English

• Antibiotics were generously applied for controlling infectious diseases and ignored their toxicity, which leading to the misuse of antibiotics and resulting in antibiotic accumulation and an abundance of multi-resistant bacteria in the water environment [1-3]. The freshwater was contaminated not only by human medicine but also by practically available antibiotics used by veterinarians. The variety of antibiotic contaminants and their concentration in the aquatic environment were steadily increasing, due to human and animal excretion with forms of antibiotic invariance and activity [4-6]. Metronidazole (MNZ) has been widely used as an antibiotic for the treatment or prevention of systemic or local infections caused by anaerobic bacteria [7,8]. However, MNZ has potential carcinogenicity, mutagenicity, and genotoxicity, it is difficult to degrade naturally in water environment owing to its characteristics of stable structure and strong antibacterial ability. Moreover, the extensive release of MNZ into the environment has led to the accumulation of toxicity in aquatic organisms and will have adverse effects on humans and ecology [9-11]. Thus, various physical and chemical processes and biological approaches have already been applied to remove antibiotics from wastewater, and there was a strong and growing desire to find more sustainable and efficient antibiotic removal methods.

  Advanced oxidation process (AOPs) has received considerable interest as one of the most popular green methods for the complete degradation of macromolecular contaminants in water by producing free radicals and destroying them into small molecules with low or no toxicity [12-14]. Activating hydrogen peroxide (H₂O₂) via homogeneous and heterogeneous Fenton processes to produce ^•OH radicals has been considered as one of the commonly available AOPs, providing an advantageous strategy for removing various hard-to-degrade pollutants [15-17]. However, the comparatively lower efficiency, limited acidic pH requirements, instability and the safety management of H₂O₂ are the drawbacks preventing their development. Consequently, the exploration of reliable oxidant alternatives to H₂O₂ was essential to realize cost-efficient water purification technologies. Persulfate (PS), including peroxymonosulfate (PMS) and peroxydisulfate (PDS), is considered to be an ideal alternative oxidant that can be rapidly activated by various methods to produce sulfate radicals (SO₄^•−) [18-22]. AOPs based on SO₄^•− are increasingly used, and SO₄^•− has a higher redox capacity (2.5–3.1 V) than ^•OH (1.8–2.7 V) [23,24]. Solid PDS is easier and safer to use, transport and store than liquid H₂O₂. However, PDS has been found to be slow to react to target contaminants at ambient temperatures. Transition metal oxides have been used as activators for the activation of PDS due to their strong activation capability and reduced energy consumption [25-27]. Photocatalysis as an AOPs technology is considered to be an opportune approach for wastewater treatment. It is worth noting that PDS can also be reactivated with photogenerated electrons produced by the photocatalyst. Accordingly, the development of Fenton-like photocatalysts for PDS activation to remove antibiotics under solar light is of great importance.

  Vanadium dioxide (VO₂) provides a unique 2D layered structure and is an excellent transition metal oxide material [28]. Various applications of VO₂ have been extensively investigated due to its special physical and chemical properties, such as catalysts, lithium-ion cells, solar cells, gas sensors, optoelectronic components, electrochromic devices and electro-optical switches [29-32]. In addition, VO₂ had a unique narrow band gap of 0.7 eV and are favorable candidates for light conversion applications [33]. The N-type narrow band gap semiconductor material VO₂ is widely used in the field of photocatalysis due to its diversified crystal forms, which can be transformed into each other at a certain temperature, the growth direction can be controlled by elemental doping, and the excellent light absorption performance [34]. In addition, recent studies have shown that ultrathin two-dimensional nanosheets with larger specific surface area, improved permeability and higher porosity have more obvious advantages in the process of ¹O₂ generation [35,36]. In comparison with other structures, nanosheets provide a shorter distance for the transfer of photogenerated charges, thus accelerating the charge transfer and the electron-hole separation [37-40]. Specifically, the high redox capacity allows VO₂ to be well adapted to PDS activation. However, the photocatalytic activity of VO₂ remains relatively low and has been modified by various methods to improve the photocatalytic performance of the system. Elemental doping of semiconductors has been found to be a valuable and effective approach to tune physicochemical properties such as structure, physical properties, composition and photocatalytic properties in various modification methods, leading to their increased suitability in the field of photocatalysis [41-43]. Furthermore, exploiting the interaction between semiconductor materials and Fe³⁺ is considered to be one of the effective methods to overcome the slow Fe³⁺/Fe²⁺ cycle kinetics of the Fenton reaction [44-46], Driven by visible light, the e⁻ provided by V⁴⁺ is accepted by Fe³⁺ to generate Fe²⁺, which activates PDS, thereby promoting the Fenton-like reaction cycle [47-49]. The loss of e⁻ converts V⁴⁺ to V⁵⁺, providing the possibility of metal self-circulation [50,51] and also providing sufficient active sites for the oxidation of antibiotics. In this case, the design of 2D ultrathin Fe-doped VO₂ nanobelts structures as photocatalysts for PDS activation under solar illumination for antibiotic removal is intriguing but challenging.

