Photocatalytic Degradation of Tetracycline by ZnO/γ-Fe2O3 Paramagnetic Nanocomposite Material

In recent years, the presence of numerous xenobiotic substances, such as antibiotics, has been detected in water environments. They can be considered as environmental contaminants, even if their effect on human health has yet to be totally understood. Several approaches have been studied for the removal of these kinds of pollutants. Among these compounds, tetracycline (TC), a broad-spectrum antibiotic, is one of the most commonly found in water due to its widespread use. In the context of reducing the presence of TC in aqueous solution, in this contribution, a composite catalyst based on zinc oxide (ZnO) and iron oxide (γ-Fe2O3) was developed and its photocatalytic properties were investigated. The catalytic materials were synthesized by a microwave-assisted aqueous solution method and characterized by Field Emission Scanning Electron Microscope (FESEM), X-Ray Fluorescence (XRF) and Brunauer−Emmett−Teller (BET) analysis. The TC concentration was evaluated by spectrophotometer measurements at specific time intervals. The performed photocatalytic experiments clearly demonstrated that the ZnO/γ-Fe2O3 composite catalyst presents significant photocatalytic activity, indeed a TC degradation efficiency of 88.52% was registered after 150 min. The presence of iron oxide in the structure of the catalyst enhances both the surface area and the pore volume, facilitating the adsorption of the analyte on the surface of nanostructures, a fundamental phase to optimize a photodegradation process. Moreover, ZnO was found to play the key role in the photocatalytic process assisted by γ-Fe2O3 which enhanced the TC degradation efficiency by 20%.


Introduction
Tetracyclines (TC) are a class of broad-spectrum antibiotics widely employed to treat various human and animal bacterial infections [1]. Due to their incomplete adsorption and partial metabolization by humans and animals, more than 70% of the TC used for pharmacological treatment are excreted by feces and urine [2] in zootechnical liquid waste and in municipal wastewaters. The presence of tetracyclines in environmental water induces several effects on soil microbial activity and aquatic photosynthetic organisms, causing an increase in antibiotic-resistant bacteria [3] and significant human

Catalyst Preparation
To synthesize flower-like ZnO structures, a microwave assisted aqueous solution method was used. zinc sulfate heptahydrate (ZnSO 4 × 7H 2 O, 100 mg, Merck, Darmstadt, Germany) were dissolved in 200 mL of deionized water and then an ammonium hydroxide solution (25% NH 3 in H 2 O, Fluka, Munich, Germany) was slowly added dropwise to the solution under continuous stirring until a pH value of 11 was achieved. The solution was heated by means of a microwave device (800 W) for 15 min resulting in the formation of a white precipitate. Then, the solution was left to cool to room temperature, the precipitate was collected by centrifugation, washed with water several times, and dried in an oven at 60 • C for 10 h. By this microwave-assisted aqueous solution method it is possible to synthesize three-dimensional structures composed of small nanoparticles. Regarding the growth mechanism, it may be assumed that ZnO crystal nuclei are first polarized by the electrical field of microwave radiation and then form nanoparticles arranged in different layers [25]. Iron oxide nanoparticles capped by polyethylene glycol (PEG) were prepared by means of coprecipitation and solvothermic synthesis according to the procedure reported by Bonfrate and coauthors [26]. Chemicals such as hexahydrate ferric chloride, tetrahydrate ferrous chloride, sodium acetate, polyethylene glycol (PEG 8000), diethylene glycol (DEG) and ethylene glycol (EG) (with a volume ratio DEG:EG = 30:10) were purchased from Merck and used without any further purification. This synthesis allows one to obtain nanostructures based on ferric oxide (γ-Fe 2 O 3 ) with paramagnetic features [26,27].
For the synthesis of ZnO/γ-Fe 2 O 3 composite material, 100 mg of the previously prepared iron oxide nanoparticles capped by PEG were mixed with 100 mg of ZnSO 4 × 7H 2 O and dissolved in 200 mL of deionized water. Then, the same microwave-assisted aqueous solution method described before for the synthesis of flower-like ZnO nanoparticles was applied, but in this case the obtained ZnO/γ-Fe 2 O 3 were simply collected by means of a neodymium magnet and washed three times in ethanol and then in deionized water. During this synthesis, it can be supposed that the iron oxide grains are placed between the ZnO aggregates and this inhibits the formation of the peculiar 3D flower-shape structures [28].

