- Email: [email protected]
Iron promotion of the TiO2 photosensitization process towards the photocatalytic oxidation of azo dyes under solar-simulated light irradiation
Materials Chemistry and Physics 129 (2011) 1176–1183
Contents lists available at ScienceDirect
Materials Chemistry and Physics journal homepage: www.elsevier.com/locate/matchemphys
Iron promotion of the TiO2 photosensitization process towards the photocatalytic oxidation of azo dyes under solar-simulated light irradiation Camilo A. Castro, Aristóbulo Centeno, Sonia A. Giraldo ∗ Centro de Investigaciones en Catálisis (CICAT), Escuela de Ingeniería Química, Universidad Industrial de Santander (UIS), A.A. 678, Bucaramanga, Colombia
a r t i c l e
i n f o
Article history: Received 24 January 2011 Received in revised form 29 April 2011 Accepted 31 May 2011 Keywords: Fe2 O3 TiO2 Photosensitization Orange II UV–vis irradiation
a b s t r a c t The photocatalytic oxidation of the azo dye Orange-II (Or-II) using Fe loaded TiO2 (Fe–TiO2 ) was studied under ultraviolet (UV), visible (vis) and simultaneous UV–vis irradiations using a solar light simulator. Photocatalysts were characterized by means of XRD, SEM-EDX, FTIR and DRS. Fe3+ species, identified in XPS analyses, were responsible of the increased absorption of visible light. Moreover, DRS analyses showed a decrease in the bandgap due to Fe3+ loading. Photocatalystic tests proved that Fe modification enhanced the TiO2 photocatalytic activity towards Or-II photodegradation under simultaneous UV–vis irradiation. Even so, the performance of the Fe–TiO2 samples towards the photodegradation of phenol, under UV irradiation, was lower than TiO2 suggesting the recombination of the UV photogenerated electron–hole pair. Therefore, results evidence a Fe3+ promotion of the electron caption in the photosensitization process of TiO2 by Or-II acting as a sensitizer. Such process leads to the Or-II photooxidation under UV–vis irradiation by losing energy in electron transferring processes to sensitize TiO2 , and, the formation of reactive oxygen species promoted by the injected electron to the TiO2 conduction band. © 2011 Elsevier B.V. All rights reserved.
1. Introduction Azo dyes and other industrial dyestuff represent an increasing environmental danger. About 1–20% of the total world production of dyes is lost during the dyeing process and is released in the textile effluents [1,2]. Due to high content of aromatic rings, conventional biodegradation treatments do not lead to an effective removal of the dye molecules from water [2]. One of the most interesting approaches for azo dyes degradation is TiO2 photocatalysis, in which the photogeneration of reactive oxygen species in the TiO2 s surface by light excitation [3], or, azo dye photosensitization [4], induce powerful oxidation processes. Among the different methodologies used in order to increase the TiO2 photoactivity doping with transition metal ions has been extensively used, but the understanding of the mechanisms of the photocatalytic process have turned into a controversy [3,5,6]. The metallic ions into the TiO2 matrix may act as traps for the photogenerated charges: electron and hole (e− –h+ ). This e− –h+ pair is the responsible for the generation of radicals useful in the degradation of water pollutants [3], as well as for the inactivation of bacteria [7,8] and viruses [9]. Therefore, it is expected that charge traps may reduce the recombination of the e− –h+ pair enhancing the production of oxidative radicals.
∗ Corresponding author. Tel.: +57 7 6344746; fax: +57 7 6459647. E-mail addresses: [email protected], [email protected] (S.A. Giraldo). 0254-0584/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.matchemphys.2011.05.082
In particular, Fe doping has been a research interest in synthesis, characterization, optical features, recombination and photocatalytic activity of the TiO2 [5,10–13]. The presence of Fe in the TiO2 increases visible (vis) light absorption [5,14], and may generate an increase in the photocatalytic activity of the material when irradiated with light of > 400 nm [14]. Nevertheless, it has been proved a decrease in photoactivity under irradiation with light of higher energy, such as, UV-A type irradiation ( < 400 nm) [5,14]. In addition, Fe3+ ions may act as either electron or hole traps [15], but even if the photogenerated pair is “trapped”, these traps may interact and promote the recombination turning into a decrease in photoactivity [14]. Fe–TiO2 photocatalysts has been synthesized using the hydrothermal method [5,11,12,16–18]. This process allows to insert Fe atoms in the TiO2 structure located in interstices or replacing Ti atoms in the Ti–O–Ti network [16] since Fe3+ has an ionic radius of 0.69 A˚ while that of Ti4+ is 0.64 A˚ [19], thus it is possible to form a solid solution in the TiO2 matrix. Our group has compared different synthesis routes suggesting the hydrothermal route as a suitable process to obtain Fe–TiO2 particles with high photocatalytic activity under solar-simulated light irradiation [17]. However, it should be noted that a wide range of optimum iron content, correspondent to the major increase in photoactivity can be found in the literature with molar percentages from 0.03% to 3% [16,18]. Its optimal value might depends on a variety of parameters, such as the synthesis conditions and the substrate target of oxidation itself as we have observed [17].
C.A. Castro et al. / Materials Chemistry and Physics 129 (2011) 1176–1183
As a result of the exposed, it is possible to predict a competition of negative and positive effects under simultaneous irradiation of UV and vis. The aim of this work is to analyze the action of Fe loaded in TiO2 particles towards the photocatalytic degradation of the azo dye Orange-II (Or-II) in order to elucidate the effect of the type of light irradiation on the photooxidation mechanisms. Moreover, the photoactivity of the materials was also analyzed towards the photooxidation of phenol which does not photosensitize the TiO2 , thus, allowing the establishment of the action of the modifying metal in the subsequent processes, after photogeneration of the electron–hole pair, such as, recombination or trapping. The analyses of the results of the photooxidation of Or-II and phenol under several light irradiation set-ups such as ultraviolet and UV–vis using a solar light simulator, in addition to the characterization of the materials, led to the understanding of the possible mechanisms of photocatalytic and photodynamic degradation of the chosen molecules by the Fe–TiO2 photocatalysts.
1177
the Fe loaded samples with Fe nominal contents of 1, 3 and 10 mol%, respectively, indicating that part of the Fe is lost during the synthesis and/or sample preparation for AAS analysis. Results of characterization by means of XRD, and DRS, are presented for all photocatalysts except for the 0.5 mol% loaded sample since its loading did not promote significant changes in the different characterized physicochemical responses of TiO2 . 2.3. Photocatalytic tests Photocatalytic tests were done using Pyrex bottles (50 mL) with aqueous suspensions of the target (Orange-II or phenol). Illumination of the suspensions was made using three different set-ups. (I) 400 W m−2 of solar-simulated light irradiation using a suntest system model CPS+ from ATLAS, with temperature and irradiation power control, and a xenon lamp emitting light with wavelengths of 300 to 800 nm, and 5% of the irradiation corresponds to UV-A. (II) 38 W m−2 of UV light irradiation using a closed chamber with a set of 5 TLD 18 W BLB Phillips lamps with an emission spectra of 330–400 nm; to evaluate the effect of Fe on the UV photogeneration charges on TiO2 : electron and hole. And (III) 60 W m−2 of vis light irradiation using a closed chamber with 5 TLD 18 W Blue Phillips lamps with an emission spectra of 400–500 nm; to evaluate any possible photosensitization and photoactivation due to visible light. The radiant flux was monitored with a Kipp & Zonen (CM3) power meter (Omni Instruments Ltd., Dundee, UK).
