Acetaminophen oxidation under solar light using Fe-BiOBr as a mild Photo-Fenton catalyst

environment. This work demonstrates that Fe – BiOBr using the solar photo-Fenton process eliminates acetaminophen at mild pH in aqueous media. Fe-BiOBr is produced using microwave-assisted solvothermal synthesis, and the formation of the BiOBr phase is confirmed with XRD. SEM and TEM demonstrated the flower-like morphology, in which crystallite size reduces as a function of the Fe loading. The chemical environment at the surface of Fe – BiOBr is investigated with XPS. The results are connected with Raman analysis, which suggests the presence of oxygen vacancies in Fe – BiOBr. Furthermore, the effect of Fe in BiOBr is assessed by determining the optical band gap with UV – Vis. The Fe-BiOBr functionality is assessed during acetaminophen degradation. Fe-BiOBr revealed excellent performance in degrading acetaminophen in the first minutes ( Q = 10 kJ m (cid:0) 2 ) under natural sunlight. Results reveal that 1% Fe content in BiOBr can degrade acetaminophen and its main byproduct (30 min, Q = 50 kJ m (cid:0) 2 ) at pH 5 and using 0.25 gL -1 of catalyst. A synergistic mechanism between heterogeneous photocatalysis and Fenton processes with primary superoxide ( • O 2 – ) radical, followed by hydroxyl ( • OH) radical and photogenerated holes ( h + ), is proposed. Our research contributes to the degradation of pharmaceuticals under mild conditions and sunlight irradiation.


