The removal of tetracycline with biogenic CeO2 nanoparticles in combination with US/PMS process from aqueous solutions: kinetics and mechanism

Antibiotics have received great attention because of their abuse and potential hazards to the human health and environment. In the current work, peroxymonosulfate (PMS) was added to a cerium oxide (CeO2)/ultrasonic (US) system for tetracycline (TC) degradation. CeO2 nanoparticles (NPs) were synthesized by a simple and cost-effective method using Stevia rebaudiana leaf extract and cerium nitrate as precursors. The as-synthesized CeO2 NPs were characterized by X-ray diffraction, field emission scanning electron microscopy, and Fourier-transform infrared spectroscopy analysis. The effects of catalyst dosage, PMS concentration, US power, initial antibiotic concentration, and pH on TC removal were investigated. The results confirmed the formation of CeO2 NPs with a fluorite structure, spherical shape, and average particle size of 29 nm. The removal efficiency of TC was 92.6% in the optimum oxidation conditions ([TC]1⁄4 15 mg/L, [PMS]1⁄4 50 mM, [CeO2]1⁄4 0.6 g/L, pH1⁄4 6, and US1⁄4 70 W) and followed the zero-order kinetics. Experiment scavenger demonstrated both sulfate and hydroxyl radicals (SO4 • , OH) were responsible for degrading antibiotics. Biogenic CeO2 NPs and ultrasound waves-activated PMS is a promising technology for water pollution caused by contaminants such as pharmaceuticals.


INTRODUCTION
Tetracycline (TC; C 22 H 24 N 2 O 8 ) is the second broad-spectrum antibiotic which is largely used in human and veterinary medicine (Hou et al. ). TC used in medical treatment is poorly absorbed by humans or animals (Xing et al. ). Around 30-80% of TC is excreted through feces and urine as an active compound into the environment and becomes one of the emerging pollutants in water (Marzbali et al. ). TC residual has been detected in influent and effluent of wastewater treatment plants and surface water at 0.52 μg/L, 0.17 μg/L, and 0.11 μg/L, respectively (United States) (Eslami et al. ). TC, as one of the highly consumed antibiotics, is extensively used for treating infectious diseases from Gram-positive and Gram-negative bacteria for both humans and animals (Jiang et al. ; Song Ma & Li ).
To eliminate the undesirable TC accumulation in aquatic environments, developing new and reliable treatment processes is necessary (Gao et al. ; Daghrir & Drogui ). In the case of TC treatment, a variety of techniques, such as adsorption (Gao et al. ), electrocatalytic oxidation (Xu et al. ), electrocoagulation (Ouaissa et al. ), sonochemical processing (Yazdani et al. ), photocatalytic degradation (Saadati Keramati & Ghazi ), biodegradation (Xiong et al. ), and ozonation (Zhu et al. ) have been reported. Common treatment processes are not efficient methods because they result in solid waste, thus creating other environmental problems requiring further treatment (Daghrir & Drogui ; Priya & Radha ).
Advanced oxidation processes (AOPs), based on the production of sulfate and hydroxyl radicals, have been considered for water and wastewater treatment due to their high oxidation power and conversion of antibiotic (Malakootian et al. a, b, c, d, ; Nasiri et al. ; Tamaddon et al. a, b) and many organic chemical compounds into minerals, water, and carbon dioxide (Pi et al. ; Yang et al. ).
Peroxymonosulfate (PMS ¼ Oxone) is considered to be an environmentally friendly oxidant because it is benign, as are most of the by-products of its reactions (Yang et al. ; Ao et al. ). The decomposition rate of organic compounds with PMS at the room temperature is low, but it can be enhanced through activation by photolysis (ultraviolet (UV) irradiation) or sonolysis (ultrasonic (US) wave; Equation (1)), or by using transition metals (Equation (2)) or heat (Equation (3)), thereby producing highly reactive SO 4 •À and • OH radicals (Yin et al. b; Ghanbari Ahmadi & Gohari ). The decomposition of TC by PMS is shown in Equations (4) and (5).
Sonocatalytic methods have been studied in various studies using metal oxides under different conditions to eliminate antibiotics (Karimi Fatehifar & Alizadeh ). Of those metal oxide catalysts, CeO 2 is a suitable activator for PMS (Shen et al. ). CeO 2 has been widely investigated because of its multiple applications such as in catalysis, as an electrolyte material for solid oxide fuel cells, as a material with a high refractive index, and as an insulating layer on silicon substrates (Wen et al. ). CeO 2 is abundant, nontoxic, and inexpensive (Saravanakumar Muthupoongodi & Muthuraj ; Xing et al. ). Therefore, CeO 2 can be used in heterogeneous sonocatalyis reactions. Today, the combination of two or three activators is very attractive for wastewater treatment. In this way, the US waves in the presence of CeO 2 nanoparticles (NPs) can accelerate PMS activation to produce sulfate radicals.
The current study aims to investigate the ability of the US/ PMS/CeO 2 sonocatalytic process to remove TC antibiotics from artificial wastewater. In this paper, we demonstrate the effectiveness of the CeO 2 NP as a catalyst for TC removal in artificial wastewater. Biosynthesis is a simple method that can be utilized for large-scale production of nanoparticle at low cost. CeO 2 nanoparticles were characterized by X-ray diffraction (XRD), scanning electron microscopy (SEM), and Fourier-transform infrared spectroscopy (FTIR) techniques. Reusability of the catalyst and mechanism were also studied. In addition, the effects of pH, catalyst dosage, PMS, TC concentrations, and US power on the removal of TC were studied. ), as the source of PMS, and cerium nitrate (99% pure) were purchased from Sigma-Aldrich Co. (USA). Hydrochloric acid (HCl) and sodium hydroxide (NaOH) were purchased from Merck, Germany, and used for pH adjustment. Ethanol (EtOH, purity: 99.8%) and tert-butyl alcohol (TBA, purity: > 99.0%), which were used as radical scavenging compounds, were also purchased from Merck.

