Ti₃AlC₂ (MAX) powder (purity >99.9%, 400 mesh) was purchased from Foshan Xinxi Technology Co. Graphite felt was purchased from Hebei Ruitong Carbon Co. Titanium plates were purchased from Baoji Jucheng Titanium Company, China. SDZ, Ferrous chloride (FeCl₂), Hydrogen fluoride (HF), C₂H₅OH, Hydrochloric acid (HCl), Sodium hydroxide (NaOH), Sodium sulfate (Na₂SO₄), Nafion, Dimethylsulfoxide (DMSO), Tertiary butyl alcohol (TBA), and 1,4-Benzoquinone (BQ) were purchased from Shanghai Macklin Chemical Co. Deionized water was used in all experiments and all chemical reagents used in the experiments were analytically pure.

MXene was prepared by HF etching. 1 g of Ti₃AlC₂ was added to 20 mL of HF (40 wt%) and stirred at room temperature for 24 h. The collected powder was washed with deionized water to pH = 7 and dried at 60°C. The powder was then placed in DMSO and sonicated in a nitrogen atmosphere for 24 h. The black powder was centrifuged and vacuum dried. It was then stirred in DMSO for 24 h and sonicated under nitrogen atmosphere for 24 h. The resulting black powder was centrifuged, washed and vacuum dried to obtain pristine MXene.

50 mg MXene powder was mixed with 25, 50, 75, 100 mL of ammonia solution (30 wt%) and a certain amount of NaBH₄, and then ultrasonicated at room temperature for 3 h in a bath ultrasonic apparatus with a power of 300 W (Table S1). The material was washed with deionized water until pH >6, and dried in a vacuum oven for 8 h. The final product was named N-MXene-x (x represents the mass ratio of MXene to ammonia, x = 1, 2, 3 and 4) (Fig. S1). To prepare for further use, N-MXene-x was dispersed in water (2.95 mL), followed by the addition of a solution containing Nafion at a concentration of 5 wt% (5 μL). This suspension was then uniformly dripped onto both sides of a graphite felt measuring dimensions of approximately 4 cm×2 cm×0.5 cm and subsequently dried in an oven set to maintain a temperature of 60°C.

The electrochemical investigations of SDZ were conducted in a 200 ml reactor (Fig. S2). The anode and cathode were composed of titanium plates and graphite plates, respectively, drop-coated with a catalyst. The distance between them was maintained at 1 cm. A 0.1 M Na₂SO₄ solution was utilized as the electrolyte, and either 0.1 M H₂SO₄ or NaOH was added to adjust the pH during the reaction process. The EF reaction is facilitated by the introduction of Fe²⁺ through the addition of FeSO₄ particles. All experiments for the electrocatalytic process were carried out under a constant DC current at a specific air flow rate to provide oxygen for the production of H₂O₂ continuously. The SDZ removal efficiency was investigated at initial pH (2.0–9.0), current density (2.0–8.0 mA/cm²), electrocatalyst dosage (2.0–8.0 mg/cm²), initial SDZ concentration (10.0–25.0 mg/L), Fe²⁺ concentration (0–1.0 mM).

The crystal structure of the prepared samples was investigated using an X-Pert PRO MPD fixed target X-ray diffractometer (XRD; PANalytical, The Netherlands), Conditions of use are Cu-Kα radiation (40 kV, 40 mA) with 2θ ranged from 10° to 90°. The morphology and fine structure of the samples were examined using a transmission electron microscope (TEM, JEM-1400 flash, Japan), SU-8020 thermal field emission scanning electron microscope (SEM; Hitachi, Tokyo), and energy dispersive X-ray spectrometer (EDS). The chemical properties of the samples were analyzed using a Fourier infrared spectrometer (FT-IR; Thermo Nicolet, USA). XPS spectra were recorded with Thermo Scientific ESCALAB 250Xi using Al-Kα radiation (1486.69 eV). The specific surface areas were examined at 77 K with N₂ gas as adsorbate using a Quantachrome QuadraWin Instruments version 5.12. ESR spectra of solution were obtained using JES-FA200 ESR spectrometer. Microwave power employed was 1mW to 10 W; sweep width ranged from 223 mT to 423 mT. Modulation frequency was 100 kHz.

