A Brief Photocatalytic Study of ZnO Containing Cerium towards Ibuprofen Degradation

Ibuprofen (IBU) is one of the most-sold anti-inflammatory drugs in the world, and its residues can reach aquatic systems, causing serious health and environmental problems. Strategies are used to improve the photocatalytic activity of zinc oxide (ZnO), and thosethat involvethe inclusion of metalhave received special attention. The aim of this work was to investigate the influence of the parameters and toxicity of a photoproduct using zinc oxide that contains cerium (ZnO-Ce) for the photodegradation of ibuprofen. The parameters include the influence of the photocatalyst concentration (0.5, 0.5, and 1.5 g L−1) as well as the effects of pH (3, 7, and 10), the effect of H2O2, and radical scavengers. The photocatalyst was characterized by Scanning Electron Microscopy-Energy Dispersive Spectroscopy, Transmission electron microscopy, Raman, X-Ray Diffraction, surface area, and diffuse reflectance. The photocatalytic activity of ibuprofen was evaluated in an aqueous solution under UV light for 120 min. The structural characterization by XRD and SEM elucidated the fact that the nanoparticle ZnO contained cerium. The band gap value was 3.31 eV. The best experimental conditions for the photodegradation of IBU were 60% obtained in an acidic condition using 0.50 g L−1 of ZnO-Ce in a solution of 20 ppm of IBU. The presence of hydrogen peroxide favored the photocatalysis process. ZnO-Ce exhibited good IBU degradation activity even after three photocatalytic cycles under UV light. The hole plays akey role in the degradation process of ibuprofen. The toxicity of photolyzed products was monitored against Artemia salina (bioindicator) and did not generate toxic metabolites. Therefore, this work provides a strategic design to improve ZnO-Ce photocatalysts for environmental remediation.


Introduction
Pharmaceutical products are consideredemerging pollutants in numerous aquatic matrices such as waste, surface, and groundwater [1]. They are considered semi-persistent pollutants due to their resistance, consumption, and non-volatile character [2]. The toxicity, high biological activity, and formation of dangerous metabolites can lead to serious and long-term problems for aquatic organisms and water quality [3]. In addition, their ability to be absorbed generates uncertainties about their effects on metabolism and persistence in the body of living beings and ecosystems [4]. The Handbook of Environmental Chemistry

Synthesis of Photocatalyst
The synthesis of ZnO-Ce was previously reported by Dias et al. [29], whereby a 3:1 (mol) solution of glycerol:urea containing Zn(NO 3 ) 2 (0.05 mol) and NH 4 (Ce) 4 NO 3 (0.01 mol) was prepared in 30 mL under stirring at 70 • C. Afterwards, 5 mL of a solution of sodium hydroxide (0.10 mol) was added slowly, and the white precipitate was filtered. Then, the material named ZnO-Ce was washed and oven dried at 60 • C.

Characterization
The semiconductors were characterized by X-Ray diffraction (XRD) using a Bruker diffractometer (D8 Advance) with Cu-Kα radiation and a scanning rate of 2 • min −1 . Morphology was examined by High-resolution transmission electronic microscopy (HRTEM) images obtained using a JEM-2100-JEOL microscope with a 0.23 nm resolution point and an 80 to 200 kV accelerating voltage. The images were analyzed using Image J software to obtain the average size of the nanoparticles. Scanning electron microscopy (SEM) images were obtained using a field emission electron microscope JEOL JSM-7401F at an acceleration tension of 5.0 kV, an SEI secondary electron detector, a working distance ranging from 3.0 mm, and a resolution of 1.5 nm, with EDS.The analysis of Diffuse Reflectance Spectroscopy (DRS) wasperformed by a Shimadzu spectrophotometer, Model UV-3600 with a diffuse reflectance accessory monitoring in the region of 200 to 800 nm. The optical properties of the synthesized semiconductors were performed on a spectrophotometer UV-Vis spectrometer Shimadzu (UV-2550) and the band gap value was determined based on the standard Kubelka-Munk model [27,30]. The textural properties of the solids were investigated by adsorption-desorption analysis of N 2 at 77 K using the Quanta chrome Autosorb-iQ Instruments equipment. Thus, before each analysis, approximately 200 mg of the sample was degassed for 4 h at 200 • C. The surface area and pore volume and diameter were calculated using the Brunauer-Emmett-Teller (BET) method based on N 2 adsorption-desorption. Raman spectra were obtained at 25 • C using a spectrophotometer, WITEC brand, Confocal model, using an Ar laser as the excitation source with 50 mW at 533 nm, and spectra were achieved in the region of 200 to 800 cm −1 .

