Degradation and mineralization of 4-tert-butylphenol in water using Fe-doped TiO2 catalysts

In the present work, the photocatalytic degradation and mineralization of 4-tert-butylphenol in water was studied using Fe-doped TiO2 nanoparticles under UV light irradiation. Fe-doped TiO2 catalysts (0.5, 1, 2 and 4 wt.%) were prepared using wet impregnation and characterized via SEM/EDS, XRD, XRF and TEM, while their photocatalytic activity and stability was attended via total organic carbon, 4-tert-butyl phenol, acetic acid, formic acid and leached iron concentrations measurements. The effect of H2O2 addition was also examined. The 4% Fe/TiO2 demonstrated the highest photocatalytic efficiency in terms of total organic carbon removal (86%). The application of UV/H2O2 resulted in 31% total organic carbon removal and 100% 4-t-butylphenol conversion, however combining Fe/TiO2 catalysts with H2O2 under UV irradiation did not improve the photocatalytic performance. Increasing the content of iron on the catalyst from 0.5 to 4% considerably decreased the intermediates formed and increased the production of carbon dioxide. The photocatalytic degradation of 4-tert-butylphenol followed pseudo-second order kinetics. Leaching of iron was observed mainly in the case of 4% Fe/TiO2, but it was considered negligible taking into account the iron load on catalysts. The electric energy per order was found in the range of 28–147 kWh/m3/order and increased with increasing the iron content of the catalyst.

Heterogeneous photocatalysis belongs to the Advanced Oxidation Processes (AOPs), which have been effectively used for the removal of recalcitrant organic compounds 16 . The efficiency of AOPs depends on the formation of highly active free radicals 17 , which are atoms or molecules with one or several unpaired electrons 18 . Among various oxidative agents, hydroxyl radicals play a central role in AOPs for effluent treatment 19 . They are used as highly reactive species in UV/H 2 O 2 , photo-Fenton/Fenton-like systems, photocatalytic oxidation, and others 20 .
Titanium dioxide has high stability to light irradiation, relatively high activity, low cost and non-toxicity among many semiconductor photocatalysts 21 . It can exist in three crystallographic forms, namely anatase, rutile and brookite 22 . Degussa P25 is a mixture of 70% anatase and 30% rutile, and improved degradation efficiencies have been reported with its use compared to other photocatalysts 21 . TiO 2 -mediated photocatalysis has been widely used in the wastewater treatment. The benefits of this process include work under ambient conditions, the absence of mass transfer limitations when nanoparticles are used as photocatalysts, highly oxidizing photogenerated holes, cheap and readily available forms of TiO 2 , and the possibility of using solar irradiation 23 . TiO 2 can oxidize various organic substances ultimately to CO 2 and H 2 O 24 .
An increasing number of studies is focused on improving the efficiency of photocatalysts by either expanding the absorption spectrum to visible light or slowing down the recombination rate between electrons and holes and increasing thus the efficiency of interfacial charge transfer. This is attempted by doping TiO 2 with metals, metal ions, non-metal atoms, and semiconducting oxides 25 .
Wang et al. 26 used templates of silver oxide octahedra and titanium tetrafluoride as precursor to prepare hollow octahedra of silver-modified titanium dioxide. The prepared catalysts showed enhanced photocatalytic performance as a result of fast electron transfer between titanium dioxide and silver nanoparticles. Moreover, Wang et al. 27 applied a combination of hydrothermal and photodeposition techniques to prepare a novel Ag/F-TiO 2 photocatalyst, which exhibited improved photocatalytic activity compared to the reference TiO 2 , F-TiO 2 , and Ag/ TiO 2 catalysts.
The dehydrogenative methane coupling to ethane was conducted with increased energy efficiency on a gold-modified TiO 2 photocatalyst, which was synthesized via the photodegradation of gold nanoparticles on the polar {001} facet of titanium dioxide in the anatase form 28 . Similarly, the efficiency of solar-to-hydrogen conversion was increased by a factor of 64 via introducing gold nanoparticles on the {001} facets of nanosheets of anatase 29 .
