English

• The intensively discharged phenolic and nitrogen-contained heterocyclic organic contaminants within the wastewater from chemical, coking and pharmaceutical industries jeopardize the health of both natural lives and human beings. Advanced oxidation processes (AOPs) relying on the powerful generated reactive species have been developed as the promising remediation technologies to completely destruct the aqueous organic contaminants [1–3]. Among AOPs, catalytic ozonation processes have attracted broad research interests and been widely exploited in the practical applications [4–6]. Metal-based heterogenous catalysts, especially the transition metal-based ones, such as MnO₂, CoFe₂O₄ and MnO₂/rGO have been proved as highly reactive for O₃ activation [7–9]. The redox cycles originated from multi-valence states and the rich surface acidic sites on the transition metal-based catalysts facilitate the adsorption of O₃ molecules and their subsequent activation to produce reactive oxygen species (ROS) for destruction of contaminants [10, 11]. However, the secondary pollution induced by metal-leaching under the oxidative environment cannot be fully avoided when metal oxides were used.

  Metal-free carbocatalysts as the capable alternatives to the transitional metal oxides circumvent the metal leaching. Additionally, the adjustable structures of the carbocatalysts offer a versatile platform for tuning the reactivities [12, 13]. Compared with bulky carbons such as activated carbon and biochar-based carbon materials with micropore-dominant structure, nanocarbons with inherent mesopores and macropores expediate the diffusion of both O₃ and the organics and thus accelerate their interactions with the active sites [14, 15]. Recent years, metal-free nanocarbons with different dimensions have been synthesized and employed as the effective ozonation catalysts for destructing the aqueous organic contaminants [5]. It is revealed that the catalytic activities of the nanocarbons are closely related with their surface chemistry. Defective sites and the surface oxygen functionalities are the key active sites for interacting with O₃ to produce ROS [16, 17]. However, like the metal-based catalysts, the surface chemistry even the carbon skeleton of the nanocarbons can be altered by the powerful oxidative species during the catalytic ozonation reactions, resulting in the irreversible passivation.

  Doping metal-free heteroatoms within the nanocarbon structure can increase the reactivity and the stability [18, 19]. Both the charge and spin density of the carbon basal plane would be redistributed owing to the differences in electronegativity and the radius between the incorporated heteroatoms and the C atoms [20]. As the result, different oxidation regimes would be initiated depending on electron density distribution [21]. The synergistic effect arising from the co-doping different heteroatoms further regulates the charge distribution. Co-doping S atoms together with N atoms enlarges the electron-sparse regions of the C atoms adjacent to the doped N species, favoring the nucleophilic attack of the O₃[22]. Furthermore, the unbalanced electron distribution also serves as the electron-mediator to protect the surface of the catalyst from over-oxidizing by the oxidative species during the catalytic reactions.

  Herein, N and S co-doped hollow-structured carbon microspheres (NSCs) were prepared from the polymers generated during persulfate wet air oxidation (PWAO) of benzothiazole [23]. During this process, both N and S atoms were in-situ doped within the polymeric microspheres and their compositions were further regulated by the N₂ annealing at different temperatures (Scheme 1). Catalytic ozonation activities of the N₂-annealed NSCs were evaluated by utilizing benzimidazole (BMZ) as the target pollutants. The influencing factors such as solution pH and inorganic anions on catalytic performance were also investigated. The potential active sites and their correlations with the catalytic activities were analyzed by a quantitative structure-activity relationship (QSAR) analysis. In addition, a combined study of quenching and electron paramagnetic resonance (EPR) tests were performed to unveil the generated ROS and their possible roles in NSCs/O₃ system for benzimidazole degradation.

  Scheme 1

  

  Scheme 1.  Synthesis route for NSCs.