2. Experimental 2.4. Photocatalytic degradation of Or-II 2.1. Photocatalysts synthesis The TiO2 sample was obtained using the hydrothermal synthesis. In this case, a volume of titanium butoxide (Ti(O-But)4 , Sigma) was added dropwise to isopropanol (Isop-OH, Merck) under vigorous magnetic stirring in a volumetric ratio Isop-OH/Ti(O-But)4 = 5. Then, 3 L of HNO3 (65%, Merck) per each 2 mL of Ti(O-But)4 were added to the previous solution. Immediately after, water was added dropwise to the reaction in a volumetric ratio H2 O/Ti(O-But)4 = 1. The formed gel was pressure treated using an autoclave for 3 h at 120 ◦ C and ∼144 kPa. The obtained crystals were manually grounded in mortar, followed by a washing step in which the powders were twice suspended in water with an intermediate centrifugation step. Finally, water was extracted in an oven at 70 ◦ C for 12 h. To synthesize the Fe modified TiO2 sample the procedure described above was used but instead of water, an aqueous solution of Fe(NO3 )3 ·9H2 O was added (Merck), in an appropriate amount to obtain Fe nominal molar percentages (mol%) of 0.5, 1, 2, 3 and 10%. Photocatalysts are labeled as Fe(x)–TiO2 , where x corresponds to the nominal mol%. 2.2. Photocatalysts characterization X-ray diffraction patterns were collected at room temperature from 2 to 70◦ in 2, using a RIGAKU model D/MAX IIIB system. The diffractometer was operated at 40 kV and 80 mA and the Cu K␣ radiation was selected using a graphite monochromator. The average crystallite size, d, was estimated from the full width at half maximum (FWHM) of the (1 0 1) peak using the Scherrer equation reported elsewhere [20]. Scanning electron microscopy (SEM) was performed on gold-coated samples using a Leo 1450VP microscope equipped with an OXFORD INCA system for energy dispersive spectroscopy analyses. X-ray photoelectron spectroscopy (XPS) analyses were carried out on an AXIS NOVA photoelectron spectrometer (Kratos analytical, Manchester, UK) equipped with a monochromatic AlK␣ (hv = 1486.6 eV) anode. The kinetic energy of the photoelectrons was determined with the hemispheric analyzer set to the pass energy of 160 eV for wide-scan spectra and 20 eV for the case of high resolution spectra. Electrostatic charge effect of the sample was overcompensated by means of the low-energy electron source working in combination with a magnetic immersion lens. The carbon C 1s line with position at 284.6 eV was used as a reference to correct the charging effect. Spectra were decomposed using the casaXPS program (Casa Software Ltd., UK) with a Gaussian/Lorentzian (70/30) product function after subtraction of a linear baseline. The assignation of different Ti peaks in Ti 2p spectra were restricted by distances between Ti peaks of 5.7 eV [21] and 13.6 eV for Fe 2p peaks [22]. UV–vis diffusive reflectance spectra (DRS) were recorded on a UV-2401PC Shimadzu spectrophotometer with an ISR 240A integrating sphere accessory. BaSO4 was used as a blank reference. The energy band gap widths of the samples were determined using the Kubelka–Munk phenomenological theory [23]. Fourier transformed infrared spectra (FTIR), in the wavenumber region 4000–600 cm−1 , were obtained on a FTIR-8400s Shimadzu spectrophotometer using the KBr pellet technique. Fe content in the 1, 3 and 10% Fe-loaded samples was established by flame atomic absorption spectrophotometry (AAS) in a Buck Scientific 210 VGP spectrometer. To determine Fe concentration in Fe–TiO2 samples, 6.4 g of (NH4 )SO4 and 16 mL of concentrated H2 SO4 were added to 0.2 g of the photocatalyst sample. The suspension was heated at 80 ◦ C under vigorous agitation until the complete dissolution of the sample was achieved. After cooling the digestion was diluted in distilled water for absorption measurements. The determined contents were 0.7, 2.1 and 8.1 mol% for
The photocatalytic degradation of Or-II (95%, Aldrich) was performed using a 20 mg L−1 solution with 0.25 g L−1 of TiO2 following the experimental conditions set by our group on azo dye photodegradation in a previous work [17]. A series of photocatalytic tests were performed. First, the Or-II photodegradation was done under solar-simulated light irradiation and was followed by UV–vis spectrophotometry using a UV–vis Hewlett-Packard 8453 equipment. To measure the concentration 3 mL samples of the Or-II solutions were taken and filtered with a 0.2 m filter membrane. The wavelength of 486 nm corresponding to the absorption of N N bond in the Or-II molecule, which is the maximum absorption peak, was used for absorption determinations. Degradation was performed using UV irradiation alone, or vis alone, in order to subtract the effect of vis, or UV irradiation, respectively, on the photoactivity of the Fe–TiO2 samples. Furthermore, since UV–vis spectrophotometry allows to analyze Or-II concentration, without taking account the formed intermediates, it was necessary to evaluate the net oxidative effect by measuring the mineralization of the Or-II molecule using the evolution of the Total Organic Carbon (TOC) concentration during reactions under solar-simulated and UV irradiations. The TOC equipment used was a Shimadzu 500 instrument provided with an auto-sampler. Fe leaching was determined for the Fe(1)–TiO2 sample since this showed interesting photoactivity features. For this purpose, the reaction media after Or-II photodegradation tests under UV–vis was filtered using a 0.45 m membrane and 5 mL of H2 SO4 were added for digestion of organics. Finally, volume was made up to 100 mL using distilled and analyzed by AAS. 2.5. Photocatalytic degradation of phenol Phenol oxidation was made under UV irradiation using a 9.4 mg L−1 phenol solution, with a photocatalyst concentration of 0.5 g L−1 following the experimental ´ et al. and Górska et al. to evaluate its oxidation conditions suggested by Sobczynski by hydroxyl radicals photoproduced on TiO2 irradiation [24,25]. Phenol concentration was followed by HPLC using a Hewlett-Packard series 1100 equipment with a reverse phase Spherisorb silica column (Macherey–Nageland) and a Diode Array Detector. Phenol detection was carried out at 220 nm. As mobile phase a mixture of acetonitrile and water in a volume ratio of 60/40 was used. Data hereby reported is expressed as C/C0 , where C stands for the concentration in mg L−1 of: Or-II, phenol or TOC. The initial concentration (C0 ) is the concentration reached after 1 h of stirring prior to the illumination. The Or-II degradation percentage (%) was calculated by means of the ratio (C0 -C)/C0 .
3. Results and discussion 3.1. Fe–TiO2 photocatalyst features Fig. 1 shows the X-ray diffractograms of the TiO2 and Fe loaded TiO2 photocatalysts. In all cases, there is presence of the anatase phase with peaks at 2 angles of 25.2◦ , 37.9◦ , 48.2◦ , 55.0◦ and 62.6◦ . As observed in Fig. 1, the increased concentration of Fe shortens and widens the anatase (A) characteristic peak at 2 = 25.2◦ , as well as increases the intensity of the peak at 31.2◦ assigned to brookite (B). Popa et al. [26] have observed a similar increase in brookite phase content due to the arrangement of the lattice when increasing the Fe concentration [5,18,26]. As it is well known, Fe solubility
1178
C.A. Castro et al. / Materials Chemistry and Physics 129 (2011) 1176–1183
Fig. 2. Effect of Fe loading on the calculated crystallite size of the Fe–TiO2 samples.