A photodeposition technique was also applied by Gao et al. 30 to prepare Sn/TiO 2 photocatalysts with considerably enhanced performance in H 2 generation compared to that of the titanium dioxide base catalyst.
Huang et al. 31 achieved a 0.56% quantum efficiency in the carbon dioxide reduction to form methane under visible light illumination by means of a novel triad photocatalyst based on a mononuclear C 5 H 5 -RuH complex oxo-bridged TiO 2 hybrid.
A graphene-titanium dioxide catalyst with magnetic properties exhibited higher removals than the base titanium dioxide during the treatment of Methylene Blue and tetrabromobisphenol A in water 32 .
The type-II nanostructured TiO 2 @Ta 2 O x N y nanorods photoanodes were shown to achieve a 12 fold-enhanced photoelectrocatalytic water splitting efficiency under solar light irradiation as well as a solar-to-chemical energy conversion efficiency of ca. 1.49% at 1.23 V vs RHE 33 .
A novel Bi-BiOI/GR composite photocatalyst prepared solvothermally showed improved performance in the oxidation of nitric oxide under visible light irradiation in relation with the pure BiOI photocatalyst 34 .
The photocatalytic activity of Fe-doped TiO 2 nanoparticles has been examined in a number of studies [35][36][37][38][39][40][41][42][43] . Choi et al. 35 studied the effect of doping 21 metal ions into TiO 2 and found that doping with Fe 3+ at 0.1-0.5% significantly increased the photoreactivity for both oxidation and reduction. Anwar et al. 41 synthesized Fe-doped TiO 2 catalyst with the concentration of 6 wt.% iron through a sol-gel method, and the Fe/TiO 2 catalyst efficiently degraded methylene blue dye under UV and visible light irradiation.
The presence of iron particles can favourably affect the photocatalytic activity, which may be due to the role of iron particles acting as h + /e − traps, thereby inhibiting the recombination rate and enhancing the photocatalytic activity 44 . However, further study of Fe-doped TiO 2 catalysts is required, and 4-t-BP, a dangerous endocrine disruptor, is an ideal target compound in water as it has not been studied previously.
In this work, Fe-doped TiO 2 catalysts with different iron content (Fe/Ti weight ratio percentage = 0.5%, 1%, 2% and 4%) were prepared and used to eliminate 4-tert-butylphenol in water under ultraviolet light irradiation. The effect of adding hydrogen peroxide in the photocatalytic system was also investigated. Although hundreds or even thousands of studies exist for target compounds in water like phenol, only a very limited number of works have been devoted to 4-tert-butylphenol, which is an endocrine disruptor and a serious emerging environmental concern. To the best of our knowledge, a study on the photocatalytic mineralization of 4-tert-butylphenol in water using Fe-doped TiO 2 catalysts has not been conducted before.

Results and Discussion
Characterization of catalysts. The XRD patterns of Fe-doped and undoped TiO 2 catalysts are shown in  45 . The X-ray diffraction pattern of anatase has a major peak at 2θ = 25.3°. The XRD patterns for all synthesized catalysts were similar to the one for the base TiO 2 catalyst. No crystalline iron-related phase was observed, which has been also reported in similar studies previously 42,46,47 . This result can be explained by the fact that crystalline forms of Fe were not formed on the material or that particles of amorphous iron oxides were too small on the surface of TiO 2 particles 47 . In addition, the peak associated with iron cannot be observed in the XRD spectra when all iron ions are either incorporated into the TiO 2 structures replacing titanium ions or are at interstitial site due to similar ionic radii (Ti (0.68 Å) and Fe (0.64 Å)) 42,46 . www.nature.com/scientificreports www.nature.com/scientificreports/ The TiO 2 crystallite sizes calculated by the Scherrer equation on the anatase diffraction peak (2θ = 25.3°) are listed in Table 1. The average particle size of Fe/TiO 2 catalysts was found in the range of 17-21 nm, which is agreement with the TiO 2 specifications.