  DownLoad: Full-Size Img PowerPoint

  During PWAO of benzothiazole under high temperature (160 ℃) and high pressure (5 MPa), hollow polymeric microspheres (NSC-Poly) were formed along with degradation, with the average diameter around 1 µm and the shell thickness of 200 nm (Fig. 1A). Annealing the NSC-Poly in N₂ at elevated temperature (800–1000 ℃) maintained the hollow structure and the size of the microspheres (Figs. 1B and C). High-resolution transmittance electron microscope (HRTEM) observation in Fig. 1D further revealed that a greater graphitization degree was achieved by annealing in N₂ at 1000 ℃ (NSC-1000) by detecting some quasi-graphene lattices formed at the edge of the carbon shell.

  Figure 1

  

  Figure 1.  SEM images of NSC-Poly (A) and NSC-1000 (B); TEM (C) and the corresponding HRTEM observations (D) of NSC-1000.

  DownLoad: Full-Size Img PowerPoint

  The improvement of the graphitic degree after high-temperature N₂ annealing also reflected in the X-ray diffraction (XRD) patterns (Fig. 2A). Compared with NSC-Poly, NSC-800 annealing at 800 ℃ possessed a discernable characteristic (101) diffraction peak for graphitic carbon. The appearance of the characteristic C-N and C=S transmittance peaks in the Fourier transform infrared spectroscopy (FTIR) suggested that both N and S atoms were in-situ doped within the graphitic carbon lattice (Fig. 2B) [22]. X-ray photoelectron spectroscopy (XPS) analysis (Table S1 and Fig. S1 in Supporting information) and elemental analysis (Fig. 2C) were then performed to reveal the variations in the elemental composition of the NSCs after N₂-annealing. Compared with NSC-Poly, both XPS full surveys and the elemental analysis suggested that N₂ annealing significantly decreased the compositions of S (from 18 wt% to 5–8 wt%), O (from 16 wt% to ~ 4 wt%), and H (from 5 wt% to trivial) within the structure and enhanced the C composition. Additionally, elevating the annealing temperature further decreased both the N and S doping amounts. Interestingly, for the N₂ annealed NSCs, the slightly decreased amount of the dopants did not reflect in the reduction of the defective level as N₂ annealing temperature increased. As illustrated in Raman spectra in Fig. 2D, NSCs annealing at different temperatures obtained similar intensity ratios of defective band to graphitic band (I_D/I_G).

  Figure 2

  

  Figure 2.  XRD patterns (A), FTIR (B), element compositions (C) and Raman (D) spectra of the as-synthesized NSCs.

  DownLoad: Full-Size Img PowerPoint

  High resolution XPS surveys on C 1s, N 1s and S 2p of the NSCs were performed and deconvoluted to investigate the detailed alterations in the surface chemistry under different treatments. For high resolution XPS surveys on C 1s, less than 10% variations were observed for sp²/sp³ ratios of N₂ annealed NSCs (Table S2 and Fig. S2 in Supporting information). Considering the large inherent error in deconvolution process, it is suggested that insignificant variations in defective levels of NSCs upon annealing at different temperatures were observed, which were in line with the Raman results. Furthermore, trivial changes in surface oxygen functionalities were discerned for the annealed NSCs. Different from C 1s species, substantial variations occurred to the doped N species for NSCs samples (Fig. 3A and Table S3 in Supporting information). After N₂ annealing, the dominant doped N species in NSCs shifted from the pyrrolic N to graphitic N. Additionally, further increasing the annealing temperature resulted in the enhancement of the graphitic N content. It is suggested that the high temperature annealing in the inert atmosphere favored the transformation of pyrrolic N into the graphitic N [24]. For doped S species, two main peaks were observed in the high resolution XPS S 2p surveys (Fig. 3B and Table S4 in Supporting information). The peak situated at 168.2 eV could be assigned to C-SO_x-C (x = 2, 3, 4) groups [25], and the other peak can be deconvoluted into two characteristic peaks, corresponding to the S 2p_1/2 and S 2p_3/2 of the thiophene-S, respectively [26]. Compared with NSC-Poly, the N₂ annealed NSCs obtained much reduced C-SO_x-C contents, validating the significant decrease in the S-doping amounts after N₂ treatment. Meanwhile, increasing the N₂ annealing temperature enhanced the composition of S 2p_1/2.

