Preparation of Mesoporous Fe 2 O 3-Supported ZSM-5 Zeolites by Carbon-Templating and their Evaluation as Photo-Fenton Catalysts to Degrade Organic Pollutant

Mesoporous Fe2O3-supported ZSM-5 zeolites were prepared by carbon-templating and subsequently evaluated as photo-Fenton catalysts to degrade a dye used as a model heavy organic pollutant. The synthesis procedure of the mesoporous ZSM-5 zeolites was performed employing a nucleating gel and carbon particles as mesopores template. Thereafter, the precursor salt of the iron oxide (Fe2O3) was impregnated and then calcined to obtain the final catalyst. For comparison purposes, a conventional Fe2O3-supported ZSM-5 zeolite was also prepared. The results showed that the amount of intracrystalline mesopores formed in the ZSM-5 crystals was influenced by the amount of carbon added into the synthesis mixture. In comparison to the conventional prepared catalyst, the mesoporous Fe2O3/ZSM-5 ones showed an improved performance in the degradation of the target organic pollutant by the photo-Fenton reaction, which was attributed to the improvement of their textural properties as consequence of the mesopores generation.


Introduction
Heterogeneous photo-Fenton reaction has proved to be an emerging and promising technology for remediation of organic pollutant containing in liquid industrial effluents 1,2 .The technique consist in the use of iron-based catalysts which, in contact with hydrogen peroxide, and in the presence of light irradiation, produce highly oxidative radicals (HO • ) in aqueous solution 3 , leading to degradation of organic molecules.The iron-based catalysts can be directly used in the chemical reaction in the form of finely divided powders or immobilized on solid supports 4 .Several works have demonstrated that the use of different supports to disperse iron oxides lead to greater efficiency in catalytic oxidation processes [5][6][7] .Among the various available supports, ZSM-5 zeolite has been widely used as catalyst support in Fenton oxidation systems due to their intrinsic properties such as thermal stability, high surface area, and uniform pores and channels [8][9][10] .However, the catalytic efficiency of ZSM-5 zeolite may be severely limited due to their microporous nature, which restricts the access of reactant molecules to active catalyst sites located inside of the zeolite crystals 11,12 .Due to this fact, the preparation of mesoporous ZSM-5 zeolites has attracted a great interest in the catalytic applications [13][14][15] .Among the several methods showed in the literature to produce mesoporous zeolites [16][17][18][19] , the carbon-templating method has demonstrated to be relatively facile and inexpensive, being that after burning of the carbon particles, zeolite crystals with intracrystalline mesoporous are obtained 16,20,21 .However, works reporting the preparation of mesoporous Fe 2 O 3 -supported ZSM-5 zeolites employing a nucleating gel and carbon particles as mesopores template in the synthesis procedure for application as photo-Fenton catalyst under visible light has not been reported yet.
Hence, this work aimed to synthesize mesoporous ZSM-5 zeolites using a nucleating gel and carbon particles as mesopores template.The obtained mesoporous ZSM-5 samples were impregnated with the precursor iron salt, being subsequently calcined to obtain the Fe 2 O 3 -supported ZSM-5 catalysts, which were evaluated in the photo-Fenton degradation of a dye molecule used as a model organic pollutant.In addition, the effect of the amount of nucleating gel as well as the amount of carbon particles added into the synthesis mixture on the textural properties of the obtained mesoporous ZSM-5 zeolites was also investigated.
The nucleating gel was prepared following the procedure described in a previous work 22 , that resulted in a mixture with a molar composition of: 1 SiO 2 : 0.3 Na 2 O: 0.05 TPA 2 O: 24 H 2 O: 0.3 OH -.Then, the mixture was charged into a PTFElined stainless autoclave, and aged for 7 days at 60 °C.The precursor gel for the synthesis of the Na-ZSM-5 zeolites was prepared using the molar composition of: 1 SiO 2 : 0.033 Al 2 O 3 : 0.3 Na 2 O: 25 H 2 O: 0.25 OH -.Then, different amounts of the nucleating gel (1, 2.5 or 5 % wt.) were added into the precursor gel, resulting in the ratios of TPAOH/ SiO 2 = 0.001, 0.0025 and 0.005, respectively.The respective mixtures were charged into PTFE-lined stainless autoclaves and submitted to a hydrothermal treatment at 170 °C for 12 h.After the crystallization process, the powders were washed with deionized water and then dried at 110 °C for 24 h.After analyzing the results shown in Figure 1, where the use of different concentrations of nucleating gel did not result in significant changes in the formation of the obtained ZSM-5 zeolites, it was chosen the mixture having 1 % wt. of SiO 2 for the subsequent preparation of the mesoporous ZSM-5 samples.The procedure described above has as advantage of the use of low amount of TPAOH, which is an expensive reagent in the synthesis of ZSM-5 zeolites 22 .
vacuum filtered, washed with deionized water and then dried at 110 °C for 24 h.Finally, the occluded carbon black was removed by calcination at 600 °C during 10 h under oxidizing atmosphere.