Introduction
Acetaminophen (ACP) is an analgesic used as a first-choice treatment for pain and fever [1]; an average of 4% can be excreted by humans through the urinary tract after intake [2], yielding high concentrations of ACP residues in wastewater [3].The environmental impact of ACP is worrisome due to its high consumption during the COVID outbreak, which nearly tripled between 2019 and 2021 [4].Some studies reported ACP concentrations in wastewater up to 22.8 μg/L in North Scotland, 51.22 μg/L in North Mexico, 105.8 μg/L in Singapore, and 667 μg/L in Colombia [5][6][7][8].In response to pharmaceutic discharges in wastewater, the European Union has implemented measures through the Water Framework Directive (WFD) to monitor and evaluate health risks associated with emerging contaminants.The Watch List (WL) is updated every two years to track the presence of contaminants and issue legal regulations, if necessary, to limit their environmental impact.Since 2020, the WL has included antibiotics like amoxicillin, ciprofloxacin, sulfamethoxazole, and trimethoprim, among others, to evaluate bacterial resistance.However, until now, ACP (Fig. 1) is not in the WL despite its high consumption and discharge [9], probably because it is not used as an antibiotic, although similar painkillers have been demonstrated to have antimicrobial properties [10,11], which might lead to creating resistance in microorganisms [12].Like European countries, other countries like Mexico have not defined a permissible limit for the ACP in water.Other organizations, such as the Minnesota Department of Health (MDH) in the U.S.A., have established permitted levels of drinking water (200 μg/L) [13].Similar guidelines, like in the U.S.A., should be followed since large concentrations of APC in water (4.46 μg/L) can lead to ecotoxicological risk in aquatic ecosystems, e.g., Daphnia magna has a half maximal effective concentration (EC 50 ) of 2.04 mg/L [14].From this view, pharmaceuticals-free water is vital since it can lead to plant and animal genotoxic, mutagenic, and ecotoxicological effects [15].
Advanced oxidation processes (AOPs) can remove pharmaceuticals from water.The process involves the generation of radical species that oxidize organic molecules, such as ACP, until mineralization, i.e., the formation of innocuous products such as CO 2 , H 2 O, and inorganic salts.Among the AOPs, the photo-Fenton process produces hydroxyl radical • OH, a very reactive and highly oxidizing species.Homogeneous photo-Fenton uses iron salts to catalyze the decomposition of hydrogen peroxide to • OH.Nevertheless, it requires highly acidic conditions (pH 2-3) to avoid the formation of hydroxides, which leads to iron deactivation and limits photocatalytic activity.This is the case of photo-Fenton degradation of ACP, which has been studied to its finest detail [16][17][18].Other photo-Fenton approaches require chelation and/or immobilization of iron (Fe) on solid surfaces, which are attractive options to overcome deactivation [19].The approaches are known as heterogeneous Fenton or photo-Fenton if the light is involved in the degradation process [20].The heterogeneous photo-Fenton degradation mechanism is a synergistic combination of photocatalysis like Fe and a semiconductor.When the light reaches the semiconductor, the photogenerated electrons in the conduction band are used to accelerate the Fenton redox reaction on the photocatalyst surface (∋ Fe) by increasing the cycle rate of Fe 3+ /Fe 2+ , which promotes the decomposition of H 2 O 2 , yielding strongly oxidizing • OH [21].For example, in Eq (1) to (6): Among the heterogeneous photo-Fenton catalysts to degrade ACP under controlled UV or visible light include Fe 3 O 4 , Fe 2 O 3 , Fe 2 O 3 -TiO 2clay, αFe 2 O 3 /g-C 3 N 4 , F-C 3 N 4 /NiFe 2 O 4 and zero-valent Fe [22][23][24][25][26].
Recently, BiOBr, a p-type semiconductor, has received attention because of its narrow band gap (2.62-2.90eV) that allows degradation under visible light [27,28].Photo-Fenton activity can increase further with the incorporation of Fe.Diverse approaches have been used to synthesize Fe-BiOBr, including solvothermal and microwave-assisted solvothermal.
A benefit of microwave-assisted solvothermal is that it is less timeconsuming (~a few minutes to 1 h) [29][30][31] than conventional solvothermal (8 h) [32][33][34].Time reduction can be related to uniform heating by electromagnetic waves and ease photocatalyst synthesis [34,35].Applications of solvothermal synthesized Fe-BiOBr include the degradation of rhodamine B and bisphenol A [20], phenol [31], and atrazine [36], achieving complete degradation within 30 and 120 min.The main degradation mechanism has been suggested to be related to • OH and h + as the main oxidizing species [31], of great similitude to the described chemical mechanism in Eq (1) to (6).
This work demonstrates that ACP can be degraded under solar illumination and mild conditions using low Fe loadings in Fe-BiOBr photo-Fenton catalyst synthesized using a less time-consuming microwaveassisted solvothermal method.The BiOBr phase in Fe-BiOBr is confirmed with XRD.SEM and TEM demonstrated the flower-like morphology, in which crystallite size reduces as a function of the Fe loading.The chemical environment at the surface of Fe-BiOBr is investigated with XPS.The results are connected with Raman analysis, which suggests the presence of oxygen vacancies in Fe-BiOBr.Furthermore, the effect of Fe in BiOBr is assessed by determining the optical band with UV-Vis.The Fe-BiOBr functionality is assessed during ACP photo-Fenton catalytic activity, carried out for 0.25 mg L -1 of catalyst load and pH 5. The latest conditions are most favorable since most international regulations allow wastewater discharges to surface water with a pH of 5.5 to 9.5 [37,38].The Fe-BiOBr catalyst showed excellent performance in degrading ACP in less than 10 min (Q = 10 kJ m − 2 ) under natural sunlight.Our results can contribute to developing a Fenton photocatalyst to degrade pharmaceuticals under natural sunlight and mild conditions.

Photocatalysts synthesis
The photocatalysts xFe-BiOBr (x = 1 and 3 wt% Fe) were synthesized as follows: 3 mmol of Bi(NO 3 ) 3 ⋅5H 2 O and appropriate amounts of Fe (NO 3 ) 3 ⋅9H 2 O (0.19 and 0.56 mmol) were dissolved in 30 mL of ethylene glycol by sonication; similarly, 3 mmol of CTAB were also dissolved in EG (30 mL).Subsequently, both clear dissolutions were mixed and stirred vigorously for 20 min and transferred to Teflon vessels of MARS 6 equipment (CEM Corp. USA) which was operated at 160 • C, 20 min, and 450 W. Later, the microwave vessels were cooled to room temperature, and solids were recuperated by centrifugation.The prepared materials were naturally cooled to room temperature, repeatedly washed with ethanol and distilled water, and dried in an oven at 80 • C for 12 h.As a reference, pristine BiOBr was prepared using the same procedure without the addition of Fe precursor.