  Figure 3

  

  Figure 3.  High resolution XPS spectra on N 1s (A) and S 2p (B) surveys and the corresponding deconvolution results.

  DownLoad: Full-Size Img PowerPoint

  Adsorption tests were performed prior to investigate the catalytic activities of the NSCs. NSCs displayed marginal adsorption capacities towards the BMZ at the ambient pH of 6.7 (Fig. S3 in Supporting information), though they obtained high specific surface areas (Table S5 and Fig. S4 in Supporting information). The charge difference between catalyst surface and the organics can significantly affect the adsorption capacity [5]. In accordance with the zeta potential measurements, the pH point of zero charges (pH_pzc) for NSC-Poly, NSC-800, NSC-900 and NSC-1000 were determined as 5.4, 4.6, 4.1, and 3.8, respectively. Decreasing the amount of surface oxygen functionalities by elevating the annealing temperature resulted in the abatement in pH_pzc, which agrees with our previous study [27]. With the ambient pH of 6.7, the surface of the NSCs was negatively charged. Therefore, the resulted electrostatic repulsion between the surface of the catalyst and the deprotonated BMZ molecules (pK_a of 5.5) hindered the adsorption process. BMZ removal by ozonation alone was also set as the benchmark (Fig. S5 in Supporting information). Ozonation alone resulted in the 80% BMZ within 60 min, giving rise to the pseudo-first order reaction rate of 0.027 min^-1 (Fig. 4A). Utilizing NSC-Poly as the catalyst slightly increased the BMZ degradation efficiency with reaction rate of 0.032 min^-1. N₂ annealing further improved the catalytic ozonation activities of NSCs. Moreover, the catalytic performance enhanced as the N₂ annealing temperature increased. For NSC-1000, complete BMZ degradation was achieved within 45 min, resulting in a rate constant of 0.06 min^-1. Compared with other nanocarbons, the special hollow structure within the microspheres facilitated the exposure of higher amount of the reactive sites by enhancing the specific surface area. Furthermore, it created a confined space which acted as the nano/micro reactors that accelerated the mass and charge transfer and boosted the reaction kinetics [28]. Catalytic ozonation activity of commercial-obtained single-walled carbon tubes (SWCNTs) for BMZ degradation was also evaluated and a pseudo-first order reaction rate constant of 0.049 min^-1 was resulted. Therefore, the as-synthesized NSC-1000 as the byproducts of the PWAO treatments outperformed the cost-intensive SWCNTs in BMZ degradation.

  Figure 4

  

  Figure 4.  (A) Pseudo-first order reaction rate constants plot for BMZ degradation by different treatments. (B) Reusability and regeneration tests for NSC-1000. (C) Linear regression fit between graphitic N contents and the pseudo-first order reaction rate constants for catalytic ozonation of BMZ with NSCs as catalysts. (D) Comparison of pseudo-first order reaction rate constants of quenching tests in catalytic ozonation by NSC-1000 utilizing BMZ as the target pollutants. EPR spectra with DMPO (E) and TEMP (F) as the spin trapping agents. Ozone flow rate: 100 mL/min; Ozone concentration: 25 mg/L; Catalyst loading: 0.1 g/L; Temperature: 25 ℃; [BMZ]₀: 50 mg/L; Concentration of the scavengers: [MeOH]: 10 mmol/L, [HCO₃⁻]=[DMSO]=[FFA]: 1 mmol/L; Initial solution pH was adjusted to 6.7 by diluted H2SO4 and NaOH.