Fig. 1. X-ray diffractograms of: (a) TiO2 and the Fe(x)–TiO2 samples with x (mol%) of: (b) 1, (c) 3, (d) 10. A: Anatase, B: Brookite.
in TiO2 reaches ∼1.4 mol%, after which it is possible to find the pseudobrookite phase (Fe2 TiO5 ) heating in air a hydrothermal synthesized Fe–TiO2 sample at 1273 K [27]. In our case, concentrations of 3 and 10 mol% did not promote the formation of Fe2 TiO5 but the formation of brookite using the hydrothermal treatment at 120 ◦ C. Brookite has an orthorhombic system in which lattice parameters are all unequal (a = / b = / c), whereas in anatase, with a tetragonal / c). Therefore, Fe insertion system, one of them is different (a = b = may lead to the replacement of Ti atoms implying the expansion of the unit cell, and thus, altering the lattice parameters leading to the arrangement of the tetragonal structure to the orthorhombic as observed in Fig. 1. Such changes in the lattice parameters due to the expansion of the unit cell were observed when loading Fe
in TiO2 rutile [28]. Thus, it is possible to suggest that in our case the increase in Fe concentration up to 10 mol% did not promote pseudobrookite formation due to the low temperature used in the synthesis process in comparison to that used by Wantala et al. at 923 K [29]. Nevertheless, in our case the replacement of Ti atoms by Fe did promote brookite formation on TiO2 . Moreover, Fig. 2 shows the calculated crystallite size for the TiO2 and Fe loaded TiO2 samples vs. the Fe concentration. As observed the increased Fe concentration reduces the crystallite size; 10% of Fe loading leads to a reduction of ∼50% of the crystallite size. This, in agreement of the exposed, suggests that the dimensional decrease was actually caused by a number of defects in the anatase crystallites, thus altering the organization of the structure and consequently obstructing crystallite growth [30,31]. Such defects are formed by partial substitution of Ti4+ sites by Fe3+ ions as observed in XRD analyses indeed promoting the anatase to brookite transformation. Additional information was obtained using SEM-EDX analyses. Fig. 3 shows the micrographs obtained for Fe(3)–TiO2 and Fe(10)–TiO2 . Inset shows the elemental EDX analysis of the sam-
Fig. 3. SEM micrographs of the: (a) Fe(3)TiO2 and (c) Fe(10)TiO2 samples, and its magnifications: (b and d), respectively. Insets to (b and d) show the EDX spectra for the 3% and 10% Fe loaded samples.
C.A. Castro et al. / Materials Chemistry and Physics 129 (2011) 1176–1183
1179
Fig. 4. XPS spectra of: (a) Ti 2p and (b) Fe 2p of the Fe–TiO2 and TiO2 samples.
ples. It is observed particles with irregular shape dimensions with sizes from 2 to 60 m and even smaller, such wide distribution is due to the manual grinding process. Magnifications (Fig. 3b and d) show a large quantity of small particles with polyhedral shapes with sizes of <1 m. The large particles with straight edges and sharp corners, as observed in Fig. 3a and c, show smooth surfaces with smaller deposits in the 10% Fe loaded sample (Fig. 3c). Such deposits seem to have a more homogeneous distribution of sizes than in the case of the 3% Fe loaded TiO2 . EDX analyses of both samples showed iron uniformly distributed between particles and within one particle, moreover, Fe concentration was increased due to the enrichment of surface layers with Fe. This is in agreement with XRD results suggests again the absence of Fe oxides in the samples and its insertion into the TiO2 structure. The surface composition analysis of the sample gives additional information. Fig. 4a shows the Ti 2p XPS spectra of the Fe(3)–TiO2 and the TiO2 samples. The Ti 2p spectrum is constituted by two peaks at: 458.9 eV and 464.55 eV, assigned to Ti 2p3/2 and Ti 2p1/2 , respectively, indicating a predominant state of Ti4+ in the surface of bare and Fe loaded samples [21]. No additional peaks for Ti species, such as: Ti3+ , Ti2+ or Ti0 , could be assigned following the information on constraints and restrictions for XPS deconvolution and curve fitting suggested by Biesinger et al. [21]. In addition, no shift was found for Ti in the Ti 2p signal for the 3 mol% Fe loaded TiO2 sample’s spectrum. Furthermore, Fig. 4b shows the Fe 2p XPS spectra of the samples: Fe(1)–TiO2 and Fe(3)–TiO2 . These spectra are constituted by Fe peaks at 711.1 and 724.4 eV of Fe 2p3/2 and Fe 2p1/2 , respectively, assigned to Fe3+ [22]. As expected, the increase in concentration leads to an increase in the signal intensity due to the enrichment of surface atomic layers with Fe as observed in EDX analyses. Probably, the low Fe concentrations did not allow to observe changes of the Ti binding energies due to the Fe substitution of Ti atoms as suggested in SEM-EDX and XRD characterizations. Fig. 5 shows the DRS spectra of the TiO2 and Fe–TiO2 samples with different Fe concentrations. The inset shows the decrease in the energy band gap due to the increased nominal concentration of Fe3+ in the TiO2 s network. These spectra show an effective increase in the visible light absorption capacity of TiO2 that increases with Fe3+ nominal concentration. As a consequence, calculated Eg values show a decrease with increasing Fe concentration. Such decrease
may arise from the overlapped TiO2 conduction band with the dorbitals of Fe3+ ions which will decrease the band gap. Furthermore, the spectra show an absorption shoulder-like peak at ∼480 nm for the Fe modified samples which is assigned to d–d transitions as suggested by He et al. [32]. The FTIR spectra of the Fe modified TiO2 and the TiO2 samples are shown in Fig. 6. The spectra show a broad band in the region 3600–2600 cm−1 attributed to the stretching vibrations of O–H with a shoulder at 3470 cm−1 . The band centered at 1660 cm−1 is assigned to H–O–H bending indicating that samples have hydroxylated surfaces. The wide band at 400–700 cm−1 which is attributed to Ti–O stretching and Ti–O–Ti bridging stretching modes [33] did not show significant changes with Fe loading. Furthermore, the peak at 1389 cm−1 is attributed to NO3 − [34] coming from the dissociation of HNO3 during the synthesis process. In fact, Fe loaded samples show an increase in intensity of this peak due to the contribution of NO3 − coming from the Fe precursor; the Fe(NO3 )3 . Moreover, no peaks were found for N–Ti interactions
Fig. 5. DRS spectra for the Fe(x)–TiO2 samples (x = 3, 2, 1 and 0 nominal mol% of Fe). In the inset, the effect of x on the Eg calculated value using the Kubelka–Munk theory.
1180
C.A. Castro et al. / Materials Chemistry and Physics 129 (2011) 1176–1183
Fig. 8. TOC evolution during the photocatalytic degradation of Or-II under UV–vis using the () TiO2 , and Fe(x)-TiO2 x = () 0.5, () 1, and (X) 3 Fe nominal mol%. Fig. 6. FTIR spectra of the Fe(x)-TiO2 samples, with x (mol%): of: (a) 10, (b) 3, and the (c) TiO2 as reference.
