [20]. While increasing the dosages of ZVI can alleviate competition, it will increase the generation of iron sludge [21]. The sulfur-modified ZVI (S-ZVI) can inhibit the reduction of NO₃⁻-N but fails to achieve its targeted removal [22]. Additionally, the indirect oxidation in ZVI-based processes relies on ZVI and its corrosion product Fe(II) to release electrons for activating dissolved oxygen/oxidants. Electron competition from NO₃⁻-N may affect oxidation pathways [2], but there is still a lack of quantitative studies on the suppression of oxidation. Notably, NH₄⁺-N as the main reduction product of NO₃⁻-N in the mZVI/O₂ process may be further oxidized to NO₃⁻-N by ^•OH or other reactive oxygen species (ROS) [23], initiating an inorganic nitrogen redox cycle, needlessly consuming iron electrons (Fig. S1 in Supporting information). This cycle not only wastes electrons and generates excessive iron sludge, but also may reduce the overall oxidation efficiency of organic contaminants. To address these challenges, it is necessary to clarify the dynamic coupling mechanism between inorganic nitrogen transformations and redox activities. Therefore, a two-stage gradient oxidation process (mZVI/O₂-Fenton) was proposed to synergize the degradation of organic contaminants and the directed transformation of NO₃⁻-N to NH₄⁺-N. However, the effect of inorganic nitrogen transformations (e.g., NH₄⁺-N oxidation) still needs to be further verified.

  This study investigates the mechanisms of inorganic nitrogen transformation in the mZVI/O₂ process (Fig. 1a) and develops a two-stage gradient oxidation process (mZVI/O₂-Fenton) (Fig. 1b) for treating high-concentration nitrogen-contained refractory organic wastewater. The inhibition of organic contaminant degradation by inorganic nitrogen in the mZVI/O₂ process was systematically investigated by regulating inorganic nitrogen species, organic contaminant types, and process parameters. The electron competition mechanism of NO₃⁻-N was elucidated through iron dissolution kinetics and ROS analysis, clarifying NO₃⁻-N interference on H₂O₂ production and ^•OH formation. Simultaneously, the morphological and compositional evaluations of mZVI powders were analyzed by scanning electron microscope (SEM), X-ray photoelectron spectroscopy (XPS), and X-ray diffraction (XRD) to explain the regulatory effects of inorganic nitrogen on mZVI corrosion and passivation layer formation. Furthermore, the migration and transformation pathways of inorganic nitrogen in the mZVI/O₂ process were tracked, and the mZVI/O₂-Fenton process was designed by integrating the selective oxidation behavior of NH₄⁺-N in Fenton/Fenton-like systems. Finally, the synergistic efficiencies of organic contaminant degradation, NH₄⁺-N generation as well as the optimization potential of electron utilization efficiency were validated by treating high-concentration nitrogen-contained refractory organic wastewaters derived from multiple chemical plants.

  Figure 1

  

  Figure 1.  Process flow diagrams of (a) mZVI/O₂ process and (b) two-stage gradient oxidation process (mZVI/O₂-Fenton).

  DownLoad: Full-Size Img PowerPoint

  The details of all chemicals, analysis and characterization methods were provided in Texts S3-S9 in Supporting information.

  The effect of inorganic nitrogen species (NO₃⁻-N, NO₂⁻-N, and NH₄⁺-N) on organic contaminant removal in the mZVI/O₂ process was investigated, with phenol (PE) was selected as the target contaminant. Fig. 2a depicts that high concentrations of NO₃⁻-N and NO₂⁻-N can significantly inhibit the mZVI/O₂ process to remove organic contaminants, whereas the effect of NH₄⁺-N was unobvious. The distinct oxidation–reduction characteristics of nitrogen species likely underlie this inhibition. Under acidic conditions (pH 3.5), NO₃⁻-N undergoes preferential reduction on the Fe⁰ surface, competing for electrons required for oxidizing organic contaminants [24]. Although the oxidizing capacity of NO₂⁻-N is weak, it can act as an electron donor and acceptor, thus significantly affecting the redox pathways. The concentration effects (50–200 mg/L) of NO₃⁻-N were further investigated due to its widespread presence in real wastewater (Fig. 2b). With the increase of NO₃⁻-N concentration, the removal of COD and TOC significantly decreased, and the k_obs for PE removal decreased to less than half of that in the absence of NO₃⁻-N (Fig. S2 in Supporting information). This inhibitory effect might be associated with the electron competition and surface passivation, which are inherent disadvantages of the mZVI/O₂ process.

  Figure 2

  

  Figure 2.  Effect of (a) nitrogen species, (b) NO₃⁻-N concentration, (c) pH, (d) mZVI dosage, (e) different organic contaminants on the performance of the mZVI/O₂ process.

  DownLoad: Full-Size Img PowerPoint

  The performance of mZVI exhibits a pH-dependent phenomenon, and the main reason is attributed to electron competition and surface passivation. The performance of mZVI/O₂ process greatly depended on pH within the range of pH 2.0–6.0, with the best performance at pH 3.5 (Fig. 2c). The corrosion of mZVI was obviously enhanced with the lowering pH, and active sites were generated to improve the contaminant removal and NO₃⁻-N reduction, while side reactions would happen when the pH is to too low [25]. Although the pH showed the same changing trend, because NO₃⁻-N is an electron acceptor, the electrons available for oxidizing organic contaminants are decreased, and the overall removal efficiency is always lower than that of the process without NO₃⁻-N [26]. The increased pH favored the formation of Fe(II)/Fe(III) (oxy)hydroxide passivation layers, and the activity of mZVI was inhibited [11]. The removal efficiency of all systems decreased systematically, regardless of the presence of NO₃⁻-N. XRD analysis showed that no obvious iron oxide peaks were detected (60 min, pH 3.5), which indicated that NO₃⁻-N mainly affected the performance by electron competition, and pH mainly affected the mZVI corrosion and active sites formation. Increasing the dosage of mZVI alleviated the inhibitory effect of NO₃⁻-N by providing more active sites and electrons (Fig. 2d, Figs. S5a and b in Supporting information) [24] and electron competition mechanism was further verified.