  Hence, a SO₄^•−-based Fenton-like photocatalysts 2D ultrathin Fe-doped VO₂ (Fe-VO₂) nanobelts were synthesized in this work by hydrothermal-calcination process. The synthesized Fe-VO₂ photocatalysts were thoroughly characterized in terms of their components and structures using various characterization methods. Meanwhile, the influence of different conditions on the performance of photocatalytic MNZ removal in the presence of PDS under solar light illumination, such as the amount of Fe-VO₂, the PDS concentrations, and the MNZ content, was also discussed. To demonstrate the photodegradation performance of Fe-VO₂ nanobelts on PDS activation in a real environment, reusability experiments and typical anions and cations (K⁺, Ca²⁺, NO₃⁻, HPO₄²⁻, SO₄²⁻, and Cl⁻), various antibiotics, different pH values and real water experiments were performed. Through the density function theory (DFT) analysis and various advanced experimental techniques, the feasible degradation pathways of MNZ were comprehensively investigated, and a plausible mechanism of Fe-VO₂ photocatalyst for MNZ removal via PDS activation was established.

  Details of the chemicals and materials, activity assessment, characterization and DFT calculation were presented in Text S1 (Supporting information).

  First, 1.82 g of V₂O₅ powder was added to a mixture of isopropanol (32 mL) and benzyl alcohol (4 mL), and heated to 328 K for 20 min to obtained the light-yellow oligomer. Then, the above oligomer was dispersed in deionized water, and FeCl₃·6H₂O was added and stirred vigorously for 10 min. The resultant mixture was transferred to a Teflon-lined autoclave and heated at 453 K for 48 h. After cooling to room temperature, the powders were separated by centrifugation, washed with deionized water and ethanol, and dried in an oven at 353 K for 30 min. Subsequently, the ground above powder was placed in a tube furnace and calcined at a heating rate of 276 K/min to 623 K for 2 h. After cooled naturally to room temperature, and ground to obtain a black powder of Fe-VO₂. The samples prepared with varying the addition amount of FeCl₃·6H₂O (1 wt%, 2 wt%, 3 wt%), and denoted as 1-Fe-VO₂, 2-Fe-VO₂, 3-Fe-VO₂, respectively. Meanwhile, the Fe contents were determined by ICP measurements, and the results were shown in Table S1 (Supporting information). Pure VO₂ was fabricated by a similar method without the addition of FeCl₃·6H₂O. The whole fabrication route was shown in Fig. 1a.

  Figure 1

  

  Figure 1.  The results of Fe-VO₂: (a) The hydrothermal synthesis process, (b) SEM image, (c) TEM image, (d) HRTEM image. Inset: the corresponding SAED pattern. (e) Mapping shows the different element with different color Fe (yellow), V (green), and O (purple). (f) AFM image and (g) height profile of the Fe-VO₂.

  DownLoad: Full-Size Img PowerPoint

  The scanning electron microscopy (SEM) and transmission electron microscopy (TEM) images of the Fe-VO₂ nanobelts were shown in Figs. 1b and c The uniformly smooth Fe-VO₂ nanobelts with lengths of a few µm, widths of 50–100 nm, and thicknesses of 1–2 nm were synthesized. The high-resolution transmission electron microscopy (HRTEM) image of labelled region shown a d-spacing of 0.361 nm, which is attributed to the (110) face of VO₂ [52]. The results were consistent with the corresponding selected area electron diffraction (SAED) image (inset of Fig. 1d), which has distinct diffraction spots, demonstrating its single-crystal character. The energy dispersive spectrometer (EDS) mapping showed a uniform distribution of Fe and V elements (Fig. 1e). The height of the samples was characterized by atomic force microscopy (AFM). As shown in Figs. 1f and g, the thickness of the Fe-VO₂ nanobelts were 0.82 nm and 1.40 nm, respectively.

  The X-Ray diffraction (XRD) patterns using different contents of Fe-doped VO₂ were shown separately in Fig. 2a. Clearly, all the diffraction peaks in Fe-VO₂ matched well with the standard diffraction pattern of VO₂ (JCPDS No. 81–2392) [53]. Meanwhile, the XRD pattern did not change significantly after Fe doping, and no characteristic peaks of Fe or iron oxide were found, which was presumably due to the low Fe content in the composites. FTIR spectra were used to determine the structural information, and the FTIR spectra of the prepared samples are shown in Fig 2b. There are spectral bands associated with V-O stretching vibrations between 450 and 1150 cm⁻¹ for VO₂ and different contents of Fe-doped VO₂ samples. The band located at about 923 cm⁻¹ is attributed to V=O and V-O-V stretching vibrations. In addition, the band appearing at 530 cm⁻¹ could be considered as in-plane and out-of-plane V-O-V bending vibrations [54].

  Figure 2

  

  Figure 2.  (a) XRD patterns and (b) FTIR spectra of the as-prepared samples, XPS spectra of (c) Fe 2p and (d) V 2p. Differential charge densities of the (e) VO₂ and (f) Fe-VO₂ using DFT calculations. The red and blue isosurfaces correspond to the increase in the number of electrons and the depletion zone, respectively. Red represents O, gray represents V, and purple represents Fe.