such as Ti–N–O–Ti at 1458 cm−1 , as proposed by Huang et al. [34], suggesting that N doping was not promoted with the presence of nitrate anion, which indeed remains in the structure, probably, as an interstitially located impurity. In fact, as observed in DRS analyses, vis light absorption of the TiO2 sample, containing NO3 − , was almost zero thus confirming Fe loading as the determinant parameter for the increase in light absorption. Therefore, characterization results show that the Fe–TiO2 photocatalyst is constituted of aggregates with anatase as the major phase component (Fig. 1) with surface-dispersed Fe3+ species (as observed on SEM-EDX analyses) and NO3 − ions in both Fe loaded and bare TiO2 samples as a local impurity (Fig. 6). In addition, Fe3+ was suggested to decrease crystallinity due to rearrangements in the TiO2 crystalline structure due to the replacing of Ti atoms inducing brookite formation and its coexistence with anatase. Indeed, such phase mixing has been observed to prop up TiO2 photoactivity in organic contaminants degradation [17,35]. Indeed, brookite XRD characteristic peak (Fig. 1) was observed to increase when increasing Fe loading. Such increase in Fe concentration also led to a decrease in crystallite size, which in fact suggests an increase in available area for photon absorption and activation. In addition, the increase in Fe loading turned into an increased light absorption capacity, thus suggesting that iron loading leads to an increase in photoactivity of the TiO2 sample. 3.2. Effect of Fe on the photooxidation activity of TiO2 under different light irradiation sources The Or-II degradation under simulated solar light using the TiO2 and the Fe–TiO2 samples is shown in Figs. 7 and 8. Fig. 7 shows
Fig. 7. Photocatalytic degradation of Or-II under UV–vis light irradiation by: () TiO2 , and Fe(x)-TiO2 with x = () 0.5, () 1, () 2, and (X) 3 nominal mol% of Fe. (♦) The photolysis blank. In the inset the Or-II conversion percentage achieved by the Fe–TiO2 samples at different Fe contents after 120 min of UV–vis irradiation.
the Or-II photodegradation followed by UV–vis spectrophotometry. The photolysis blank in Fig. 7 evidences almost zero degradation due to UV–vis light alone. What is more, photoactivity of TiO2 is increased due to the presence of Fe. The inset to Fig. 7 shows the effect on the Or-II degradation percentage, after 120 min of reaction, of the increase in Fe concentration. As observed, the TiO2 s photoactivity is increased by increasing Fe concentration up to 1 mol%, after which, there is a decrease in the photodegradation of Or-II. Higher concentrations of Fe result in the promotion of recombination centers for the photogenerated charges [36,37]. Thus, Fe concentration is a determinant parameter in the design of Fe modified TiO2 photocatalysts. The suggested increase in photoactivity due to: (i) decrease in crystallite size, (ii) the coexistence of anatase and brookite phases, and (iii) the increased Vis light absorption capacity, as previously suggested for Fe loaded TiO2 samples, were surpassed by the Fe negative effect after 1 mol% loading. Consequently, recombination of the photogenerated electron–hole pair (e− –h+ ) seems to be promoted using 2 and 3 mol% of Fe loading. However, before proposing a mechanism of interaction between e− –h+ and Fe3+ species it was necessary to evaluate the photooxidative action of the Fe(x)–TiO2 photocatalysts on direct mineralization of the Or-II target using TOC analysis. The evolution of TOC concentration during the photocatalytic degradation of Or-II under solar light irradiation is shown in Fig. 8. In accordance to this, Fe enhances the photooxidation activity of the TiO2 sample. When comparing velocities of Figs. 7 and 8 it is seen that discoloration is faster than mineralization as observed in TOC decay analyses. This is due to mineralization of Or-II by reactive oxygen species that undergoes through attack of the N N bond as first step, thus decreasing absorption at 486 nm. Proposed oxidation mechanisms reported in the literature show numerous intermediates formed during Or-II degradation, such as, HSO4 − , NH4 + , NO3 − , 4-hydroxybenzenesulfonic acid, nitrogen and sulfocontaining products, benzensulfonate, carboxylic and dicarboxylic acids and their anions [38]. However, the TOC evolution of the photocatalytic degradation of Or-II under UV irradiation shows contrary results as presented in Fig. 9. In this case, the performances of the Fe modified TiO2 samples were opposite to the obtained results under simulatedsolar light. This implies that under UV–vis light there possibly is a beneficial effect of vis irradiation for the photooxidation of the dye. This effect is not due to the promotion of e− to the TiO2 conduction band since this would increase the photoactivity which by the contrary is diminished. Moreover, Or-II photooxidation tests were done using a set of visible light emitting lamps; results (not presented) showed no significant activity for all of the photocatalysts during 6 h of visible light irradiation. In fact, nitrogen presence in Fe(x)–TiO2 photocatalysts could be related to an increase in visible light absorption and thus to an
C.A. Castro et al. / Materials Chemistry and Physics 129 (2011) 1176–1183
1181
Fig. 9. Or-II photocatalytic mineralization under UV followed by TOC evolution using the () TiO2 , and Fe(x)-TiO2 samples, x = Fe nominal mol% of: () 0.5, () 1, and (X) 3.
increase in photoactivity [34]. However, zero photoactivity under vis irradiation discarded any beneficial effect due to NO3 − impurities observed in FTIR analyses. In addition, it is possible to reject any given interaction of the dye with leached Fe3+ that could promote oxidation of Or-II. Such possibility arises from the Fe3+ complexing capacity with organic compounds in solution under UV irradiation [39]. Nevertheless, that possible leaching did not increase the dye oxidation under the different light irradiation systems used here. Therefore, the photooxidation pathway of the Fe–TiO2 photocatalyst is due to a photodynamic effect in which Or-II photosensitizes Fe–TiO2 particles; process promoted by the presence of Fe3+ . 3.3. Fe promoted electron caption mechanism on Or-II photooxidation by Fe–TiO2 under UV–vis irradiation To understand the photooxidation mechanism it is important to evaluate the recombination of the photogenerated charges and its interaction with Fe3+ . The photocatalytic oxidation of phenol using TiO2 has been proved to be due to the indirect oxidation with • OH radicals [24,25,40]. Therefore, it is possible to get indirect information of the recombination when degrading phenol since • OH is directly related to the availability of the photoproduced holes. Fig. 10 shows the photocatalytic degradation of phenol after 2 h of UV by the Fe–TiO2 and TiO2 samples. In this case the Fe modification of TiO2 decreases photoactivity, thus, under UV irradiations there is a negative main effect of the Fe3+ in the TiO2 matrix. Consequently, during the TiO2 excitation by UV light (Eq. (1)), Fe ions may act as traps for the photogenerated e− –h+ (Eqs. (1) and (2)) and interact promoting the recombination (Eq. (3)), thus decreasing the TiO2 photoactivity. Furthermore, the performance of the photocatalysts towards phenol photooxidation under 6 h of vis light irradiation (data
Fig. 10. Photocatalytic degradation of phenol during 2 h of UV irradiation by the () TiO2 , and, Fe(x)-TiO2 samples, x = Fe nominal mol% of: () 0.5, () 1, and (X) 3.
Fig. 11. Energy band diagram for the suggested mechanism of oxidation of Or-II using the Fe–TiO2 photocatalyst under solar light irradiation.