  In the mZVI/O₂ process, the concentration variation and COD removal of different organic contaminants were significantly different. (Fig. 2e and Fig. S6 in Supporting information). The reduction removal of nitrobenzene (NB) [27] and PNP [14] is not affected by NO₃⁻-N, which is attributed to the electron-withdrawing capability of -NO₂, enabling NB and PNP to preferentially act as electron acceptors rather than NO₃⁻-N in the mZVI/O₂ process [28]. However, COD removal was significantly reduced with NO₃⁻-N. Meanwhile, PE [29], aniline (AN) [27], benzoic acid (BA) [30], tetrakis(hydroxymethyl)phosphonium sulfate (THPS) [31], and rhodamine B (RhB) [32] were removed by indirect oxidation, weak reduction pathways, or flocculation, which significantly decreased their elimination with NO₃⁻-N. This indicates that NO₃⁻-N also affects the oxidation of organic pollutants by the mZVI/O₂ process. NO₃⁻-N may inhibit the Fenton reaction-driven oxidation process by competing with O₂ for abstracting electrons released from mZVI.

  It has been reported that NO₃⁻-N as an iron-reducible anion, curbs the reductive removal of organic contaminants via competition for active site or mZVI passivation [26]. This study suggests NO₃⁻-N further reduces removal efficiency by impairing the indirect oxidation of mZVI through electron competition.

  Under aerobic conditions, stepwise oxidation from Fe⁰ to Fe(II) and Fe(III) during mZVI corrosion continuously donates electrons driving O₂ reduction to various ROS. The mZVI/O₂ process achieves organic contaminant removal through two main stages: (1) mZVI corrosion generates Fe(II) and releases electrons competitively captured by oxidizing species (O₂, NO₃⁻-N, and electron-withdrawing groups of organic contaminants); (2) ROS generation via Fenton reaction involving Fe(II) and H₂O₂. The whole process is controlled by the rapid escape of iron ions from the mZVI surface [11]. Although NO₃⁻-N suppresses organic contaminant removal through electron competition, it can also paradoxically enhance mZVI corrosion. To analyze this, the total dissolved iron (T_Fe) was determined both with and without nitrogen species (Fig. 3a). NO₃⁻-N increased T_Fe by 25.0% (from 2.8 g/L to 3.5 g/L), whereas NO₂⁻-N decreased T_Fe by 42.9% (from 2.8 g/L to 1.6 g/L), with negligible impact from NH₄⁺-N. It suggests direct redox interaction of NO₃⁻-N with mZVI, which promotes the Fe(II) release via enhanced corrosion [22]. XPS spectra confirmed NO₃⁻-N induced greater exposure of reactive Fe⁰ sites, accelerating Fe(II) oxidation products generation (Fig. S7 in Supporting information).

  Figure 3

  

  Figure 3.  Total dissolved iron concentration in the reaction solution. Effect of (a) nitrogen species, (b) pH, (c) different organic contaminants, (d) time series data for total dissolved iron, (e) NO₃⁻-N concentration, (f) iron electron utilization in different organic contaminants.

  DownLoad: Full-Size Img PowerPoint

  In acidic conditions, H⁺ and NO₃⁻-N synergistically accelerate mZVI corrosion, contributing to the maintenance of a high value of T_Fe. T_Fe can be increased by 17.5% to 65.7% with NO₃⁻-N, which also enhances different organic pollutants removal by mZVI/O₂ (Figs. 3b–d). It increased further with increasing NO₃⁻-N dosing (Fig. 3e). The standard electrode potential of NO₃⁻/NO₂⁻ (0.01 V) exceeds Fe(II)/Fe⁰ (−0.44 V) enabling NO₃⁻-N to function as a strong electron acceptor that promotes Fe⁰ oxidation to Fe(II), enhancing iron dissolution [33]. Under neutral conditions (pH 5.0-6.0), T_Fe decreases sharply (pH 6.0, 0.08 g/L with NO₃⁻-N and 0.03 g/L without NO₃⁻-N). This stems from Fe(II) and Fe(III) hydrolysis forming dense passivation layers comprising iron (hydroxide)oxides that function as diffusion barriers, inhibiting further iron dissolution (Figs. S4 and S8 in Supporting information). Conversely, NO₃⁻-N addition increased surface Fe(II) content (Fig. S9 in Supporting information), demonstrating its capacity to continuously drive Fe⁰ oxidation even at mild pH, partially counteracting passivation layer inhibition and sustaining mZVI corrosion activity across wide pH ranges.