  DownLoad: Full-Size Img PowerPoint

  The valence state and electronic structure of the Fe-VO₂ nanobelts were confirmed by X-ray photoelectron spectroscopy (XPS) and DFT calculations. The chemical valence states of the samples were examined by XPS. Compared with pure VO₂, the XPS spectra of Fe-VO₂ contained V, Fe and O (Fig. S1a in Supporting information). As shown in Fig. S1b (Supporting information), the O 1s spectrum of pure VO₂ and Fe-VO₂ existed three peaks centered at 529.76, 531.09 and 533.00 eV. The binding energy of 529.76 eV was attributed to O²⁻ in the VO₂ lattice. The binding energy of 531.09 eV was suggested to be the presence of an oxygen absorption peak on the surface of VO₂, indicating the adsorption of free hydroxyl groups or H₂O [55]. The peak at 533.00 eV was ascribed to the oxygen vacancies created in the VO₂ nanobelts, which were shifted with the addition of Fe³⁺ by 1.91 and 1.68 eV, respectively, assumably resulting from the formation of Fe-O bonds. As shown in Fig. 2c, the Fe²⁺ and Fe³⁺ peaks of Fe-VO₂ were located at 711.4 and 714.28 eV, ascribed to the stronger oxidizing potential of Fe³⁺ that will oxidize V⁴⁺ during the hydrothermal reaction, leading to the appearance of Fe²⁺. Meanwhile, the corresponding satellite peaks of Fe³⁺ and Fe²⁺ appeared at 718.18 and 725.58 eV, respectively [56]. The increase of Fe³⁺ leads to the increase of Fe²⁺ conversion during the hydrothermal reaction. In V 2p spectrum (Fig. 2d), V 2p_3/2 and V 2p_1/2 appeared at binding energies of 516.58 and 523.68 eV, respectively, where 516.24 eV corresponded to V(Ⅳ) in VO₂ and the peak at 517.16 eV to V(Ⅴ) [57]. With the continuous addition of Fe ions, V(Ⅳ) and V(Ⅴ) are positively shifted by 0.17 and 0.07 eV, respectively, which can be attributed to the interaction between metals during the hydrothermal process. Some strong oxidizing Fe(Ⅲ) oxidizes V(Ⅳ) to V(Ⅴ) [58-60].

  The charge density distribution between the dispersed Fe and VO₂ nanobelts was further investigated by DFT calculations. As shown in Fig. 2e, the red isosurfaces around the V atoms indicated that the electrons were released from the V atom and transferred to the nearby O atom. As for Fe-VO₂ (Fig. 2f), the doped Fe atoms had mushroom-like charge density distributions. The Fe atoms dispersed in the VO₂ nanobelts carry positive charges, which was consistent with the XPS results in Fig. 2c. Supported by the VO₂ nanobelts, the electron delocalizes on the Fe atoms and moves toward the neighboring O atoms. The red isosurfaces around the Fe atoms (Fig. 2f) indicate the asymmetrically distributed residual partial charge densities on the isolated Fe atoms.

  The optical properties of as-fabricated photocatalysts were presented in Text S2 (Supporting information).

  The photocatalytic degradation performance of the as-constructed Fe-VO₂ materials for MNZ was investigated. Fig. 3a demonstrated that individual light, persulfate, or photocatalyst are not effective for the removal of MNZ, whereas when all three were used in combination, the removal rate of MNZ was significantly increased. The 2-Fe-VO₂/PDS/visible light system exhibited the best photodegradation efficiency of MNZ, where almost 98% degradation was achieved within 40 min, outperforming other photocatalysts in the recent reports (Table S2 in Supporting information). Excitation of VO₂ under light generated photogenerated e⁻, which promoted the conversion of neighboring Fe³⁺ to Fe²⁺. Subsequently, Fe²⁺ participated in PDS activation to generate SO₄^•− [61], which reduced the recombination rate of the photogenerated e⁻-h⁺ pairs, demonstrating the synergistic effect of photocatalytic technology and PDS activation. The excess Fe³⁺ would react with PDS to form S₂O₈^•−, which had lower oxidizing ability than SO₄^•−, leading to a reduction in photodegradation efficiency [62]. The mineralization efficiency of MNZ in the 2-Fe-VO₂/PDS/Vis system was 81% in the TOC test (Fig. S3 in Supporting information). In summary, the 2-Fe-VO₂/PDS/Vis system presented higher photocatalytic activity and mineralization ability compared to other systems.

  Figure 3

  

  Figure 3.  (a) Control experiments with different Fe-VO₂ on MNZ degradation. (b) Effects of PDS dosage, (c) effects of catalyst dosage, (d) effects of pH on MNZ removal over Fe-VO₂. (e) Effects of coexisting cation and anions on MNZ removal over Fe-VO₂ photocatalyst. (f) Degradation of MNZ in simulated natural water by Fe-VO₂. (g) The specific location longitude and latitude of the natural water, Charge density difference induces by PDS on VO₂ (h) and Fe-VO₂ (i). Electron excess and deficiency are represented as yellow and purple isosurfaces, respectively. [MNZ] = 20 mg/L, [PDS] = 0.5 g/L, [Catalyst] = 0.5 g/L, pH 6. Only one element was changed in the respective experiment, while the other parameters were held constant in accordance with the previous description.

  DownLoad: Full-Size Img PowerPoint

  The effect of various environmental factors on the photocatalytic activity of the Fe-VO₂/PDS/Vis system was investigated. Since it was necessary to consider cost control in practical applications, the effects of adding different doses of PDS and photocatalyst to the Fe-VO₂/PDS/Vis system on MNZ degradation were investigated. The dosage of PDS in the range of 0.2–0.5 g/L enhanced the photodegradation rate with increasing PDS concentration (Fig. 3b). However, the excessive amount of PDS caused SO₄^•− to react with S₂O₈²⁻ to form S₂O₈^•−, resulting in the decrease in the degradation rate. The oxidative capacity of the system increased continuously when the addition amount was increased from 0.2 g/L to 0.5 g/L (Fig. 3c), which could be attributed to the increase in the activation sites of the reaction due to the increased addition of photocatalyst. However, when the photocatalyst dosage was increased to 0.6 g/L, the degradation rate decreased significantly, due to the excess Fe³⁺ reacting with PDS to form S₂O₈^•−. Finally, the optimal PDS dosage and photocatalyst loading were 0.5 g/L and 0.5 g/L to obtain fast radical generation and high degradation efficiency.