not shown) was almost zero, suggesting no activation of the photocatalyst by this type of irradiation. Fe3+ + h+ → Fe4+ −
3+
+ e → Fe
2+
4+
Fe Fe
+ Fe
(1)
2+ −
(2) +
→ (e –h ) + Fe Fe3+
3+
(3)
leaching, which was ∼28% of the Fe content in In addition, the Fe(1)–TiO2 sample, did not promote the phenol degradation in Fenton like reactions promoted by H2 O2 production during TiO2 irradiation. Therefore, this result also discards the effect of leached Fe in the photocatalytic process. As observed above, in contrast to the diminished activity of Fe–TiO2 towards the phenol degradation, there was an increase in photoactivity of the Fe modified TiO2 photocatalyst towards the photodegradation of Or-II under UV–vis light irradiation, thus suggesting that photoactivity is due to the simultaneous irradiation of the system under vis and UV irradiation. This assumption implies a possible photodynamic mechanism of photooxidation in which there is an interaction between the Or-II molecule and the Fe–TiO2 surface. Therefore, it is possible to suggest that Fe3+ increase the TiO2 s electron caption capacity from a photosensitizer. Hence, vis-light-photoexcited Or-II (Or-II*; LUMO state) injects an e− to the Fe–TiO2 particles promoting the formation of reactive oxygen species, such as, superoxide radical (O2 •− ) to efficiently oxidize the Or-II molecule. Thus, the photooxidation of Or-II is due to a double oxidation pathway promoted by Fe3+ in the TiO2 matrix: (i) the auto-photodegradation of the molecule due to the lost energy in the injection of the e− to the TiO2 , and (ii) the oxidation caused by the photoproduced oxidative species in the Or-II photosensitization. Fig. 11 shows a scheme of the double oxidation pathway proposed for the photodegradation of the Or-II due to UV–vis irradiation of the Fe–TiO2 photocatalyst. Thermodynamics dictate that the couple Fe3+ /Fe2+ is located just below the conduction band. Electrons in this state cannot reduce O2 to form O2 •− since the O2 /O2 •− redox potential, at ∼−0.51 V vs. the saturated calomel electrode (SCE), lies above the conduction band of TiO2 at ∼0.3 V vs. SCE [41]. Moreover, the Or-II’s LUMO state is positioned at ∼−1.5 V vs. SCE [42], thus, it can inject one electron to the Fe3+ forming Fe2+ . This process is not related to the formation of oxidative radicals. Subsequently, this e− may stabilize the positive vacancy (h+ ) in the TiO2 s valence band. This process leaves free the UV-photogenerated e− in the TiO2 s conduction band to effectively reduce O2 to form O2 •− , thus also involving a photocatalytic oxidation pathway. This process could explain that simultaneous irradiation of UV and vis is strictly necessary to promote the photoactivity of Fe–TiO2 since the Or-II excitation occurs under visible
1182
C.A. Castro et al. / Materials Chemistry and Physics 129 (2011) 1176–1183
light, while the promotion of electrons in the TiO2 s CB is possible under UV irradiation. As a result, these materials have a potential as heterogeneous photocatalysts in the degradation of azo dyes, or to promote an electron flux in dye photosensitized TiO2 for artificial photosynthesis due to its increased electron caption capacity by embedded Fe3+ in TiO2 . 4. Conclusions The chosen hydrothermal synthesis route leads to the formation of TiO2 particles constituted of anatase with embedded Fe3+ . This photocatalyst has an enhanced photooxidation activity towards the degradation of azo dyes. Mineralization tests of Orange II (Or-II) dye by the Fe–TiO2 photocatalysts, under different illumination set-ups, suggests that Fe3+ promotes the electron caption in TiO2 photosensitization by Or-II under solar-simulated light irradiation. Indeed, the photocatalytic oxidation of Or-II was observed under strict simultaneous irradiation of ultraviolet (UV) and visible (vis) of the Fe–TiO2 photocatalyst using a solar-simulated light source. Moreover, Fe3+ effectively promotes visible light absorption of TiO2 , even so, enhancing the recombination of the photogenerated electron–hole pair, as observed in the photocatalytic degradation of phenol under UV irradiation. Therefore, there is a remarked effect of the type of light irradiation source on the coupled photodynamic and photocatalytic degradation of Or-II. Such interaction leads to a double oxidation pathway of the OrII molecule: (i) the auto-degradation due to the charge injection to the TiO2 from its excited state and (ii) the oxidation caused by oxidative species produced in the photosensitization of the Fe–TiO2 photocatalyst.
Acknowledgements Authors wish to present their gratitude to COLCIENCIAS and SENA for the financial support of the project with code: 110234119419. C. Castro thanks also to the named Colombian government entities and to the UIS in Bucaramanga, Colombia, for their financial support to his PhD studies. Special thanks to the Group of Electrochemical Engineering in the EPFL, Switzerland, for the HPLC and TOC measurements.
References [1] T. Lazar, Color chemistry: synthesis, properties, and applications of organic dyes and pigments, Color. Res. Appl. 30 (2005) 313–314. [2] P. Pitter, J. Chudoba, Biodegradability of Organic Substances in the Aquatic Environment, CRC Press, Boca Raton, 1990. [3] O. Carp, C.L. Huisman, A. Reller, Photoinduced reactivity of titanium dioxide, Prog. Solid State Chem. 32 (2004) 33–177. [4] I.K. Konstantinou, T.A. Albanis, TiO2 -assisted photocatalytic degradation of azo dyes in aqueous solution: kinetic and mechanistic investigations. A review, Appl. Catal. B: Environ. 49 (2004) 1–14. [5] Z. Li, W. Shen, W. He, X. Zu, Effect of Fe-doped TiO2 nanoparticle derived from modified hydrothermal process on the photocatalytic degradation performance on methylene blue, J. Hazard. Mater. 155 (2008) 590–594. [6] C. Hu, Y. Lan, J. Qu, X. Hu, A. Wang, Ag/AgBr/TiO2 visible light photocatalyst for the destruction of azodyes and bacteria, J. Phys. Chem. B 110 (2006) 4066–4072. [7] D. Gumy, C. Morais, P. Bowen, C. Pulgarín, S.A. Giraldo, R. Hajdu, J. Kiwi, Catalytic activity of commercial TiO2 powders for the abatement of the bacteria (E. coli) under solar simulated light: influence of the isoelectric point, Appl. Catal. B: Environ. 63 (2006) 76–84. [8] A.G. Rincón, C. Pulgarin, Photocatalytic inactivation of E. coli: effect of (continous-intermittent) light intensity and of (suspended-fixed) TiO2 concentration, Appl. Catal. B: Environ. 44 (2003) 263–2849. [9] L. Zan, W. Fa, T. Peng, Z. Gong, Photocatalysis effect of nanometer TiO2 and TiO2 -coated ceramic plate on Hepatitis B, J. Photochem. Photobiol. B: Biol. 86 (2007) 165–169. [10] M.I. Litter, J.A. Navío, Photocatalytic properties of iron doped-titania semiconductors, J. Photochem. Photobiol. A: Chem. 98 (1996) 171–181.