XRF analysis revealed the iron loading of the synthesized 4% Fe/TiO 2 catalyst as Fe 2 O 3 /TiO 2 = 3.74%. The results of XRF analysis are presented in Table 2. The chemical composition of the synthesized catalyst revealed that it was consisted of TiO 2 (96.174%), Fe 2 O 3 (3.597%) and CaO (0.078%), SiO 2 (0.011%), Cl (0.139%) and the ratio of Fe 2 O 3 /TiO 2 was 3.597%/96.174% = 3.74%. These results confirmed the existence of iron as dopant in the sample and showed that iron was in its oxide form.
Elemental mappings (SEM/EDS) for all Fe/TiO 2 catalysts are presented in Figs. 2 and 3. SEM/EDS provided information on the elements present and their quantities. A well-distributed iron phase on the surface of TiO 2 was   Fig. 4. It was found that the particle size was around 20 nm. The optical properties of fresh catalysts were measured using a UV-Vis spectroscopy in the wavelength range of 200 to 750 nm (Fig. 5). The prepared Fe/TiO 2 catalysts had relatively the same absorption as the base TiO 2 , and only the 4% Fe/TiO 2 showed a significantly higher absorption in the range of 200-365 nm.
The shift of the absorption range of 4% Fe/TiO 2 can be explained by the transition of charge transfer between the d-electrons of the iron ion and the conduction or valence band of TiO 2 48 . Photocatalytic treatment of 4-t-BP. Titanium dioxide semiconductor absorbs ultraviolet light and generates hydroxyl radicals according to the following reactions 49 : Oxidative degradation of organic compounds can occur through their reactions with hydroxyl and peroxide radicals, valence band holes, and reductive splitting through their reactions with electrons 23 .   The presence of iron particles can favourably affect the photocatalytic activity, which may be due to the role of iron particles acting as h + /e − traps, thereby inhibiting the recombination rate and enhancing the photocatalytic activity 44 . Equations (4)(5)(6)(7)(8) show the detailed reaction steps 50,51 :    4 3 However, when the concentration of iron is high, iron ions can also act as recombination centres for h + /e − pairs, according to Eqs. (9) and (10) 50 : In addition, an excess of deposited iron on TiO 2 can form Fe (OH) 2+ particles with higher light absorption in the range of 290-400 nm compared to TiO 2 . The competition in the absorption of photons subtracts the photon to TiO 2 and can reduce thus the photocatalytic activity of Fe/TiO 2 catalysts 52 .
The overall reaction of photocatalytic degradation of 4-tert-butylphenol can be represented by Eq. (11): The initial pH of the solution was around 6.3, and it decreased to about 3.8 at the end of the experiments. 4-tert-butylphenol was effectively removed by all studied catalysts, as it is shown in Fig. 6. The maximum removal of 4-t-BP was observed with the base TiO 2 catalyst reaching 97%. All Fe-doped TiO 2 catalysts led to 92-93% removal of 4-t-BP even though they exhibited superior adsorption capacity.  www.nature.com/scientificreports www.nature.com/scientificreports/ For the estimation of adsorption capacity, 0.25 g of each catalyst was added to 0.25 L of aqueous solution containing 4-tert-butylphenol (30 mg L −1 ). The solution was then stirred for 1 hour in the dark, before UV irradiation, so that the system reached adsorption equilibrium. The equilibrium adsorption capacity was calculated according to Eq. (12): The results obtained are presented in Fig. 7. Increasing the iron content of the catalyst led to increasing q e values, which however were low compared to values reported elsewhere 53 .
Adsorption can be either beneficial or unfavourable to photocatalysis. For example, Lu Lin et al. 54 studied the adsorption and photocatalytic oxidation of ibuprofen using nanocomposites of TiO 2 nanofibers with boron nitride nanosheets, and they reported that the adsorption of the target compound molecules resulted in increased photocatalytic degradation rates due to better transfer of photogenerated radicals on the catalytic surface in some cases, while for some catalysts the photocatalytic process was adversely affected as a result of the screening of light access to the catalysis by adsorbed molecules.