  Increased iron dissolution provides more redox active sites (Fig. 2d and Fig. S10 in Supporting information) [2]. However, although T_Fe was consistently higher with NO₃⁻-N, organic contaminant degradation efficiency was significantly lower. Analysis of COD removal/mZVI consumption ratios (ΔCOD/ΔFe) revealed the ratio was significantly higher without NO₃⁻-N (Fig. S11 in Supporting information). It indicates that NO₃⁻-N resulted in a decrease in the amounts of organic contaminants removed per unit mass of dissolved iron. To further characterize electron donor efficiency, the concept of electron efficiency (EE) was applied (Eq. 1) [34]. N_o represents electrons for organic contaminants removal (determined by COD removal). N_Fe denotes electrons from mZVI corrosion (quantified via T_Fe) (Text S10 in Supporting information). NO₃⁻-N decreased EE by 0.3%−17.8% across different organic contaminants (Fig. 3f). This suggests that while NO₃⁻-N promoted iron dissolution and particle corrosion, as visually confirmed by significant fragmentation of residual mZVI particles in SEM images (Fig. 4), it concurrently reduced the electron efficiency for organic contaminant removal [2]. Ultimately, elevated T_Fe cannot offset NO₃⁻-N-induced electron competition, wherein NO₃⁻-N preferentially captures the electrons released by mZVI and inhibits the transformation from O₂ to ROS, blocking the Fenton oxidation pathway.

  $ \mathrm{EE}=N_{\mathrm{o}} / N_{\mathrm{Fe}} \times 100 \% $

  (1)

  Figure 4

  

  Figure 4.  SEM of mZVI before and after reaction at pH 3.5. (a) Pristine mZVI, (b) without nitrogen, (c) with NO₃⁻-N, (d) with NO₂⁻-N.

  DownLoad: Full-Size Img PowerPoint

  The accelerated iron dissolution by NO₃⁻-N, however, does not translate to improved organic contaminants removal. This discrepancy necessitates an investigation into ROS dynamics under NO₃⁻-N interference. To clarify the inhibition mechanism of ROS generation in the mZVI/O₂ process from NO₃⁻-N via electron competition, H₂O₂ quantification, EPR analysis, radical scavenging assays, and chemical probe tests were used for systematic validation. Theoretically, the reaction of Fe⁰ and O₂ can produce H₂O₂ [11], but concurrent Fenton-driven H₂O₂ consumption results in dynamic fluctuations of H₂O₂ concentrations between 40.7–54.9 μmol/L (mean concentration: 43.2 μmol/L). Increasing NO₃⁻-N dosage progressively lowers equilibrium H₂O₂ levels, exemplified by a reduction to 10.2–32.9 μmol/L (mean concentration: 20.0 μmol/L) at 600 mg/L NO₃⁻-N (Fig. 5a). EPR was used to characterize ROS because 5,5-dimethyl-1-pyrroline-N-oxide (DMPO) reacts specifically with ^•OH to form DMPO-OH [35]. The presence of a typically quartet line of DMPO-OH (α^N = α^H = 14.9 G, 1:2:2:1) in Fig. 5b confirms the generation of ^•OH in the process [27]. NO₃⁻-N addition weakened the DMPO-OH adduct peak intensity, suggesting that NO₃⁻-N inhibits the cascade pathway of O₂ activation to produce H₂O₂ and ^•OH by competing for electrons.

  Figure 5

  

  Figure 5.  (a) H₂O₂ generation, (b) EPR analysis for radicals, (c) TBA inhibition on PE removal, (d) hydroxylated products pathway, (e) and (f) BA as chemical probe for analyzing hydroxylation products.

  DownLoad: Full-Size Img PowerPoint

  Radical scavenging experiments verified the dominant role of ^•OH in the mZVI/O₂ process and the inhibitory effect of NO₃⁻-N on its generation. Tert‑butyl alcohol (TBA) was used as a radical scavenging reagent to selectively screen ^•OH (k(^•OH, TBA) = 6.0 × 10⁸ L mol⁻¹ s⁻¹) [36]. The presence of TBA decreased PE removal by 41.9% (from 56.5% to 14.6%) without NO₃⁻-N and by 30.4% (from 36.8% to 6.4%) with NO₃⁻-N (Fig. 5c). This confirms that NO₃⁻-N not only reduces ^•OH content but may also diminish secondary oxidation pathways (e.g., direct surface oxidation). BA hydroxylation is an effective method for ^•OH quantification via H-abstraction producing characteristic products (Fig. 5d) [37]. Figs. 5e and f show the rapid hydroxylation of BA to 4-hydroxybenzoic acid (4-HBA) (35.6 μmol/L) and further oxidation to produce 3,4-dihydroxybenzoic acid (3,4-DHBA) (78.6 μmol/L) without NO₃⁻-N, indicating that the mZVI/O₂ process can continuously produce ^•OH and have the ability to oxidize organic contaminants in a multistep process. In contrast, the addition of NO₃⁻-N resulted in slower BA degradation, a lower rate of 4-HBA production, and a decrease in the accumulation of 3,4-DHBA to 56.9 μmol/L, further indicating that NO₃⁻-N weakened the degradation efficiency of the mZVI/O₂ process for organic contaminants by inhibiting the production of ^•OH.

  Overall, NO₃⁻-N significantly regulates ROS generation and organic contaminants removal in the mZVI/O₂ process through electron competition. By competing with O₂ for electrons released during mZVI corrosion, NO₃⁻-N markedly suppresses the oxygen activation cascade. This leads to reduced generation intensity of the critical oxidant ^•OH, thereby inhibiting multi-step oxidation processes and limiting the oxidation of organic contaminants. Combined analysis of rules of iron dissolution and ROS generation, NO₃⁻-N has a dual role in the mZVI/O₂ process: (1) As an electron acceptor enhancing Fe° corrosion, increasing T_Fe by 17.5%−65.7%; (2) Inhibiting the cascade pathway of O₂ activation to produce H₂O₂ and thus ^•OH by competing for electrons, which results in a decrease of COD removal efficiency per unit of iron consumed of up to 57.9%.