Catalyst Characterization
X-ray diffraction profiles of the synthetized nanostructures were recorded by a Miniflex diffractometer (Rigaku, Tokyo, Japan) from 20 to 80 • . The step was 0.020 • and the scan speed 0.25 • /min. A CuKα radiation at 30 kV and 100 mA was used as X-rays source.
The morphology of the synthesized structures was analyzed by a Field Emission Scanning Electron Microscope (FESEM) on a Sigma VP microscope (Zeiss, Oberkochen, Germany) equipped with a Gemini electron column.
To determine the stoichiometric ratio of Zn and Fe in the composite catalyst, X-Ray Fluorescence (XRF) spectroscopic measurements were performed by using a M4 TORNADO Micro-XRF spectrometer (Bruker, Billerica, MA, USA).
Brunauer−Emmett−Teller (BET) method was used to determine the specific surface area and porosity of the photocatalytic materials. The analysis was carried out by a Quantachrome NOVA 2200e series surface analyzer (Boynton Beach, FL, USA) using N 2 adsorption/desorption at 77 K. Before performing the measurements, the samples were degassed under a N 2 atmosphere for 12 h at 120 • C.

Photocatalysis Experiments
In each experiment, the opportune amounts of ZnO/γ-Fe 2 O 3 catalyst were added in 20 mL of an aqueous solution of tetracycline (Merck, Darmstadt, Germany) with a concentration of 30 mg/L and the system was maintained under continuous stirring (250 rpm) at room temperature and at pH equal to 6.7. The effect of different nanocomposite catalyst dosage (5, 10, 15 mg) was tested and 10 mg (0.5 mg/mL) of ZnO/γ-Fe 2 O 3 catalyst was used for the most of measurements carried out. It will be pointed out that this value represents the optimum amount of catalyst for TC photodegradation. The TC degradation efficiency of ZnO flower-like and ZnO/γ-Fe 2 O 3 catalysts were also compared. The photocatalytic activity of the synthesized materials was evaluated by time-monitoring the TC decomposition under UV-visible light irradiation. Before the illumination, the system was kept in dark for about 15 min to reach the adsorption-desorption equilibrium, then was irradiated by simulated solar light using a Solar S class A (100 mW cm −2 light intensity) halogen lamp (LOT-Oriel Italia Srl, Rome, Italy). The illumination experiments were carried out for 150 min because no significant improvements of degradation efficiency were observed at greater times. Every 30 min, an aliquot (2 mL) of TC solution was collected and separated by catalyst to measure the corresponding concentration of TC by an UV-Vis spectrophotometer (Lambda 650 UV/VIS Spectrometer, PerkinElmer, Waltham, MA, USA). The absorbance value at the characteristic band of about 360 nm was considered.
To calculate the TC degradation efficiency (E), the following Equation (1) was used: where C 0 and C t are the concentrations of tetracycline (mg/L) at the initial and degradation time.
The kinetics of photodegradation reaction of TC were studied by first-order kinetics model expressed in Equation (2), according to the Langmuir-Hinshelwood kinetics model [14]: where k and t are respectively the kinetic rate constant and irradiation time. The k value indicates the photocatalytic activity and it was calculated by the linear plot slope. In order to investigate the main photodegradation mechanism, the production of radicals (·OH and O 2 − ) and holes (h + ) was tracked by adding some radical oxygen species (ROS) scavengers, such as isopropanol (IPA; ·OH scavenger), L-ascorbic acid (LAA; ·O 2 − scavenger) and ethylenediamine-tetraacetic acid disodium salt (EDTA-Na 2 ; h + scavenger). Each aqueous solution of scavenger with a final concentration of 1 mM was added to TC solution (30 mg/L) in presence of ZnO/γ-Fe 2 O 3 nanocomposite catalyst. All measurements were performed in triplicate to ensure the reproducibility of the results and the mean values were reported.
To probe the recyclability of the catalyst, consecutive photocatalytic experiments were performed using the same ZnO/γ-Fe 2 O 3 . After the first photocatalytic process, the catalyst was used three times in subsequent TC photodegradation reactions. In each experiment, the catalyst was separated from the solution by a neodymium magnet, washed carefully with deionized water, dried and reused in the next degradation cycle under the same conditions.