Preparation of mesoporous Fe 2 O 3 -supported ZSM-5 zeolites
The mesoporous Fe 2 O 3 -supported ZSM-5 zeolites were prepared by wet impregnation, using a procedure described in a previously published work 23 , which involves three stages: impregnation, dispersion and heat treatment.The impregnation step was performed by adding 30 mL per gram of zeolite of an aqueous solution of iron salt in isopropyl alcohol per gram of zeolite, in order to result in a sample containing 7 wt.% of Fe 2 O 3 .Subsequently, the suspension was submitted to an ultrasound process (Bransonic Ultrasonic Cleaner 2510R-MT -100 W, 42 KHz ± 6) at 60 °C for 40 min, in order to promote the dispersion of iron on the ZSM-5 particles and simultaneously to promote the solvent evaporation.Then, the formed solids were submitted to a heat treatment at 250 °C for 4 h under oxidizing atmosphere.The obtained samples were named ZP-Fe, ZC1-Fe, ZC2-Fe, ZC3-Fe, ZC4-Fe and ZC5-Fe, where ZP-Fe corresponds to the ZSM-5 sample prepared without the presence of carbon particles.

Characterization
The samples were characterized by X-ray diffraction (XRD) using a Rigaku Miniflex model 300 diffractometer being operated with Cu-Kα radiation (λ = 1.5418Å), 30 kV, 10 mA, step size of 0.03º and a count time of 0.5 s per step.The textural properties were determined using a Micromeritics ASAP 2020 apparatus.The morphology and chemical analysis of the samples were obtained by scanning electron microscopy (SEM) using a FEI Inspect S 50 apparatus coupled to an auxiliary Energy Dispersive X-ray Spectroscopy (EDS) detector.For the EDS analysis, the powder samples were put onto the sample holder, which was isolated by a carbon tape.In order to observe the dispersion of Fe 2 O 3 particles on the support (ZC5-Fe sample), a SEM analysis was done using a Philips XL 30 FEG instrument operated at 25 kV with a secondary electron detector.The sample was suspended in acetone (99.5 vol %) by sonication and the suspension was dropped on a metal grid.Previous to the analysis, a thin coating of gold was deposited onto the sample.