To investigate the kinetics of 4-t-BP photocatalytic degradation, a pseudo-first order and a pseudo-second order kinetic model were tested as represented by the following equations, respectively: where k 1 and k 2 are the corresponding reaction rate coefficients. The experimental data were best fitted by the pseudo-second order model as presented in Fig. 8. The reaction rate coefficients calculated are shown in Table 3.  www.nature.com/scientificreports www.nature.com/scientificreports/ TOC removals for all catalysts are shown in Fig. 9. After 60 min of UV irradiation, 73% of the initial TOC was eliminated with the base TiO 2 catalyst. The TOC removal increased with increasing the iron doping content from 0.5% to 4%, resulting in 65% and 86% removal, respectively. The 2% Fe/TiO 2 catalyst led to the same TOC removal as TiO 2 , whereas 4% Fe/TiO 2 had higher removal efficiency than TiO 2 . Controversial results are reported in literature regarding the effect of iron doping concentration. Specifically, according to Zhu et al. 38 , the 0.40% Fe-TiO 2 showed a higher photoactivity than both the undoped TiO 2 and the commercial photocatalyst Degussa P25 under UV irradiation. Much more oxygen vacancies in the crystal lattice and on TiO 2 surface were introduced by doping with iron. Zhao et al. 47 studied the photocatalytic degradation of 4-nitrophenol using Fe-doped (1, 3, 5 and 8 wt.% Fe) TiO 2 catalysts. The catalysts with the lowest Fe content (1%) showed a considerably better behaviour than TiO 2 and the TiO 2 catalysts with higher Fe contents. On the other hand, Nikolaki et al. 43 showed that the 10% Fe/TiO 2 catalyst efficiently degraded 1,3-dichloro-2-propanol in water. Anwar et al. 41 synthesized 6 wt.% Fe/TiO 2 catalyst, which exhibited high performance in the degradation of methylene blue under UV and visible light irradiation.
The higher TOC removals observed compared to 4-t-BP removals can be attributed to the formation of 4-t-BP intermediates, which were more resistant to the photocatalytic process than the parent compound.
In the irradiated solution, there is competition between the parent pollutant and the formed intermediate compounds for oxidizing agents 55 . The base TiO 2 catalyst seems to have favoured the conversion of 4-t-BP, while the 4% Fe/TiO 2 catalyst clearly promoted the carbon oxidation to CO 2 .
Effect of H 2 o 2 addition. The combination of catalysts and hydrogen peroxide under UV irradiation was also investigated. The purpose was to examine whether the photocatalytic degradation of 4-t-BP could be enhanced as a result of heterogeneous photo-Fenton reaction 47 .
The Fenton method is the oldest and most used chemical AOP, in which the Fenton's reagent, a mixture of a soluble iron(II) salt (catalyst) and hydrogen peroxide (oxidant), is used to destroy recalcitrant organic compounds 56,57 . The classical Fenton reaction under UV irradiation is called photo-Fenton process (Eq. 15), which enhances the catalytic reduction of Fe 3+ into Fe 2+ in H 2 O 2 aqueous solutions, thereby increasing the generation of •OH radicals 58 : The heterogeneous photo-Fenton degradation of 4-t-BP can have a beneficial effect due to an increase in the concentration of generated hydroxyl radicals 47 :  Table 3. Reaction rate coefficients considering pseudo-second order kinetics for 4-tert-butylphenol photocatalytic degradation.  The photochemical treatment by means of UV/H 2 O 2 was also studied for comparison purposes. 100% 4-t-BP degradation (after only 30 minutes) and 31% TOC removal were obtained, as shown in Figs. 10 and 11, respectively. Although H 2 O 2 led to complete conversion of 4-t-BP faster than all catalysts tested, it showed much less activity in the oxidation of carbon to CO 2 . In the presence of hydrogen peroxide, the 4-t-BP degradation was adversely affected by increasing the iron content on the catalyst; the 4-t-BP removal decreased from 95% for 0.5% Fe/TiO 2 catalyst to 87% for 4% Fe/TiO 2 .