SDZ and degradation products were analyzed by reversed-phase high-performance liquid chromatography (RP-HPLC, Shimadzu LC-20 A, Japan) equipped with a UV-visible detector (SPD-20A) and a C18 column (4.6×250 mm) with λ = 269 nm, using 25% acetonitrile and 75% water as the mobile phase, at a flow rate of 1.0 mL/min. The accumulation of H₂O₂ was measured using the potassium titanium oxalate spectrophotometric method. ^•OH concentration was determined using fluorescence photometry. Determination of Fe²⁺ and Fe³⁺ concentrations using spectrophotometry. The concentrations of hydroxyl radicals, Fe²⁺ and Fe³⁺ were determined using a UV-visible spectrophotometer (UV 2600; 106 Shimadzu, Japan). The electrochemical properties were investigated using an electrochemical workstation with a three-electrode system (CHE 760e, Shanghai Chenhua Instrument Co., Ltd.), with glassy carbon electrode (GCE) or modified GCE as the working electrode, AgCl as the reference electrode, and platinum wire as the counter electrode. The electrochemical characterization of MXene and N-MXene-3 was investigated using cyclic voltammetry (CV) in solutions containing Fe(CN₆)^3-/4-. In this case, CV was measured over the potential range of −0.2 V to 0.6 V with a scan rate of 0.1 V/s.

The morphology of the prepared MXene and N-MXene-x was observed by SEM. As shown in Fig. S3, Ti₃AlC₂ was etched by HF, the lamellar structure was preserved and the obtained MXene was accordion-shaped. After ultrasonication, the morphology of the prepared N-MXene-1, N-MXene-2 and N-MXene-4 still retained the laminar structure with almost no structural collapse and the surface became rough due to N doping [25].

The morphology of N-MXene-3 was observed using high-resolution SEM and TEM, as depicted in Fig. 1. Following sonication in ammonia, the prepared N-MXene-3 maintained its morphology without any structural collapse (Fig. 1a). TEM images revealed a multilayer structure consistent with the SEM results, while HRTEM images show that N-MXene-3 has a lattice-stripe multilayer structure with a spacing of 0.21 nm (the lattice-stripe spacing of MXene is usually 0.8–1.0 nm), demonstrating successful doping with N elements (Fig. 1b and 1c) [26]. The elemental mapping diagrams further confirmed the uniform distribution of C, N, O, and Ti elements within the N-MXene-3 nanomaterials (Fig. 1d).

The structures of MXene and N-MXene-x nanomaterials were initially analyzed and investigated using XRD, as depicted in Fig. S4a. The diffraction pattern reveals the presence of four distinct peaks at 8.8°, 37.4°, 43°, and 60.8°, which can be attributed to the crystal planes (002), (0010), (0012), and (110) of MXene [27]. Subsequent sonication in ammonia solution resulted in a noticeable shifting the 002 peak to lower 2q value for N-MXene-x, indicating an increased interlayer distance resulting from successful doping of N atoms during the reaction [28].

The chemical composition and bonding information of the samples were determined using FTIR spectroscopy and the corresponding results are shown in Fig. S4b. The prominent and robust absorption peak at 3448 cm⁻¹ was attributed to the stretching vibration of the hydroxyl (-OH) terminals, while the peak observed at 1635 cm⁻¹ signifies the presence of carboxylic acid (-COOH) groups. Additionally, the peak at 1395 cm⁻¹ can be attributed to the presence of C-N bonds [29–31]. These FTIR findings further confirm successful N atom doping, which aligns well with XRD results.