Materials
Solutions were prepared daily before the experiments. Distilled water was used throughout the experiment.
Preparing Stevia rebaudiana S. rebaudiana leaves were collected from the surrounding areas of Bam, Kerman, Iran, and washed carefully to remove dust particles. Extraction was done by soaking in water. Then, 10 g S. rebaudiana leaf powder was dissolved in 100 mL of distilled water and placed on a rotary shaker at 1.6 s À1 for 24 h. The solution was then filtered with Whatman filter paper No. 42. Distilled water was used as a solvent for all described experiments.

Synthesis and characterization of CeO 2 NPs
CeO 2 NPs were synthesized using cerium nitrate as a salt and prepared plant extract. Briefly, 0.05 M cerium nitrate solution was prepared in deionized water. Then, 50 mL of S. rebaudiana leaf extract was added to 50 mL from this solution and stirred for 5 h at 70 C. Then, the solvent was dried in an oven at 90 C. Last, the solution was calcinated for 5 h at 300, 400, 500, 600, and 700 C, leaving yellow-colored CeO 2 NPs. The absorption spectra of biosynthesized CeO 2 NPs were recorded by a UV-Vis spectrophotometer in the spectral range of 200 to 850 nm. FTIR analysis was carried out in the range of 400 cm À1 to 4,000 cm À1 (Perkin Elmer). The surface morphology of the samples (CeO 2 NPs) was observed by using a TESCAN field emission scanning electron microscope (FESEM,Mira3,Czech Republic). FESEM was used to analyze the morphology and particle size of the as-synthesized CeO 2 NPs. XRD analysis using CuKα radiation (λ ¼ 1.54060 Å) was also performed to show the crystallinity and phase composition of the samples (Philips XRD, Model: PW1730, Holland).