[11] M.A. Khan, S.I. Woo, O. Yang, Hydrothermally stabilized Fe(III) doped titania active under visible light for water splitting reaction, Int. J. Hydrogen Energy 33 (2008) 5345–5351. [12] S. Liu, Y. Chen, Enhanced photocatalytic activity of TiO2 powders doped by Fe unevenly, Catal. Commun. 10 (2009) 894–899. [13] E. Piera, M.I. Tejedor, M.E. Zorn, M.A. Anderson, Relationship concerning the nature and concentration of Fe(III) species on the surface of TiO2 particles and photocatalytic activity of the catalyst, Appl. Catal. B: Environ. 46 (2003) 671–685. [14] Y. Wu, J. Zhang, L. Xiao, F. Chen, Preparation and characterization of TiO2 photocatalyst by Fe3+ doping together with Au deposition for the degradation of organic pollutants, Appl. Catal. B: Environ. 88 (2008) 525–532. [15] K. Mizushima, M. Tanaka, S. Lida, Energy levels of iron group impurities in TiO2 , J. Phys. Soc. Jpn. 32 (1972) 1519–1524. [16] M. Asiltürk, F. Sayilkan, E. Arpac¸, Effect of Fe3+ ion doping to TiO2 on the photocatalytic degradation of Malachite Green dy under UV and vis-irradiation, J. Photochem. Photobiol. A. Chem. 203 (2009) 64–71. [17] C.A. Castro-López, A. Centeno, S.A. Giraldo, Fe modified TiO2 photocatalysts for the oxidative degradation of recalcitrant water contaminants, Catal. Today 157 (2010) 119–124. [18] Z. Ambrus, N. Balázs, T. Alapi, G. Wittmann, P. Sipos, A. Dombi, K. Mogyorósi, Synthesis, structure and photocatalytic properties of Fe(III)-doped TiO2 prepared from TiCl3 , Appl. Catal. B: Environ. 81 (2008) 27–37. [19] J. Maier, Physics and Chemistry of Ionic Materials: Ions and Electrons in Solids, John Wiley & Sons Ltd., London, 2004. [20] K. Naeem, F. Ouyang, Preparation of Fe3+ -doped TiO2 nanoparticles and its photocatalytic activity under UV light, Physica B 405 (2010) 221–226. [21] M.C. Biesinger, L.W. Lau, A.R. Gerson, R.S. Smart, Resolving surface chemical states in XPS analysis of first row transition metals oxides and hydroxides: Sc, Ti, V, Cu and Zn, Appl. Surf. Sci. 257 (2010) 887–898. [22] K.M. Parida, G.K. Pradhan, Sol-gel synthesis and characterization of mesoporous iron-titanium mixed oxide for catalytic application, Mater. Chem. Phys. 123 (2010) 427–433. [23] N. Todorova, T. Giannakopoulou, G. Romanos, T. Vaimakis, J. Yu, C. Trapalis, Preparation of fluorine-doped TiO2 photocatalysts with controlled crystalline structure, Int. J. Photoenergy 2008 (2008) 1–9. ´ ´ [24] A. Sobczynski, L. Duczman, W. Zmudzinsky, Phenol destruction by photocatalysis on TiO2 : an attempt to solve the reaction mechanism, J. Mol. Catal. A 213 (2004) 225–230. [25] P. Górska, A. Zaleska, J. Hupka, Photodegradation of phenol by UV/TiO2 and vis/N,C–TiO2 processes: comparative mechanistic and kinetic studies, Sep. Purif. Technol. 28 (2009) 90–96. [26] M. Popa, L. Diamandescu, F. Vasiliu, C.M. Teodorescu, V. Cosoveanu, M. Baia, M. Feder, L. Baia, V. Danciu, Synthesis, structural characterization, and photocatalytic properties of iron-doped TiO2 aerogels, J. Mater. Sci. 44 (2009) 358–364. [27] D. Gazzoli, G. Minelli, M. Valigi, The FeOx –TiO2 system. An X-ray diffraction, thermogravimetric and magnetic susceptibility study, Mater. Chem. Phys. 21 (1989) 93–106. [28] G.K. Li, X.Q. Zhang, H.Y. Wu, W.G. Huang, J.L. Jin, Y. Sun, Z.H. Cheng, Electron spin resonance investigation of the subsitution of Fe3+ for Ti4+ in rutile TiO2 single crystal, Chin. Phys. Soc. 18 (2009) 3551–3554. [29] K. Wantala, L. Laokiat, P. Khemthong, N. Grisdanurak, K. Fukaya, Calcination temperature effect on solvothermal Fe–TiO2 and its performance under visible light irradiaiton, J. Taiwan Inst. Chem. Eng. 41 (2010) 612–616. [30] N.A. Jamalluddin, A.Z. Abdullah, Reactive dye degradation by combined Fe(III)/TiO2 catalyst and ultrasonic irradiation: effect of Fe(III) loading and calcination temperature, Ultrason. Sonochem. 18 (2010) 669–678. [31] A.R. Liu, S.M. Wang, Y.R. Zhao, Z. Zheng, Low-temperature preparation of nanocrystalline TiO2 photocatalyst with a very large specific surface area, Mater. Chem. Phys. 99 (2006) 131–134. [32] Y.P. He, Y.M. Miao, C.R. Li, S.Q. Wang, L. Cao, S.S. Xie, G.Z. Yang, B.S. Zou, Size and structure effect on optical transitions of iron oxide nanocrystals, Phys. Rev. B 71 (2005) 125411–125419. [33] J.A. Navio, G. Colon, M. Macias, C. Real, M.I. Litter, Iron-doped titania semiconductor powders prepared by a sol–gel method. Part I. Sythesis and characterization, Appl. Catal. A: Chem. 177 (1999) 111–120. [34] D. Huang, S. Liao, S. Quan, L. Liu, Z. He, J. Wan, W. Zhou, Synthesis and characterization of visible light responsive N-TiO2 mixed crystal by a modified hydrothermal process, J. Non-Cryst. Solids 354 (2008) 3965–3972. [35] S. Ardizzone, C. Bianchi, G. Cappelletti, S. Gialanella, C. Pirola, V. Ragaini, Tailored anatase/Brookite nanocrsytalline TiO2 . The optimal particle features for liquid and gas-phase photocatalytic reactions, J. Phys. Chem. C 111 (2007) 13222–13231. [36] J. Li, J. Xu, W.L. Dai, K. Fan, Direct hydro-alcohol thermal synthesis of special core–shell structured Fe-doped titania microspheres with extended visible light response and enhanced photoactivity, Appl. Catal. B: Environ. 85 (2009) 162–170. [37] B. Xin, Z. Xin, H. Ren, Hu, X. Zhang, C. Dong, K. Shi, L. Jing, H. Fu, Photocatalytic activity and interfacial carrier transfer of Ag–TiO2 nanoparticle films, Appl. Surf. Sci. 252 (2005) 2050–2055. [38] J. Bandara, J. Kiwi, Fast kinetic spectroscopy, decoloration and production of H2 O2 induced by visible light in oxygenated solutions of the azo dye Orange II, New J. Chem. 23 (1999) 717–724. [39] W. Du, Y. Xu, Y. Wang, Photoinduced degradation of Orange II on different Iron (Hydr)oxides in aqueous suspension: rate enhancement on addition of
C.A. Castro et al. / Materials Chemistry and Physics 129 (2011) 1176–1183 hydrogen peroxide, silver nitrate and sodium fluoride, Langmuir 24 (2008) 175–181. [40] C.M. Miller, R.L. Valentine, Mechanistic studies of surface catalyzed H2 O2 decomposition and contaminant degradation in the presence of sand, Water Res. 33 (1999) 2805–2816.
1183
[41] M. Gratzel, Energy Resources Through Photochemistry and Catalysis, Academic Press Inc., New York, 1983. [42] N. Helaïli, Y. Bessekhouad, A. Bouguelia, M. Trari, Visible light degradation of Orange II using xCuy Oz /TiO2 heterojuncions, J. Hazard. Mater. 168 (2009) 484–492.
Contents lists available at ScienceDirect
Materials Chemistry and Physics journal homepage: www.elsevier.com/locate/matchemphys
Iron promotion of the TiO2 photosensitization process towards the photocatalytic oxidation of azo dyes under solar-simulated light irradiation Camilo A. Castro, Aristóbulo Centeno, Sonia A. Giraldo ∗ Centro de Investigaciones en Catálisis (CICAT), Escuela de Ingeniería Química, Universidad Industrial de Santander (UIS), A.A. 678, Bucaramanga, Colombia
a r t i c l e
i n f o
Article history: Received 24 January 2011 Received in revised form 29 April 2011 Accepted 31 May 2011 Keywords: Fe2 O3 TiO2 Photosensitization Orange II UV–vis irradiation
a b s t r a c t The photocatalytic oxidation of the azo dye Orange-II (Or-II) using Fe loaded TiO2 (Fe–TiO2 ) was studied under ultraviolet (UV), visible (vis) and simultaneous UV–vis irradiations using a solar light simulator. Photocatalysts were characterized by means of XRD, SEM-EDX, FTIR and DRS. Fe3+ species, identified in XPS analyses, were responsible of the increased absorption of visible light. Moreover, DRS analyses showed a decrease in the bandgap due to Fe3+ loading. Photocatalystic tests proved that Fe modification enhanced the TiO2 photocatalytic activity towards Or-II photodegradation under simultaneous UV–vis irradiation. Even so, the performance of the Fe–TiO2 samples towards the photodegradation of phenol, under UV irradiation, was lower than TiO2 suggesting the recombination of the UV photogenerated electron–hole pair. Therefore, results evidence a Fe3+ promotion of the electron caption in the photosensitization process of TiO2 by Or-II acting as a sensitizer. Such process leads to the Or-II photooxidation under UV–vis irradiation by losing energy in electron transferring processes to sensitize TiO2 , and, the formation of reactive oxygen species promoted by the injected electron to the TiO2 conduction band. © 2011 Elsevier B.V. All rights reserved.
