New Porphyrin/Fe-Loaded TiO2 Composites as Heterogeneous Photo-Fenton Catalysts for the Efficient Degradation of 4-Nitrophenol

A new class of porphyrin(Pp)/Fe co-loaded TiO2 composites opportunely prepared by impregnation of [5,10,15,20-tetra(4-tert-butylphenyl)] porphyrin (H2Pp) or Cu(II)[5,10,15,20-tetra(4-tert-butylphenyl)] porphyrin (CuPp) onto Fe-loaded TiO2 particles showed high activities by carrying out the degradation of 4-nitrophenol (4-NP) as probe reaction in aqueous suspension under heterogeneous photo-Fenton-like reactions by using UV-visible light. The combination of porphyrin-Fe-TiO2 in the presence of H2O2 showed to be more efficient than the simple bare TiO2 or Fe-TiO2.


Introduction
Nowadays, due to the increasing presence of refractory molecules in the wastewater streams, it is important to develop new technologies to degrade such recalcitrant pollutant molecules into smaller innocuous ones. For this reason efficient oxidation processes operating under environmentally friendly conditions are needed [1]. As well known, Fenton chemistry encompasses reactions of hydrogen peroxide in the presence of iron to generate highly reactive species such as the hydroxyl radical and possibly others.
In the last few years, Fenton-like reactions, in combination with other advanced oxidation processes, are assuming fundamental and practical perspectives in water treatment processes [2,3]. e combination of various technologies, in fact, is oen effective to achieve a complete mineralization of the pollutant(s) present in the starting effluents because many stable products of environmental concern can be persistent aer the treatment by Fenton reaction.
Recently, the utilization of TiO 2 as catalyst for the photooxidation of organic pollutants in water is becoming a relevant topic in view of a possible application in economically advantageous and environmentally friendly processes not only performed with the aim to abate pollutants but also for synthetic purposes [4][5][6][7][8].
In this work the design of novel composites metal free or Cu-porphyrin/Fe co-loaded TiO 2 as well as their application as catalytic systems for photoassisted heterogeneous Fentonlike reactions has been reported. In particular, we demonstrated that the presence of porphyrins and Fe species coloaded onto the TiO 2 surface along with H 2 O 2 in the reacting 2. Experimental 2.1. Materials. 4-Nitrophenol, used without further puri�cation, Fe(NO 3 ) 3 ⋅9H 2 O, and hydrogen peroxide solution (30% wt.) were purchased from Aldrich. Solutions were prepared dissolving the required quantity of 4-NP in water obtained by a New Human Power I water puri�cation system. TiO 2 in the microcrystalline phase of anatase, speci�c surface area 8 m 2 g −1 , was kindly provided by Tioxide Huntsman.

Preparation of the Hybrid H 2
Pp-Fe-TiO 2 and CuPp-Fe-TiO 2 Catalysts. e 1% wt Fe-TiO 2 powder, successively indicated as Fe-TiO 2 , was prepared by wet impregnation of TiO 2 with aqueous solutions of Fe(NO 3 ) 3 ⋅9H 2 O by an incipient wetness impregnation followed by a drying process at 393 K and �nal calcination at 350 ∘ C for 5 h as described in a previous work [21].
Fe-loaded TiO 2 powder impregnated with functionalized metal-free porphyrin and Cu(II)-porphyrin Fe-TiO 2 , successively indicated as H 2 Pp-Fe-TiO 2 and CuPp-Fe-TiO 2 , used as photocatalytic systems, were prepared by impregnation of Fe-TiO 2 powders with 6 mol/g of sensitizer (H 2 Pp or CuPp) per gram of TiO 2 . e opportune amount of sensitizers was dissolved in 15 ml of CHCl 3 (or CH 2 Cl 2 ), and 2 g of �nely ground Fe-TiO 2 was added to this solution.
e mixture was stirred for 3-4 h, and the solvent was removed under vacuum.

2.3.
Characterizations. e morphology of the Fe-TiO 2 photocatalysts was studied by using a scanning electron microscopy (SEM) Zeiss Evo 40. X-ray diffraction patterns of all of the samples were performed by using a powder diffractometer (model Ultima + Rigaku) equipped with CuK radiation from 20 ∘ to 80 ∘ . e accelerating voltage and current used were 40 kV and 26 mA, respectively. e diffuse re�ectance spectra (DRS) of photocatalysts were recorded in the range 200-800 nm by using a Varian CARY 100 Scan UVvis spectrophotometer equipped with a diffuse re�ectance integration sphere.