XPS was employed to analyze the elemental composition and surface state of N-MXene-x. Fig. S5 shows four different peaks at 285.5 eV, 401 eV 454.8 eV, and 530.6 eV, corresponding to C 1s, N 1s, Ti 2p, and O 1s respectively [32]. The Ti 2p spectra are illustrated in Fig. 2a, where the three peaks fitted at 455.9, 459, and 461.7 eV can be attributed to the states of Ti(II) 2p3/2, Ti(III) 2p1/2, and Ti(IV) 2p1/2 respectively [3]. The C1s spectrum of N-MXene-3 is depicted in Fig. 2b, illustrating the characteristic peaks at 281.7 eV, 284.7 eV, 286.5 eV, and 288.9 eV corresponding to Ti-C, C-C, C-O, and C-F bonds respectively [33]. The high-resolution energy spectrum of the N-MXene-3 sample O1s is depicted in Fig. 2c, wherein the peaks observed at 531.9 eV and 529.7 eV correspond to -OH (groups) and Ti-O bonds, respectively [34]. The high-resolution N1s spectrum of N-MXene-3 is depicted in Fig. 2d, revealing distinct peaks at specific binding energies. Notably, the peak centered at 400.1 eV corresponds to pyridine-N, while the peak at 400.9 eV represents pyrrole-N. Additionally, the presence of a peak at 401.9 eV indicates graphite-N bonding, and another peak observed at 402.6 eV signifies the formation of O-N-C bonds [35]. Remarkably, a discernible peak appears at 399.2 eV corresponding to Ti-N bond formation as well, suggesting that nitrogen atoms not only interact with carbon atoms but also form chemical bonds with titanium [36].

In addition, the effect of different doping amounts on the N component of the catalyst was explored by analyzing the N 1s spectrum in N-MXene-x, and the results are shown in Fig. S6. It can be observed that the peaks of N 1s in the four catalysts are basically not shifted. The proportions of different components of N in the four catalysts were obtained by integrating the different peak areas, and the results are shown in Table S2. After the enhancement of N doping, the percentage of pyrrolic nitrogen in the catalyst first increases and then decreases, reaching a maximum of 47.54% in N-MXene-3. The studies reported so far have shown that pyridine N is easily protonated in acidic solutions, which in turn selectively catalyzed the 2e⁻-ORR reaction, but 4e⁻-ORR is more favored under alkaline conditions [37]. Furthermore, the vast majority of studies have shown that the graphite N site corresponds to a weak adsorption of oxygen molecules and its ORR activity is poor [38]. Whereas pyrrole N as a carbon defective site is likely to 2e⁻-ORR catalytically active site [39].

The electrochemical properties of catalysts are typically influenced by their specific surface area and pore size distribution. Analysis of the N₂ adsorption-desorption isotherms revealed that both MXene and N-MXene-3 exhibited type IV isotherms with distinct hysteresis loops, indicating their mesoporous nature [40] (Fig. S7a). The N-MXene-3 exhibits a higher surface area and a larger distribution of pore volume compared to MXene (Fig. S7b). The nitrogenating behavior of NH₄⁺ within the intercalation between the N-MXene surface layers is further substantiated, thereby resulting in the delamination of multilayer films [41].

The electrochemical characterization of glassy carbon electrode (GCE) modified with prepared materials was conducted using CV and Tafel in 5 mM Fe(CN)₆^3-/4- and 0.1 M KCl solutions. As depicted in Fig. S8a, the oxygen reduction peak of N-MXene-x modified GCE is significantly higher than that of MXene modified GCE, and N-MXene-3 has the highest oxygen reduction peak. This suggests that N doping is favorable to enhance the electron transfer of redox probes [3]. To investigate the electron transfer rate, the free corrosion potential of the catalysts was measured using the Tafel polarization curve. In Fig. S8b, N-MXene-3 has the lowest free corrosion potential, indicating a higher electron transfer rate and better catalytic activity.

The ORR kinetics and electron transfer number of N-MXene-3 were probed using RDE experiments, the RDE experiments were measured in the range of rotational speeds from 400 rpm to 2025 rpm and the results are shown in Fig. S9a. The diffusion current density increases with increasing rotational speed due to the decrease in the diffusion distance at high rotational speed conditions. And the number of electron transfer in the experiment was calculated using the Koutecky-Levich (K-L) equation [42], which is shown in Eq. (1):

where J is the measured current density (mA/cm²), J_k is the kinetic current density (mA/cm²), ω is the number of revolutions (rpm) of the rotating disc electrode, and B is the Levich slope [43].