Photocatalytic Test
The photocatalytic activity of ZnO-Ce was investigated by degradation of the drug ibuprofen in an aqueous solution at 25 ± 1 • C under UV light (125 W Hg without bulb). In a typical experiment, IBU (20 ppm) was tested at photocatalyst concentrations (0.5 gL −1 ; 1.0 gL −1 ; and 1.5 gL −1 ) under magnetic agitation. The potency of the lamp was monitored by a radiometer (HANNA-HI 97500-Luxmeter). The suspension was removed at different times (0, 5,10,15,30,45,60,90, and 120 min). After irradiation, the samples were immediately centrifuged, then absorbance measurements were conducted in a CARY 300 model spectrophotometer. The degradation rate of IBU was determined using Equation (1): A 0 represents the absorbance in initial time; A is the absorbance of IBU after irradiation. The kinetic study was conducted using the Langmuir-Hinshelwood model, and the rate constant, k, was determined under pseudo-first order [31].

OperationalParameters in Photocatalytic Test
To monitor photocatalytic activity, we investigated the effect of pH and H 2 O 2 , and the scavengers were investigatedto understand the role of radicals in the photocatalytic process. The pH was around 3.7 to 10, and methyl alcohol (3.4 × 10 −3 mg/L), silver nitrate (5.0 × 10 −4 mol/L), and EDTA (2.4 × 10 −6 mol/L) were used as scavengers of hydroxyl radicals, electrons, and holes, respectively [27,32]. The amounts and concentrations of scavengers used were established according to previous studies [27,33]. The effect of hydrogen peroxide on the photodegradation of IBU in the 20 mgL −1 solution was studied by adding 0.5 mmol/L of H 2 O 2 using 0.5 gL −1 of the catalyst. The tests were carried out under the same conditions as the photocatalytic tests, only adding the scavengers or H 2 O 2 at the described concentrations.

Artemia salina Bioassays
The toxicity of the IBU solution irradiated in bioassays with Artemia salina wasdetermined [34] with adaptations according to the procedure described by Araújo et al. [27,33]. The microcrustaceans were obtained after 48 h of cultivation in a synthetic saline solution under continuous illumination and oxygenation. Synthetic saline was obtained by adding salts to ultrapure water as described in Materials. The nauplius wasadded to asolution containing irradiated IBU and synthetic saline solution (1:1 v/v), and the mortality of microcrustaceans was evaluated after 24 h and 48 h.

Characterization
To verify the phase formation, X-ray diffraction measurements were performed on the samples and the results are shown in Figure 1 [35,36]. No secondary or undesirable phases were observed for the ZnO-Ce sample, suggesting its incorporation in the ZnO hexagonal structure. For both samples, the lattice constants a and c were calculated using Equation (2) [37]: The dhkl values for each (hkl) family can be calculated by the Bragg formula, λ = 2d sin(θ), where θ is the Bragg angle and λ is the incident X-ray wavelength. The lattice parameters a and c for the ZnO sample are 3.2274 Å and 5.1708 Å, respectively, while for the ZnO-Ce sample, they are 3.2843 Å and 5.2420 Å, respectively. Comparing the lattice constant values of the ZnO sample with the ZnO-Ce sample, there was a significant increase, which may be associated with the differences between the ionic radii of Zn (0.74 Å) and Ce (0.97 Å) [38]. On the other hand, the average crystallite sizes (D) and the lattice strain (ε) of samples were estimated using the Williamson-Hall equation [37]. For ZnO and ZnO-Ce samples, the D and ε values are 22 nm and 12 nm and 0.21 × 10 −3 % and 0.28 × 10 −3 %, respectively. Evidently, the presence of Ce in the ZnO hexagonal structure causes a decrease in the average crystallite size, which may be related to the nucleation rate modification during the ZnO crystallization. If the retarding force is greater than the driving force for growth, a decrease in the average crystallite size is expected [39].  Figure S1).
The Raman spectrum of ZnO-Ce is shown in Figure 3a. ZnO-Ce exhibited an intense absorption band centered at 438 cm −1 (E2H), which is attributed to the main vibration mode of the ZnO phase with a hexagonal wurtzite structure [40], indicating good crystallization and a defined peak. Additional bands are found at 332 cm −1 , 382 cm −1 , 538 cm −1 , 584 cm −1 , and 677 cm −1 , also related to active polar modes in Raman attributed to polar modes in ZnO [40]. The peak at 332 cm −1 is attributed to second-order Raman scattering, coming from the phonon limit zone 3E2H-E2L, while the peak at 382 cm −1 can be attributed to the A1T mode [40]. There is a weak band around 584 cm −1 (phonon E1L), which may be related to the formation of different defects, such as oxygen vacancies and interstitial Zn [41,42]. Usually, low-intensity bands are identified in materials containing metal due to the network distortion promoted by the presence of metal.  Some properties directly influence photocatalytic processes, such as the band gap value, because it determines the excitation region of the material. The analysis of absorption spectroscopy in the ultraviolet and visible region (Figure 3b) allows the calculation of the band gap, which is a very important physical characteristic of materials as it affects their electrical behavior [43]. The band gap value of ZnO is one of the most important oxides due to its unique physical characteristics of wide and direct gap (~3.44 eV) with a large exciton binding energy [44]. The Eg value obtained for ZnO-Ce was estimated at 3.31 eV with no significant changes compared to ZnO. ZnO-Ce has a band around 377 nm in the UV-Vis region, with this being the region with the highest absorption energy referring to the transitions constituted mainly by the 2p orbital of oxygen (O) and the 4f orbital of cerium (Ce), resulting in a difference between the energies of each orbital and band gap [45].
Another factor is that ZnO-Ce has a surface area of 24.807 m 2 g −1 , which is higher than ZnO, according to Zhang et al. [46] and other works [47][48][49]. ZnO-Ce has a pore volume of 0.103 cm 3 g −1 and has anaverage pore size of 9.369 nm; thus, ZnO-Ce exhibits more active sites, providing greater amounts of reactions, consequently improving photocatalytic performance. Table 1 shows the specific surface of ZnO-Ce and other materials.