  The initial concentration of pollutants in water is an important indicator of water quality, and the effect of MNZ concentration on the Fe-VO₂/PDS/Vis system was investigated. The higher the concentration of MNZ, the lower the degradation efficiency of MNZ by the Fe-VO₂/PDS/Vis system (Fig. S4 in Supporting information). The degradation rate of MNZ remained above 96% at initial concentrations of 10 mg/L and 20 mg/L, while the degradation efficiency of MNZ decreased slowly with increasing MNZ concentrations from 20 mg/L to 50 mg/L. The degradation rate in the Fe-VO₂/PDS/Vis system was limited by the reactive radicals with high concentrations of MNZ.

  The effect of initial pH on the degradation process of MNZ was explored as shown in Fig. 3d. According to the pKa value of MNZ (pK_a = 2.58), the molecule was almost in the protonated form at pH ≤ 2 and amphipathic at pH ≥ 4. The PDS activation was facilitated by the prolonged presence of Fe²⁺ in solution under acidic conditions. Therefore, the system exhibited excellent degradation efficiency in the pH range of 3–9 with MNZ in amphipathic ionic form with positive surface charge, promoting electrostatic interactions between MNZ and SO₄^•−. At pH 12, the highly alkaline conditions enabled the release of V ions from VO₂ into solution and inhibited the transfer process of e⁻ from Fe-VO₂ to PDS, resulting in a decrease in the degradation effect of MNZ.

  The effects of different anions and cations on the photodegradation process of MNZ by the Fe-VO₂/PDS/Vis system were shown in Fig. 3e. The experimental results demonstrated that metal ions, such as K⁺ and Ca²⁺, and anions, such as NO₃⁻, HPO₄²⁻, SO₄²⁻, and Cl⁻, had less influence on the photodegradation process of MNZ. For HCO₃⁻, the photodegradation process was mainly inhibited by increasing the solution pH and promoting the conversion of SO₄^•− to ^•OH. The HCO₃⁻could regulate the pH of the system, but excess HCO₃⁻ would react with SO₄^•− as well as ^•OH to form ^•HCO₃ [63], which the redox potential was lower than that of SO₄^•− and ^•OH, leading to a decrease in MNZ removal. In addition, the HCO₃⁻ would also react with Fe³⁺ in the solution, thus inhibiting the PDS activation process. After the addition of SCN⁻, the photodegradation efficiency of MNZ reduced to 23%, which was attributed to the reaction of SCN⁻ with Fe³⁺ to form Fe(SCN)₃ [64], resulting in the loss of active sites for persulfate activation and increased complexation of photogenerated e⁻-h⁺ pairs.

  In order to verify the practicality of the Fe-VO₂/PDS/Vis system, actual water samples were collected in two different areas and MNZ degradation was investigated (Fig. 3g). The effects of different water qualities, such as underground water, surface water, tap water, distilled water and purified water, on the photodegradation ability of the Fe-VO₂/PDS/Vis system were investigated. As shown in Fig. 3f, the photodegradation rates of MNZ in underground water, surface water, tap water, distilled water and purified water were 34%, 98%, 63%, 94% and 97%, respectively. The Fe-VO₂/PDS/Vis system was confirmed to be resistant to environmental perturbations.

  The reusability experiments of as-fabricated photocatalysts were presented in Text S3 (Supporting information.).

  In VO₂ (Fig. 3h), the V lost electron upon the adsorption of PDS, indicating that V was a decomposition site that allowed PDS to react with V atoms. Meanwhile, the DFT calculations in Fig. 3i demonstrated that the Fe lost electron upon the adsorption of PDS on the Fe-VO₂ surface, suggesting that Fe exists as an active site on the surface of the material during the reaction. Moreover, a pattern of lost electrons can be found on nearby V atoms, also suggesting that V can transfer internal electrons within the material to provide enough electrons for the cyclic reaction on Fe. Thus, the system achieved rapid and stable catalytic system reaction of the material through the joint action of the two active sites for PDS activation.