Structural and morphological characterization
The photocatalysts xFe-BiOBr (x = 1 and 3 wt% Fe) morphology was analyzed using scanning electron microscopy (SEM, JEOL JSM6510-LV) equipped with an energy-dispersive X-ray detector (EDX) and highresolution transmission electron microscope (HR-TEM), using FEI TITAN G2 80--300 (operated at 300 keV).X-ray diffraction (XRD) determined the catalysts' crystalline phase and crystallite size.XRD was carried out using a Bruker AXS Model D2 PHASER diffractometer (Cu Kα radiation, λ = 1.5406Å) at a scan rate of 0.1 • /s over a diffraction angle 2θ from 5 to 80 • .The average crystallite sizes of the photocatalysts have been calculated according to the classical Scherrer's equation: where D is crystallite size (nm), K is a constant equal to 0.9, λ is X-ray wavelength (1.5406 Å), β is a half-high width of the diffraction peak (1 1 0), and θ B is the 2θ diffraction angle [39].

Chemical composition and band gap determination
Chemical species in Fe-BiOBr were obtained using a Raman Thermo Scientific microscope with a laser diode as a radiation source (780 nm).Chemical species present at the surface of BiOBr and Fe-BiOBr were investigated with X-ray photoelectron spectroscopy (XPS).The spectra were recorded using a Thermofisher Nexsa G2 (Al-Kα, 1486.6 eV, 120 W).The amounts of Fe in the prepared photocatalysts were determined using atomic absorption spectroscopy (AAS, SpectrAA 220FS, Varian); before the AAS analysis, 0.015 g of each powder was digested in a mixture of HNO 3 and HCl (4:1) by microwave heating (MARS 6, CEM) for 20 min at 180 • C and 800 W. UV-Vis diffuse reflectance spectroscopy (UV-Vis/DRS) was recorded using a spectrophotometer Lambda 365 (Perkin Elmer) equipped with an integrating sphere using BaSO 4 as reference material.From the acquired spectra, the band gap (E g ) of the prepared catalysts was estimated by extrapolation from the plot of [F(R) hν] 1/n vs. hν, where F(R) is the Kubelka-Munk function obtained from the UV-Vis diffuse reflectance data, hν is the photon energy described by the Planck constant (h = 4.135 × 10 -15 eV·s), and the light frequency (ν) given by the ratio of the speed of light (3 × 10 8 m/s) and the wavelength (m); n depends on transition characteristics of the semiconductor, a value of n = 2 was used due to BiOBr have indirect transitions [40].

Textural analysis
The specific surface area (SSA) of the catalysts was estimated by the Brunauer-Emmett-Teller method (BET), and pore sizes by the Barret-Joyner-Halenda method (BJH) using an N 2 adsorption-desorption equipment (TriStar II Plus, Micromeritics).