  DownLoad: Full-Size Img PowerPoint

  Mineralization rates of BMZ were further monitored by time-dependent total organic carbon (TOC) abatements (Fig. S6 in Supporting information). NSC-Poly demonstrated a poor structural stability under the oxidative environment created by O₃ and the generated ROS, and an increased TOC was observed with the prolonged treatment time. N₂ annealing helped the construction of an outer graphitic shell which protected the attack of the oxidative species and thus increased the structural integrity of the catalysts. For NSC-1000, around 60% of the initial TOC was mineralized after 90 min. The robust structure of the N₂ annealed NSCs was also reflected by a high recyclability (Fig. 4B). A minor passivation was observed in the 3^rd usage when NSC-1000 as employed and the catalytic activity could be recovered by a facile low-temperature annealing process.

  Defective sites within the nanocarbons have been regarded as the potential active sites in catalytic reactions because they disturb the homogeneity of the carbon skeleton and tune the electronic status by the dangling bonds with unpaired electrons [5]. In this study, although N₂ annealing created higher number of defective sites on NSCs compared with NSC-Poly, the N₂-annealed samples at different temperatures obtained similar I_D/I_G ratios, suggesting the defective level might not be the predominant factor influencing the catalytic activities of the N₂-annealed NSCs. Incorporating N atoms within the carbon skeleton can drastically alter the charge and spin density distribution around the doped area because of the large electronegativity difference between N and C (3.04 vs. 2.55)[29]. The edge-located pyridinic N and the graphitic N doping zone were the potential active sites facilitating the O₃ adsorption and its subsequent dissociation into ROS [21, 27]. And the catalytic activities of the N-doped nanocarbons often demonstrated positive correlations with the relative contents of the pyridinic N and/or graphitic N. In this study, the surface N contents decreased with the increase of the annealing temperature. Additionally, deconvolution results of the high resolution XPS surveys on N 1s suggested that a higher amount of graphitic N was formed at the elevated temperature. By correlating the relative graphitic N contents with the pseudo-first order reaction rate constants, a good linearity was observed for the fitted line, suggested the key role of graphitic N (Fig. 4C). For S-doping, the relatively small difference in electronegativity between S and C (2.58 vs. 2.55) resulted in an insignificant charge redistribution; however, the S atom with a large radius could induce the deformation of the carbon skeleton when incorporated, resulting in the positively charged in the adjacent C atoms, especially for the thiophene-S [26]. The positively charged adjacent C atoms favored the nucleophilic attack of the peroxide-based oxidants and resulted in the nonradical pathway. The detailed ROS generation will be discussed in the following section. Therefore, N and S doping sites could initiate the synergistic effect that greatly accelerated the electron transfer process between the surface of the catalyst and the adsorbed O₃ molecules and promoted the catalytic activity.

  To investigate the possibly generated ROS and their contributions to BMZ degradation, a combined scavenging and in-situ electron paramagnetic resonance (EPR) tests were performed. Methanol (MeOH) and bicarbonate anion (HCO₃^-) were selected as the ^·OH scavengers to quench the ^·OH in both bulk solution and on the surface of the catalyst (k_OH^· = 9.7 × 10⁸ L mol^-1 s^-1 and 8.5 × 10⁶L mol^-1 s^-1, respectively)[30-33]. As displayed in Fig. 4D, marginal quenching effects were observed with the presence of these scavengers. Further increasing the MeOH dosage only resulted in a slight increase in inhibitory effect (Fig. S7 in Supporting information). These observations revealed the minor contribution of free and surface-confined ^·OH to BMZ degradation. After from ^·OH-based oxidation, some recent studies also reported the nonradical oxidations of the organics based on the surface-adsorbed activated species such as surface-absorbed O (*O_ad) and surface-activated O₃ complex [4]. Dimethyl sulfoxide (DMSO) possessing the high affinity towards the surface-adsorbed activated species but recalcitrant to O₃ attack (0.42 L mol^-1 s^-1) was added to evaluate their possible contributions to BMZ degradation [34]. A similar reaction rate to that of addition of MeOH was achieved when DMSO was involved within the catalytic oxidation, suggesting that the insignificant role of surface-adsorbed activated species in degrading BMZ. Singlet oxygen (¹O₂) with the moderate reduction potential (~ 0.85 V) obtained the selective oxidation behavior towards phenolic and unsaturated heterocyclic organics [35, 36]. Addition of furfuryl alcohol (FFA) which obtains a fast reaction kinetic with ¹O₂ (1 × 10⁸ L mol^-1 s^-1) resulted in a significant decrease in the reaction rate [37]. Noted that FFA is also an effective ^·OH scavenger. Considering the minor inhibitory effect brought by both the organic and inorganic ^·OH scavengers, the strong suppression effect with the addition of FFA mainly originated from the effect of ¹O₂.