Photocatalytic Degradation of IBU
Preliminarily, the photolysis of IBU was investigated ( Figure 4). The electronic absorption spectra of IBU (20 ppm) showed a degradation rate of 37% after 120 min under UV light.
The photodegradation of IBU was inversely proportional to the photocatalyst concentration-the lower the ZnO-Ce concentration, the higher the degradation rate. The photocatalyst promoted a higher number of active sites, consequently increasing the concentrations of hydroxyl radicals and superoxides [55]. However, the radicals can recombine and inhibit the photocatalytic process, and a large amount of suspended material can cause the inhibition of light passage, implying a decrease in the photocatalytic efficiency [56]. Similar behavior was reported by Wang et al. [9].  The material can transfer conduction band (CB) electrons from ZnO to cerium due to the larger relative band gap potential of ZnO. This decreases photocatalytic recombination and consequently increases the half-life of the electron species/photo-transferred holes to the photocatalyst surface and contributes to redox reactions. Furthermore, the excessive addition of the photocatalyst has been shown to exceed the synergistic interactions between cerium and ZnO, slowing the process [57].
When ZnO-Ceis irradiated under UV light, the electron excitation favors the formation of photoinduced electrons and holes that generate hydroxyl radicals (HO • ) and peroxyls (O 2 •− ) in the photoreactional process.
The main radical reactions generated from the photocatalytic process using ZnO-Ce in the degradation of IBU are presented in Figure 6. recombination and consequently increases the half-life of the electron species/photo-transferred holes to the photocatalyst surface and contributes to redox reactions. Furthermore, the excessive addition of the photocatalyst has been shown to exceed the synergistic interactions between cerium and ZnO, slowing the process [57].
When ZnO-Ceis irradiated under UV light, the electron excitation favors the formation of photoinduced electrons and holes that generate hydroxyl radicals (HO • ) and peroxyls (O2 •− ) in the photoreactional process.
The main radical reactions generated from the photocatalytic process using ZnO-Ce in the degradation of IBU are presented in Figure 6. The pH of heterogeneous photocatalysis is one of the key factors inits operability and yield [55]. Ahigh or low concentration of protons (or hydroxyls) can act on the functional groups of the molecule and the surface of the catalystto be degraded. This can interfere with the adsorption of pollutants, the photogeneration of hydroxyl radicals, and the presence of clusters [59].
To evaluate the effect of pH on IBU degradation, photoreactions were conducted at pH 3, 7, and 10 ( Figure 7). The pH of the initial IBU solution measured was around pH 6.0 (natural solution, without pH adjustment). The degradation rate increased with the increase in the number of protons, following the order pH 10 (22%), pH 7 (41%), and pH 3 (47%); therefore, it increased by about 25% compared to the alkaline medium. The low degradation at pH 10 can be attributed to competition between the surface charges of the photocatalyst, which are slightly altered and may be protonated/deprotonated, consuming the hydroxyl radical and consequently decreasing the degradation rate