Catalyst Characterization
ZnO/γ-Fe 2 O 3 nanostructures were characterized by means of XRD in order to identify the crystalline phase of the two components. As reported in Figure 1, ZnO shows the typical diffraction pattern of the zinc oxide in the wurtzite form (JPCS card 036-1451) and the iron oxide seems to maintain the original maghemite phase [29].
Nanomaterials 2020, 10, x FOR PEER REVIEW 4 of 12 where C0 and Ct are the concentrations of tetracycline (mg/L) at the initial and degradation time.
The kinetics of photodegradation reaction of TC were studied by first-order kinetics model expressed in Equation (2), according to the Langmuir-Hinshelwood kinetics model [14]: where k and t are respectively the kinetic rate constant and irradiation time. The k value indicates the photocatalytic activity and it was calculated by the linear plot slope. In order to investigate the main photodegradation mechanism, the production of radicals (•OH and O2 − ) and holes (h + ) was tracked by adding some radical oxygen species (ROS) scavengers, such as isopropanol (IPA; •OH scavenger), L-ascorbic acid (LAA; •O2 − scavenger) and ethylenediamine-tetraacetic acid disodium salt (EDTA-Na2; h + scavenger). Each aqueous solution of scavenger with a final concentration of 1 mM was added to TC solution (30 mg/L) in presence of ZnO/γ-Fe2O3 nanocomposite catalyst. All measurements were performed in triplicate to ensure the reproducibility of the results and the mean values were reported.
To probe the recyclability of the catalyst, consecutive photocatalytic experiments were performed using the same ZnO/γ-Fe2O3. After the first photocatalytic process, the catalyst was used three times in subsequent TC photodegradation reactions. In each experiment, the catalyst was separated from the solution by a neodymium magnet, washed carefully with deionized water, dried and reused in the next degradation cycle under the same conditions.

Catalyst Characterization
ZnO/γ-Fe2O3 nanostructures were characterized by means of XRD in order to identify the crystalline phase of the two components. As reported in Figure 1, ZnO shows the typical diffraction pattern of the zinc oxide in the wurtzite form (JPCS card 036-1451) and the iron oxide seems to maintain the original maghemite phase [29].  FESEM measurements were performed to investigate the morphological details of the synthesized catalysts. In Figure 2a, it is possible to clearly observe a three-dimensional flower shaped ZnO structure with a width of about 2.5-3 µm. Moreover, the magnified FESEM image inserted in Figure 2a highlights that the surface of the flower-shaped structures appears very rough, an interesting feature for a photocatalyst. Nanomaterials 2020, 10, x FOR PEER REVIEW 5 of 12 FESEM measurements were performed to investigate the morphological details of the synthesized catalysts. In Figure 2a, it is possible to clearly observe a three-dimensional flower shaped ZnO structure with a width of about 2.5-3 μm. Moreover, the magnified FESEM image inserted in Figure 2a highlights that the surface of the flower-shaped structures appears very rough, an interesting feature for a photocatalyst.
However, in the ZnO/γ-Fe2O3 composite material, the characteristic flower shape is not preserved because of the iron oxide nanoparticles that are located both inside and outside of the ZnO microstructure. FESEM image of ZnO/γ-Fe2O3, shown in Figure 2b, reveals a surface densely covered by paramagnetic spherical grains with a diameter size of about 80-100 nm, in good agreement with the literature [26]. The weight percentage of Zn and Fe oxides in the structure of the synthetized composite catalyst was measured by XRF analysis. The amount of the oxides in the composite calculated from the XRF spectra ( Figure S1) was found to be 40.07 (wt. %) for ZnO and 59.32 (wt. %) for Fe2O3 as reported in Table S1. The amount of ZnO present in the composite material will be used to calculate the opportune dosage of ZnO/γ-Fe2O3 required in the catalytic processes.
In Figure 3a the N2 adsorption−desorption isotherm of ZnO flower-like and ZnO/γ-Fe2O3 nanocomposite catalysts are reported. The specific surface area was calculated by the multipoint BET method from the N2 adsorption branch of the isotherm, the pore diameter and pore volume were calculated from the desorption branch of isotherm using the Barret−Joyner−Halenda (BJH) method ( Figure 3b). However, in the ZnO/γ-Fe 2 O 3 composite material, the characteristic flower shape is not preserved because of the iron oxide nanoparticles that are located both inside and outside of the ZnO microstructure. FESEM image of ZnO/γ-Fe 2 O 3 , shown in Figure 2b, reveals a surface densely covered by paramagnetic spherical grains with a diameter size of about 80-100 nm, in good agreement with the literature [26].
The weight percentage of Zn and Fe oxides in the structure of the synthetized composite catalyst was measured by XRF analysis. The amount of the oxides in the composite calculated from the XRF spectra ( Figure S1) was found to be 40.07 (wt. %) for ZnO and 59.32 (wt. %) for Fe 2 O 3 as reported in Table S1. The amount of ZnO present in the composite material will be used to calculate the opportune dosage of ZnO/γ-Fe 2 O 3 required in the catalytic processes.
In Figure 3a the N 2 adsorption−desorption isotherm of ZnO flower-like and ZnO/γ-Fe 2 O 3 nanocomposite catalysts are reported. The specific surface area was calculated by the multipoint BET method from the N 2 adsorption branch of the isotherm, the pore diameter and pore volume were calculated from the desorption branch of isotherm using the Barret−Joyner−Halenda (BJH) method ( Figure 3b). Nanomaterials 2020, 10, x FOR PEER REVIEW 6 of 12 The respective values reported in Table 1 emphasize that both the surface area and pore volume of the ZnO/γ-Fe2O3 nanocomposite are significantly greater than those ones of the ZnO flower-like catalyst. This substantial difference is very probably due to the iron oxide grains that are placed not only on the external surface of composite material, but also between the ZnO particles.