  In-situ EPR experiments utilizing 5, 5-dimethyl-1-pyrroline (DMPO) and 4-hydroxyl-2, 2, 6, 6-tetramethyl-4-piperidine (TEMP) as spin trapping agents were then conducted to unravel the types of the generated ROS. Trace DMPO-^·OH adducts with the hyperfine splitting couplings a_N = a_H = 14.9 G were discerned for ozonation alone (Figs. 4E and Fig. S8 in Supporting information). The involvement of NSC-1000 as the catalyst enhanced the signal intensities of the DMPO-^·OH adducts, yet they were still significantly lower than those produced by α-MnO₂ which is a well-recognized catalyst initiating ^·OH-based oxidations in catalytic ozonation [38]. Furthermore, the much inferior catalytic ozonation activity of α-MnO₂ to that of NSC-1000 indicated that ^·OH was not the dominant ROS in NSC-1000/O₃ system for BMZ degradation (Fig. S9 in Supporting information). Superoxide radical (HO₂^·/O₂^·-) was also detected with the appearance of characteristic signal of DMPO-^·OOH adducts. However, the recalcitrant BMZ is unlikely to be degraded by HO₂^·/O₂^·- because of its weak oxidation capacity [39]. HO₂^·/O₂^·- is more possibly to act as the key radical chain reaction intermediate for generating and evolving into other ROS. 4-hydroxyl-2, 2, 6, 6-tetramethyl-4-piperidine (TEMP) prepared in MeOH solution was used as the spin trapping agent for ¹O₂ detection by eliminating the influence of ^·OH [40]. Compared with ozonation, catalytic ozonation with the presence of NSC-1000 drastically boosted the signal intensity of the triplet TEMPO signals (a_N = 16.9 G), suggesting a greater amount of ¹O₂ was produced (Fig. 4F). Additionally, the as-resulted TEMPO signal intensity for NSC-1000/O₃ system was higher than that in α-MnO₂/O₃ system. Combining with the results of the quenching tests, it was thus proposed that ¹O₂ was the dominant ROS for BMZ degradation.

  Previous efforts suggested that graphitic N would induce a nonradical oxidation process based on the electron transfer between the adsorbed O₃ and the organic pollutants [21, 41]. In this study, the co-doped S might further influence the charge and spin distribution of the carbon basal plane, creating a more electron-deficient zone on the adjacent C atoms due to the electronegativity and radius differences between the doped heteroatoms and the C atoms. Therefore, O₃ favored the nucleophilic attack on the electron-deficient zone and resulted in the formation of ¹O₂ [42]. On the other hand, other active sites with unpaired and lone-pair electrons such as the defective sites and the pyridinic N sites can also initiate the radical-based oxidation pathways, though at low yield.

  Apart from HCO₃^-, effects of the other inorganic anions which extensively exist in the real wastewater on catalytic performance were investigated (Fig. S10). The presence of Cl^- (10 mmol/L) as the ^·OH scavenger (k_OH^· = 4.3 × 10⁹ L mol^-1 s^-1) suppressed the BMZ degradation efficiency and the resulted rate constant was less than those with the addition of MeOH and HCO₃^-. It is suggested that Cl^- can also react with ¹O₂ and O₃ to form the secondary chlorine species such as Cl^· and Cl₂^·- with inferior oxidation abilities and thus reduced the degradation efficiency [43]. SO₄^2- has also been reported as the ^·OH quenching agent and the addition of which gave rise to similar decreased rate constant to that of HCO₃^-[43]. The presence of PO₄^3- and NO^3- slightly hindered the degradation efficiency. Apart from acting as the radical scavengers, these inorganic anions increased the ionic strength of the reaction solution and declined both solubility and mass transfer rate of O₃.

