Visible-light-induced peroxymonosulfate activation over ZnFe 2 O 4 fine nanoparticles for ofloxacin degradation

Widespread overuse and misuse of antibiotics has led to unintended consequences, and it is necessary to find effective ways to remove antibiotics. In this study, a visible-light-response photocatalyst zinc ferrite (ZnFe 2 O 4 ) was synthesized via a hydrothermal method. Meanwhile, the X-ray diffraction, Brunauer– Emmett–Teller, scanning electron microscope, X-ray photoelectron spectroscopy, and Fourier transform infrared spectra analysis were applied to characterize the structure, morphology, and physicochemical properties of the ZnFe 2 O 4 . The results indicated that the ZnFe 2 O 4 was circular granular morphology with a particle size of approximately 3 0 –5 0 nm and the noticeable intergranular agglomeration. The specific surface area, pore volume, and pore diameter of the ZnFe 2 O 4 were determined to be 126.8655 m 2 /g, 0 .2 0 46 cm 3 /g, and 64.519 0 Å, respectively, representing that the ZnFe 2 O 4 had a large specific surface area. Moreover, the enhancement of degradation efficiency of ofloxacin (OFL) by peroxymonosulfate (PMS) under the visible light (Vis) was systematically evaluated.The results exhibited that the ZnFe 2 O 4 achieved the relatively optimum catalytic activity with 8 0 .9% of OFL degradation efficiency in 3 0 min at pH 6. 0 under the PMS concentration of 1 00 mg/L and the corresponding pseudo-first-order kinetic constant of OFL degradation was 0 . 0 438 min –1 . In addition, the effects of ZnFe 2 O 4 dosage, PMS concentration, initial OFL concentration, solution pH, and water matrix on the OFL degradation were comprehensively investigated in the Vis/PMS/ZnFe 2 O 4 process. Furthermore,the ZnFe 2 O 4 exhibited excellent stability and reusability for OFL degradation. The Vis/PMS/ZnFe 2 O 4 process would be a reliable alternative for the degradation of OFL-like antibiotics to solve the increasingly serious problem of antibiotic pollution.