  The transformation of NO₃⁻-N is dominated by the direct reduction to NH₄⁺-N as the dominant pathway in the mZVI/O₂ process [38]. It has been shown that NO₃⁻-N is preferentially reduced to NH₄⁺-N by H⁺-driven electron transfer through the surface of mZVI, and the TN decreased slightly after 30 min of reaction, with NH₄⁺-N accounting for > 98.2% of the reduction products (Fig. 6a). The production of minor NO₂⁻-N (0.2 mg/L) and the decrease in TN suggest the existence of an indirect reduction pathway: a little NO₃⁻-N can be further transformed to NH₄⁺-N and N₂ via the NO₂⁻-N intermediate state. The addition of NO₂⁻-N in the control experiment resulted in 100.0% conversion, 53.0% TN removal and 40.8% product NH₄⁺-N selection (Fig. 6b) [38]. It is further demonstrated that in the mZVI/O₂ process, most of NO₃⁻-N can be directly reduced into NH₄⁺-N, and a little NO₃⁻-N will be reduced to NO₂⁻-N first through an indirect reduction pathway and then converted to NH₄⁺-N or N₂ (Text S2) [39]. Significantly, NH₄⁺-N is chemically inert in this process and neither affects the removal of organic contaminants nor participates in redox reactions (Figs. 2a and 6b).

  Figure 6

  

  Figure 6.  Conversion of nitrogen in the mZVI/O₂ process. (a) NO₃⁻-N transformations, (b) transformation of different nitrogen species, (c) effect of pH on NO₃⁻-N conversion in the mZVI/O₂ process, (d) oxidation of NH₄⁺-N by peroxymonosulfate (PMS) systems.

  DownLoad: Full-Size Img PowerPoint

  The conversion efficiency of NO₃⁻-N is significantly regulated by pH and organic contaminants species (Fig. 6c and Fig. S12 in Supporting information), independent of the initial NO₃⁻-N concentration (Fig. S13 in Supporting information). The conversion of NO₃⁻-N in acidic conditions (pH 2.0–3.0) is 61.0%−70.2% within 60 min by accelerating the Fe° corrosion and enhancing the H⁺ supply. When the pH increases, it leads to a slowing down of Fe° corrosion and a gradual decrease in the conversion rate [40,41]. In addition, the electron demand characteristics of coexisting organic contaminants can affect the conversion rate of NO₃⁻-N. However, the NO₃⁻-N reduction pathway cannot be altered regardless of changing environmental conditions, and NH₄⁺-N is always stabilized as the end product. Consequently, in the mZVI/O₂ process, NO₃⁻-N is predominantly converted to NH₄⁺-N via a direct 8-electron reduction pathway, with the presence of the NH₄⁺-N product exhibiting virtually no adverse impact on treatment efficacy.

  NO₃⁻-N addition promotes T_Fe dissolutionin in the mZVI/O₂ process and is essentially in the form of Fe(II) (Fig. S14 in Supporting information). Therefore, the residual Fe(II) from the mZVI/O₂ process was coupled with extra added oxidation to construct the subsequent Fenton/Fenton-like oxidation process to deeply oxidize organic contaminants [42]. Considering the potential oxidation of NH₄⁺-N by ROS generated in these Fenton/Fenton-like processes, where such oxidation may impact contaminant degradation and nitrogen resource recovery, the study systematically evaluated the NH₄⁺-N oxidation capabilities of four Fenton or Fenton-like processes. In the conventional Fenton system (Fe(II)/H₂O₂) [43], both H₂O₂ and its activated product ^•OH (2.7 V vs. NHE) exhibited negligible oxidation of NH₄⁺-N under acidic conditions due to extremely slow reaction kinetics (Fig. S15a in Supporting information). It is consistent with Figs. 2a and 6b that the presence of NH₄⁺-N do not affect the mZVI/O₂ process. Within the persulfate activation systems (Fe(II)/PS) [44], peroxydisulfate (PDS) and PMS activated by Fe(II) can generate SO₄^•− (2.5-3.1 V vs NHE), Fe(IV) (1.3-2.0 V vs. NHE), and ^•OH (Fig. 6d and Fig. S15b in Supporting information) [45,46]. These ROS struggled to oxidize positively charged NH₄⁺-N owing to their electrophilic attack mechanisms. However, PMS itself slowly oxidized NH₄⁺-N via a non-radical pathway. The peracetic acid (PAA) system (Fe(II)/PAA) primarily produced Organic Radicals (RO^• 1.5–2.3 V vs. NHE), Fe(IV), and ^•OH, all displaying minimal oxidation efficiency toward NH₄⁺-N (Fig. S15c in Supporting information) [47]. Collectively, although ^•OH, SO₄^•−, Fe(IV), and RO^• exhibit high redox potentials, their reaction mechanisms predominantly rely on electrophilic attack (Table S1 in Supporting information) [48,49]. The nitrogen atom in NH₄⁺-N is positively charged with low electron density, exhibiting electron-deficient properties which make it resistant to oxidation by electrophilic radicals.