  In order to further investigate the degradation pathway of MNZ by Fe-VO₂/PDS/Vis system, MNZ intermediates were detected by HPLC-MS (Fig. S6 in Supporting information). Moreover, the frontier orbitals and Fuki Index of MNZ were computed by DFT to clarify the photodegradation pathway. The HOMO and LUMO orbits of MNZ were revealed by the light stability of their molecular structures (Figs. 4a and b), so it is difficult to self-degrade under natural conditions. The Fukui index describes radical attack (f ⁰) of the atoms on the MNZ molecule which was calculated by DFT (Table S3 in Supporting information). The atoms with larger f ⁰ were vulnerable to attack by reactive species, such as C (10), C (2), C (6), O (11), N (3), O (14), O (13), and C (1) marked in Table S3 (Supporting information). Therefore, the C—N bond and the imidazole ring can be broken easily by attack. Combined with computational outcomes, and the detected intermediates by HPLC-MS, three photodegradation pathways of MNZ was proposed (Fig. 4c). Pathway Ⅰ, MNZ was converted to m/z = 172.06 after structural rearrangement, which was further converted to m/z = 128.04 by attacking the C—N bond by reactive radicals, and the imidazole ring-opened to form acetylglycine with m/z = 117.06 during the subsequent reaction. Pathway Ⅱ, MNZ turned hydroxyl into carboxyl group generating m/z = 194.11 under the attack of h⁺, replacing hydroxyl and nitro groups on imidazole ring to form m/z = 140.06 as by-products, while the O₂^•− can attack imidazole ring during degradation to make m/z = 117.06 as ionic fragments. Pathway Ⅲ, MNZ denitrification generates m/z = 142.06, and the hydroxyethyl group was removed under the attack of reactive groups to form m/z = 101.07 [65,66]. MNZ molecules and intermediates may be oxidized to other small molecules and ultimately mineralized to H₂O and CO₂. These results agreed with the TOC analyses (Fig. S3 in Supporting information).

  Figure 4

  

  Figure 4.  (a) HOMO and (b) LUMO orbit images of MNZ. (c) The proposed pathways for MNZ photodegradation.

  DownLoad: Full-Size Img PowerPoint

  To confirm that the Fe-VO₂ activated sulfite to produce active free radicals, the selective reactive oxygen species (ROS) quenching tests and the ESR test were carried out. As shown in Fig. 5a, the furfuryl alcohol, t-BuOH, sodium oxalate, BQ and MeOH are used to capture ¹O₂, ^•OH, h⁺, O₂^•−, and SO₄^•−, respectively. After the addition of MeOH, the degradation of the MNZ was suppressed, and its degradation efficiency was reduced from 98% to 71%, indicating that SO₄^•− was one of the main active radicals. When sodium oxalate and BQ were added, the degradation of MNZ was inhibited, indicating that h⁺ and O₂^•− were also the main active radicals. The degradation rate of the MNZ also declined when the Furfuryl alcohol and t-BuOH were added, suggesting that ¹O₂ and ^•OH also partially contribute to the degradation of MNZ. As shown in Figs. 5b and c, SO₄^•−, ^•OH and ¹O₂ appeared in the Fe-VO₂ system, indicating the excellent oxidizing ability of Fe-VO₂/PDS under visible light. In addition, O₂^•− was detected under light conditions (Fig. 5d), suggesting that O₂^•− also played a role in the degradation of MNZ. The ESR results were in agreement with the selective ROS quenching tests.

  Figure 5

  

  Figure 5.  (a) Free radical quenching experiment on photocatalytic degradation of MNZ. ESR spectra of Fe-VO₂ sample for (b) DMPO− SO₄^•− and ^•OH, (c) TMPO−¹O₂, and (d) DMPO−O₂^•−.

  DownLoad: Full-Size Img PowerPoint

  Based on the above results, the possible degradation mechanism of the Fe-VO₂/PDS/Vis system was illustrated in Fig. 6, according to the Eqs. 1-12:

  (1)

  (2)

  (3)

  (4)

  (5)

  (6)

  (7)

  (8)

  (9)

  (10)

  (11)

  (12)

  Figure 6

  

  Figure 6.  Possible mechanism of Fe-VO₂/PDS/Vis degradation of MNZ.

  DownLoad: Full-Size Img PowerPoint

  VO₂ was excited by visible light to generate e⁻-h⁺ pairs, in which the jump of e⁻ to the conduction band induced the conversion of Fe³⁺ to Fe²⁺, and the Fe²⁺ in solution promoted the activation of PDS to produce SO₄^•−, constructing an external cycle of Fe ions. And the electron-loss of V⁴⁺ was reconverted to V⁴⁺ through the oxidation of MNZ by V⁵⁺, forming an inner cycle of V ions. When the conduction band potential of the semiconductor was lower than O₂/O₂^•− (−0.33 eV), then the reduction reaction converted the dissolved O₂ to O₂^•−, and the conduction band potential of VO₂ was −0.23 eV, thus generating O₂^•− radicals. Additionally, PDS, which acts as an electron acceptor, can also formed O₂^•−, ¹O₂, and ^•OH together through electron reactions contributing to the degradation process of MNZ.

  In summary, an innovative strategy was developed to synthesize atomically dispersed Fe-doped VO₂ nanobelt catalysts by a simple hydrothermal calcination method, which effectively and stably removed MNZ. The significant electronegativity difference between Fe and V drove the electron transfer from V to Fe within the lattice. The dual cycling between Fe atoms and VO₂ nanobelts allowed the more efficient charge carriers separation and transfer in photocatalysis, and rapid activation of PDS to generate free radicals such as SO₄^•− for the removal of various antibiotics. This work provided new ideas for single atoms anchored into the oxide lattice for sustainable and efficient practical water remediation.

  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

  Shukun Le: Conceptualization, Investigation, Writing – original draft. Peng Wang: Data curation, Formal analysis. Yuhao Liu: Methodology. Mutao Xu: Project administration. Quansheng Liu: Resources, Validation. Qijie Jin: Investigation, Visualization. Jie Miao: Investigation. Chengzhang Zhu: Methodology, Resources, Supervision. Haitao Xu: Resources, Writing – review & editing.