Introduction
Antibiotics are often detected in surface water, groundwater, and various sediments due to their widespread usage all over the world (Li et al., 2020a). The presence of antibiotics with low degradability and concentration in the environment, particularly the entire urban water cycle and food chain, poses a serious public health problem (Christou et al., 2017;Liu et al., 2019;Girijan et al., 2020). In addition, contaminants in the soils are partly mobile and easily get into the air or water, becoming a secondary pollutant via natural/uncontrollable process typical of leaching and weathering . Ofloxacin (OFL) is a typical fluoroquinolone antibiotic administered to both humans and animals, and after administration, approximately 78% of OFL is excreted (Tong et al., 2011). As the biodegradation of OFL is difficult, sewage treatment plants (STPs) have a low removal rate, and the OFL concentrations in the STP effluents of Beijing, Hangzhou, and Vancouver have been determined to be between 0.6 and 1405 ng/L (Xiao et al., 2008;Tong et al., 2011).
Generally, the advanced oxidation processes (AOPs), such as the Fenton or Fenton-like reaction, ozonation or catalytic ozonation, photocatalytic oxidation, electrochemical oxidation, and ionizing radiation, have been widely used for antibiotics degradation in recent years (Anjali and Shanthakumar, 2019;Liu et al., 2020b;Wang and Zhuan, 2020). Moreover, peroxymonosulfate (PMS), the latest in situ chemical oxidation oxidant, has been receiving widespread attentions due to its numerous advantages including safety, stability, and relatively high redox potential of the corresponding intermediate sulfate radicals (SO 4 -) with a long lifetime and strong oxidizing ability (Hu and Long, 2016). Moreover, PMS can be readily activated by heat, ultraviolet (UV) light, alkaline materials, ultrasound, and appropriate transition metal ions (Fe 2þ , Co 2þ , and Mn 2þ ; Zhu et al., 2016).
The AOPs involving the production of SO 4by the activation of persulfate have been widely exploited for the degradation of organic pollutants in water (Wacławek et al., 2017). Under normal conditions, the degradation effect of persulfate on organic matter is insignificant; however, under certain conditions after activation, more SO 4with strong oxidizing activity can be produced. Compared with the hydroxyl radical (OH), the SO 4with a higher standard reduction potential (2.5-3.1 V) would be more selective toward target pollutants and require relatively moderate reaction conditions and exhibit a stronger mineralization ability (Yin et al., 2018;Yao et al., 2021).
The nano zinc ferrite (nano-ZnFe 2 O 4 ) is a spinelstructured material with good magnetic and catalytic properties, which has been commonly used in the fields of catalysis, magnetic fluid research, and magnetic photocatalysis (Tang et al., 2013). Once the nano-ZnFe 2 O 4 absorbs photons, the electrons in its valence band become excited and jump to the conduction band. Moreover, the valence band forms strong oxidizing holes, which can oxidize the organic matter at the interface, as well as other substances, through stationary adsorption under the action of diffusion; otherwise, the holes could continue to exert an oxidation-induced electron-transfer effect (Cai et al., 2016).
Photocatalysis, which occurs under exposure to UV light, is also a common method for the environmental pollutant elimination (Ahmed et al., 2014). The conventional photocatalysis utilizes mostly UV from sunlight, which accounts for only 4% of the solar energy. However, through the introduction of catalysts, the utilization rate of sunlight can be effectively improved. The ZnFe 2 O 4 semiconductor possesses a relatively narrow bandgap of 1.9 eV; thus, its electrons can be excited by visible light (Vis) without any modification (Yao et al., 2014). However, the ZnFe 2 O 4 exhibited the low pollutant degradation efficiency due to the rapid recombination of photoinduced electron-hole pairs (Fan et al., 2012). To overcome the low-efficiency problem of the photocatalysis, the development of a more efficient catalyst system that would effectively improve the catalytic oxidation efficiency and overcome the existing limitations is important.
The catalytic activity of the catalyst can be effectively improved by modulating its surface area, preparation method, and changing its properties and structures Ren et al., 2015;Mady et al., 2019;Zhu et al., 2020). Meanwhile, the preparation of the ZnFe 2 O 4 catalyst by using the hydrothermal method for PMS activation under Vis has not yet been systematically investigated. Therefore, in this study, the parameters of the catalytic system for the PMS process were regulated, with OFL serving as the representative pollutant, PMS as the oxidant, and ZnFe 2 O 4 as the catalyst. The ZnFe 2 O 4 catalyst was synthesized using a hydrothermal method. Under visible-light exposure, the key steps and rates of the catalytic process were investigated with varying catalyst dosage, pollutant concentration, and reaction factors, such as pH, to explore the catalytic activity and degradation mechanism. We believe that the findings of this study will provide technical and valuable insights for the degradation of trace contaminants in wastewater. . All reagents were of analytical grade and used without any further purification. Ultrapure water (>18.2 MO cm), used for all the experimental solutions, was produced from a Milli-Q system (Burlington, USA).

Synthesis of ZnFe 2 O 4
Zn (NO 3 ) 2 Á6H 2 O (2.082 g) was dissolved in 35 mL of deionized water in a 250 mL beaker, after which 1 g of CTAB was added as a dispersant to form a uniform solution (Solution A). Thereafter, 5.656 g of Fe (NO 3 ) 3 Á9H 2 O was completely dissolved in 35 mL of deionized water to afford Solution B; Solution B was added to Solution A to yield Solution C. Solution C was placed in a supersonic machine and mixed for 30 min at 25 C. An NaOH solution (8 M) was added to Solution C dropwise to achieve a pH of 12.24. The resulting solution was subjected to ultrasonic stirring in the water bath (30 C) for 30 min, after which it was transferred to an autoclave and maintained at 130 C for 24 h. After the reactor was naturally cooled to room temperature, the precipitate was centrifuged and washed with deionized water to a neutral pH of approximately 6.0-7.0. Subsequently, the resultant was dried in a vacuum oven at 60 C for 24 h and ground to afford the ZnFe 2 O 4 . The molar ratio among the constituents of the prepared ZnFe 2 O 4 was Fe:Zn:CTAB ¼ 0.014:0.007:0.0027.