Regardless of the Fe doping concentration, both 0.5% and 4% Fe/TiO 2 catalysts showed final TOC removals in the range of 59-61%.
The results showed that the addition of hydrogen peroxide was not beneficial for the photocatalytic treatment in terms of TOC removal. Previously reported results on the efficiency of combining hydrogen peroxide with heterogeneous catalysts under UV irradiation are contradictory. Some studies reported that this combination increased the efficiency of the process, since UV light was combined with both the oxidant and the photocatalyst 59 , while others claimed that the efficiency of the process was reduced due to the competition for ultraviolet irradiation between the oxidant and the photocatalyst 60 or due to H 2 O 2 adsorption on the surface of catalytic particles, which reduced the activity of the catalyst 61 . In addition, it could be suggested that hydrogen peroxide was consumed at the beginning of the process, and the rest of the organic matter was removed by the catalyst 55 .

Formation of intermediates.
Acetic and formic acid were quantified, as they are the last intermediates during the decomposition of organic compounds before the formation of CO 2 and H 2 O. The formation of acetic acid and formic acid increased with increasing the iron content on the catalyst. The concentrations of these acids increased also with the addition of H 2 O 2 .
The measurements of the concentrations of the parent compound, acetic and formic acid, and of organic carbon, allowed the indirect calculation of the carbon corresponding to the intermediates formed (other than acetic and formic acid) during the photocatalytic degradation of 4-t-BP. Specifically, the concentration of intermediates and CO 2 at the end of the process can be calculated according to the following material balances expressed in terms of organic carbon (OC), respectively:   The results obtained are presented in Fig. 12. Although the 4-t-BP conversion was practically the same, the increase in iron content on the catalyst from 0.5% to 4% decreased considerably the carbon of intermediates, increasing in parallel the formation of CO 2 . It is noteworthy that the photochemical degradation via H 2 O 2 led quickly to 100% removal of 4-t-BP without increasing the formation of intermediates and acids. The use of iron catalysts led again to decreased concentrations of intermediates and acetic and formic acids, while it enhanced the generation of CO 2 .
The path of the photo-degradation of 4-t-BP in aqueous solutions has not been studied in detail. Wu et al. 15  Stability of Fe/TiO 2 catalysts. The stability of the catalysts prepared was examined by determining the iron released in the solution at the end of photocatalytic experiments. Fe leaching was observed for 4% Fe/TiO 2 during all experiments. Specifically, 0.01 ppm of Fe were detected for 0.5-2% Fe/TiO 2 catalysts and 0.14 ppm for the 4% Fe/TiO 2 catalyst. The presence of hydrogen peroxide contributed to iron leaching, since 0.04 ppm of Fe were detected for the 0.5% Fe/TiO 2 catalyst and 0.40 ppm for the 4% Fe/TiO 2 catalyst. In the case of 4% Fe/TiO 2 , the iron concentrations measured in the solution corresponded to 0.35% and 1% of the Fe amount on the catalyst in the absence and in the presence of hydrogen peroxide, respectively. So, the extent of Fe leaching can be considered insignificant compared to the total amount of iron deposited on the surface of TiO 2 , and therefore it did not play any role in the performance of the catalysts.