Effect of ZnO/γ-Fe2O3 Catalyst Dosage
The effect of dosage of ZnO/γ-Fe2O3 composite material on the TC degradation was investigated by adding 5, 10 and 15 mg of catalyst, respectively, to 20 mL of TC solution (30 mg/L). The results shown in Figure 4 indicate that initially the increase of ZnO/γ-Fe2O3 dosage enhanced the TC degradation efficiency due to the intensification of the photocatalytic activity. The respective values reported in Table 1 emphasize that both the surface area and pore volume of the ZnO/γ-Fe 2 O 3 nanocomposite are significantly greater than those ones of the ZnO flower-like catalyst. This substantial difference is very probably due to the iron oxide grains that are placed not only on the external surface of composite material, but also between the ZnO particles.

Effect of ZnO/γ-Fe 2 O 3 Catalyst Dosage
The effect of dosage of ZnO/γ-Fe 2 O 3 composite material on the TC degradation was investigated by adding 5, 10 and 15 mg of catalyst, respectively, to 20 mL of TC solution (30 mg/L). The results shown in Figure 4 indicate that initially the increase of ZnO/γ-Fe 2 O 3 dosage enhanced the TC degradation efficiency due to the intensification of the photocatalytic activity.  The respective values reported in Table 1 emphasize that both the surface area and pore volume of the ZnO/γ-Fe2O3 nanocomposite are significantly greater than those ones of the ZnO flower-like catalyst. This substantial difference is very probably due to the iron oxide grains that are placed not only on the external surface of composite material, but also between the ZnO particles.

Effect of ZnO/γ-Fe2O3 Catalyst Dosage
The effect of dosage of ZnO/γ-Fe2O3 composite material on the TC degradation was investigated by adding 5, 10 and 15 mg of catalyst, respectively, to 20 mL of TC solution (30 mg/L). The results shown in Figure 4 indicate that initially the increase of ZnO/γ-Fe2O3 dosage enhanced the TC degradation efficiency due to the intensification of the photocatalytic activity.  When 15 mg of ZnO/γ-Fe 2 O 3 were added to TC solution, a slight rise in degradation efficiency was initially observed, however at 150 min of irradiation no further increase of TC degradation occurred, probably due to the lack of tetracycline molecules. Consequently, all measurements were performed using 20 mL of TC solutions at concentration of 30 mg/L in which were added 10 mg of ZnO/γ-Fe 2 O 3 composite catalyst.