Photocatalytic activity
The experiments were carried out under natural sunlight in Apodaca City in Nuevo Leon, Mexico (25 • 45′ N, 100 • 7′ O).The photocatalytic activity of the catalysts was evaluated on the degradation of ACP solution (15 mgL -1 ) (ACP, p-acetaminophen, C 8 H 9 NO 2 ) prepared using local drugs tablets (Medimart, 500 mg) in distilled water.The experiments were carried out in 100 mL of ACP solution on a Pyrex reactor, the pH was adjusted at 3 or 5 using HCl 0.1 M, and the catalyst added (0.25 gL -1 ) was stirred for 30 min in darkness to allow the adsorption-desorption equilibrium between ACP and catalyst.After that, 102 μL of H 2 O 2 (30% Fisherbrand, 10 mM) was added, and the reaction was conducted outdoors using natural sunlight.At regular intervals, samples were collected and filtered through a 0.45 μm nylon syringe filter to remove the photocatalyst.In solar degradation, once the light exposition started, the accumulated solar radiation was measured with a Delta OHM HD2102.2 radiometer (range: 315-400 nm), and samples were collected as needed, reaching a total of 300 kJ m − 2 for each experiment.ACP quantification was performed by liquid chromatography on HPLC Agilent Technologies 1260 Infinity equipment with diode array detector using a column Thermo Scientific Accucore C18 (150 × 4.6 mm).The mobile phase was a mixture of 25:75 v/v methanol (HPLC grade, Tedia) and 4% (v/v) acetic acid prepared in water, and the flow rate was 1.5 mL min − 1 .The injection volume was 25 μL, and the detection wavelength (λ) was 242 nm.Similar tests were carried out under simulated solar radiation until reaching an accumulated energy of 300 kJ m − 2 (Solar simulator Suntest, XLS + Model Atlas, Germany, equipped with a daylight filter that emits radiation from 300 to 800 nm).The mineralization degree was determined by the total organic carbon (TOC) diminished on a Shimadzu TOC-V CSH analyzer with an ASI-V model autosampler.Finally, to determine the main reactive species involved in ACP degradation, different scavengers of radicals were added in degradation tests under the same studied conditions.For this, terbutanol (TBA, 5 mM), p-benzoquinone (BQ, 0.5 mM), and sodium oxalate (OXA, 5 mM) were added as • OH, • O 2 -and h + quenchers.All experiments were carried out in duplicates.The experiments have been carried out in duplicates.The difference between each pair of experiments was less than 10%; in the end, all reached total ACP degradation.

Results and discussions
A BiOBr photocatalyst with Fe is synthesized via the microwaveassisted solvothermal method.The synergy between Fe and BiOBr is investigated morphologically, structurally, chemically, and optically.Fe-BiOBr functionality is assessed during the photocatalytic degradation of ACP under solar irradiation.Finally, a photocatalytic action mechanism is proposed.

Morphological and structural characteristics of Fe-BiOBr
The morphology of the synthesized BiOBr and BiOBr loaded with 1 wt% (1Fe-BiOBr) and 3 wt% (3Fe-BiOBr) of Fe is investigated with SEM, as shown in Fig. 2. In Fig. 2 (a), BiOBr has spherical flower-like morphology, which is assembled by nanosheets in the form of interlocking petals, with interstitial spaces between them [32].No apparent morphological differences exist between BiOBr and 1Fe-BiOBr and 3Fe-BiOBr in Fig. 2 (b) and 2 (c), showing the same morphology and behavior as BiOBr and Fe-BiOBr materials reported by others [31,41,42].Fig. 2 (d) shows an SEM-EDX map of the 1Fe-BiOBr image to compare with (d1) Bi, (d2) Br, and (d3) Fe elemental composition.This analysis reveals that the elements present in the samples are homogeneously distributed, confirming that the microwave-assisted solvothermal synthesis method is suitable for this material.
The morphological and structural properties are further investigated with TEM.Fig. 3 shows TEM, STEM-HAADF, and HR-TEM images for (a, b, c) BiOBr, (d, e, f) 1Fe-BiOBr, and (g, h, i) 3Fe-BiOBr.TEM images (a, d, and g) reveal that the structure retained a flower-like shape with a size of approximately 1.5 µm and did not show morphological changes as the Fe content increased.However, it seems that the 3Fe-BiOBr is less dense than BiOBr and 1Fe-BiOBr.STEM-HAADF images (b, e, and h) confirm the intercalated sheet-like morphology characteristics of BiOBr.Nevertheless, 3Fe-BiOBr has a bright core and darker spike-like features in a different sample area; bright dark regions can be associated with regions with heavier atoms, most probably from Bi. HR-TEM images (c, f, and i) show the presence of crystals with sizes around 10 nm.The crystal size decreases with increasing iron content, which is consistent with the XRD results in Fig. 4a.In Fig. 3, high crystallinity and lattice fringes with an interplanar lattice spacing of 0.36, 0.23 and 0.27 nm are estimated and assigned to BiOBr (1 0 1), (1 1 2), and (1 1 0) crystallographic planes.1Fe-BiOBr and 3Fe-BiOBr exhibit slightly smaller crystallites compared to BiOBr.
The optical properties of BiOBr, 1Fe-BiOBr, and 3Fe-BiOBr are measured using UV-Vis DRS in Fig. 4b to estimate and compare the bandgap.BiOBr has a steep increase of absorption at wavelengths shorter than 413 nm, which can be assigned to the intrinsic band gap of pure BiOBr (~3.00 eV) [45].The spectra of 1Fe-BiOBr, and 3Fe-BiOBr exhibit a redshift and increased photoabsorption in the visible light range and near-infrared region [31,32], which can indicate the Fe   loading.This observation is also reflected in the color change of these materials from white to reddish yellow.A possible mechanism for such redshift may be related to the transition of electrons between the conduction band or valence band of BiOBr and Fe ions or the internal transfer charge between Fe ions (Fe 3+ + Fe 3+ → Fe 4+ + Fe 2+ ) [32,46,47].The band gap (E g ) for 1Fe-BiOBr and 3Fe-BiOBr are in the order of 2.73 and 2.75 eV (Table 1).
Raman analysis in Fig. 4c shows the characteristic bands of BiOBr observed at 56.8, 95.0, 112.4,and 163.0 cm − 1 [48], which are assigned to the internal Bi-Br stretching modes [49].Likewise, a weak and broad signal corresponding to the motion of oxygen atoms is observed at 385 cm − 1 [50], and the signal at 86 cm − 1 is ascribed to the formation of oxygen vacancies (OVs) [51].It should be noted that the Raman signal for BiOBr is the most intense, indicating its higher crystallinity than the 1Fe-BiOBr and 3Fe-BiOBr.Furthermore, the decrease in the intensity of the Raman signal is attributed to the formation of oxygen vacancies [52].These results are coherent with TEM (Fig. 3) and XRD (Fig. 4a) analyses.Compared to BiOBr, a blue shift for 1Fe-BiOBr and 3Fe-BiOBr is observed.These variations could be associated with changes in the structural and chemical environment in BiOBr due to the incorporation of Fe.