The Levich slope can be calculated from Eq. (2):

where n is the number of electrons transferred, F is the Faraday constant (96485 C/mol), CO₂ is the concentration of oxygen in the system, DO₂ is the oxygen diffusion coefficient (1.9*10⁻⁵ cm²/s), and ν is the kinetic viscosity of the electrolyte (0.01 cm²/s).

The results of the computational fitting are shown in Fig. S9b, where it can be seen that the K-L curve shows a good linear relationship, which indicates that the ORR exhibits first order reaction kinetics. The values of electron transfer number (n) at potentials of −0.90 V, −0.95 V, −1.00 V, and −1.05 V were 1.58, 1.61, 1.63, and 1.64, respectively, indicating that O₂ was mainly reduced to H₂O₂ at the N-MXene-3 electrode.

To optimize the degradation of SDZ by N-MXene-x nanocomposites, the following parameters were varied one at a time: the type of N-MXene-x, the applied current density, the catalyst loading, the Fe²⁺ concentration, the initial concentration of SDZ and the initial pH of the solution. Unless otherwise specified, the parameters were consistently optimized using a catalyst loading of 4.0 mg/cm², current density of 6.0 mA/cm², Fe²⁺ concentration of 0.50 mM, and an initial pH value of 3.0.

In the EF system, the amount of N doping affects the electron transfer rate and the number of active sites exposed in the catalyst. The degradation efficiency of SDZ by the EF system was investigated by varying the mass ratio of MXene/N (MXene/N= 2:1, 1:1, 1:1.5, and 1:2), and the results are shown in Fig. 3a. When the mass ratio of MXene/N was 2:1, 1:1, and 1:1.5, the removal efficiency of SDZ increased to 49%, 65%, and 97% within 60 min, respectively. However, when the mass ratio of N/MXene was increased to 1:2, the removal rate of SDZ did not increase further, and the degradation curve roughly coincided with that of N/MXene at 1:1.5 and the removal rate was slightly lower. This may be due to the fact that the N doping was saturated and the saturated N atoms would stack in the MXene layer and affect the number of active sites of the catalyst, resulting in a decrease in the removal rate [44]. In summary, the optimal N/MXene mass ratio should be 1:1.5, and therefore the N-MXene-3 catalyst was selected for the subsequent experiments.

In the EF system, solution pH plays a crucial role in determining the state of existence of Fe²⁺ and H₂O₂. The degradation performance of SDZ was investigated using N-MXene-3 as a cathodic catalyst over the pH range of 3.0 to 9.0 (Fig. 3b). The highest removal efficiency of SDZ, reaching 97%, was achieved at pH 3.0 within a duration of 60 min. However, the observed decrease in SDZ removal at pH values as low as 2.0 can be attributed to the generation of H₃O₂⁺ from H₂O₂ in the acidic environment. This protonated form hinders the generation of ^•OH radicals with ferrous ions [45]. In an alkaline solution, the precipitation reaction generates a greater number of insoluble iron hydroxide complexes, resulting in a reduced availability of ferrous ions for activating H₂O₂ generation. Consequently, this leads to a decrease in ^•OH and subsequently hampers the removal efficiency of SDZ [46]. The optimal pH for the EF reaction in this study was determined to be 3.0.