  Effect of solution pH on BMZ degradation efficiency was also investigated utilizing the most reactive NSC-1000 as the catalyst (Fig. S11 in Supporting information). A general trend can be observed that increasing the solution pH resulted in the increase of both ozonation and catalytic ozonation treatment efficiency. The rate constant ratio of catalytic ozonation to ozonation alone enhanced from 1.27 to 2.83 as the solution pH elevated. Meanwhile, greater TOC mineralization rates were achieved at alkaline solution pH (Fig. S11 in Supporting information). A "substrate-dependent" oxidation behavior of the generated ROS during the catalytic ozonation for degradation of the organics has been revealed by the previous studies [27, 41]. The produced ¹O₂ with moderate oxidation ability would selectively attack the phenolic and the heterocyclic organics with unsaturated structures, while the generated ^·OH favored the destruction of aliphatic acids and was responsible for the abatement of TOC. Therefore, the increased TOC mineralization rates at elevated pH ranges can be ascribed to the enhanced production of ^·OH originated from alkaline activation of O₃. Furthermore, the boosted generation of ^·OH might also induce the synergistic effect for the enhanced production of other ROS and thus improved the degradation efficiency.

  In summary, N and S co-doped hollow carbon microspheres were successfully synthesized via catalytic polymerization process of benzothiazole during the PWAO and the compositions of doped N and S were tuned by the N₂ annealing. Compared with the NSC-Poly, the N₂-annealed NSCs obtained improved catalytic ozonation activities for BMZ degradation and greater structural stabilities. The synergistic effect between the graphitic N and the thiophene-S redistributed the charge density of the carbon basal plane and induced the ¹O₂-dominant oxidation pathway.

  Declaration of competing interest

  The authors declare no competing financial interest.

  Acknowledgments

  The authors greatly appreciate the financial supports from the National Natural Science Foundation of China (No. 21978324), Tianjin Municipal Science and Technology Bureau (No. 18YFYSZ00170), Beijing Natural Science Foundation (No. 8192039), and Science Foundation of China University of Petroleum, Beijing (No. 2462020YXZZ034).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2021.07.055.

  
  
• 1. [1]

     B.C. Hodges, E.L. Cates, J.H. Kim, Nat. Nanotechnol. 13(2018) 642-650. doi: 10.1038/s41565-018-0216-x

  2. [2]

     Y. Shang, X. Xu, B. Gao, S. Wang, X. Duan, Chem. Soc. Rev. 50(2021) 5281-5322. doi: 10.1039/D0CS01032D

  3. [3]

     M. Du, Q. Yi, J. Ji, et al., Chin. Chem. Lett. 31(2020) 2803-2808. doi: 10.1016/j.cclet.2020.08.002

  4. [4]

     G. Yu, Y. Wang, H. Cao, H. Zhao, Y. Xie, Environ. Sci. Technol. 54(2020) 5931-5946. doi: 10.1021/acs.est.0c00575

  5. [5]

     Y. Wang, X. Duan, Y. Xie, H. Sun, S. Wang, ACS Catal. 10(2020) 13383-13414. doi: 10.1021/acscatal.0c04232

  6. [6]

     J. Wang, H. Chen, Sci. Total Environ. 704(2020) 135249. doi: 10.1016/j.scitotenv.2019.135249

  7. [7]

     Y. Wang, Y. Xie, H. Sun, et al., J. Hazard. Mater. 301(2016) 56-64. doi: 10.1016/j.jhazmat.2015.08.031

  8. [8]

     F. Nawaz, H. Cao, Y. Xie, et al., Chemosphere 168(2017) 1457-1466. doi: 10.1016/j.chemosphere.2016.11.138

  9. [9]

     C. Cai, X. Duan, X. Xie, et al., J. Hazard. Mater. 410(2021) 124604. doi: 10.1016/j.jhazmat.2020.124604

  10. [10]