Sonocatalytic activity of biogenic CeO 2 NPs
The sonocatalytic potential of the as-prepared CeO 2 NPs was investigated by degrading TC antibiotics under the US wave. The experiments were performed using the batch flow mode Erlenmeyer flasks (working volume: 500 mL). To measure the US irradiation emitted, a US probe with a titanium tip (Ti-6Al-4 V, 1 cm diameter) on a digital sonicator (IKA@RW 20), which had a constant frequency of 20 kHz and varied power, was placed at the reactor center 3 cm from the bottom. During sonication, the temperature of solution was kept at 25 ± 2 C thereby circulating cooling water. The reactor contents were mixed using a mechanical mixer. In this test, PMS as an oxidant and CeO 2 as a catalyst was added to the solution. The experimental set-up was covered to avoid the photocatalysis. The residual of TC concentration was analyzed by a spectrophotometer (Shimadzu, Japan) at the wavelength of 261 nm. The removal efficiency and the adsorption capacity of biogenic CeO 2 NPs were calculated using the following equation: where A o ¼ sample adsorption rate before testing; and A ¼ sample adsorption rate after the test (Saadati Keramati & Ghazi ).
In this work, the effective parameters in the US/PMS/ CeO 2 process were studied separately, including pH, PMS concentration, CeO 2 dosage, antibiotic concentration, and US power to determine the optimum conditions. Moreover, the effect of scavenging agents TBA and EtOH and various chemical oxidants on the degradation rate of TC over US/ PMS/CeO 2 were evaluated under optimal operating conditions.
The reusability of CeO 2 was examined for up to five catalytic runs at the optimal parameter. After each run, the used catalysts were filtered and separated using centrifuged (100 s À1 , 30 min), and then washed with distilled water and dried in an oven at 60 C for 6 h. Then, the catalyst was reused with a fresh TC aqueous solution for the next run. Herein, a tentative mechanism was proposed for the sonocatalytic system and the generation of reactive oxidizing species. Aqueous solubility (mmol/L) 0.52-117
A histogram of the diameter size distribution of the assynthesized CeO 2 NP is illustrated in Figure 3. Most particles have a diameter in the range of 1-50 nm (in the defined range for nanoparticles). Most of the as-generated particles were smaller than 300 nm, indicating the high surface area along with more reactive sites for generating • OH radicals to catalytically decompose TC.
FTIR spectra of the synthesized CeO 2 NPs are shown in Figure 4. FTIR spectroscopy was used to identify the functional group and also atomic and molecule vibration. The functional group is one of the many factors that affect adsorption efficiency. The FTIR spectrum showed several absorption peaks in the interval of the wavelength of 400 cm À1 to 4,000 cm À1 . The absorption peak at 3,480.29 cm À1 confirmed the presence of CeO 2 NPs. The absorption range of 3,800-3,000 cm À1 corresponded to the O-H stretching frequency. As CeO 2 readily traps atmospheric CO 2 , its corresponding peaks were observed at 2,426 cm À1 and 1,385 cm À1 . The absorption band at 1,542 cm À1 and 1,342 cm À1 was related to OH-adopted water molecule. An absorption peak at 492 cm À1 associated with the vibration of Ce-O indicated the formation of pure CeO 2 phase as observed by XRD analysis (Liying et al. ). The Ce-O stretching frequency was expected below 400 cm À1 , but in this study, it was observed at 452 cm À1 , indicating the formation of CeO 2 .

Effect of pH solution
In the sonocatalytic system, the rate of decomposition depended on the pH because the antibiotic uptake capacity on a catalyst is an important factor. The increase in the number of molecules absorbed on the catalyst was due to the increase in the decomposition of the molecule (Zargar Pourreza & Samadifar ; Wang et al. ). In this study, in order to investigate the effect of pH on sonocatalytic decomposition (condition experiment: 0.6 g/L CeO 2 NP dosage, 50 mM PMS concentration, 15 mg/L TC concentration, and US power ¼ 50 W), pH ranges acidic (3), neutral (6), and alkaline (9) were adjusted by adding appropriate amounts of NaOH and HCl (0.1 M). TC degradation performance at different pH is shown in Figure 5(a). The efficiency of TC decomposition at pH values of 3, 6, and 9  after 120 min reaction was achieved as 77.4, 82.1, and 85.7%, respectively. The pKa value of TC was 3.3, and thus, at pH < 3.3, its molecules were negatively charged, and at pH > 3.3, its molecules were positively charged (Zhu et al. ). The pH value dependency can be explained based on the metal oxide zero point charge (pH zpc ). The pH zpc of CeO 2 was 6.8. For pH values lower than pH zpc , the surface became positively charged; for pH values higher than pH zpc , the CeO 2 surface was negatively charged (Zargar Pourreza & Samadifar ).
Regarding these conditions, therefore, TC removal can occur effectively when pK a TC < pH < pH zpc can be ascribed to a negatively charged catalyst surface, and is more available for the adsorption of the cationic form of TC molecules (Zhang et al. ). Heterogeneous AOPs can be operated in a broad pH range, but always have better performance in neutral conditions (Nurhasanah Gunawan & Sutanto ). This has been previously reported in other studies. According to the results, pH 6.0 was selected as the optimal value for subsequent experiments due to the similarity to the neutral environment, which allowed the process to be used in a real-time destruction process without pre-pH adjustment.