  Acknowledgments

  The authors gratefully acknowledge the financial supports from the National Key Research and Development Program of China (Nos. 2021YFB3500600, 2021YFB3500605, 2022YFB3504100), Key R&D Program of Jiangsu Province (No. BE2022142), Ministry of Education Chunhui plan international cooperation project (No. 202200554), National Natural Science Foundation of China (No. grant 22208170), Natural Science Foundation of Inner Mongolia (No. 2021BS02016), Jiangsu International Cooperation Project (No. BZ2021018), Nanjing Science and Technology Top Experts Gathering Plan, and Open Foundation of State Environmental Protection Key Laboratory of Soil Environmental Management and Pollution Control (No. SEMPC2023004), Cooperation Foundation for the Chunhui Plan Program of Ministry of Education of China (No. 202200554), Open Project Program of Key Laboratory of Optic-electric Sensing and Analytical Chemistry for Life Science (No. M2024-7), MOE.

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.110087.

  
  
• 1. [1]

     H.J. Wang, J. Zhang, P. Wang, Chin. Chem. Lett. 31 (2020) 2789–2794. doi: 10.1016/j.cclet.2020.07.043

  2. [2]

     S.Y. Yu, Z.H Xie, X. Wu, et al., Chin. Chem. Lett. 35 (2024) 108714. doi: 10.1016/j.cclet.2023.108714

  3. [3]

     X.D. Yang, J. Duan, X. Zhang, et al., Chin. Chem. Lett. 33 (2022) 3792–3796. doi: 10.1016/j.cclet.2021.11.031

  4. [4]

     P.M. Sathya, H. Mohan, J. Venkatachalam, K.K. Seralathan, Chemosphere 313 (2023) 137485. doi: 10.1016/j.chemosphere.2022.137485

  5. [5]

     F. Chen, Y.H. Zhang, H.W. Huang, Chin. Chem. Lett. 34 (2023) 107523. doi: 10.1016/j.cclet.2022.05.037

  6. [6]

     Z.Q. Zhang, J.L. Liang, W. Zhang, et al., Appl. Catal. B: Environ. 330 (2023) 122621. doi: 10.1016/j.apcatb.2023.122621

  7. [7]

     C. Wang, H.Y. Liu, G.F. Wang, et al., J. Alloy. Compd. 917 (2022) 165507. doi: 10.1016/j.jallcom.2022.165507

  8. [8]

     Z.X. Yang, Z.Z. Wang, J.T. Wang, Y. Li, G.K. Zhang, Environ. Sci. Technol. 56 (2022) 18008–18017. doi: 10.1021/acs.est.2c06571

  9. [9]

     M. Alsaidi, F.A. Azeez, L.A. Al-Hajji, A.A. Ismail, J. Environ. Manage. 314 (2022) 115047. doi: 10.1016/j.jenvman.2022.115047

  10. [10]

      C. Lykos, T. Kourkouta, I. Konstantinou, Sci. Total Environ. 870 (2023) 161877. doi: 10.1016/j.scitotenv.2023.161877

  11. [11]

      L. Äystö, N. Vieno, P. Fjäder, J. Mehtonen, T. Nystén, Ecotox. Environ. Safe. 262 (2023) 115149. doi: 10.1016/j.ecoenv.2023.115149

  12. [12]

      W.Y. Peng, Y. Fu, L.L. Wang, et al., Chin. Chem. Lett. 32 (2021) 2544–2550. doi: 10.1016/j.cclet.2021.01.028

  13. [13]

      K. Tian, L.M. Hu, L.M. Li, et al., Chin. Chem. Lett. 33 (2022) 4461–4477. doi: 10.1016/j.cclet.2021.12.042

  14. [14]

      S.K. Le, C.Z. Zhu, Y.W. Cao, et al., Appl. Catal. B: Environ. 303 (2022) 120903. doi: 10.1016/j.apcatb.2021.120903

  15. [15]

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

  16. [16]

      M.X. Ran, H. Xu, Y. Bao, et al., Angew. Chem. Int. Ed. 135 (2023) e202303728. doi: 10.1002/ange.202303728

  17. [17]

      A. López-Francés, F. Bernat-Quesada, M. Cabrero-Antonino, et al., Appl. Catal. B: Environ. 336 (2023) 122924. doi: 10.1016/j.apcatb.2023.122924

  18. [18]

      F. Ye, Y.M. Su, R.P. Li, et al., Appl. Catal. B: Environ. 337 (2023) 122992. doi: 10.1016/j.apcatb.2023.122992

  19. [19]

      H.Y. Zhu, M.T. Liu, G. Wang, et al., Rare Met. 43 (2024) 1686–1701. doi: 10.1007/s12598-023-02499-3

  20. [20]

      M.T. Liu, H.Y. Zhu, R.R. Du, et al., J. Environ. Chem. Eng. 11 (2023) 109939. doi: 10.1016/j.jece.2023.109939

  21. [21]

      S.Q. Yang, W.X. Zhang, M.T. Liu, et al., J. Environ. Chem. Eng. 11 (2023) 109190. doi: 10.1016/j.jece.2022.109190

  22. [22]

      R.R Du, H.Y. Zhu, H.Y. Zhao, et al., Environ. Res. 222 (2023) 115365. doi: 10.1016/j.envres.2023.115365

  23. [23]

      X. Li, S.W. Wang, P. Chen, et al., Appl. Catal. B: Environ. 325 (2023) 122401. doi: 10.1016/j.apcatb.2023.122401

  24. [24]