Chemical species at the surface of Fe-BiOBr
XPS analysis confirmed the elemental compositions of BiOBr, 1Fe-BiOBr, and 3Fe-BiOBr.The XPS spectra are shown in Fig. 5; the signals present in samples are Br 3d (a), Bi 4f (b), O 1 s (c), and Fe 2p (d).The high-resolution XPS Fe 2p core level spectra illustrate the increasing iron concentration in BiOBr, 1Fe-BiOBr, and 3Fe-BiOBr (Fig. 5d).The Fe 2p3/2 and Fe 2p1/2 contributions located at ca. 711.5 eV and ca.725.1 eV suggest the insertion of Fe 3+ in BiOBr [31,53,54].The presence of Fe in 1Fe-BiOBr before and after the reaction is shown in Fig. S2; no major differences in Bi, O, Br, and Fe content have been found.However, in Table S1, a slight reduction in the Fe content in the used catalyst is observed, which can be explained by an increase in the carbon content due to the pollutant adsorption or a low iron leaching [55,56].For all studied samples, the high-resolution XPS spectra of the Br 3d (Fig. 5a)   and Bi 4f core levels (Fig. 5b) revealed the contribution of doublet pairs, i.e., Br 3d5/2 at ca. 68.7 eV and Br 3d3/2 at ca. 69.7 eV and Bi 4f7/2 at ca. 159.4 eV and Bi 4f5/2 at ca. 164.8 eV (indicating the presence of Bi 3+ in the materials) [57,58].The high-resolution XPS O 1 s core level spectra all exhibit a dominant contribution at ca. 530.5 eV attributed to O in BiOBr (Fig. 5c).Additional contributions, ascribed to surfaceadsorbed oxygen, H 2 O, and -OH groups [54,57,59], are also detected at higher binding energies (ca.532.0-534.0eV).