Effect of initial PMS concentration
To obtain the optimal concentration of PMS on the removal efficiency, the PMS concentrations of 2, 10, 20, 50, and 80 mM (condition experiment: 0.6 g/L CeO 2 , 15 mg/L TC, time ¼ 120 min, US power ¼ 50 W) were investigated. As shown in Figure 5(b), the efficiency of TC removal increased rapidly with increasing the concentration of PMS from 2 to 50 mM. After 120 min of contact time, the TC removal efficiency was about 83% when the PMS concentration was 50 mM. However, by increasing the concentrations of PMS from 50 to 80 mM, the removal efficiency decreased. Excessive chemical oxidant dosage has been reported to have an unfavorable effect on the performance of AOPs. The maximum removal efficiency was obtained at the PMS concentration of 50 mM. Increasing the removal efficiency by increasing the concentration of PMS was due to the fact that, by increasing the concentration of PMS in the solution, the sulfate radical that plays a major role in the decomposition of antibiotics and other pollutants also increased (Cao et al. ; Li et al. ). In some studies, increasing the sulfate radical beyond a certain concentration does not affect the removal efficiency, which has been attributed to the recombination of sulfate radicals. In other words, when concentration exceeds a certain limit, the PMS removes the radical hydroxyl (by recombining them) and produces low reactivity radicals (Wang & Wang ; Luo et al. ).

Effects of CeO 2 NP dosage
In this study, the sonocatalytic degradation of TC at different doses of CeO 2 NPs between 0.2 and 0.8 g/L was investigated. Figure 5(c) shows that by increasing the CeO 2 NPs to 0.6 g/L, the removal efficiency of TC increased and, in quantities greater than 0.6 g/L, the removal efficiency decreased. An increase in the rate of antibiotic degradation may be due to  an increase in the number of active sites available on the surface of CeO 2 , which could expedite reactions to produce more free radicals, resulting in an increase in the number of antibiotic molecules absorbed at the catalyst level (Yi et al. ; Guan et al. ). At a concentration greater than 0.6 g/L of catalyst, the rate of antibiotic degradation decreased due to density, reduction of penetration depth, and scattering of efficiency radiation. Also, the lower degradation efficiency for higher CeO 2 dosage (0.6 g/L) can be explained by insufficient consumption of PMS due to the agglomeration of catalysts (Shen et al. ). Hence, the optimum catalyst dosage was selected as 0.6 g/L for TC degradation. To prevent unnecessary catalyst loss, it is important to avoid adding more than the optimal amount of catalyst for sonocatalytic analysis (Tizhoosh et al. ). The stabilization of the removal efficiency of pollutants with a further increase in CeO 2 catalyst may also be because the Ce 2þ plays the role of scavenger for sulfate radicals at high concentrations. Pouretedal et al. (Pouretedal & Kadkhodaie ) found that the CeO 2 activated sunlight irradiation was effective at 1 g/L with 90.6% methylene blue (20 mg/L) removal after 125 min.

Effect of initial TC concentration
The effect of the initial concentration of TC (15-70 mg/L) on the efficiency of the process is shown in Figure 5(d).
With increasing the TC concentration, the removal efficiency of this process declined. Generally, the rate of degradation decreased with increasing initial TC concentration. The removal efficiency obtained was 83, 72, 68, and 59% for 15, 30, 50, and 70 mg/L TC, respectively, after 120 min of treatment. Three factors reduced the efficiency of degradation by increasing the antibiotic concentration: (1) the increase in the number of antibiotic molecules adsorbed on the surface of the catalyst led to a decrease in the number of active sites that generated hydroxyl and sulfate radicals (Yi et al. ); (2) the production rate of free reactive species in the system was constant, therefore, they would not be able to degrade all the pollutant (Ghanbari Ahmadi & Gohari ); and (3) intense competition was formed between primary compounds and by-products to react with oxidizing species at higher concentrations of pollutants (Isari et al. ).

Effect of US power
The effect of intensity of US irradiation using the US/PMS/ CeO 2 system on TC degradation rate as another affecting operational factor was evaluated at 30, 50, and 70 W in optimal conditions. The results are presented in Figure 5(e). An increasing trend in antibiotic degradation was observed with enhancement in intensity of the US. When the US power was increased from 30 to 70 W, the decontamination percentage accelerated substantially from 74.1 to 92.6%. As shown in the literature, improvement in the intensity of ultrasound can effectively increase the amount of US energy transferred to the system and result in enhancing the collapse of cavitation bubbles and forming more reactive oxidizing radicals (Hou Zhang & Xue ). On the other hand, this result can be described by the fact that the improvement of the US intensity increases the catalytic activities and available catalyst surface area due to cleansing the catalyst surface from aggregated particles by microstreaming of US irradiation (Nasseri et al. ). According to the maximum removal percentage, 70 W of US power was used as their optimum intensity for further experiments.
Some studies show that increasing the US power beyond the optimal amount reduces the removal efficiency. It can be deduced that at high US power, a portion of US power would be consumed and converted into heat because of the scattering effect (Malakotian et al. ).