Photocatalytic
Measurements. e set-up used for the photocatalytic experiments is reported in Figure 1 and consists of a 500 ml glass Pyrex reactor containing 4-NP solution/photocatalyst suspension placed in the center of a wood box and irradiated from the top with a 300 W UVvisible lamp (SANOLUX HRC) emitting in the wavelength range 300-900 nm. e lamp was housed in the upper window of the box at 14 cm distance from the reactor, and the radiant �ux measured by a DELTA OHM Photo-Radiometer HD 9221, equipped with a sensor LP 9221 PHOT, was 340 W/m 2 in the 200-950 nm range. e emission spectrum  of the lamp is reported in Figure 2. Oxygenation was ensured by bubbling air in the suspension during the experiments. e novel hybrid composite photocatalysts based on the metal free and Cu porphyrins onto the Fe-loaded TiO 2 have been used to test the degradation of 4-NP as a probe pollutant molecule.
e removal of 4-NP during the reaction processes has been evaluated as the ratio of the concentrations / 0 versus time. t and 0 were calculated measuring the absorbance values and 0 of 4-NP at 317 nm at time and at the initial time 0 , respectively, by means of a UV-vis spectrophotometer (Cary 100 Scan, VARIAN).
e extent of mineralization of the 4-NP was determined on the basis of total organic carbon measurement using a TOC analyzer (IL550 TOC-TN, HACH-LANGE). e amount of Fe 3+ in solution was measured according to the UNI-EN-ISO 11885 method using an ICP spectrometer THERMO SCIENTIFIC iCAP 6000 SERIES.
Also, the Fe-TiO 2 composite, used as the support for the sensitizers H 2 Pp and CuPp, was prepared by a wet impregnation process followed by dryness and calcination as described in a previous work [21].
Further, the novel composites used as the photocatalysts in this work were prepared by impregnation of the Fe-TiO 2 powder with 6 mol/g of sensitizers (H 2 Pp or CuPp) per gram of Fe-TiO 2 as described in the experimental section, and they were indicated, respectively, as H 2 Pp-Fe-TiO 2 and CuPp-Fe-TiO 2 .
Analysis of SEM picture ( Figure 3) shows that the Fe-TiO 2 (Figure 3(b)) and CuPp-Fe-TiO 2 (Figure 3(c)) samples have a higher number of irregular shaped particles than bare TiO 2 (Figure 3(a)). However, the sizes of the Fe-loaded particles, consisting of aggregates of tiny crystals, are smaller compared to that of the bare TiO 2 sample. e presence of Fe 3+ ions seems to hamper the growth of TiO 2 particles. Figure 4 shows the X-ray diffractograms of selected samples. It can be noticed that no modi�cation of the starting anatase phase of the bare TiO 2 supports occurred aer the impregnation treatments as no additional lines attributable to the presence of other phases can be observed. Figure 5 shows the diffuse re�ectance spectra in air of the bare TiO 2 , Fe-TiO 2 , and CuPp-Fe-TiO 2 recorded in the range 200-800 nm.
e spectrum of bare TiO 2 clearly shows an absorption starting at about 380 nm which is typical of bare titania in the anatase phase.
An improvement of light absorption in the visible range can be observed for the Fe-TiO 2 and CuPp-Fe-TiO 2 metal loaded samples, due to the presence of both iron and porphyrin systems producing a modest shi of the band gap edge in the case of CuPp-Fe-TiO 2 sample. Typical absorption bands centered at, respectively, 417 nm (Soret band) and 540 nm (Q band), due to the presence of the porphyrinic moiety, have been observed. Hence, the presence of iron onto the TiO 2 surface enhances the light absorption capability in the visible region which is a prerequisite for the better utilization of the visible light for the photocatalytic processes. e band gap values (Eg) of such unsupported materials were determined from their diffuse re�ectance spectra by using the Kubelka-Munk equation [22]. is equation is based assuming that the re�ectance at any wavelength is de�ned as 2 /2 , where is the measured diffuse re�ectance (%).
A plot of the modi�ed Kubelka-Munk function [ h /2 versus the energy of absorbed light h is shown in Figure 6. All materials are considered to be indirect semiconductors, as TiO 2 .
e results obtained afford band gap energies of 3.20, 3.09, and 3.05 eV for bare TiO 2 , Fe-TiO 2 and CuPp-Fe-TiO 2 samples, respectively. Iron-induced band gap narrowing of 0.11 eV was observed for Fe-loaded titania.    Figure 1 was used as irradiation system. Figure 7 shows the changes in 4-NP concentrations occurring under these experimental conditions. e results obtained in the case of H 2 Pp-Fe-TiO 2 or CuPp-Fe-TiO 2 -in the presence of hydrogen peroxide and under the experimental condition reported above-were more satisfactory than those performed by using bare TiO 2 or Fe-TiO 2 under UV-visible light irradiation. e total organic carbon (TOC) analyses showed complete mineralization of 4-NP aer ca. 60 min of irradiation for both samples loaded with porphyrins (H 2 Pp-Fe-TiO 2 and CuPp-Fe-TiO 2 ). On the contrary residual amounts of TOC (40-50% of abatement) were found aer the same irradiation time when bare TiO 2 or Fe-TiO 2 samples were used.