  Considering the high salinity characteristics of some industrial organic wastewater, chloride ions (Cl⁻) were introduced into the system to explore the effects on the nitrogen transformation pathways and reaction kinetics under saline conditions. The results showed that the Fe(II)/PMS/Cl⁻ system could completely oxidize of NH₄⁺-N, and most of the NH₄⁺-N was transformed to N₂. This phenomenon was mainly attributed to the action of reactive chlorine radicals (e.g., ClO^•, Cl^•) [50]. Despite the relatively lower redox potentials of chlorine radicals (e.g., Cl^•: 2.4 V; ClO^•: 1.5–1.8 V vs. NHE), they exhibit enhanced substrate-specific oxidative capacity compared to conventional radicals [51]. The reaction mechanisms of these chlorine-derived radicals predominantly involve nucleophilic attack, which enables efficient oxidation of NH₄⁺-N by targeting its electron-deficient nitrogen atom and weakening N-H bonds [52]. Accordingly, the nitrogen conversion pathway in the mZVI/O₂ process is proposed: NO₃⁻-N competes with O₂ for the electrons released from Fe⁰, and is directly reduced to NH₄⁺-N through a one-step 8-electron pathway. NH₄⁺-N is difficult to be oxidized not only by ^•OH generated in Fenton system, but also by SO₄^•−, RO^• or Fe(Ⅳ) from other Fenton-like systems. However, ClO^•, Cl^• and other reactive chlorine radicals can effectively oxidize it to N₂ or NO₃⁻-N (Fig. 7).

  Figure 7

  

  Figure 7.  Inorganic nitrogen transformation pathways in the mZVI-based process.

  DownLoad: Full-Size Img PowerPoint

  Based on the previously established electron efficiency (EE) concept (Eq. 1), this study further defines a modified electron efficiency (EE_N) to explicitly account for electrons consumed in NO₃⁻-N reduction (Eq. 2). EE_N considers both N_o and N_n (electrons required for NO₃⁻-N reduction) as an efficient use of electrons (Text S10). The results demonstrate that the addition of NO₃⁻-N can increase the EE_N (increased by 8.2%−53.8%), both in the case of treating different organic contaminants (Fig. 8a). It has been shown that NO₃⁻-N enhances the effective utilization of iron electrons and reduces the generation of side reactions [22], which in turn may enhance the overall efficiency of wastewater treatment in the mZVI/O₂ process.

  $ \mathrm{EE}_{\mathrm{N}}=\left(N_{\mathrm{n}}+N_{\mathrm{o}}\right) / N_{\mathrm{Fe}} \times 100 \% $

  (2)

  Figure 8

  

  Figure 8.  (a) Iron electron utilization in different organic contaminants, (b) NO₃⁻-N conversion and NH₄⁺-N selection with different iron dosages, (c) and (e) organic contaminants removal by two-stage gradient oxidation process, (d) and (f) NO₃⁻-N transformation in a two-stage gradient oxidation process.

  DownLoad: Full-Size Img PowerPoint

  Based on the characteristics that after degradation of organic contaminants and reduction of NO₃⁻-N by the mZVI/O₂ process: (1) Excess Fe(II) in the system is underutilized, (2) NO₃⁻-N can be converted to NH₄⁺-N directionally, (3) NH₄⁺-N is difficult to be oxidized by ^•OH, (4) NO₃⁻-N enhances the effective utilization of iron electrons. The two-stage gradient oxidation process of mZVI/O₂-Fenton has been proposed (Fig. 1b). In the 1^st stage, NO₃⁻-N is used to promote Fe° corrosion to realize the pre-oxidation of high concentration of organic contaminants and the initial utilization of iron electrons to convert NO₃⁻-N to NH₄⁺-N directionally. At the same time, a large amount of Fe(II) is released. In the 2nd stage, H₂O₂ is added to produce ^•OH via Fenton reaction to strengthen organic contaminant oxidation and achieve further utilization of iron electrons. The mZVI/O₂-Fenton two-stage gradient oxidation process is used to treat high-concentration NO₃⁻-N-contained organic wastewater, synergistically realizing highly efficient removal of organic contaminants and targeted recovery of nitrogen. It improves wastewater treatment efficiency and reduces wastewater treatment operating costs [53].

  Through adjusting factors such as pH, iron dosage [39], and reaction time, the 1st stage achieves 93.1% NO₃⁻-N reduction efficiency with nearly 100% NH₄⁺-N selectivity under optimal conditions (Fig. 8b). More than 91.0% COD removal was achieved in the two-stage gradient oxidation process for treatment of organic wastewater with high concentration of organic contaminants (COD₀ 2000 mg/L, Fig. 8c). Surprisingly, COD removal was better in the presence of NO₃⁻-N. This is the opposite of the removal effect in the 1st stage. Furthermore, the NO₃⁻-N conversion rate of 76.8% with 72.7% NH₄⁺-N selectivity was achieved, demonstrating significant potential for nitrogen resource recovery (Fig. 8d). Acute toxicity test with luminescent bacteria revealed that the mZVI/O₂-Fenton two-stage gradient oxidation process significantly reduces toxicity when NO₃⁻-N is present (a more significant decrease in IR value) (Fig. S17, Table S2 in Supporting information) [54,55]. To evaluate the potential application of the process in actual wastewater treatment, four organic industrial wastewaters containing high concentrations of NO₃⁻-N (including paint wastewater, primary explosive wastewater, tetrazene wastewater and pharmaceutical wastewater) were selected for treatment experiments (Table S3 in Supporting information). The results demonstrated significant organic contaminant removal coupled with effective nitrogen transformation across all wastewaters after two-stage gradient oxidation treatment with mZVI/O₂-Fenton, despite variations in treatment efficiency among the different types. The COD removal rate was stable between 79.8% and 93.7% for all four wastewaters, with conversion of NO₃⁻-N and selectivity of NH₄⁺-N higher than 64.7% (Figs. 8e and f, and Fig. S18 in Supporting information). In addition, UV–vis was used to monitor the degradation of pollutants before and after wastewater treatment (Fig. S19 in Supporting information) [53]. The results further confirmed that the mZVI/O₂-Fenton two-stage gradient oxidation process has significant degradation and detoxification effects on organic industrial wastewater with high NO₃⁻-N. The 8-h continuous flow experiment confirmed the long-term operational stability of the process (Text S11 in Supporting information). For paint wastewater containing high concentrations of NO₃⁻-N, the effluent COD removal remained consistently above 88.1%, with ammonia nitrogen selectivity exceeding 69.9% (Fig. S20 in Supporting information). The mZVI/O₂-Fenton two-stage gradient oxidation process takes advantage of the high electron utilization of NO₃⁻-N and synergizes the directed NO₃⁻-N transformation and the high efficiency removal of organic contaminants. It provides an innovative solution based on efficient treatment and nitrogen recovery for refractory organic wastewater with high concentration of NO₃⁻-N.