Photocatalytic activity of Fe-BiOBr
The solar photo-Fenton is conducted under natural sunlight to oxidate ACP using Fe-BiOBr is showcased.Fig. 6 shows the degradation of ACP [mg/L] as a function of the solar accumulated energy Q [kJ/m 2 ].For 1Fe-BiOBr and 3Fe-BiOBr in Fig. 6a and 6b, ACP is degraded in the first minutes (Q = 10 kJ m − 2 ) (closed circles and squares).Concurrently, the ACP byproduct is formed (open circles and squares) and progressively degraded.For BiOBr, ACP degradation is not achieved until 240 kJ/m 2 (i.e., 2.5 h) in Fig. S3.Fig. 6a and 6b show that 15 mg/L of ACP and its main byproduct is fully degraded in 30 min under natural solar light and 10 mM of H 2 O 2 .It is important to highlight that under solar irradiation, 1Fe-BiOBr and 3Fe-BiOBr degrade ACP at mild pH conditions, i.e., close to pH 5.At pH 5, more byproduct (Bypr) is produced (Fig. 6a and 6b), similar to the obtained by homogeneous solar photo-Fenton, where hydroxylated species are initially formed as intermediates, improving the degradation of ACP.Such species can lead to an improvement in the reduction of Fe 3+ to Fe 2+ in comparison to more recalcitrant intermediates produced, such as acetamide, hydroquinone, or benzoquinone (Fig. S4) [17,60].Regarding the mineralization percentage, this has been measured using TOC (Fig. 6c).The 1Fe-BiOBr achieved the highest mineralization of ACP ca.58% at pH 5, similar to what was reported for atrazine and bisphenol A, under visible light, but at pH 3 using similar catalysts [31,36].The results represent a great advantage of our system under natural solar light and pH 5, considered mild conditions.
The results demonstrate that 1Fe-BiOBr promotes the degradation of ACP.Since the estimated Eg are similar (Fig. 4b), variation in the product degradation can be related to the higher degree of crystallinity of 1Fe-BiOBr shown in Fig. 3f and Fig. 4a.Higher degree of crystallinity of 1Fe-BiOBr can promote the necessary pathway to charge carriers to be readily available, possibly degrading ACP side-products more efficiently [61].Moreover, the effect of OVs in 1Fe-BiOBr should not be disregarded.
We further essay APC degradation under controlled solar irradiance (i.e., solar simulator), using 1Fe-BiOBr at pH 3 and pH 5. The results compare with ACP photo-oxidation without H 2 O 2 (photocatalysis) and with H 2 O 2 (photo-Fenton) in Fig. 7.The photocatalytic degradation of ACP shows a significantly better performance in the presence of H 2 O 2 (Fig. 7a).At pH 5 ACP oxidation without H 2 O 2 is slightly better; the results are in accordance with other studies and relate to the surface interaction between ACP and the catalyst [62].The results of the heterogeneous photo-Fenton process (with H 2 O 2 ) show that the pH has a low effect since similar outcomes are obtained at pH 3 and 5 (Fig. 7a).Nevertheless, pH 5 is attractive for the process because it is well-known that the limitation of the Fenton reaction often needs more acidic conditions (near pH 3) to achieve maximum performance and avoid iron precipitation at higher values.On the other hand, iron leaching in an acidic medium has been widely reported and depends on the catalyst's stability.However, it has been observed that low iron leaching and the presence of chelating agents (such as organic acids or byproducts) could contribute to contaminant elimination through the homogeneous Fenton reaction [55,56].This phenomenon could explain the slight degradation increase at pH 3 in Fig. 7c compared to these observed at pH 5.
The effectiveness of natural solar illumination is contrasted with controlled solar irradiance.Compared with solar irradiation (Fig. 6), results in Fig. 7a show longer times for ACP degradation, close to 2 h (Q = 150 kJ m − 2 ).Furthermore, for the byproducts, a longer time is needed for degradation, ca. 4 h (Fig. 7b).The main difference between the photocatalytic experiments under solar light and controlled solar irradiance is the solar energy contribution, thus affecting the degradation times.In other words, the solar spectrum has UV, Vis, and IR contributions, while controlled solar irradiance might not fully include UV contributions (Fig. S5).The results explain the benefit of natural solar light-induced reactions, which use a significant portion of the solar spectrum.
It is well known that heterogeneous photo-Fenton results from a synergistic combination of photocatalysis and Fenton reaction processes.Therefore, it is important to compare the mineralization with and without H 2 O 2 (Fig. 7c).Here, higher mineralization is observed in heterogeneous photo-Fenton than in photocatalysis.In the latest, the • OH is mainly produced by splitting water in the valence band, while in heterogeneous photo-Fenton, the possibility of producing -OH is greatly increased, and other oxidizing species may also play an important role.To clarify the main species involved in ACP degradation by the synergistic combination of photocatalytic and Fenton processes using 1Fe-BiOBr, p-benzoquinone (BQ), sodium oxalate (OXA), and tert-butanol (TBA) are used as radical scavengers of superoxide radicals ( • O 2 -), h + and • OH, respectively [30].As seen in Fig. 8, total ACP degradation is achieved using 1Fe-BiOBr + H 2 O 2 without adding any quencher.In contrast, when TBA and OXA are added during degradation, diminishes 20% and 30% of ACP, which proved that • OH and h + play a significant role in the heterogeneous photo-Fenton process.However, the ACP degradation dropped abruptly (60 %) in the presence of BQ, indicating that • O - 2 is the most important active species in the synergistic degradation process.The results demonstrate that • O 2 -, followed by h +, and • OH contribute to the degradation of ACP, prominently produced under