Photoactivity of the
Interestingly, despite the fact that the observed initial photoreaction rate was higher when CuPp instead of H 2 Pp was used as sensitizer, the maximum of degradation was obtained by using H 2 Pp-Fe-TiO 2 photocatalyst; in fact, 4-NP disappeared completely within 45 minutes of irradiation time.
Negligible photoactivity was observed for all of the samples when carried out under dark. is suggests that the photoexcitation, together with presence of H 2 O 2 , is essential for inducing the photodegradation of 4-NP processes.
e photostability and the reusability of the photocatalysts are important parameters for practical application. In this work we have observed that all the composites, freshly prepared, that is, Fe-TiO 2 , H 2 Pp-Fe-TiO 2 and CuPp-Fe-TiO 2 , can be recycled at least three times without any appreciable decrease of photoactivity.

Journal of Catalysts 5
In the light of the above results the bene�cial effect of porphyrin-based sensitizers for the photodegradation of 4-NP has been con�rmed [20,23].
e porphyrins used as sensitizers (Sens) can be excited by visible light to produce electron-hole pairs (an electron in the excited singlet or triplet state of Pps and a hole in the ground state of Pps; see (1) and (3)): Photoexcitation with UV light of energy greater than the TiO 2 band gap promotes an electron from the valence band to the conduction band and leaves an electronic vacancy or hole (h + ) in the valence band (2).
As shown in Figure 6 the band gap energies for bare TiO 2 , Fe-TiO 2 , and CuPp-Fe-TiO 2 samples are, respectively, 3.20, 3.09, and 3.05 eV. us minor amount of energy is required for the generation of an electron-hole pair photoexcitation of the photocatalysts Fe-TiO 2 and CuPp-Fe-TiO 2 .
e Pp transfers electron into the conduction band of TiO 2 according to (3). TiO 2 works as an electron trapper and hinders the hole-electron recombination. In addition, Pp rapidly transfers excited electrons to TiO 2 and enhances the separation of holes and electrons, signi�cantly improving the photoefficiency.
In a cooperative manner, loading with Fe 3+ ion can enhance the photocatalytic activity due to the charge trapping effect of Fe 3+ , which prevents the recombination of e CB − and h VB + according to the following reactions: In order to better establish the role of the iron ions to try the distinction between a heterogeneous or a homogeneous process we have measured the amount of Fe 3+ in solution by ICP analyses. As result of these measurements, very low amounts of Fe 3+ ions (1-3 ppb) were detected in solutions at the end of each experiment. ese amounts can be considered negligible compared with 4 ppm of Fe 3+ loaded onto TiO 2 surface dispersed in the solution. Hence, although a possible contribution of the homogeneous Fenton reaction occurring in the process cannot be excluded, this contribution can be considered negligible compared with the contribution of the heterogeneous photo-Fenton process.
According to the crystal �eld theory, Fe 2+ (d 6 ) is relatively unstable compared to Fe 3+ (d 5 ). erefore, a release of trapped electron becomes easy to return to Fe 3+ . However, the Fe 2+ /Fe 3+ energy level lies close to the Ti 3+ /Ti 4+ level. As a result of this proximity, the trapped electron in Fe 2+ can be easily transferred to a neighbouring super�cial Ti 4+ and combines with the oxygen molecule to form O 2 •− and �nally • OH [24]. e heterogeneous photo-Fenton degradation of 4-NP occurring in the presence H 2 O 2 can be attributed to the increase of the concentration of hydroxyl radicals generated by photolytic peroxidation efficiently generated as shown in the following equation (5): In order to assess the role of dissolved O 2 during the photocatalytic degradation process, N 2 was bubbled through the suspension to remove O 2 from the solution. Figure 8 shows the photodegradation of 4-NP under N 2 , air, or pure dioxygen bubbling. It is possible to observe that the degradation of 4-NP occurs also under dinitrogen atmosphere.
In addition to the role described previously [24], the presence of dioxygen could be also important during the process due to the possible generation of singlet oxygen ( 1 O 2 ) or • O 2 − species according to (10). e generation of 1 O 2 in a heterogeneous system, where porphyrins are present, has been highlighted by Zebger and coworkers [25]:
e synergistic effect of these porphyrinic structures (H 2 Pp and Cu-Pp) and iron co-loaded onto TiO 2 powders has been studied for the photodegradation of 4-NP in aqueous suspension under UV-visible light irradiation in the presence of H 2 O 2 . To the best of our knowledge this complex system porphyrin-Fe-TiO 2 + H 2 O 2 , that showed to be more performant than the simpler bare TiO 2 , Fe-TiO 2 , porphyrin-Fe-TiO 2 , H 2 O 2 -TiO 2 , H 2 O 2 -Fe-TiO 2 systems, has been studied for the �rst time.