1. Introduction Azo dyes and other industrial dyestuff represent an increasing environmental danger. About 1–20% of the total world production of dyes is lost during the dyeing process and is released in the textile effluents [1,2]. Due to high content of aromatic rings, conventional biodegradation treatments do not lead to an effective removal of the dye molecules from water [2]. One of the most interesting approaches for azo dyes degradation is TiO2 photocatalysis, in which the photogeneration of reactive oxygen species in the TiO2 s surface by light excitation [3], or, azo dye photosensitization [4], induce powerful oxidation processes. Among the different methodologies used in order to increase the TiO2 photoactivity doping with transition metal ions has been extensively used, but the understanding of the mechanisms of the photocatalytic process have turned into a controversy [3,5,6]. The metallic ions into the TiO2 matrix may act as traps for the photogenerated charges: electron and hole (e− –h+ ). This e− –h+ pair is the responsible for the generation of radicals useful in the degradation of water pollutants [3], as well as for the inactivation of bacteria [7,8] and viruses [9]. Therefore, it is expected that charge traps may reduce the recombination of the e− –h+ pair enhancing the production of oxidative radicals.
∗ Corresponding author. Tel.: +57 7 6344746; fax: +57 7 6459647. E-mail addresses: [email protected], [email protected] (S.A. Giraldo). 0254-0584/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.matchemphys.2011.05.082
In particular, Fe doping has been a research interest in synthesis, characterization, optical features, recombination and photocatalytic activity of the TiO2 [5,10–13]. The presence of Fe in the TiO2 increases visible (vis) light absorption [5,14], and may generate an increase in the photocatalytic activity of the material when irradiated with light of > 400 nm [14]. Nevertheless, it has been proved a decrease in photoactivity under irradiation with light of higher energy, such as, UV-A type irradiation ( < 400 nm) [5,14]. In addition, Fe3+ ions may act as either electron or hole traps [15], but even if the photogenerated pair is “trapped”, these traps may interact and promote the recombination turning into a decrease in photoactivity [14]. Fe–TiO2 photocatalysts has been synthesized using the hydrothermal method [5,11,12,16–18]. This process allows to insert Fe atoms in the TiO2 structure located in interstices or replacing Ti atoms in the Ti–O–Ti network [16] since Fe3+ has an ionic radius of 0.69 A˚ while that of Ti4+ is 0.64 A˚ [19], thus it is possible to form a solid solution in the TiO2 matrix. Our group has compared different synthesis routes suggesting the hydrothermal route as a suitable process to obtain Fe–TiO2 particles with high photocatalytic activity under solar-simulated light irradiation [17]. However, it should be noted that a wide range of optimum iron content, correspondent to the major increase in photoactivity can be found in the literature with molar percentages from 0.03% to 3% [16,18]. Its optimal value might depends on a variety of parameters, such as the synthesis conditions and the substrate target of oxidation itself as we have observed [17].
C.A. Castro et al. / Materials Chemistry and Physics 129 (2011) 1176–1183
As a result of the exposed, it is possible to predict a competition of negative and positive effects under simultaneous irradiation of UV and vis. The aim of this work is to analyze the action of Fe loaded in TiO2 particles towards the photocatalytic degradation of the azo dye Orange-II (Or-II) in order to elucidate the effect of the type of light irradiation on the photooxidation mechanisms. Moreover, the photoactivity of the materials was also analyzed towards the photooxidation of phenol which does not photosensitize the TiO2 , thus, allowing the establishment of the action of the modifying metal in the subsequent processes, after photogeneration of the electron–hole pair, such as, recombination or trapping. The analyses of the results of the photooxidation of Or-II and phenol under several light irradiation set-ups such as ultraviolet and UV–vis using a solar light simulator, in addition to the characterization of the materials, led to the understanding of the possible mechanisms of photocatalytic and photodynamic degradation of the chosen molecules by the Fe–TiO2 photocatalysts.
1177
the Fe loaded samples with Fe nominal contents of 1, 3 and 10 mol%, respectively, indicating that part of the Fe is lost during the synthesis and/or sample preparation for AAS analysis. Results of characterization by means of XRD, and DRS, are presented for all photocatalysts except for the 0.5 mol% loaded sample since its loading did not promote significant changes in the different characterized physicochemical responses of TiO2 . 2.3. Photocatalytic tests Photocatalytic tests were done using Pyrex bottles (50 mL) with aqueous suspensions of the target (Orange-II or phenol). Illumination of the suspensions was made using three different set-ups. (I) 400 W m−2 of solar-simulated light irradiation using a suntest system model CPS+ from ATLAS, with temperature and irradiation power control, and a xenon lamp emitting light with wavelengths of 300 to 800 nm, and 5% of the irradiation corresponds to UV-A. (II) 38 W m−2 of UV light irradiation using a closed chamber with a set of 5 TLD 18 W BLB Phillips lamps with an emission spectra of 330–400 nm; to evaluate the effect of Fe on the UV photogeneration charges on TiO2 : electron and hole. And (III) 60 W m−2 of vis light irradiation using a closed chamber with 5 TLD 18 W Blue Phillips lamps with an emission spectra of 400–500 nm; to evaluate any possible photosensitization and photoactivation due to visible light. The radiant flux was monitored with a Kipp & Zonen (CM3) power meter (Omni Instruments Ltd., Dundee, UK).