  This study reveals the dual regulatory mechanism of NO₃⁻-N in the mZVI/O₂ process. NO₃⁻-N acts as an electron acceptor to promote Fe° corrosion, increasing iron ion dissolution by 17.5%−65.7% under various pH conditions compared to the system without NO₃⁻-N. And meanwhile, it inhibits ROS generation pathways through electron competition, thereby reducing organic contaminants removal efficiency. Analysis of nitrogen transformation pathways demonstrates that NO₃⁻-N is predominantly converted to NH₄⁺-N via a direct 8-electron reduction pathway, with the resultant NH₄⁺-N exhibiting chemical inertness in the mZVI/O₂ process. Based on these findings, a novel mZVI/O₂-Fenton two-stage gradient oxidation process was developed: 1st stage leverages the corrosion-enhancing effect of NO₃⁻-N to achieve targeted NH₄⁺-N reduction (76.8% NO₃⁻-N conversion, 72.7% NH₄⁺-N selectivity), while 2nd stage employs the Fenton reaction to efficiently degrade organic contaminants, resulting in 91.0% removal of COD. Critically, NO₃⁻-N addition promotes iron dissolution through direct redox reactions with mZVI, elevating effective iron electron utilization efficiency (EE_N increased by 8.2%−53.8%) and subsequently liberating more Fe(II) for 2nd stage. The process achieves > 79.8% COD removal for four actual wastewaters (including paint wastewater, primary explosive wastewater, tetrazene wastewater and pharmaceutical wastewater), with NO₃⁻-N conversion rates and NH₄⁺-N selectivity higher than 64.7%, demonstrating its application potential for treating a variety of refractory industrial wastewaters. Therefore, the corrosion-promoting effect of NO₃-N for mZVI was utilized to overcome the bottleneck of electron competition in the treatment of nitrogen-contained refractory organic wastewater by the mZVI/O₂ process, and meanwhile, the efficient degradation of organic contaminants and simultaneous recovery of nitrogen resources were realized by the mZVI/O₂-Fenton two-stage gradient oxidation process.

  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

  Yan Zhang: Writing – review & editing, Writing – original draft, Visualization, Investigation, Formal analysis, Data curation. Shuo Chen: Visualization, Formal analysis. Huihui Peng: Visualization, Formal analysis. Ankang Wang: Formal analysis. Yiming Sun: Formal analysis. Chenying Zhou: Visualization, Formal analysis. Shuang Meng: Visualization, Formal analysis. Chuanshu He: Supervision. Peng Zhou: Writing – review & editing, Writing – original draft, Visualization, Project administration, Investigation, Formal analysis, Data curation. Bo Lai: Writing – review & editing, Supervision, Project administration, Funding acquisition.

  Acknowledgments

  This research was supported financially by National Natural Science Foundation of China (No. U24A20561) and Sichuan Science and Technology Program (No. 2024NSFTD0014).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2025.112209.

  
  
• 1. [1]

     P. Jiang, T. Zhou, J. Bai, et al., Water Res. 235 (2023) 119914. doi: 10.1016/j.watres.2023.119914

  2. [2]

     X. Wang, J. Xin, M. Yuan, F. Zhao, Water Res. 183 (2020) 116060. doi: 10.1016/j.watres.2020.116060

  3. [3]

     M. Farghali, Z. Chen, A.I. Osman, et al., Environ. Chem. Lett. 22 (2024) 2699–2751. doi: 10.1007/s10311-024-01768-6

  4. [4]

     X. Huang, S. Liu, G. Liu, et al., Appl. Catal. B 323 (2023) 122134. doi: 10.1016/j.apcatb.2022.122134

  5. [5]

     Y. Yuan, Z. Zhou, X. Zhang, et al., Chin. Chem. Lett. 34 (2023) 107932. doi: 10.1016/j.cclet.2022.107932

  6. [6]

     Z. Jiang, L. Lv, W. Zhang, et al., Water Res. 45 (2011) 2191–2198. doi: 10.1016/j.watres.2011.01.005

  7. [7]

     K. Wei, Y. Wan, M. Liao, et al., Water Res. 218 (2022) 118453. doi: 10.1016/j.watres.2022.118453

  8. [8]