ZnFe 2 O 4 characterization
The specific surface area and microstructure of the samples were tested and analyzed using the Micromeritics ASAP 2010 porosimeter (Norcross, USA). The phase composition and microstructure of the ZnFe 2 O 4 were studied by the PANalytical Empyrean X-ray powder diffraction (XRD) characterization (Etten Leur, the Netherlands) (Liu Art. 9(1) page 2 of 10 Sun et al: Visible-light-induced PMS activation over ZnFe 2 O 4 for ofloxacin degradation et al., 2020a, 2021). Hitachi FlexSEM 1000 Scanning electron microscopy (Tokyo, Japan) was employed to characterize the surface morphology of the ZnFe 2 O 4 . The composition and chemical property of the ZnFe 2 O 4 were characterized by Thermo Fisher VG Multi-lab2000 X-ray photoelectron spectroscopy (XPS; Oregon, USA).

Photocatalytic experiment of ZnFe 2 O 4
Certain amounts of the OFL, PMS solutions, and the ZnFe 2 O 4 were mixed in the reactor, which was placed on a magnetic stirrer at 20 C. The NaOH or HCl solution (0.1 M) was used to adjust the pH of the reaction. Afterward, a dark adsorption reaction was conducted for 30 min before illumination. The Xenon lamp was turned on to initiate the reaction. At a certain time interval, samples were taken out, quenched with excess Na 2 S 2 O 3 , centrifuged at 10,000 rpm, filtered by a 0.45 mm filter, and further determined. The concentration of OFL was measured by Waters e2695 HPLC instrument (Milford, USA) equipped with a Sunfire C18 column (150 mm Â 4.6 mm, 5 mm) and the UV-Vis detector set at 294 nm. The mobile phase was a mixture of water and ACN (v: v ¼ 30:70) at a flow rate of 1.0 mL min -1 . The total organic carbon (TOC) of the samples was monitored by the Analytikjena multi N/C 3100 (Jena, Germany).

Results and discussion
3.1. Characterization of the ZnFe 2 O 4 The XRD characterization was employed to investigate the crystal structure and chemical composition of ZnFe 2 O 4 composites, and their diffraction patterns are shown in Figure 1. As revealed in Figure 1, there were distinct diffraction peaks at 2y ¼ 18.  Figure 2. It can be observed that the ZnFe 2 O 4 was circular granular morphology with a particle size of approximately 30-50 nm and the noticeable intergranular agglomeration. This was mainly attributed to the interaction forces among the ZnFe 2 O 4 particles originating from the magnetic property. Moreover, there also existed some distinct porous structures due to the intergranular agglomeration of the ZnFe 2 O 4 ( Figure 2).
The specific surface area and pore structure of the catalyst are important parameters, which could impact the performance of the catalyst, reaction order, and reaction activation energy. The specific surface area, pore volume, and pore diameter of ZnFe 2 O 4 were determined to be 126.8655 m 2 /g, 0.2046 cm 3 /g, and 64.5190 Å, respectively, indicating that the ZnFe 2 O 4 had a large specific surface area. Meanwhile, the N 2 adsorption/desorption isotherm and the pore size distribution of ZnFe 2 O 4 were displayed in  could be speculated that the large specific surface in ZnFe 2 O 4 would be beneficial for the adsorption, desorption, and diffusion of reactants and products, further greatly accelerating the pollutants degradation (Mady et al., 2019). The surface element composition and metal state of the ZnFe 2 O 4 were thoroughly probed through XPS analysis, and the results were illustrated in Figure 4. The Fe 2p spectrum exhibited two peaks at around 711.5 and 723.7 eV, which were ascribed to Fe 2p 3/2 and Fe 2p 1/2 in Figure 4a, suggesting the presence of Fe 3þ (Yao et al., 2014;Malakootian et al., 2019;Liu et al., 2020b). The two symmetric peaks at binding energies of 1021.3 and 1045.5 eV shown in Figure 4b are assigned to Zn 2p 3/2 and Zn 2p 1/2, suggesting the existence of Zn 2þ in the ZnFe 2 O 4 (Yao et al., 2014). The spectrum of O 1s (Figure 4c) could be deconvoluted into four peaks with binding energy at 529.18, 529.78, 531.68, and 533.08 eV (Yao et al., 2014). In brief, the peak at 529.18 eV was assigned to the lattice oxygen. The peaks at 529.78 eV and 531.68 eV corresponded to the oxygen-containing functional groups (Mady et al., 2019). The peak at 533.08 eV was assigned to the surface-adsorbed hydroxyl groups (-OH).