Catalytic evaluation
For the photo-Fenton assays, a dye (CI: Reactive Red 141 dye, CAS number 61931-52-0; chemical formula is C 52 H 34 O 26 S 8 Cl 12 N 14 ; molecular weight = 1,952 g mol -1 ; average molecular size of 2.3 nm 24 ) was used as a model organic Then, different amounts of carbon black (Black Pearls ® 2000, Cabot Corporation) were added into the precursor gel in order to generate the following carbon/silica (C/SiO 2 ) ratios: 0, 0.125, 0.25, 0.5, 1.0 and 2.0.The respective samples were named as ZP, ZC1, ZC2, ZC3, ZC4, ZC5, where ZP corresponds to sample synthesized without carbon particles.The resulting mixture was homogenized during 30 min using an ultrasound equipment.Subsequently, the respective mixture was charged into PTFE-lined stainless autoclaves and submitted to a hydrothermal treatment at 170 °C for 12 h.After the crystallization step, the formed powders were Preparation of Mesoporous Fe 2 O 3 -Supported ZSM-5 Zeolites by Carbon-Templating and their Evaluation as Photo-Fenton Catalysts to Degrade Organic Pollutant molecule, which it is widely applied in textile industries.
The experimental conditions used in the experiments were based on a previous work 25 .The experiments were carried out at 25 °C using a glass reactor with 0.25 L of capacity.The irradiation was provided by a commercial fluorescent lamp (EMPALUX, 85 W, luminous efficacy of irradiation = 65 lumens/W, emission at wavelength above 400 nm) fixed 10 cm above the aqueous dye solution.In a typical experiment, the reactor was charged with 100 mL of solution containing an initial dye concentration of 50 mg L -1 , at pH adjusted at 3.0, employing an aqueous solution of sulfuric acid 0.1 mol L -1 , under continuous magnetic stirring.After addition of the catalyst (0.5 g L -1 ), the medium was maintained under stirring for 60 min without irradiation, in order to reach the adsorption equilibrium.Subsequently, the suspension was irradiated by the lamp and the hydrogen peroxide (8 mM) was added to initiate the reaction.Samples were withdrawn using a syringe attached to a filter (PVDF membrane, 0.45 µm).The dye concentration was determined using a UV-Vis spectrophotometer (Bel Photonics, SP1105), with a maximum wavelength of 543 nm.Decolorization efficiency (DE, %) was determined by the following equation: , where: A t is the absorbance after a reaction time t, and A 0 is the initial absorbance before the reaction.Catalytic assays were also performed to evaluate the conventional Fenton reaction (without irradiation), and also with the system under visible light irradiation, but without the presence of catalyst.The stability of the catalysts was evaluated by determination of the iron-leached amount in the reaction media after the photo-Fenton degradation, which was measured by atomic absorption spectroscopy (Agilent Technologies, 200 series AA).