Energy consumption. An assessment of the energy consumption of the ultraviolet lamp is important because
it increases the operating cost of wastewater treatment. The electric energy per order, E EO (kWh/m 3 /order), is commonly used, which can be estimated using Eq. (22) for a batch reactor 63 : TOC values were used for C o and C f , and the results obtained are shown in Table 4. The values obtained were in the range 28-147 kWh/m 3 /order. It is obvious that the effect of the process applied on the energy consumed was important. The addition of hydrogen peroxide affected E EO resulting in higher energy consumptions. Increasing the iron doping of TiO 2 catalysts decreased the energy consumption. The results are in accordance with previously www.nature.com/scientificreports www.nature.com/scientificreports/ reported ones. Specifically, Foteinis et al. 64 estimated E EO values in the range of 4-958 kWh/m 3 /order in the photochemical oxidation of an endocrine disrupting micro-pollutant. The authors reported a high dependence of energy consumption on the process used and the fact that with the addition of small amounts of oxidative agents, the environmental impact can be significantly reduced.

Materials and Methods
Reagents. 4-tert-butylphenol (99%) used as target pollutant, and titanium (IV) oxide (P25, nanopowder, 21 nm primary particle size, ≥99.5%) used as the base photocatalyst, were supplied by Sigma-Aldrich. The ultrapure water used in all experiments was obtained by means of Direct-Q 3UV water purification system. Iron (II) chloride (98%) purchased from Sigma-Aldrich was used for doping the TiO 2 photocatalyst. Hydrogen peroxide solution (37.6% w/w) received from Skat-Reactiv company was used as source of hydroxyl radicals. All chemical reagents were used without further purification.
Preparation of 4-tert-butylphenol solution. 1000 mL stock solution of 4-tert-butylphenol with a concentration of 300 ppm was prepared in ultrapure water. The solutions used were prepared by further suitable dilution of the stock solution to the desired concentration of 30 ppm of 4-t-BP. The initial 4-t-BP concentration was measured via High-Pressure Liquid Chromatography as 30 ± 1 ppm. The initial total organic carbon of 4-tert-butylphenol (C 10 H 14 O, MW = 150.22 g mol −1 , carbon present = 79.88% w/w) in the aqueous solution was measured via Total Organic Carbon analysis as 24.5 ± 0.5 mg L −1 , which is close to the calculated theoretical value of 23.96 mg L −1 . The stock solution was prepared on weekly basis and it was stored at 5.6 °C.
Synthesis of Fe-doped TiO 2 catalysts. Iron-doped TiO 2 catalysts were synthesized with dopant concentrations of 0.5, 1, 2 and 4 wt.% using the wet impregnation method 65 . Specifically, 3 g of TiO 2 (P25) were suspended in 100 ml of ultrapure water, and then the required amount of FeCl 2 was added. The obtained mixture was constantly stirred for 24 hours and washed three times with distilled water to remove any precursor of physical adsorption before drying in an air oven at 80 °C for 12 hours. The dried solids were ground in a mortar and calcined at 500 °C for 6 hours in a muffle furnace.
Characterization of Fe-doped TiO 2 catalysts. The catalysts were characterized using X-ray powder diffraction (XRD, SmartLab automated multipurpose X-ray Diffractometer purchased from Rigaku), scanning electron microscopy (SEM, Auriga Cross Beam 540, Carl Zeiss) and transmission electron microscopy (TEM, JEM-2100 from Jeol Ltd., Japan equipped by EDS system, Inca Energy 350, Oxford Instruments PLC, UK). The XRD pattern was recorded using Cu Kα radiation, and the 2θ ranged from 10 to 80°. The average size of TiO 2 nanoparticles was calculated based on the Scherrer equation 65 : where D is the crystallite size of the catalyst, λ is the X-ray wavelength (1.54060 Å), β is the full width at half maximum of the diffraction peak and θ is the diffraction angle. The semi-quantitative (standard-less) chemical composition analysis of 4% Fe-TiO 2 was carried out using X-ray fluorescence spectrometer (XRF, PANalytical Axios, UK). The sample weight 0.5 g and was prepared as homogeneous powder, which was analysed under He atmosphere.
In addition, UV-VIS spectroscopy analysis (Evolution 60 S UV-Visible spectrophotometer, Thermo Fisher Scientific) of all catalysts was carried out in the wavelength range from 200 to 750 nm. Prior to the UV-Vis analysis, the catalysts were sonicated in ethanol for 5 minutes.