Degradation of OFL in different systems
The OFL degradation efficiencies in different processes including Vis alone, PMS alone, ZnFe 2 O 4 alone, PMS/ ZnFe 2 O 4 , Vis/ZnFe 2 O 4 , Vis/PMS, and Vis/PMS/ZnFe 2 O 4 processes were evaluated in Figure 5. It is noted from Figure 5a that the efficiency of OFL removal by the direct adsorption of ZnFe 2 O 4 was negligible (<3%). Besides, the OFL was slightly removed (<5%) via Vis alone, PMS alone, Vis/ZnFe 2 O 4 , and Vis/PMS processes. Meanwhile, the OFL removal efficiencies in PMS/ ZnFe 2 O 4 and Vis/PMS/ZnFe 2 O 4 processes were 54.2% and 86.5% within 60 min, respectively. The pseudofirst-order rate constant (k obs ) were 0.0125 and 0.0343 min -1 in PMS/ZnFe 2 O 4 and Vis/PMS/ZnFe 2 O 4 processes ( Figure 5b). Evidently, the presence of the ZnFe 2 O 4 and PMS enhanced the degradation efficiency of OFL under the Vis, and the best OFL degradation performance was obtained in Vis/PMS/ZnFe 2 O 4 process.

The factors affecting the catalytic activity of ZnFe 2 O 4 3.3.1. ZnFe 2 O 4 dosage
Here, the effects of the ZnFe 2 O 4 dosages (0-1,000 mg/L) on the OFL degradation efficiencies were investigated in this study, and the results were revealed in Figure 6. As shown in Figure 6a, as the ZnFe 2 O 4 dosage ranged from 100 mg/L to 750 mg/L, the OFL degradation efficiency increased from 35.4% to 80.9% within 30 min reaction. When the ZnFe 2 O 4 dosage further increased to 1,000 mg/ L, the OFL degradation efficiency decreased to 70.4%. The k obs values corresponding to different ZnFe 2 O 4 dosages were 0.0014, 0.0141, 0.0208, 0.0390, 0.0438, and 0.0349 min -1 , respectively (Figure 6b). With the increasement of the ZnFe 2 O 4 dosage, the degradation of OFL was enhanced simultaneously, probably because more active surface sites for oxidation, further to induce more reactive radical species initiation. While at high dosage (1,000 mg/ L), the self-aggregation of ZnFe 2 O 4 would partially affect the surface reactive sites and inhibit the penetration of Vis into the solution as well (Li et al., 2020b).

PMS dosage
The effects of PMS concentrations on the OFL degradation were studied, and the results were represented in Figure  7. It is investigated from Figure 7a that the OFL degradation efficiency gradually increased from 24.3% to 88.6% (60 min) with the PMS concentration varied from 25 mg/L to 100 mg/L and decreased to 75.3% when the PMS concentration further raised to 250 mg/L. The catalytic effect on OFL was optimal at the PMS concentration of 100 mg/L. The k obs values of the different PMS concentration from 25 mg/L to 250 mg/L were 0.0054, 0.0193,  (Zhu et al., 2016). Based on the results, 100 mg/L of PMS was used in subsequent experiments.

OFL concentration
The OFL degradation at initial concentration varying from 2.5 mg/L to 25 mg/L is displayed in Figure 8a. The OFL degradation efficiencies during 60 min reaction obviously decreased from 87.5% to 45.5% when the initial OFL concentration rose from 2.5 mg/L to 25 mg/L (Figure 8a). More OFL and its intermediates in the aqueous solution could occupy the surface reactive sites of ZnFe 2 O 4 and also can compete with OFL for the available oxidants in Vis/ZnFe 2 O 4 /PMS system (Li et al., 2020b).