Effect of radical scavengers
In order to investigate free radicals on TC degradation using the US/PMS/CeO 2 process, tests in the presence of TBA and EtOH scavengers were performed. Results with the quenching agents are shown in Figure 6. Two quencher agents, including TBA and EtOH, were used at a concentration of 100 mM. The scavenging rate constants for EtOH varied from 1.2 to 2.8 × 10 9 M À1 s À1 for • OH and from 1.6 to 7.7 × 10 7 M À1 s À1 for SO 4 •À . For TBA, the rate constant was approximately 1,000-fold greater for • OH (3.8 to 7.6 × 10 8 M À1 s À1 ) than that for SO 4 •À (4 to 9.1 × 10 5 M À1 s À1 ) (Yin et al. a).
The ratio of scavenging rates of oxidant radicals by EtOH and TBA varied between 1.3 and ∼3.5 for • OH radicals and between ∼18 and 200 for SO 4 • À radicals. TBA was used as a specific scavenger of holes and • OH, while EtOH was applied to quench • OH and dioxygen (O 2 ) (Li et al. ; Yan et al. ). From these results, holes, • OH, SO 4 •À , and O 2 species, were contributed during TC degradation in the US/PMS/CeO 2 system. TC degradation in the presence of TBA and EtOH was reported as 68% and 46%, respectively. TC degradation was much higher in the presence of TBA than for EtOH. Therefore, in the sonocatalytic US/PMS/CeO 2 system, the sulfate, and hydroxyl radicals were the oxidizing and destroying agents of TC. PMS can also produce sulfate radicals during the direct sonolysis reaction, which allows TC to absorb free radical species to produce intermediate products and complete the mineralization.

Comparison of the efficiency of alternative processes for TC degradation
In order to verify the efficiency of the process of US/PMS/ CeO 2 against US alone, PMS alone, CeO 2 alone, PMS/US, and CeO 2 /US, the CeO 2 /PMS processes were tested for tetracycline antibiotic removal under similar conditions (Figure 7). The results showed that the use of the US alone, PMS alone, and CeO 2 alone had a slight effect on TC removal, which could be related to the absence of hydroxyl and sulfate radical production. The results showed that US waves alone could remove only 12.3% of antibiotics because of the low production of the hydroxyl radical ( • OH). The application of PMS alone (50 mM) induced about 21% antibiotic removal as the result of direct oxidation by PMS. In the CeO 2 -only process (0.6 g/ L), about 30.12% antibiotic removal was obtained due to adsorption on the catalyst surface. The combined process of US waves with catalysts (US/ CeO 2 ) had low efficiency (58%) due to its low ability to produce reactive radicals. The US/PMS process had the removal efficiency of 40.11%, which was due to the breaking of the O-O band in PMS and producing radicals by US waves (Equations (7)-(9)) (Feizi et al. ). When PMS was combined with CeO 2 , it was able to remove 25% of the antibiotic. In fact, Ce 2þ was the main activator of PMS for sulfate radical production. The highest removal efficiency was associated with the US/PMS/CeO 2 process with a removal efficiency of 92.6%. This high performance of the US/PMS/CeO 2 system can be explained by: (1) the simultaneous existence of both catalytic oxidation and adsorption in the system, (2) the presence of several PMS activators (i.e. US and Ce 2þ ), which could lead to the efficient decomposition of PMS molecules and generation of more reactive species (hydroxyl and sulfate radicals), and (3) the synergistic effect among the techniques applied for PMS activation (Ghanbari Ahmadi & Gohari ). In relation to the high adsorption capacity and the excellent catalytic potential of the as-synthesized catalyst in coupling the US into PMS activation, the US/PMS/CeO 2 process was selected as the best method for the subsequent experiments of TC degradation.
Effect of reusability of CeO 2 for sonocatalytic degradation Recycling tests were performed to evaluate the reusability potential of an as-prepared catalyst (biogenic CeO 2 ) in the US/PMS/CeO 2 system for five consecutive cycles without any chemical/physical modification. For supported heterogeneous catalysts, reusability is of fundamental importance in practical applications. Figure 8 shows the effects of the reusability of CeO 2 NPs in the sonocatalytic process of TC degradation. As shown in Figure 8, CeO 2 exhibited a similar catalytic activity during five cycles and more than 86% of TC was removed after five cycles. The study showed that the rate of sonocatalytic analysis of CeO 2 in the first stage was higher than that in the subsequent stages, which can cause a decrease in activity due to the accumulation and deposition of antibiotics around CeO 2 (Fadzeelah et al. ). Today, one of the strategies for industrial wastewater treatment is to focus on developing environmentally friendly technology. Reusing CeO 2 is a good and suitable suggestion for treating sustainable wastewater because it has been  demonstrated that the catalyst can be reused after the sonocatalytic treatment process.