     J. Yu, X. Hou, X. Hu, et al., Appl. Catal. B Energy 256 (2019) 117876. doi: 10.1016/j.apcatb.2019.117876

  9. [9]

     L.J. Matheson, P.G. Tratnyek, Environ. Sci. Technol. 28 (1994) 2045–2053. doi: 10.1021/es00061a012

  10. [10]

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

  11. [11]

      X. Guan, Y. Sun, H. Qin, et al., Water Res. 75 (2015) 224–248. doi: 10.1016/j.watres.2015.02.034

  12. [12]

      J. Wu, J. Zhao, J. Hou, et al., Environ. Sci. Technol. 53 (2019) 8105–8114. doi: 10.1021/acs.est.8b06834

  13. [13]

      J. Xie, Y. Zheng, Q. Zhang, et al., Appl. Catal. B: Environ. Energy 320 (2023) 121935. doi: 10.1016/j.apcatb.2022.121935

  14. [14]

      Z. Xiong, B. Lai, Y. Yuan, et al., Chem. Eng. J. 302 (2016) 137–145. doi: 10.1016/j.cej.2016.05.052

  15. [15]

      Z. Xiong, B. Lai, P. Yang, et al., J. Hazard. Mater. 297 (2015) 261–268. doi: 10.1016/j.jhazmat.2015.05.006

  16. [16]

      Y. Wang, L. Li, X. Guo, et al., Water Res. 268 (2025) 122648. doi: 10.1016/j.watres.2024.122648

  17. [17]

      F. Fu, D.D. Dionysiou, H. Liu, J. Hazard. Mater. 267 (2014) 194–205. doi: 10.1016/j.jhazmat.2013.12.062

  18. [18]

      J.M. Rodriguez-Maroto, F. Garcia-Herruzo, A. Garcia-Rubio, et al., Chemosphere 74 (2009) 804–809. doi: 10.1016/j.chemosphere.2008.10.020

  19. [19]

      T. Suzuki, M. Moribe, Y. Oyama, M. Niinae, Chem. Eng. J. 183 (2012) 271–277. doi: 10.1016/j.cej.2011.12.074

  20. [20]

      Q. Lu, S. -W. Jeen, L. Gui, R.W. Gillham, Water Res. 112 (2017) 48–57. doi: 10.1016/j.watres.2017.01.031

  21. [21]

      H. Liu, Q. Wang, C. Wang, X. Li, Chem. Eng. J. 215-216 (2013) 90–95.

  22. [22]

      H. Qin, X. Guan, P.G. Tratnyek, Environ. Sci. Technol. 53 (2019) 9744–9754. doi: 10.1021/acs.est.9b02419

  23. [23]

      J. Wang, X. Li, S. Shen, et al., ACS Catal 13 (2023) 8783–8791. doi: 10.1021/acscatal.3c01855

  24. [24]

      Y.H. Huang, T.C. Zhang, P.J. Shea, S.D. Comfort, J. Environ. Eng. 140 (2014) 04014029. doi: 10.1061/(ASCE)EE.1943-7870.0000849

  25. [25]

      Y. Sun, J. Li, T. Huang, X. Guan, Water Res. 100 (2016) 277–295. doi: 10.1016/j.watres.2016.05.031

  26. [26]

      K. Ritter, M.S. Odziemkowski, R. Simpgraga, et al., J. Contam. Hydrol. 65 (2003) 121–136. doi: 10.1016/S0169-7722(02)00234-6

  27. [27]

      C. Qin, J. Zhang, C. Zhang, et al., Environ. Sci. Technol. 55 (2021) 6828–6837. doi: 10.1021/acs.est.1c00653

  28. [28]

      C. Le, J.H. Wu, S.B. Deng, et al., Water Sci. Technol. 63 (2011) 1485–1490. doi: 10.2166/wst.2011.392

  29. [29]

      J.A. Donadelli, L. Carlos, A. Arques, F.S. García Einschlag, Appl. Catal. B: Environ. Energy 231 (2018) 51–61. doi: 10.1016/j.apcatb.2018.02.057

  30. [30]

      M.P. Rayaroth, C.S. Lee, U.K. Aravind, et al., Chem. Eng. J. 315 (2017) 426–436. doi: 10.1016/j.cej.2017.01.031

  31. [31]

      Y. Xiang, Y. Liu, B. Cong, et al., Water Res. 267 (2024) 122494. doi: 10.1016/j.watres.2024.122494

  32. [32]

      C. Xue, Y. Peng, A. Chen, et al., J. Colloid Interface Sci. 582 (2021) 22–29. doi: 10.1016/j.jcis.2020.08.036

  33. [33]

      W. Yin, J. Wu, P. Li, et al., Chem. Eng. J. 184 (2012) 198–204. doi: 10.1016/j.cej.2012.01.030

  34. [34]

      P. Fan, L. Li, Y. Sun, et al., Water Res. 159 (2019) 375–384. doi: 10.1016/j.watres.2019.05.037

  35. [35]

      S. Pei, S. You, J. Ma, et al., Environ. Sci. Technol. 54 (2020) 13333–13343. doi: 10.1021/acs.est.0c05287

  36. [36]

      P. Zhou, Y. Yang, W. Ren, et al., Appl. Catal. B 319 (2022) 121916. doi: 10.1016/j.apcatb.2022.121916

  37. [37]

      Y. Sun, P. Zhou, P. Zhang, et al., Chem. Eng. J. 450 (2022) 138423. doi: 10.1016/j.cej.2022.13842305287

  38. [38]