      J. Ma, C. Wang, H. He, Appl. Catal. B: Environ. 201(2017) 503-510. doi: 10.1016/j.apcatb.2016.08.050

  11. [11]

      Y. Wang, Z. Chi, C. Chen, et al., Appl. Catal. B: Environ. 272(2020) 118972. doi: 10.1016/j.apcatb.2020.118972

  12. [12]

      L. Zhang, Y. Shi, Y. Wang, N.R. Shiju, Adv. Sci. 7(2020) 1902126. doi: 10.1002/advs.201902126

  13. [13]

      H. Qiu, P. Guo, L. Yuan, G. Sheng, Chin. Chem. Lett. 31(2020) 2614-2618. doi: 10.1016/j.cclet.2020.08.014

  14. [14]

      Z. Liu, J. Ma, Y. Cui, B. Zhang, Appl. Catal. B: Environ. 92(2009) 301-306. doi: 10.1016/j.apcatb.2009.08.007

  15. [15]

      J. Peng, Y. He, C. Zhou, S. Su, B. Lai, Chin. Chem. Lett. 32(2021) 1626-1636. doi: 10.1016/j.cclet.2020.10.026

  16. [16]

      J. Wang, S. Chen, X. Quan, H. Yu, Chemosphere 190(2018) 135-143. doi: 10.1016/j.chemosphere.2017.09.119

  17. [17]

      Z. Song, M. Wang, Z. Wang, et al., Environ. Sci. Technol. 53(2019) 5337-5348. doi: 10.1021/acs.est.9b01361

  18. [18]

      Y. Shang, X. Xu, Q. Yue, B. Gao, Y. Li, Environ. Sci. Nano 7(2020) 1444-1453. doi: 10.1039/D0EN00050G

  19. [19]

      Z. Zhou, L. Miao, H. Duan, et al., Gan, Chin. Chem. Lett. 31(2020) 1226-1230. doi: 10.1016/j.cclet.2020.02.026

  20. [20]

      Z. Song, Y. Zhang, C. Liu, et al., Chem. Eng. J. 357(2019) 655-666. doi: 10.1016/j.cej.2018.09.182

  21. [21]

      Z. Sun, L. Zhao, C. Liu, Y. Zhen, J. Ma, Environ. Sci. Technol. 53(2019) 10342-10351. doi: 10.1021/acs.est.9b02745

  22. [22]

      X. Duan, K. O'Donnell, H. Sun, Y. Wang, S. Wang, Small 11(2015) 3036-3044. doi: 10.1002/smll.201403715

  23. [23]

      L. Zhou, Y. Xie, H. Cao, et al., Chem. Eng. J. 370(2019) 208-217. doi: 10.1016/j.cej.2019.03.201

  24. [24]

      W. Ren, G. Nie, P. Zhou, et al., Environ. Sci. Technol. 54(2020) 6438-6447. doi: 10.1021/acs.est.0c01161

  25. [25]

      H. Li, Y. Su, W. Sun, Y. Wang, Adv. Funct. Mater. 26(2016) 8345-8353. doi: 10.1002/adfm.201601631

  26. [26]

      Y. Qiu, J. Huo, F. Jia, B.H. Shanks, W. Li, J. Mater. Chem. A 4(2016) 83-95. doi: 10.1039/C5TA06039G

  27. [27]

      Y. Wang, N. Ren, J. Xi, et al., ACS ES&T Eng. 1(2021) 32-45.

  28. [28]

      J. Qian, X. Gao, B. Pan, Environ. Sci. Technol. 54(2020) 8509-8526. doi: 10.1021/acs.est.0c01065

  29. [29]

      L. Zhang, C. Lin, D. Zhang, et al., Adv. Mater. 31(2019) 1805252. doi: 10.1002/adma.201805252

  30. [30]

      Y. Sun, J.J. Pignatello, Environ. Sci. Technol. 29(1995) 2065-2072. doi: 10.1021/es00008a028

  31. [31]