The degradation efficiency of the EF reaction system is significantly influenced by the current density, as it plays a pivotal role in regulating the production of H₂O₂. As the current density increased from 2.0 to 6.0 mA/cm², a gradual increase in the degradation rate of SDZ was observed (Fig. 3c). However, further augmentation of the applied current density to 8.0 mA/cm² did not yield a substantial enhancement in the degradation rate of SDZ. This is because when the current density is less than 6.0 mA/cm², the continuous increase in current density can promote the generation of H₂O₂ by oxygen adsorption reduction near the cathode, which in turn accelerates the generation of ^•OH. When the current density is continuously increased to 8.0 mA/cm², the system has to provide a higher voltage to maintain the higher current density, which makes the hydrogen precipitation reaction and side reactions to be accelerated and inhibits the main reaction. Moreover, excess H₂O₂ would react with ^•OH and affect the mineralization of organic pollutants [47].

In the EF reaction, Fe²⁺ plays a dual role as both an initiator and scavenger of ^•OH, thus emphasizing the criticality of controlling its concentration in the EF process. The degradation rate of SDZ after reaction was observed to be 29%, 52%, 97%, 98%, and 95% at initial Fe²⁺ concentrations of 0, 0.25, 0.50, 0.75, and 1.00 mM respectively (Fig. 3d). This is because at the beginning of the reaction, as the Fe²⁺ concentration increases, the yield of ^•OH increases, which contributes to SDZ degradation. However, excess Fe²⁺ can depletes ^•OH, thus preventing the mineralization of pollutants [48]. In addition, excess Fe²⁺ may prevent the production of H₂O₂ as well as adsorption of active sites, thus in turn limiting ^•OH formation [49]. Therefore, the most suitable initial concentration of Fe²⁺ in this study was 0.50 mM.

Based on the initial concentration of SDZ, the range of pollutants that can be effectively treated at a given time can be determined. Therefore, the effect of the initial concentration of SDZ on the degradation efficiency of the EF system was investigated at SDZ concentrations of 10~25 mg/L, as shown in Fig. 3e. When the concentration of SDZ was increased from 10 mg/L to 25 mg/L, the degradation rate of SDZ decreased from 99.2% to 76.5%. The decrease in the degradation effect can be attributed to two factors: firstly, the higher concentration of SDZ leads to increased generation and consumption of intermediate products; secondly, a greater amount of SDZ and its intermediate products may adsorb onto the cathode surface, thereby impeding the reaction between active sites and pollutants.

In the EF system, catalyst loading plays a crucial role in ^•OH formation. The effect of four different loadings (2.0, 4.0, 6.0, and 8.0 mg/cm²) on the degradation of SDZ in the EF reaction was investigated as shown in Fig. 3f. When the catalyst loading was increased from 2.0 to 6.0 mg/cm², the degradation rate of SDZ also increased from 35.6% to 97%. And when the catalyst loading was further increased to 8.0 mg/cm², the SDZ degradation rate was not significantly promoted, indicating that the loading had reached saturation at this time. Therefore, the most suitable catalyst loading in this study was 6.0 mg/cm².

To gain a deeper understanding of the reaction mechanism of MXene and N-MXene-3 as cathodic catalysts for the degradation of SDZ, the accumulation of H₂O₂ in the MXene and N-MXene-3 electrodes was measured using iodination in the absence of the addition of SDZ (the other reaction conditions were kept consistent with the degradation of SDZ) (Fig. 4a). The H₂O₂ accumulation of both MXene and N-MXene-3 electrodes exhibited a gradual increase over 60 min. Notably, the H₂O₂ accumulation of N-MXene-3 demonstrated a time-dependent decrease, which can be attributed to its enhanced catalytic ability in activating H₂O₂. To investigate this phenomenon, the reaction was carried out under nitrogen atmosphere (to prevent O₂ from entering the reaction system), and then 50 mg/L H₂O₂ was added to the system. Under non-electrolytic conditions, the activation ability of N-MXene-3 for H₂O₂ was significantly stronger than that of MXene, and meanwhile, under electrolytic conditions, the catalytic activity of N-MXene-3 was further enhanced. The above experiments show that the catalysts prepared by N doping have excellent activation ability for H₂O₂ [50].