Proposed mechanism for the photo-Fenton system
From our results, the proposed reaction mechanism for ACP degradation by heterogeneous photo-Fenton using Fe-BiOBr considers the synergy between photocatalysis and Fenton processes.The mechanism from Fig. 9 considers solar irradiation, which promotes photoelectrons (e -) generation at the CB and holes (h + ) at the VB; the e -is moved to the surface of BiOBr and is used to reduce Fe 3+ (Fe(III)) to Fe 2+ (Fe(II)).This leads the Fe(III)/Fe(II) cycle over the BiOBr surface and promotes the active generation of • OH in the presence of H 2 O 2 .Within this cycle, Fe (II) can react with oxygen O 2 in the presence of protons (H + ) to generate Fe (III), which can then be reduced and participate in further cycles of the photo-Fenton reaction.In addition, the e -available in the OVs can react with the H 2 O 2 to form • OH.On the other hand, the h + in the VB contributes to the • OH generation by splitting the H 2 O molecule.Also, h + can degrade ACP by direct oxidation.It is worth highlighting the importance of OVs in the process.OVs involve releasing two electrons per removed oxygen on the surface [63] that can act as an electron pump, leading to the formation of • O 2 -radicals.It is proposed that • O 2 -is the most important oxidant species for this system, in which the generation takes place in three ways: i) traditionally, the dissolved oxygen reacts with e-photogenerated or those in the OV to form These mechanisms inhibit the recombination of photogenerated h + / e -pairs and favor the Fenton degradation process.The proposed mechanism can explain the observed effect for 1Fe-BiOBr, which shows a higher ACP degradation.

Conclusions
Fe-BiOBr has been successfully synthesized by microwave-assisted solvothermal method and exhibited excellent degradation activity in the solar photo-Fenton system, which allowed the total and almost immediate ACP removal at mild conditions.The results highlight that 1% Fe content in BiOBr is enough to degrade ACP and its main byproduct (30 min, Q = 50 kJ m − 2 ) at pH 5 and using 0.25 gL -1 of catalyst loading.The incorporated Fe in the BiOBr with a spherical flower-like shape decreased the recombination of photocarriers e -/h + and favored the   generation of OVs.The solar photo-Fenton yield for ACP degradation is attributed to the synergistic combination of heterogeneous photocatalysis and Fenton reaction processes.This leads the Fe(III)/Fe(II) cycle over the surface catalyst in the presence of light and H 2 O 2 .The main active species involved in this process were confirmed to be superoxide radicals ( • O 2 -), followed by photogenerated holes (h + ) and hydroxyl radicals ( • OH).Our research reliably contributes to the development of Fenton photocatalysts for degrading pharmaceuticals under mild conditions and natural sunlight irradiation.

Fig. 9 .
Fig. 9. Proposed reactions mechanism by heterogeneous photo-Fenton for ACP degradation under solar light.