      G. Wang, K. Wang, Z.Y. Liu, et al., Appl. Catal. B: Environ. 325 (2023) 122359. doi: 10.1016/j.apcatb.2023.122359

  25. [25]

      Y.A. Li, J. Wei, N. Cui, et al., Adv. Funct. Mater. 33 (2023) 2301229. doi: 10.1002/adfm.202301229

  26. [26]

      M.P. Zhu, J.C.E. Yang, D.Q. Guan, et al., Appl. Catal. B: Environ. 329 (2023) 122570. doi: 10.1016/j.apcatb.2023.122570

  27. [27]

      S.S. Li, M.H. Zhou, H.Z. Wu, et al., Appl. Catal. B: Environ. 338 (2023) 123064. doi: 10.1016/j.apcatb.2023.123064

  28. [28]

      B. Ko, J.Y. Chae, T. Badloe, et al., ACS Appl. Mater. Interfaces 14 (2022) 1404–1412. doi: 10.1021/acsami.1c19223

  29. [29]

      M.X. Bai, J.F. Gao, Z.H. He, J.F. Hou, L.B. Kong, J. Electroanal. Chem. 907 (2022) 116039. doi: 10.1016/j.jelechem.2022.116039

  30. [30]

      B.B. Shuai, C. Zhou, Y.Q. Pi, X. Xu, ACS Appl. Energ. Mater. 5 (2022) 6139–6145. doi: 10.1021/acsaem.2c00540

  31. [31]

      Z.P. Zheng, Y. Luo, H. Yang, et al., Phys. Chem. Chem. Phys. 24 (2022) 8846–8853. doi: 10.1039/d2cp01070d

  32. [32]

      R. Shi, Y.H. Wu, Z.Q. Xin, et al., Angew. Chem. Int. Ed. 62 (2023) e202301421. doi: 10.1002/anie.202301421

  33. [33]

      M. Jourshabani, M.R. Asrami, B.K. Lee, Small 19 (2023) 2300147. doi: 10.1002/smll.202300147

  34. [34]

      X.Y. Li, Z.C. Zhuang, J. Chai, et al., Nano Lett. 23 (2023) 2905–2914. doi: 10.1021/acs.nanolett.3c00256

  35. [35]

      F. Yang, X.Y. Chu, J.H. Sun, et al., Chin. Chem. Lett. 31 (2020) 2784–2788. doi: 10.1016/j.cclet.2020.07.033

  36. [36]

      D.C. Yang, Y.C. Xu, K. Pan, et al., Chin. Chem. Lett. 33 (2022) 378–384. doi: 10.1016/j.cclet.2021.06.035

  37. [37]

      J. Wang, B. Gao, D.L. Liu, et al., Nanomaterials 13 (2023) 443. doi: 10.3390/nano13030443

  38. [38]

      Y. Chen, P.M. Wang, Z.T. Chen, B. Li, ACS Appl. Mater. Inter. 15 (2023) 46982–46994. doi: 10.1021/acsami.3c09985

  39. [39]

      C. Yang, S.S. Shang, Y. Fan, et al., Appl. Catal. B: Environ. 325 (2023) 122344. doi: 10.1016/j.apcatb.2022.122344

  40. [40]

      C.Z. Zhu, Q.Y. He, T.Y. Sun, et al., Chem. Eng. J. 464 (2023) 142704. doi: 10.1016/j.cej.2023.142704

  41. [41]

      J.Y. Qin, Y. Pei, Y. Zheng, D.Q. Ye, Y. Hu, Appl. Catal. B: Environ. 325 (2023) 122346. doi: 10.1016/j.apcatb.2022.122346

  42. [42]

      S.W. Du, S.Q. Lin, K.K. Ren, C.H. Li, F.X. Zhang, Appl. Catal. B: Environ. 328 (2023) 122503. doi: 10.1016/j.apcatb.2023.122503

  43. [43]

      W.Q. Li, S.Q. Xia, Z.Y. Wang, et al., Appl. Catal. B: Environ. 325 (2023) 122358. doi: 10.1016/j.apcatb.2023.122358

  44. [44]

      Y.X. Deng, M.Y. Xing, J.L. Zhang, Appl. Catal. B: Environ. 211 (2017) 157–166. doi: 10.1016/j.apcatb.2017.04.037

  45. [45]

      Y. Ji, Z.L. Song, Y. Xu, Y.L. Zhang, J. Alloy. Compd. 925 (2022) 166655. doi: 10.1016/j.jallcom.2022.166655

  46. [46]

      J.S. Hu, P.F. Zhang, W.J. An, et al., Appl. Catal. B: Environ. 245 (2019) 130–142. doi: 10.1016/j.apcatb.2018.12.029

  47. [47]

      C.V. Reddy, R.R. Kakarla, J. Shim, T.M. Aminabhavi, Chemosphere 308 (2022) 136414. doi: 10.1016/j.chemosphere.2022.136414

  48. [48]

      A. Mancuso, N. Morante, M. De Carluccio, et al., Chem. Eng. J. 450 (2022) 138107. doi: 10.1016/j.cej.2022.138107

  49. [49]

      A. Zhou, C. Zhao, J.C. Zhou, et al., J. Mater. Chem. A 11 (2023) 12950–12957. doi: 10.1039/d2ta09236k

  50. [50]

      C.H. Gao, J.J. Liu, M.L. Jia, G. Sa, A.J. Xu, Dye. Pigment. 212 (2023) 111115. doi: 10.1016/j.dyepig.2023.111115