Solution pH
As shown in Figure 8b, the effects of solution pH on OFL degradation in Vis/ZnFe 2 O 4 /PMS system were investigated from 3.0 to 9.0. Notably, OFL was rarely removed (less than 5.0%) by direct ZnFe 2 O 4 adsorption under all the experimental pH conditions (data not shown). When the solution pH varied from 3.0 to 6.0, the Vis/ZnFe 2 O 4 / PMS process achieved an excellent OFL degradation performance and more than 80% of OLF could be removed during 30 min reaction. The k obs values of the OFL degradation were 0.0565, 0.0484, 0.0443, and 0.0438 min -1 at the conditions of 3.0, 4.0, 5.0, and 6.0, respectively ( Figure 8b). The OFL degradation was highly pHdependent, and the OFL degradation efficiencies were distinctly declined under neutral and alkaline conditions. The k obs value of the OFL degradation was decreasing to 0.0091 min -1 at pH 9.0. Thus, it can be seen that the optimal pH value for achieving the maximum catalytic activity of ZnFe 2 O 4 was 3.0. However, considering the adverse effect of metal-ion dissolution at pH 3.0, the experimental pH value was chosen as 6.0 in the following experiment.

Temperature
Here, the effects of solution temperature (10 C-40 C) on the catalytic activity of ZnFe 2 O 4 were explored, and the results were displayed in Figure 9a. As shown in Figure  9a, as the temperature increased from 10 C to 40 C in the Vis/ZnFe 2 O 4 /PMS process, the k obs constants of OFL degradation were 0.040, 0.0438, 0.0334, and 0.0322 min -1 , respectively. The results indicated that the reaction temperature had little impact on the OFL degradation.   There were no obvious effects on the OFL degradation by the inorganic ions (such as Cland NO 3 -) at ambient concentrations, while the presence of CO 3 2and HCO 3 greatly inhibited the OFL degradation in the Vis/ZnFe 2 O 4 / PMS process. The introduction of CO 3 2affected most obviously on the OFL degradation due to its strong capability for scavenging OH and SO 4and then formed carbonate radical (CO 3 -), which was more selective and less oxidizing than OH and SO 4 -. Moreover, the degradation of OFL in Vis/ZnFe 2 O 4 /PMS process was significantly inhibited by HA (TOC was 5 mg/L) and lake water ( Figure  9b). Note that there were various complex matrix materials coexisting with relatively high concentrations in the surface water. These matrix compounds could compete with OFL for the available oxidants (PMS, OH, and SO 4 -), thus considerably affecting the OFL degradation, for example, the rate constants for the reactions of NOM with OH and SO 4were reported to be 1.4 Â 10 4 and 6.8 Â 10 3 (mgC L -1 ) -1 s -1 , respectively (Lutze et al., 2015).

ZnFe 2 O 4 photocatalytic cycle experiment
To further understand the catalyst activity and the value of reuse, the photocatalytic cyclic experiments of ZnFe 2 O 4 were conducted. The recovered ZnFe 2 O 4 was centrifuged  and dried at 60 C for 24 h in the recycling experiment and then reused for a subsequent run. As depicted in Figure  10, there was no distinct loss of catalytic properties after the third run, and still 62.2% of the OFL could be removed during the fifth cycling experiment in the Vis/ZnFe 2 O 4 / PMS process. The ZnFe 2 O 4 represented good stability and reusability, which showed a good application potential.

Conclusions
In conclusion, the physicochemical properties and the photocatalytic performance of ZnFe 2 O 4 for OFL degradation were systematically investigated in this study. About 80.9% of OFL removal efficiency within 30 min was achieved in the Vis/ZnFe 2 O 4 /PMS process, and the corresponding k obs constant of OFL degradation was 0.0438 min -1 . The characterization results of ZnFe 2 O 4 on physicochemical properties exhibited that the excellent photocatalytic performance of ZnFe 2 O 4 was probably attributed to the high specific surface area, pore volume, the abundant functional groups of the surface, and possibly reactive radicals generation. Furthermore, the ZnFe 2 O 4 represented great reusability and stability in the photocatalytic process. This study may propose a new idea to design an efficient ZnFe 2 O 4 catalyst for organic pollutants removal.

Funding
This study was financially supported by the Central Government Guidance for Local Science and Technology