Effect of pH
The pH of heterogeneous photocatalysis is one of the key factors inits operability and yield [55]. Ahigh or low concentration of protons (or hydroxyls) can act on the functional groups of the molecule and the surface of the catalystto be degraded. This can interfere with the adsorption of pollutants, the photogeneration of hydroxyl radicals, and the presence of clusters [59].
To evaluate the effect of pH on IBU degradation, photoreactions were conducted at pH 3, 7, and 10 ( Figure 7). The pH of the initial IBU solution measured was around pH 6.0 (natural solution, without pH adjustment). The degradation rate increased with the increase in the number of protons, following the order pH 10 (22%), pH 7 (41%), and pH 3 (47%); therefore, it increased by about 25% compared to the alkaline medium. The low degradation at pH 10 can be attributed to competition between the surface charges of the photocatalyst, which are slightly altered and may be protonated/deprotonated, consuming the hydroxyl radical and consequently decreasing the degradation rate [56,60]. Wang et al. [9] observed that the degradation of IBU was directly affected by the pH of the solution, with the highest degradation rate (49.45%) in acidic pH values. The best photocatalytic yield for degradation at an acidic pH has been attributed to the adsorption of IBU on the surface of the catalysts due to the increase in electrostatic interactions between IBU and the photocatalyst [61].

Effect of H 2 O 2
Hydrogen peroxide is an oxidizing agent that, when reacting with photogenerated electrons on the semiconductor surface, forms hydroxyl radicals ( • OH), and its use as an aid in the photodegradation process of pollutants is very promising [55]. However, conducting photodegradation with H 2 O 2 produced an insufficient yield (52%) for total degradation (Figure 8). The degradation of IBU in this process is given by the reaction of hydroxyl radicals generated after the addition of hydrogen peroxide, as described by Equations (8)-(10) [9]. The IBU degradation rate increased with the use of ZnO-Ce and ZnO-Ce in the presence of H 2 O 2 systems (ZnO-Ce + H 2 O 2 ) reaching about 60% and 70%, respectively. The improvement of the photochemical process can be linked to the interaction between the semiconductor and H 2 O 2 , increasing the photocatalyst sites and consequently facilitating the chemical interaction with IBU [61].

Analyses of Reactive Oxidizing Species
The degradation rate of IBU under the action of inhibitors was evaluated (Figure 9). In the presence of EDTA (h + , hole sequestrant), the decrease in degradation indicates the fundamental role of the participation of the holes, considering the substantialinhibition of the process (23%). The addition of AgNO 3 (an electron scavenger) and CH 3 OH (an • OH scavenger) reduced the degradation rate by 43% and 35% respectively. Therefore, the IBU degradation photoprocess seems to be more activated by the greater availability of h + in VB, as well as other processes that help in the formation of the H 2 Oand • OH radical [23,33]. These results contrast with the work of Xuet al. [54], who observed that the activity of the holes to generate oxidative species was less efficient in the IBU photodegradation mechanism. In the presence of Ag + ions, the photocatalysis had the smallest decrease in total degradation, possibly due tothe deposition of Ag 0 on the surface allowing better photocatalytic activity through the capture of photogenerated electrons avoiding recombination [32,62]. Overall, the order of photochemical activity in the photodegradation of IBU follows the order h + > • OH > e − .