  51. [51]

      D. Kuzyaka, C. Uzun, I. Yıldız, M. Kaya, B. Akata, Micropor. Mesopor. Mat. 323 (2021) 111256. doi: 10.1016/j.micromeso.2021.111256

  52. [52]

      X. Yang, J. Zou, J. Alloy. Compd. 940 (2023) 168868. doi: 10.1016/j.jallcom.2023.168868

  53. [53]

      W.L. Zhou, X.J. Li, X. Li, et al., Chem. Eng. J. 470 (2023) 144146. doi: 10.1016/j.cej.2023.144146

  54. [54]

      A. Shamloufard, S. Hajati, A.A. Youzbashi, et al., Appl. Surf. Sci. 590 (2022) 153118. doi: 10.1016/j.apsusc.2022.153118

  55. [55]

      B.L. Lin, Z.Y. Guo, J.H. Li, et al., Chem. Eng. J. 458 (2023) 141443. doi: 10.1016/j.cej.2023.141443

  56. [56]

      S. Ibraheem, G. Yasin, A. Kumar, et al., Appl. Catal. B: Environ. 304 (2022) 120987. doi: 10.1016/j.apcatb.2021.120987

  57. [57]

      A.K.S. Kumar, R.G. Compton, ACS Catal. 12 (2022) 4754–4764. doi: 10.1021/acscatal.2c00849

  58. [58]

      D. Kuzyaka, C. Uzun, I. Yıldız, M. Kaya, B. Akata, Micropor. Mesopor. Mat. 323 (2021) 111256. doi: 10.1016/j.micromeso.2021.111256

  59. [59]

      B. Janani, M.K. Okla, S.S. Al-Amri, et al., Chemosphere 296 (2022) 134012. doi: 10.1016/j.chemosphere.2022.134012

  60. [60]

      N. Kumar, S. Upadhyay, S. Chetana, et al., Mater. Lett. 335 (2023) 133810. doi: 10.1016/j.matlet.2022.133810

  61. [61]

      H.C. Yan, C. Lai, S.Y. Liu, et al., Water Res. 234 (2023) 119808. doi: 10.1016/j.watres.2023.119808

  62. [62]

      P. Huang, P. Zhang, C.P. Wang, J.C. Tang, H.W. Sun, Appl. Catal. B: Environ. 303 (2022) 120926. doi: 10.1016/j.apcatb.2021.120926

  63. [63]

      M.D. Jiang, J.H. Lu, Y.F. Ji, D.Y. Kong, Water Res. 116 (2017) 324–331. doi: 10.1016/j.watres.2017.03.043

  64. [64]

      Q.Y. Wu, Y. Zhang, H. Liu, et al., Water Res. 224 (2022) 119022. doi: 10.1016/j.watres.2022.119022

  65. [65]

      M.L. Tran, C.C. Fu, R.S. Juang, Environ. Sci. Pollut. Res. Int. 26 (2019) 11846–11855. doi: 10.1007/s11356-019-04683-4

  66. [66]

      M.L. Tran, C.C. Fu, R.S. Juang, Environ. Sci. Pollut. Res. Int. 25 (2018) 28285–28295. doi: 10.1007/s11356-018-2848-7

• 

  Figure 1  The results of Fe-VO₂: (a) The hydrothermal synthesis process, (b) SEM image, (c) TEM image, (d) HRTEM image. Inset: the corresponding SAED pattern. (e) Mapping shows the different element with different color Fe (yellow), V (green), and O (purple). (f) AFM image and (g) height profile of the Fe-VO₂.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  (a) XRD patterns and (b) FTIR spectra of the as-prepared samples, XPS spectra of (c) Fe 2p and (d) V 2p. Differential charge densities of the (e) VO₂ and (f) Fe-VO₂ using DFT calculations. The red and blue isosurfaces correspond to the increase in the number of electrons and the depletion zone, respectively. Red represents O, gray represents V, and purple represents Fe.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  (a) Control experiments with different Fe-VO₂ on MNZ degradation. (b) Effects of PDS dosage, (c) effects of catalyst dosage, (d) effects of pH on MNZ removal over Fe-VO₂. (e) Effects of coexisting cation and anions on MNZ removal over Fe-VO₂ photocatalyst. (f) Degradation of MNZ in simulated natural water by Fe-VO₂. (g) The specific location longitude and latitude of the natural water, Charge density difference induces by PDS on VO₂ (h) and Fe-VO₂ (i). Electron excess and deficiency are represented as yellow and purple isosurfaces, respectively. [MNZ] = 20 mg/L, [PDS] = 0.5 g/L, [Catalyst] = 0.5 g/L, pH 6. Only one element was changed in the respective experiment, while the other parameters were held constant in accordance with the previous description.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  (a) HOMO and (b) LUMO orbit images of MNZ. (c) The proposed pathways for MNZ photodegradation.

  下载: 全尺寸图片 幻灯片
  

  Figure 5  (a) Free radical quenching experiment on photocatalytic degradation of MNZ. ESR spectra of Fe-VO₂ sample for (b) DMPO− SO₄^•− and ^•OH, (c) TMPO−¹O₂, and (d) DMPO−O₂^•−.

  下载: 全尺寸图片 幻灯片
  

  Figure 6  Possible mechanism of Fe-VO₂/PDS/Vis degradation of MNZ.

  下载: 全尺寸图片 幻灯片
•