      M.J. Alowitz, M.M. Scherer, Environ. Sci. Technol. 36 (2002) 299–306. doi: 10.1021/es011000h

  39. [39]

      Y. Zhang, G.B. Douglas, L. Pu, et al., Sci. Total Environ. 598 (2017) 1140–1150. doi: 10.1016/j.scitotenv.2017.04.071

  40. [40]

      Y.H. Huang, T.C. Zhang, Water Res. 38 (2004) 2631–2642. doi: 10.1016/j.watres.2004.03.015

  41. [41]

      H. Hu, Y. Bai, C. Zhou, et al., Environ. Sci. Technol. 58 (2024) 9804–9814. doi: 10.1021/acs.est.4c00195

  42. [42]

      Z. Yang, C. Shan, Y. Mei, et al., Chem. Eng. J. 334 (2018) 2255–2263. doi: 10.1016/j.cej.2017.12.005

  43. [43]

      L. Huang, L. Li, W. Dong, et al., Environ. Sci. Technol. 42 (2008) 8070–8075. doi: 10.1021/es8008216

  44. [44]

      Y. Gao, Y. Luo, Z. Pan, et al., Water Res. 249 (2024) 120967. doi: 10.1016/j.watres.2023.120967

  45. [45]

      Z. Wang, J. Jiang, S. Pang, et al., Environ. Sci. Technol. 52 (2018) 11276–11284. doi: 10.1021/acs.est.8b02266

  46. [46]

      P. Wang, H. Zhang, Z. Wu, et al., Chin. Chem. Lett. 34 (2023) 108722. doi: 10.1016/j.cclet.2023.108722

  47. [47]

      J. Kim, T. Zhang, W. Liu, et al., Environ. Sci. Technol. 53 (2019) 13312–13322. doi: 10.1021/acs.est.9b02991

  48. [48]

      Y. Tong, X. Wang, Y. Zhang, et al., Water Res. 272 (2025) 122917. doi: 10.1016/j.watres.2024.122917

  49. [49]

      X. Fan, Y. Zhou, G. Zhang, et al., Appl. Catal. B 244 (2019) 396–406. doi: 10.1016/j.apcatb.2018.11.061

  50. [50]

      Q. Guo, Z. Xu, W. Jin, Sep. Purif. Technol. 342 (2024) 127024. doi: 10.1016/j.seppur.2024.127024

  51. [51]

      D. Song, B. Qiao, X. Wang, et al., Environ. Sci. Technol. 57 (2023) 9416–9425. doi: 10.1021/acs.est.3c02025

  52. [52]

      M. Kurban, Y. Zhang, Y. Wang, et al., J. Environ. Chem. Eng. 12 (2024) 112630. doi: 10.1016/j.jece.2024.112630

  53. [53]

      Y. Yuan, B. Lai, P. Yang, Y. Zhou, J. Taiwan Inst. Chem. Eng. 65 (2016) 286–294. doi: 10.1016/j.jtice.2016.05.021

  54. [54]

      L. Hu, M. Rossetti, J.F. Bergua, et al., ACS Appl. Mater. Interfaces. 16 (2024) 30636–30647. doi: 10.1021/acsami.4c01744

  55. [55]

      E. Fulladosa, J.C. Murat, I. Villaescusa, Chemosphere 58 (2005) 551–557. doi: 10.1016/j.chemosphere.2004.08.007

• 

  Figure 1  Process flow diagrams of (a) mZVI/O₂ process and (b) two-stage gradient oxidation process (mZVI/O₂-Fenton).

  下载: 全尺寸图片 幻灯片
  

  Figure 2  Effect of (a) nitrogen species, (b) NO₃⁻-N concentration, (c) pH, (d) mZVI dosage, (e) different organic contaminants on the performance of the mZVI/O₂ process.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  Total dissolved iron concentration in the reaction solution. Effect of (a) nitrogen species, (b) pH, (c) different organic contaminants, (d) time series data for total dissolved iron, (e) NO₃⁻-N concentration, (f) iron electron utilization in different organic contaminants.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  SEM of mZVI before and after reaction at pH 3.5. (a) Pristine mZVI, (b) without nitrogen, (c) with NO₃⁻-N, (d) with NO₂⁻-N.

  下载: 全尺寸图片 幻灯片
  

  Figure 5  (a) H₂O₂ generation, (b) EPR analysis for radicals, (c) TBA inhibition on PE removal, (d) hydroxylated products pathway, (e) and (f) BA as chemical probe for analyzing hydroxylation products.

  下载: 全尺寸图片 幻灯片
  

  Figure 6  Conversion of nitrogen in the mZVI/O₂ process. (a) NO₃⁻-N transformations, (b) transformation of different nitrogen species, (c) effect of pH on NO₃⁻-N conversion in the mZVI/O₂ process, (d) oxidation of NH₄⁺-N by peroxymonosulfate (PMS) systems.

  下载: 全尺寸图片 幻灯片
  

  Figure 7  Inorganic nitrogen transformation pathways in the mZVI-based process.

  下载: 全尺寸图片 幻灯片
  

  Figure 8  (a) Iron electron utilization in different organic contaminants, (b) NO₃⁻-N conversion and NH₄⁺-N selection with different iron dosages, (c) and (e) organic contaminants removal by two-stage gradient oxidation process, (d) and (f) NO₃⁻-N transformation in a two-stage gradient oxidation process.

  下载: 全尺寸图片 幻灯片
•