Comparison Between ZnO Flower-Like and ZnO/γ-Fe 2 O 3 Composite Catalysts
In Figure 5a the values of TC degradation efficiency in function of the irradiation time are reported and the photocatalytic activity of ZnO flower-like and ZnO/γ-Fe 2 O 3 composite catalysts is compared. To perform the comparison between the catalysts, the same quantity of ZnO (0.2 mg/mL) was used considering that the ZnO/γ-Fe 2 O 3 composite material is made up of about 40% in weight of ZnO, as detected by XRF measurements. In the absence of catalyst no degradation of tetracycline was observed, whereas the γ-Fe 2 O 3 grains induced a very weak TC degradation. However, in the presence of ZnO flower-like and ZnO/γ-Fe 2 O 3 composite catalysts a remarkable photocatalytic activity was observed during the light irradiation. After 150 min of irradiation, the TC degradation efficiency values were 68.28% and 88.52% in the presence of ZnO and ZnO/γ-Fe 2 O 3 catalyst, respectively. This is coherent and in good agreement with ZnO nanorods/magnetite systems used to degrade tetracycline in aqueous solution [30]. With the addition of iron oxide into the composite material, the degradation efficiency of TC was 1.30 times higher than that ZnO flower-like catalyst. The photocatalytic activity enhancement in presence of iron oxide will be considered in the Discussion section.

Comparison Between ZnO Flower-like and ZnO/γ-Fe2O3 Composite Catalysts
In Figure 5a the values of TC degradation efficiency in function of the irradiation time are reported and the photocatalytic activity of ZnO flower-like and ZnO/γ-Fe2O3 composite catalysts is compared. To perform the comparison between the catalysts, the same quantity of ZnO (0.2 mg/mL) was used considering that the ZnO/γ-Fe2O3 composite material is made up of about 40% in weight of ZnO, as detected by XRF measurements. In the absence of catalyst no degradation of tetracycline was observed, whereas the γ-Fe2O3 grains induced a very weak TC degradation. However, in the presence of ZnO flower-like and ZnO/γ-Fe2O3 composite catalysts a remarkable photocatalytic activity was observed during the light irradiation. After 150 min of irradiation, the TC degradation efficiency values were 68.28% and 88.52% in the presence of ZnO and ZnO/γ-Fe2O3 catalyst, respectively. This is coherent and in good agreement with ZnO nanorods/magnetite systems used to degrade tetracycline in aqueous solution [30]. With the addition of iron oxide into the composite material, the degradation efficiency of TC was 1.30 times higher than that ZnO flower-like catalyst. The photocatalytic activity enhancement in presence of iron oxide will be considered in the Discussion section.
The photocatalytic decomposition of TC was plotted as first-order fitting (Figure 5b) and the obtained value of kinetic rate constants k for ZnO and ZnO/γ-Fe2O3 were 7.47 × 10 −3 min −1 and 13.21 × 10 −3 min −1 respectively, as reported in Table 2. The kinetics study of TC photodegradation also confirms the improvement of photocatalytic performance in presence of the γ-Fe2O3 grains into the structure and on the surface of ZnO with a k value higher by 1.77 times. Moreover, it is commonly accepted that the catalyst morphology affects the photocatalytic properties because the degradation reactions take place at the interface of catalyst where the produced radicals interact with pollutants molecules [31]. Consequently, the large specific surface area of the ZnO/γ-Fe2O3 makes this composite material more favorable than ZnO for the photocatalytic activity.  The photocatalytic decomposition of TC was plotted as first-order fitting (Figure 5b) and the obtained value of kinetic rate constants k for ZnO and ZnO/γ-Fe 2 O 3 were 7.47 × 10 −3 min −1 and 13.21 × 10 −3 min −1 respectively, as reported in Table 2. The kinetics study of TC photodegradation also confirms the improvement of photocatalytic performance in presence of the γ-Fe 2 O 3 grains into the structure and on the surface of ZnO with a k value higher by 1.77 times. Moreover, it is commonly accepted that the catalyst morphology affects the photocatalytic properties because the degradation reactions take place at the interface of catalyst where the produced radicals interact with pollutants molecules [31]. Consequently, the large specific surface area of the ZnO/γ-Fe 2 O 3 makes this composite material more favorable than ZnO for the photocatalytic activity.