      T.M. El-Morsi, W.R. Budakowski, A.S. Abd-El-Aziz, K.J. Friesen, Environ. Sci. Technol. 34(2000) 1018-1022. doi: 10.1021/es9907360

  32. [32]

      J. Ma, N.J.D. Graham, Water Res. 34(2000) 3822-3828. doi: 10.1016/S0043-1354(00)00130-5

  33. [33]

      J.L. Acero, U. v. Gunten, Ozone Sci. Eng. 22(2000) 305-328. doi: 10.1080/01919510008547213

  34. [34]

      J.J. Wu, M. Muruganandham, S.H. Chen, J. Hazard. Mater. 149(2007) 218-225. doi: 10.1016/j.jhazmat.2007.03.071

  35. [35]

      A.A. Frimer, Chem. Rev. 79(1979) 359-387. doi: 10.1021/cr60321a001

  36. [36]

      W.H. Koppenol, D.M. Stanbury, P.L. Bounds, Free Radic. Biol. Med. 49(2010) 317-322. doi: 10.1016/j.freeradbiomed.2010.04.011

  37. [37]

      F.E. Scully, J. Hoigné, Chemosphere 16(1987) 681-694. doi: 10.1016/0045-6535(87)90004-X

  38. [38]

      F. Nawaz, Y. Xie, H. Cao, et al., Catal. Today 258(2015) 595-601. doi: 10.1016/j.cattod.2015.03.044

  39. [39]

      Y. Nosaka, A.Y. Nosaka, Chem. Rev. 117(2017) 11302-11336. doi: 10.1021/acs.chemrev.7b00161

  40. [40]

      P. Bilski, K. Reszka, M. Bilska, C.F. Chignell, J. Am. Chem. Soc. 118(1996) 1330-1338. doi: 10.1021/ja952140s

  41. [41]

      Y. Wang, L. Chen, C. Chen, et al., Appl. Catal. B: Environ. 254(2019) 283-291. doi: 10.1016/j.apcatb.2019.05.008

  42. [42]

      X. Chen, W. -D. Oh, Z. Hu, et al., Appl. Catal. B: Environ. 225(2018) 243-257. doi: 10.1016/j.apcatb.2017.11.071

  43. [43]

      G.V. Buxton, C.L. Greenstock, W.P. Helman, A.B. Ross, J. Phys. Chem. Ref. Data 17(1988) 513-886. doi: 10.1063/1.555805

• 

  Scheme 1  Synthesis route for NSCs.

  下载: 全尺寸图片 幻灯片
  

  Figure 1  SEM images of NSC-Poly (A) and NSC-1000 (B); TEM (C) and the corresponding HRTEM observations (D) of NSC-1000.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  XRD patterns (A), FTIR (B), element compositions (C) and Raman (D) spectra of the as-synthesized NSCs.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  High resolution XPS spectra on N 1s (A) and S 2p (B) surveys and the corresponding deconvolution results.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  (A) Pseudo-first order reaction rate constants plot for BMZ degradation by different treatments. (B) Reusability and regeneration tests for NSC-1000. (C) Linear regression fit between graphitic N contents and the pseudo-first order reaction rate constants for catalytic ozonation of BMZ with NSCs as catalysts. (D) Comparison of pseudo-first order reaction rate constants of quenching tests in catalytic ozonation by NSC-1000 utilizing BMZ as the target pollutants. EPR spectra with DMPO (E) and TEMP (F) as the spin trapping agents. Ozone flow rate: 100 mL/min; Ozone concentration: 25 mg/L; Catalyst loading: 0.1 g/L; Temperature: 25 ℃; [BMZ]₀: 50 mg/L; Concentration of the scavengers: [MeOH]: 10 mmol/L, [HCO₃⁻]=[DMSO]=[FFA]: 1 mmol/L; Initial solution pH was adjusted to 6.7 by diluted H2SO4 and NaOH.

  下载: 全尺寸图片 幻灯片
•