Since there is no direct correlation between H₂O₂ production and pollutant degradation, we evaluated the ability of different electrodes to activate H₂O₂, and the results are shown in Fig. 4b. The amount of ^•OH produced by the N-MXene-3 electrode was 0.23 mM, which was significantly higher than the amount of ^•OH produced by the MXene electrode (0.035 mM). This may be due to the fact that although the amount of H₂O₂ accumulated on the N-MXene-3 electrode was small, the total amount produced was quite large, and the active substances on the surface of the catalyst continuously converted H₂O₂ into ^•OH [51].

The catalytic activity of the cathode is significantly influenced by the reduction capacity of trivalent iron ions (Fig. S10). As the reaction progresses, there is a decrease in the total amount of Fe in the system, and N-MXene-3 exhibits lower levels of Fe³⁺ compared to MXene system. This phenomenon may arise due to the promotion of Fe³⁺ reduction by N doping, thereby facilitating and sustaining a more robust degradation process.

To elucidate the degradation mechanism of SDZ by reactive radicals in the N-MXene-3/EF system, free radical quenching experiments were conducted in the experimental setup by introducing a substantial amount of TBA (100 mM) and BQ (4 mM) (Fig. 4c). TBA acts as a potent quencher that effectively suppresses the formation of ^•OH within the reaction system, while BQ serves as a capturing agent for ^•O₂⁻ generation during the course of the reaction [52]. The degradation of SDZ was reduced to 79% and 57% upon the addition of excess BQ and TBA, respectively. Therefore, the ^•OH significantly contributes to the degradation of SDZ in the N-MXene-3/EF system. The generation of ^•OH in the system was qualitatively analyzed using ESR spectroscopy with DMPO employed as a spin-trapping agent. The DMPO-^•OH adduct, as depicted in Fig. 4d, exhibits a characteristic quadruple peak with an intensity ratio of 1:2:2:1, providing conclusive evidence for the presence of ^•OH within the system [53]. However, no DMPO-^•O₂⁻ adducts were detected in the system, which may be due to the fact that ^•O₂⁻ is extremely unstable and difficult to be captured in the reaction.

Based on these results, the catalytic mechanism of N-Mxene-3 as a cathodic electrocatalyst in the EF process was proposed, as shown in Fig. S11. Firstly, the oxygen molecules adsorbed on the electrode surface undergo an in-situ reduction process, leading to the formation of H₂O₂ through a two-electron pathway (Eq. (4)). The in situ generated H₂O₂ subsequently undergoes a reaction with externally applied Fe²⁺ to yield ^•OH (Eq. (5)). The presence of Fe³⁺ metal species on the cathode ensures the perpetuation of the catalytic reaction through a continuous redox cycle via reduction (Eq. (6)). The regeneration of Fe³⁺ through electron acquisition from the cathode significantly mitigates H₂O₂ consumption, thereby facilitating enhanced utilization of H₂O₂. During the EF process, in addition to ^•OH, a small amount of ^•O₂⁻ is generated through direct electron acquisition by oxygen molecules at the electrode surface (Eq. (7)). Finally, the generated free radicals undergo a series of reactions with the SDZ molecules, resulting in their gradual degradation and complete mineralization into CO₂, H₂O, and inorganic ions (Eqs. (8) and (9)).

The reusability of the N-MXene-3 catalyst was evaluated by conducting five consecutive cycles of SDZ degradation under identical experimental conditions utilizing the N-MXene-3 cathode. After each operation, the cathode electrode was sequentially rinsed with deionized water and ethanol. The anodes were subjected to high-temperature boiling with oxalic acid to eliminate the oxide film formed on their surfaces, followed by drying of the cathode anodes in an oven for subsequent test cycle preparation. After five consecutive cycles of reaction, the removal efficiency of SDZ and the mineralization efficiency of TOC remained high at 74% and 30.1%, respectively. Although a gradual decay trend of the N-MXene-3 cathode material was observed, this result still proves that the N-MXene-3 catalyst possesses good reusability. Moreover, it further indicates that the loss of active sites on the catalyst surface during degradation is not significant. (Fig. S12). The present study is compared with other EF systems degrading SDZ in Table 1, demonstrating the promising potential of N-MXene-3/EF for practical applications.