Characterization
In Figure 1 are illustrated the X-ray diffractograms (XRD) of the ZSM-5 samples prepared with different amounts of nucleating gel related to the fixed concentration of the precursor gel.It can be observed that all the analyzed samples show the characteristic peaks of the MFI type structure 26,27 , confirming the formation of the ZSM-5 zeolite.In addition, it is observed that the peaks intensity of the samples is very similar, indicating that the different concentrations of the nucleating gel used in this study did not influenced the crystallinity of the obtained ZSM-5 zeolites.Therefore, the lower content of nucleating gel (that resulted in a value of 1 wt.% of SiO 2 ) was chosen for the subsequent preparation of ZSM-5 samples with different carbon amounts in the synthesis mixture (Figure 2).
Figure 2 shows the XRD patterns of the samples synthesized without and in the presence of different amounts of carbon particles.It can be seen that the crystalline phase corresponding to the MFI structure is observed for all the obtained samples.However, it is observed that the intensity of the XRD peaks decreases with the increasing of the carbon amount used in the synthesis, resulting in a gradual decrease of the relative crystallinity, whose values (shown in Table 1) were calculated according to a previous work 28 , using the ZP sample as reference.These results demonstrate that the carbon presence influences the crystallization process of the ZSM-5 structure.Similar behavior has been found by other researchers 16,21 .It is important to notice that the SiO 2 /Al 2 O 3 ratio of the conventional ZP sample was 31.5 and remained practically unchanged for samples synthesized in the presence of different amounts of carbon particles (Table 1).The X-ray diffractograms shown in Figure 3 correspond to the prepared Fe 2 O 3 -supported ZSM-5 zeolites.As can be evidenced, no peaks corresponding to the presence of Fe 2 O 3 can be identified in the diffractograms, indicating that this component must be present as very small particles highly dispersed on the support.These findings are consistent with other previously reported works 29,30 .From Figure 3, it is also noticed that no important difference appear in the zeolite diffractograms after the impregnation process.The amount of Fe 2 O 3 impregnated on ZSM-5 was about 7 wt.% for all the prepared samples, which was confirmed by EDS analysis (not shown).Figure 4 shows the nitrogen adsorption-desorption isotherms for the obtained zeolites.It can be observed that the isotherms of the ZSM-5 zeolite synthesized without the addition of carbon (ZP sample) can be categorized as type I, according to the International Union of Pure and Applied Chemistry (IUPAC) classification, showing a greater pore volume variation at lower relative pressures (P/P 0 < 0.4) and without hysteresis loop, characteristic of microporous materials.On the other hand, the ZSM-5 samples synthesized in the presence of carbon particles exhibit a combination of type I and type IV isotherms with H2 hysteresis loop, characteristic of solids having microporous and mesoporous structure 31,32 .The corresponding pore size distribution curves (Figure 5) were measured from the adsorption branches of the isotherms according to the Barrett-Joyner-Halenda (BJH) method.It can be seen that the ZSM-5 zeolites synthesized in the presence of carbon particles display a broader pore size distribution compared to sample prepared without the addition of carbon (ZP sample), with a highest peak centered in the mesoporous region (between 15 and 23 nm).In addition, a smaller peak can be observed along the mesoporous region, revealing a material with a bimodal pore size distribution 33 .These features are of great importance for catalytic purposes because it allows greater accessibility and diffusion of large molecules within the zeolite pores.
In order to better analyze the formation of mesoporous ZSM-5 zeolites employing a nucleating gel and carbon particles as mesopores template, the textural properties are shown in Table 2.As can be observed, a slight increase occurs in the specific surface area of the samples synthesized in the presence of carbon compared to the conventional ZP sample.From Figure 6, it can be also observed that an increase in the C/SiO 2 ratio results in a decrease of the micropores volume and as expected in an increase of the mesopores volume.In addition, all the ZSM-5 samples maintained high values of the BET area, total porous volume and mesopores volume after the impregnation with iron.The values of specific surface area (BET) and pore volume obtained in this work are similar to those ones published by others researchers 18,21,34 , using carbon nanoparticles as mesopores template for the ZSM-5 synthesis.
Figure 7 shows SEM images of samples ZP-Fe (Figure 7a), ZC4-Fe (Figure 7b), ZC5-Fe (Figure 7c) and the distribution of Fe 2 O 3 particles on the ZC5-Fe sample (Figure 7d).All samples exhibit similar zeolite particle shapes, demonstrating that the use of carbon black in the synthesis of ZSM-5 had no significant effect on this characteristic.In addition, the particle size for all the obtained ZSM-5 zeolites was practically the same, around 5 µm.From Figure 7d, it is possible to observe a good dispersion of Fe 2 O 3 particles on the support.