Reuse and Stability of the Photocatalyst
The stability of a photocatalyst during application and after successive applications is extremely important [63,64]. Stability was assessed via consecutive reapplication of IBU degradation cycles. As shown in Figure 10a, after the third cycle, the compoundexhibited 44% degradation activity and maintained significant stability compared to the starting material (i.e., first degradation), reaching only 16% total loss, that is 11% (firstand secondrun) and 5% (second and thirdrun), which allows considering good kinetic stability. The equivalent loss value demonstrates that the combination of ZnO-Ce has a significantly active photocatalytic performance; however, any structural and morphological changes are important to observe. To verify the stability of the photocatalyst, it was examined before and after the photocatalytic reaction through the evaluation of crystallinity (XRD) and microscopy (SEM) (Figure 10b-e). After the photocatalytic cycle, ZnO-Ce exhibited characteristic diffraction peaks, indicating that the precursor photocatalyst maintained its parent crystal structure with small displacements relative to ZnO and similar toZnO-Ce before testing (observation discussed in Section 3.1).
Microscopic images were also recorded to assess morphological stability (Figure 10c,d). The typical morphology of ZnO-Ce with irregular shapes and agglomerates prevailed even after the photocatalytic test, ensuring stability viathe cerium that maintains the disorganization of the spherical structure observed only in ZnO.
XRD and SEM studies confirmed that the interaction between IBU and ZnO-Ce (under UV light) resulted in a strong interaction that guaranteed photocatalytic activity; thus, they are considered promising and stable for future applications in the photodegradation of different classes of pollutants. A previous study reinforces that recyclability tests have been used in several studies to direct such photocatalysts towardsenvironmental applications [64].
Previousauthors have evaluated material stability, such as Araújo et al., who investigated material stability through reuse/recycles in photocatalytic tests using Zno synthesized in the presence of polysaccharides [33]. Sa et al. [24] evaluated this parameter inthe degradation of a bentazone herbicide. In order to estimate the photocatalytic activity of ZnO and Ce nanorods under the same conditions with the best 1% CZ sample, Kardes et al. [65] observed three consecutive runs of color removal in AR88 azo dye. They observed a slight decrease in the percentage performance in each run, and the reason for the decrease was investigated by the ratio of ZnO leaching in the dye solution.

Toxicological Monitoring of Aqueous Solutions by Artemia salina
Artemia salinais abioindicator widely used for the toxicological evaluation of several pollutants [67]. Toxicity was evaluated in solutions before and after the photocatalytic treatment to observe the toxicological effect of ZnO-Ce as an environmental photocatalyst, that is, one that promotes both the degradation of the drug and the reduction of toxicity. Figure 11 presents the percentages found in the toxicity evaluation through the percentage of live Artemia, which guarantees the survival of the nauplii in the solutions. First, the control trial demonstrated 100% survival of Artemias at 24 h and 80% after 48 h. A reduced percentage of mortality was observed for all irradiated solutions, such as the photolytic solution (15%-24 h; 50%-8 h), the photocatalytic solution or first run (20%-24 h; 50%-48 h), and the reused solutions (20%-24 h; 25% at 48 h-second run) and (30%-24 h; 45%-48 h-third run). This behavior may be due to the derivation of photo-generated by-products during the irradiation process, which meantthe solution was able to cause mortality. However, the formation of these by-products of an unknown nature did not significantly compromise the survival of the bioindicator, since the solution only presented 45% toxicity and/or mortality 48 h after the third cycle, which ensured standardization within the lethal dose (LD 50), even if only the photogenerated solution is evaluated. Given the 50% survival in the last stage tested, it is probablethat no more toxic metabolites were generated in relation to the starting drug. According to Reddy et al. [68] and Zakari et al. [69], effluents alter their physicochemical propertieswhen released into aquatic matrices, allowing the effluent to be more or less miscible depending on its concentration in the water. Therefore, ZnO-Ce has photocatalyst potential for the decontamination of drug-containing effluents and did not show toxicity inthe photogenerated solutions. Figure 11. Bioassay using A. Salina, with control corresponding to microcrustaceans cultivated in synthetic saline solution (38 gL −1 ), followed by photolytic, photocatalytic (firstrun), and after successive cycles (secondand thirdruns).

Conclusions
X-ray diffraction showed that the photocatalyst is crystalline and has a hexagonal wurtzite structure, and micrographs illustrated that the morphology of ZnO-Ceis spherical in shape. ZnO-Ce exhibited high photocatalytic performance towards IBU degradation under UV light. The best experimental condition for IBU photodegradation was obtained in an acidic pH using 0.50 gL −1 of ZnO-Ce in a solution of 20 ppm of IBU. The effect of pH can be attributed to the protonation/deprotonation of the catalyst surface, altering the generation and activity of the hydroxyl radical and, consequently, decreasing the degradation rate. The presence of hydrogen peroxide favored the photocatalysis process. In addition, the degradation efficiency of IBU was decreased in the presence of EDTA, indicating that the hole plays a key role in the degradation process of ibuprofen. The IBU degradation rate decreased by 16% after three photochemical cycles, and XRD analysis indicated structural stability of ZnO-Ce at the end of the photoreactions. The non-toxicity of the irradiated solutions evaluated against Artemia salina produced a low percentage of mortality. A deeper understanding of the photogenerated by-products and the proposal ofan appropriate mechanism based on the identification of intermediates is an important step that needs to be investigated in future work, as well as the antimicrobial activity of this photocatalytic material. In this study, anindication of the formation of fewertoxic by-products was shown, as presented in the toxicity test, indicating an interesting system for future applications in the treatment of wastewater.