Degradation kinetics
The effect of experimental operating variables on the TC degradation by the US/PMS/CeO 2 process was performed in a batch environment. Different types of kinetic models for the US/PMS/CeO 2 process were studied ( Table 2). The best fit was considered for the model with the highest value of linear regression coefficient (R 2 ). The experimental results indicated that the TC degradation using the US/ PMS/CeO 2 process followed the zero-order kinetic model. To evaluate the removal coefficient, Equation (10) was used for the zero-order model: where C 0 , C t , and k obs represent the initial TC concentration (mg/L), TC concentration (mg/L) at a specific reaction time, and constant reaction rate (mg/L.min), respectively (Nasseri et al. ).
Comparison of PMS with persulfate and hydrogen peroxide PMS (HSO 5 À ), hydrogen peroxide (HP), and persulfate (PS, S 2 O 8 2À ) have been widely used for degrading and treating several organic pollutants and contaminated water. In this step, the performance of the US/CeO 2 system in TC degradation was evaluated in the presence of 50 mM of various chemical oxidants. The experiments were conducted under optimum conditions over 120-min reaction time. The process efficiency was calculated and compared as follows: The removal efficiency of antibiotics by process US/ PMS/CeO 2 , US/PS/CeO 2 , and US/HP/CeO 2 was 92.6, 89.1, and 83%, respectively. This significant increase can be expressed by a significant increase in the production rate of reactive species in the system and, consequently, the further degradation of contaminant molecules. Here, additional free radicals can be formed by: (1) decomposition of oxidants by US irradiations (Equation (11)) and (2) reaction between transition metal ions (Ce 2þ ) on CeO 2 NPs and oxidants (Equation (12) Probable mechanism for sonodegradation of TC using synthesized CeO 2 NPs The sonocatalytic reaction mechanism of biosynthesized CeO 2 NPs is shown in Equations (13)-(24). When the surface of the CeO 2 NPs irradiate US waves, the electron (e À ) from the valence band of CeO 2 moves to the conduction band leaving a hole (h þ ) in the valence band. The holes (h þ ) act as an oxidizing agent and oxidize the pollutant directly, or they may react with water to provide hydroxyl radicals. The electron (e À ) in the conduction band performs as a reducing agent to reduce the oxygen adsorbed on the surface of the CeO 2 sonocatalyst (Pouretedal & Kadkhodaie ; Niu et al. ). Furthermore, after irradiation, the antibiotic becomes excited and the excited antibiotic injects an electron into the conduction band of CeO 2 and is scavenged by pre-adsorbed oxygen to form active oxygen radicals. These generated active radicals drive the sono degradation process. The CeO 2 NPs play an important role as an electron carrier. Such assisted sonocatalytic processes provide an attractive path for treating antibiotics under US wave (Fadzeelah et al. ). The reasonable mechanism for the sonocatalytic degradation of TC antibiotics is schematically Second-order 0.9315 0.0017 (L/min.mg) y ¼ 0.0017x þ0.0365 shown in Figure 9.

CONCLUSION
The results obtained in this study indicated that the AOP using PMS and biogenic CeO 2 NPs under US waves was a feasible treatment method for TC elimination from the aqueous phase. It is found that the percentage of TC removal increased with a decreased initial TC concentration, and at the higher oxidant and catalyst concentration than its optimum value. The removal efficiency of TC was 92.6% under the optimum oxidation conditions (PMS ¼ 50 mM, TC concentration ¼ 15 mg/L, CeO 2 dosage ¼ 0.6 g/L, US power ¼ 70 W, and 120 min time). Biogenic CeO 2 was synthesized, suggesting that excellent activity and reusability characteristics were found during the antibiotic decomposition process. Scavenger tests showed that sulfate and hydroxyl radicals were equally responsible for antibiotic degradation. The TC degradation under the US/PMS/CeO 2 system followed the zero-order reaction kinetics under optimum conditions.