Discussion
As it is well-known [16], photocatalytic phenomena are influenced by both the surface porosity and the energetic levels of the catalyst. When a catalyst is irradiated by light of opportune wavelength, it adsorbs photon energy (hν) equal to its band gap, and the electrons (e − ) in the valence band (VB) are promoted to the empty conduction band (CB) leaving the positive charged holes in the VB. The holes can react with the water molecules to promote the formation of ·OH radicals whereas the photogenerated electrons react with dissolved O 2 to form superoxide species (·O 2 − ) and subsequently produce ·OH [31,32]. From Figure 5a, it is possible to observe that the degradation efficiency of the ZnO/γ-Fe 2 O 3 can be considered as the algebraic sum of degradation efficiency of ZnO and γ-Fe 2 O 3 separately. This suggests that the two oxides do not compete or cooperate in the photodegradation process. The predominant photodegradation is due to the zinc oxide, but the γ-Fe 2 O 3 contributes up to 20% in the total photodegradation efficiency. Furthermore, the paramagnetic features of γ-Fe 2 O 3 are preserved when the composite nanostructure with ZnO is formed. This is a crucial point since it allows one to quickly separate the nanostructures from water solutions. Thus, iron oxides play a dual role: they improve the photodegradation efficiency and increase the ease of removal of the photocatalyst from aqueous samples.
Again, to clarify the photodegradation mechanism of tetracycline, radicals and hole scavengers were used. As shown in Figure 6a, the TC photocatalytic degradation in presence of ZnO/γ-Fe 2 O 3 nanoparticles was considerably inhibited in presence of LAA, moderately reduced by IPA, whereas the addition of EDTA-Na 2 did not suppress the photocatalytic degradation efficiency. These results suggested that the TC degradation should be mainly influenced from the ·O 2 − and ·OH radical oxidative species which can oxidize the TC molecules in other mineralization products. In particular, when IPA is used as ·OH radical scavenger, it can be supposed that h + in valence band are not used in the degradation process reducing the lifetime of the charge-separated state and, then, increasing the charge recombination rate [33]. This phenomenon reduces, of course, the photodegradation efficiency. On the other side, when h + scavenger is used, the main mechanism of hydrogen peroxide formation, is inhibited even though this process appears not favorable from an energetic point of view [34].
In fact, the VB of ZnO and γ-Fe 2 O 3 are located at 2.7 and 2.3 eV, respectively. Conduction bands of the zinc oxide and maghemite are reported to be at −0.5 and 0.1 eV, respectively. These evidences exclude the possibility of a charge separation at the ZnO and γ-Fe 2 O 3 interface, confirming that the two nanostructures cannot cooperate in the photodegradation process and the H 2 O 2 formation by water oxidation is unlikely. Finally, as reported in the literature [17], the two absorption bands at about 275 and 360 nm of an aqueous solution of tetracycline can be ascribed to the conjugated double-bond structures with two carbonyl groups and enolic groups, respectively [35]. The tetracycline absorption spectra, in the presence of ZnO/γ-Fe 2 O 3 catalyst, at the initial time and after 150 min of light irradiation are shown in Figure 6b. After the photocatalysis process an important decrease and a slight hypsochromic shift of the absorption peak at greater wavelengths were observed. These results suggest that under the oxidation process conditions, tetracycline molecules were mineralized and at the same time intermediate products were formed [36,37]. In Figure 7 the TC degradation efficiency (%) in the initial photocatalytic process and in the next three reuse cycles were compared. It is possible to affirm that the ZnO/γ-Fe2O3 composite material is quite stable during the first two reuse cycles, showing a good recyclability of the catalysts [20,37]. The recyclability of a catalytic material is a very important feature for the feasibility and cheapness of the process studied.