Catalytic evaluation
The activity of mesoporous Fe 2 O 3 -supported ZSM-5 catalysts for the decolorization of the target dye molecule as a function of the reaction time is shown in Figure 8.The catalytic activity of the samples in the conventional Fenton process (without light irradiation) and the dye degradation without the presence of catalyst (with light irradiation) were also evaluated and the decolorization results were below of 17 and 5 %, respectively.However, as can be seen from Figure 8 significant decolorization was obtained with the employ of the photo-Fenton process.With the conventional ZP-Fe sample it was reached 80 % of decolorization for a reaction time of 60 min, whereas the ZSM-5 zeolites synthesized using the mesopores template (carbon black) showed superior degradation efficiency, behavior that it is attributed to the presence of intracrystalline mesoporosity (Table 2) that improved the diffusion of the dye molecules.Furthermore, the ZC4-Fe sample showed the highest efficiency, with total decolorization at 60 min, due to the presence of the highest mesoporous volume (Table 2) that may have contributed to a better dispersion of the active phase.As mentioned, the superior catalytic behaviour of the Fe 2 O 3 -supported-ZSM-5 catalysts synthesized in the presence of carbon particles is clearly understood from the improved diffusion of the big dye molecule (2.30 nm 24 ) in the created mesopores with mean diameter of about 20 nm.Thus the ratio between the mesopore diameter and the dye molecular size is 8.6.So this implies that the mesopore diameter of the catalyst could accommodate about eight dye molecules, and therefore, favoring to a high catalytic activity.
For comparison purposes regarding to the catalytic activity aiming the degradation of the Reactive Red 141 dye used in this study, some catalysts have been used and reported in the literature.Stringhini et al. 35 observed a total dye removal at 120 min of reaction using ZnAl 2 O 4 catalyst under UV irradiation, whereas Foletto et al. 24 obtained total removal at 180 min using the same catalyst, but under sunlight.ZnFe 2 O 4 catalyst prepared by microwave irradiation showed 90% of dye removal at 60 min of reaction under visible light 36 .Therefore, the results obtained in this work demonstrate that the mesoporous Fe 2 O 3 -supported ZSM-5 zeolites are efficient catalysts for the removal of Reactive Red 141 dye from aqueous solution.
Moreover, in order to evaluate the stability of the catalysts, the iron-leached amount in the solution after reaction was determined, resulting in a value below of 1 mg L -1 for all the studied catalysts.According to the Brazilian environmental legislation 37 , the value for iron disposal in effluents is 15 mg L -1 .This result indicates a satisfactory stability of the prepared Fe 2 O 3 -supported ZSM-5 zeolites for the use as catalyst in heterogeneous photo-Fenton reactions for degradation of organic pollutants, such as dyes present in industrial wastewaters.

Conclusions
The technique employing nucleating gel and carbon particles as mesopores template was very promising for the production of microporous-mesoporous ZSM-5 zeolites.Different nucleating gel concentrations used in this work had no effect in the formation of the ZSM-5 zeolite structure.Increasing the amount of carbon particles used in the synthesis of ZSM-5 resulted in a decrease of the micropores volume and, as expected, in an increase in the mesopores volume.The mesoporous Fe 2 O 3 -supported ZSM-5 zeolites showed higher photo-Fenton activity to degrade the Reactive Red 141 dye compared to the conventional Fe 2 O 3 -supported ZSM-5 zeolite.Therefore, the studied Fe 2 O 3 -supported ZSM-5 zeolites possessing a microporous-mesoporous structure are promising catalysts to degrade organic pollutants in industrial wastewaters.

Figure 1 :
Figure 1: X-ray diffratograms of ZSM-5 samples prepared with different concentrations of nucleating gel.

Figure 2 :
Figure 2: X-ray diffratograms of ZSM-5 samples prepared with different concentrations of carbon particles and sample prepared without the presence of carbon (ZP sample).

Figure 3 :
Figure 3: X-ray diffratograms of ZSM-5 samples prepared with different concentrations of carbon particles and impregnated with 7 wt.% of Fe 2 O 3 .ZP-Fe: sample prepared without carbon and also impregnated with 7 wt.% of Fe 2 O 3 .

Figure 5 :
Figure 5: Pore size distribution plots of ZSM-5 zeolites synthesized with different concentrations of carbon particles (Log scale graph).ZP: sample prepared without carbon.

Figure 6 :
Figure 6: Influence of the C/SiO 2 ratio employed in the synthesis gel on the microporous and mesoporous volume of the zeolites.

Figure 8 :
Figure 8: Decolorization efficiency (DE, %) of ZSM-5 samples prepared with different concentrations of carbon particles and impregnated with 7 wt.% of Fe 2 O 3 .ZP-Fe: sample prepared without carbon and also impregnated with 7 wt.% of Fe 2 O 3 .