Reactor configuration and experimental procedure. Experiments were carried out using the apparatus shown in Fig. 13. The total volume of the treated solution was 250 mL. An annular photoreactor of 56.8 mL irradiated volume, operated in batch recycle mode, was used. The solution was continuously pumped through the photoreactor at the rate of 178 mL min −1 using a peristaltic pump drive 5006 purchased from Heidolph. An Osram lamp with a power of 6 W, placed inside the photoreactor, emitted ultraviolet radiation of 254 nm. The non-irradiated part of the solution was constantly stirred with a magnetic stirrer (Bibby Scientific, United Kingdom). A Mettler Toledo LE409 electrode was immersed in the solution for continuous pH measurement.  www.nature.com/scientificreports www.nature.com/scientificreports/ TiO 2 catalyst (P25) was used as base catalyst for both preparing doped photocatalysts and comparison purposes. Each separate experiment was repeated twice, and the average was used in plots using the standard error of the mean. The catalysts were sonicated in water for 10 minutes before use by means of a FB15055 ultrasonic bath from Fisher Scientific. Then, the aqueous solution containing the 4-tert-butylphenol with concentration 30 mg L −1 was mixed with the catalyst (1 g L −1 ) under constant magnetic stirring. Before UV irradiation, the solution was stirred for 1 hour in the dark so that the system reached adsorption equilibrium. The start of the UV lamp corresponded to zero time.
Each photocatalytic experiment lasted 60 minutes, and samples were periodically withdrawn and sent directly to analysis. A Vitlab 1000 μL automated pipette was used to take samples. 88.31 mg L −1 of H 2 O 2 was used for the experiments on the effect of hydrogen peroxide addition on the photocatalytic performance.
Analytical methods. The photocatalytic activity of the prepared catalysts was assessed via pH measurements, Total Organic Carbon (TOC), and High-Pressure Liquid Chromatography (HPLC) analysis. TOC analysis was performed using the Multi N/C 3100 instrument by Analytik Jena AG (Germany).
The 4-tert-butylphenol in the solution was quantified by HPLC Agilent 1290 Infinity II system. The concentrations of acetic and formic acid in final solutions were determined using ion chromatography (IC) (930 Compact IC Flex supplied by Metrohm).
The iron content in the solution was determined using the atomic absorption spectrometer AAnalyst 400 from Perkin Elmer.
Prior to TOC and AAS analyses, all aqueous samples were filtered using Agilent Captiva premium syringe filters with a 0.45 µm regenerated cellulose (RC) membrane. For HPLC and IC analyses, the catalysts were separated from the solution by filtration through RC membranes with a pore size of 0.2 μm (Agilent Captiva premium syringe filters).
The removal efficiency was calculated according to Eq. (24): where C t is the concentration after time t and C o is the initial concentration.

Conclusions
4-tert-butylphenol is an endocrine disruptor and considered as an emerging pollutant and a serious water contaminant because it persists in the environment, has acute and chronic toxicity, and estrogenic activity as well. In this study, Fe-doped TiO 2 catalysts with different dopant concentrations (0.5, 1, 2 and 4 wt.%) were successfully prepared using the wet impregnation method, and their photocatalytic activities were tested in the degradation and mineralization of 4-tert-butylphenol in water under UV irradiation. The catalysts prepared were characterized via SEM, TEM, XRD and XRF, while their photocatalytic performance and stability was evaluated via TOC, HPLC, AAS, and IC analyses. The main conclusions are: (1) A well-distributed iron phase on the surface of TiO 2 was observed for all Fe-doped TiO 2 catalysts during SEM/EDS analysis. (2) 4-t-BP was rather easily degraded using UV/Fe/TiO 2 or UV/H 2 O 2 but lower TOC removals were achieved due to the formation of 4-t-BP degradation intermediates. The 4% Fe/TiO 2 catalyst exhibited the highest carbon mineralization (86%) among the catalysts tested.