Conclusions
Tetracycline aqueous solution photodegradation was promoted using a nanostructure formed by ZnO and maghemite under solar irradiation. The presence of iron oxide nanoparticles strongly influences the morphology of the ZnO even though it was highlighted that the crystalline phase of ZnO is preserved (wurtzite form) as well as that of the iron oxide (γ-Fe2O3) nanostructures. The paramagnetic properties of γ-Fe2O3 nanoparticles allow one to efficiently remove the photocatalyst from aqueous solutions by simply applying a weak external magnetic field. This substantially reduces the waiting time due to the gravimetric separation of the nanostructures from the aqueous samples proposing the nanocomposite ZnO/γ-Fe2O3 for application in large scale and continuous water treatment processes. Furthermore, it was demonstrated that the presence of γ-Fe2O3 increases the porosity and the adsorption properties of the nanostructures if compared to the bare ZnO obtained by the same synthetic protocol. This is an interesting achievement since, as it is known, the In Figure 7 the TC degradation efficiency (%) in the initial photocatalytic process and in the next three reuse cycles were compared. It is possible to affirm that the ZnO/γ-Fe 2 O 3 composite material is quite stable during the first two reuse cycles, showing a good recyclability of the catalysts [20,37]. The recyclability of a catalytic material is a very important feature for the feasibility and cheapness of the process studied. In Figure 7 the TC degradation efficiency (%) in the initial photocatalytic process and in the next three reuse cycles were compared. It is possible to affirm that the ZnO/γ-Fe2O3 composite material is quite stable during the first two reuse cycles, showing a good recyclability of the catalysts [20,37]. The recyclability of a catalytic material is a very important feature for the feasibility and cheapness of the process studied.

Conclusions
Tetracycline aqueous solution photodegradation was promoted using a nanostructure formed by ZnO and maghemite under solar irradiation. The presence of iron oxide nanoparticles strongly influences the morphology of the ZnO even though it was highlighted that the crystalline phase of ZnO is preserved (wurtzite form) as well as that of the iron oxide (γ-Fe2O3) nanostructures. The paramagnetic properties of γ-Fe2O3 nanoparticles allow one to efficiently remove the photocatalyst from aqueous solutions by simply applying a weak external magnetic field. This substantially reduces the waiting time due to the gravimetric separation of the nanostructures from the aqueous samples proposing the nanocomposite ZnO/γ-Fe2O3 for application in large scale and continuous water treatment processes. Furthermore, it was demonstrated that the presence of γ-Fe2O3 increases the porosity and the adsorption properties of the nanostructures if compared to the bare ZnO obtained by the same synthetic protocol. This is an interesting achievement since, as it is known, the

Conclusions
Tetracycline aqueous solution photodegradation was promoted using a nanostructure formed by ZnO and maghemite under solar irradiation. The presence of iron oxide nanoparticles strongly influences the morphology of the ZnO even though it was highlighted that the crystalline phase of ZnO is preserved (wurtzite form) as well as that of the iron oxide (γ-Fe 2 O 3 ) nanostructures. The paramagnetic properties of γ-Fe 2 O 3 nanoparticles allow one to efficiently remove the photocatalyst from aqueous solutions by simply applying a weak external magnetic field. This substantially reduces the waiting time due to the gravimetric separation of the nanostructures from the aqueous samples proposing the nanocomposite ZnO/γ-Fe 2 O 3 for application in large scale and continuous water treatment processes. Furthermore, it was demonstrated that the presence of γ-Fe 2 O 3 increases the porosity and the adsorption properties of the nanostructures if compared to the bare ZnO obtained by the same synthetic protocol. This is an interesting achievement since, as it is known, the analyte adsorption represents the first step of the photodegradation process. Again, it was demonstrated that γ-Fe 2 O 3 does not affect the ZnO-mediated photodegradation, actually enhancing the catalytic performances of about 20%. In conclusion, the ZnO/γ-Fe 2 O 3 is able to photodegrade more than the 88% of the water-dissolved TC and, by means of the paramagnetic γ-Fe 2 O 3 it is possible to quickly remove the photocatalyst and to re-use it, showing its good recyclability features.
Author Contributions: Conceptualization and draft editing, G.G. and S.B.; investigation, writing-original draft preparation and editing, P.S.; catalysts synthesis, S.S.; XRF and BET analysis, S.P. and A.L.; FESEM measurements, F.M. and N.L.; supervision, L.V. All authors have read and agreed to the published version of the manuscript.

Conflicts of Interest:
The authors declare no conflict of interest.