The practicality of the catalyst is a key factor in assessing its application potential. In order to further investigate the performance of the N-MXene-3/EF system in practical applications, two typical real water samples were selected for testing, including lake water (LW) and medical wastewater (MW). The water quality parameters of the typical water samples are shown in Table S3. The collected lake water itself does not contain SDZ, and the medical wastewater contains low levels of SDZ. Therefore, the real water matrix containing 20 mg/L SDZ was prepared by dissolving appropriate amounts of SDZ in the collected real water. Using ultrapure water as the control sample group, the degradation rate of SDZ in lake water and medical wastewater reached 94. 79% and 71%, respectively, within 60 min, and the corresponding reaction rate constants were 0.025 and 0.021 min⁻¹, which were lower than those of ultrapure water (Fig. S13). This may be due to the fact that the organic content of lake water and medical wastewater is much higher than that of ultrapure water in the aqueous matrix, which would consume more active radicals.

Frontier orbital theory is widely employed for the prediction of reaction mechanisms and determination of pathways for electron transfer [57]. The SDZ structure was first geometrically optimized through the B3LYP basis group and solvation model in the DMo1³ module, as shown in Fig. 5a. The highest occupied orbital (HOMO) of the SDZ molecule suggests enhanced electron escape ability and susceptibility to electrophilic attack by ^•OH; conversely, the lowest occupied orbital (LUMO) of the SDZ molecule indicates reduced electron escape ability and resistance to electrophilic attack by ^•OH [58]. From Fig. 5c and 5d, it can be observed that HOMO and LUMO are mainly located at the S atoms attached to the benzene ring, the N atoms on the phenyl ammonia, and the N atoms attached to the heterocyclic roots, which are usually the reactive regions where electrons are received and given. Based on the energies of HOMO and LUMO of the SDZ molecule, the ionization energy, electron affinity, energy gap, hardness, chemical potential and electrophilic index of SDZ can be calculated as shown in Table S4. It can be seen that the chemical potential of SDZ is −3.882 eV, indicating that SDZ is stable under normal conditions and does not undergo self-decomposition reactions. The electrostatic potential distribution on the van der Waals surface of the SDZ molecule is depicted in Fig. 5b. The atoms exhibiting negative electrostatic potential (ESP) values indicate regions of high electron density, while the atoms displaying positive ESPs signify regions with low electron density. ^•OH is a highly reactive electrophilic radical with a redox potential ranging from 1.8–2.7 eV and an electrophilic index of 8.55 eV, enabling it to readily engage in electrophilic attacks on organic compounds. The Fukui index was calculated to elucidate the reactivity of individual atoms in the SDZ molecule, thereby providing further insights into the active site and degradation pathway of SDZ (Fig. 5e). The results revealed C3, C5, N17, and N10 as its electrophilic reactive sites.

Meanwhile, the intermediates during SDZ degradation were examined by LC-MS, where the corresponding liquid phase mass spectra are shown in Fig. S14, where the structural formulae, molecular formulae and mass/charge ratios of SDZ and its intermediates are listed in Table S5. The feasible pathways of SDZ conversion in the EF system are illustrated in Fig. 6. Four potential routes for SDZ removal were identified, including (I) C-S bond cleavage; (II) hydroxylation reaction of heterocyclic roots; (III) heterocyclic root ring opening reaction; and (IV) amino oxidation. Subsequently, the aromatic intermediates undergo benzene ring cleavage, leading to the formation of propionic and acetic acids, which are ultimately mineralized into CO₂.

The developmental toxicity and bioaccumulation factors of SDZ and its degradation intermediates were assessed in Fig. S15. Significantly reduced developmental toxicity was observed for the majority of degradation intermediates generated by the N-MXene-3/EF system during SDZ degradation. In terms of bioaccumulation factors, all degradation intermediates, except P6, exhibited lower bioaccumulation factors compared to SDZ [56]. The findings suggest that the environmental risk associated with the end products of SDZ is comparatively lower in the N-MXene-3/EF system.