2. Experimental 2.4. Photocatalytic degradation of Or-II 2.1. Photocatalysts synthesis The TiO2 sample was obtained using the hydrothermal synthesis. In this case, a volume of titanium butoxide (Ti(O-But)4 , Sigma) was added dropwise to isopropanol (Isop-OH, Merck) under vigorous magnetic stirring in a volumetric ratio Isop-OH/Ti(O-But)4 = 5. Then, 3 L of HNO3 (65%, Merck) per each 2 mL of Ti(O-But)4 were added to the previous solution. Immediately after, water was added dropwise to the reaction in a volumetric ratio H2 O/Ti(O-But)4 = 1. The formed gel was pressure treated using an autoclave for 3 h at 120 ◦ C and ∼144 kPa. The obtained crystals were manually grounded in mortar, followed by a washing step in which the powders were twice suspended in water with an intermediate centrifugation step. Finally, water was extracted in an oven at 70 ◦ C for 12 h. To synthesize the Fe modified TiO2 sample the procedure described above was used but instead of water, an aqueous solution of Fe(NO3 )3 ·9H2 O was added (Merck), in an appropriate amount to obtain Fe nominal molar percentages (mol%) of 0.5, 1, 2, 3 and 10%. Photocatalysts are labeled as Fe(x)–TiO2 , where x corresponds to the nominal mol%. 2.2. Photocatalysts characterization X-ray diffraction patterns were collected at room temperature from 2 to 70◦ in 2, using a RIGAKU model D/MAX IIIB system. The diffractometer was operated at 40 kV and 80 mA and the Cu K␣ radiation was selected using a graphite monochromator. The average crystallite size, d, was estimated from the full width at half maximum (FWHM) of the (1 0 1) peak using the Scherrer equation reported elsewhere [20]. Scanning electron microscopy (SEM) was performed on gold-coated samples using a Leo 1450VP microscope equipped with an OXFORD INCA system for energy dispersive spectroscopy analyses. X-ray photoelectron spectroscopy (XPS) analyses were carried out on an AXIS NOVA photoelectron spectrometer (Kratos analytical, Manchester, UK) equipped with a monochromatic AlK␣ (hv = 1486.6 eV) anode. The kinetic energy of the photoelectrons was determined with the hemispheric analyzer set to the pass energy of 160 eV for wide-scan spectra and 20 eV for the case of high resolution spectra. Electrostatic charge effect of the sample was overcompensated by means of the low-energy electron source working in combination with a magnetic immersion lens. The carbon C 1s line with position at 284.6 eV was used as a reference to correct the charging effect. Spectra were decomposed using the casaXPS program (Casa Software Ltd., UK) with a Gaussian/Lorentzian (70/30) product function after subtraction of a linear baseline. The assignation of different Ti peaks in Ti 2p spectra were restricted by distances between Ti peaks of 5.7 eV [21] and 13.6 eV for Fe 2p peaks [22]. UV–vis diffusive reflectance spectra (DRS) were recorded on a UV-2401PC Shimadzu spectrophotometer with an ISR 240A integrating sphere accessory. BaSO4 was used as a blank reference. The energy band gap widths of the samples were determined using the Kubelka–Munk phenomenological theory [23]. Fourier transformed infrared spectra (FTIR), in the wavenumber region 4000–600 cm−1 , were obtained on a FTIR-8400s Shimadzu spectrophotometer using the KBr pellet technique. Fe content in the 1, 3 and 10% Fe-loaded samples was established by flame atomic absorption spectrophotometry (AAS) in a Buck Scientific 210 VGP spectrometer. To determine Fe concentration in Fe–TiO2 samples, 6.4 g of (NH4 )SO4 and 16 mL of concentrated H2 SO4 were added to 0.2 g of the photocatalyst sample. The suspension was heated at 80 ◦ C under vigorous agitation until the complete dissolution of the sample was achieved. After cooling the digestion was diluted in distilled water for absorption measurements. The determined contents were 0.7, 2.1 and 8.1 mol% for
The photocatalytic degradation of Or-II (95%, Aldrich) was performed using a 20 mg L−1 solution with 0.25 g L−1 of TiO2 following the experimental conditions set by our group on azo dye photodegradation in a previous work [17]. A series of photocatalytic tests were performed. First, the Or-II photodegradation was done under solar-simulated light irradiation and was followed by UV–vis spectrophotometry using a UV–vis Hewlett-Packard 8453 equipment. To measure the concentration 3 mL samples of the Or-II solutions were taken and filtered with a 0.2 m filter membrane. The wavelength of 486 nm corresponding to the absorption of N N bond in the Or-II molecule, which is the maximum absorption peak, was used for absorption determinations. Degradation was performed using UV irradiation alone, or vis alone, in order to subtract the effect of vis, or UV irradiation, respectively, on the photoactivity of the Fe–TiO2 samples. Furthermore, since UV–vis spectrophotometry allows to analyze Or-II concentration, without taking account the formed intermediates, it was necessary to evaluate the net oxidative effect by measuring the mineralization of the Or-II molecule using the evolution of the Total Organic Carbon (TOC) concentration during reactions under solar-simulated and UV irradiations. The TOC equipment used was a Shimadzu 500 instrument provided with an auto-sampler. Fe leaching was determined for the Fe(1)–TiO2 sample since this showed interesting photoactivity features. For this purpose, the reaction media after Or-II photodegradation tests under UV–vis was filtered using a 0.45 m membrane and 5 mL of H2 SO4 were added for digestion of organics. Finally, volume was made up to 100 mL using distilled and analyzed by AAS. 2.5. Photocatalytic degradation of phenol Phenol oxidation was made under UV irradiation using a 9.4 mg L−1 phenol solution, with a photocatalyst concentration of 0.5 g L−1 following the experimental ´ et al. and Górska et al. to evaluate its oxidation conditions suggested by Sobczynski by hydroxyl radicals photoproduced on TiO2 irradiation [24,25]. Phenol concentration was followed by HPLC using a Hewlett-Packard series 1100 equipment with a reverse phase Spherisorb silica column (Macherey–Nageland) and a Diode Array Detector. Phenol detection was carried out at 220 nm. As mobile phase a mixture of acetonitrile and water in a volume ratio of 60/40 was used. Data hereby reported is expressed as C/C0 , where C stands for the concentration in mg L−1 of: Or-II, phenol or TOC. The initial concentration (C0 ) is the concentration reached after 1 h of stirring prior to the illumination. The Or-II degradation percentage (%) was calculated by means of the ratio (C0 -C)/C0 .
3. Results and discussion 3.1. Fe–TiO2 photocatalyst features Fig. 1 shows the X-ray diffractograms of the TiO2 and Fe loaded TiO2 photocatalysts. In all cases, there is presence of the anatase phase with peaks at 2 angles of 25.2◦ , 37.9◦ , 48.2◦ , 55.0◦ and 62.6◦ . As observed in Fig. 1, the increased concentration of Fe shortens and widens the anatase (A) characteristic peak at 2 = 25.2◦ , as well as increases the intensity of the peak at 31.2◦ assigned to brookite (B). Popa et al. [26] have observed a similar increase in brookite phase content due to the arrangement of the lattice when increasing the Fe concentration [5,18,26]. As it is well known, Fe solubility
1178
C.A. Castro et al. / Materials Chemistry and Physics 129 (2011) 1176–1183
Fig. 2. Effect of Fe loading on the calculated crystallite size of the Fe–TiO2 samples.
Fig. 1. X-ray diffractograms of: (a) TiO2 and the Fe(x)–TiO2 samples with x (mol%) of: (b) 1, (c) 3, (d) 10. A: Anatase, B: Brookite.
in TiO2 reaches ∼1.4 mol%, after which it is possible to find the pseudobrookite phase (Fe2 TiO5 ) heating in air a hydrothermal synthesized Fe–TiO2 sample at 1273 K [27]. In our case, concentrations of 3 and 10 mol% did not promote the formation of Fe2 TiO5 but the formation of brookite using the hydrothermal treatment at 120 ◦ C. Brookite has an orthorhombic system in which lattice parameters are all unequal (a = / b = / c), whereas in anatase, with a tetragonal / c). Therefore, Fe insertion system, one of them is different (a = b = may lead to the replacement of Ti atoms implying the expansion of the unit cell, and thus, altering the lattice parameters leading to the arrangement of the tetragonal structure to the orthorhombic as observed in Fig. 1. Such changes in the lattice parameters due to the expansion of the unit cell were observed when loading Fe
in TiO2 rutile [28]. Thus, it is possible to suggest that in our case the increase in Fe concentration up to 10 mol% did not promote pseudobrookite formation due to the low temperature used in the synthesis process in comparison to that used by Wantala et al. at 923 K [29]. Nevertheless, in our case the replacement of Ti atoms by Fe did promote brookite formation on TiO2 . Moreover, Fig. 2 shows the calculated crystallite size for the TiO2 and Fe loaded TiO2 samples vs. the Fe concentration. As observed the increased Fe concentration reduces the crystallite size; 10% of Fe loading leads to a reduction of ∼50% of the crystallite size. This, in agreement of the exposed, suggests that the dimensional decrease was actually caused by a number of defects in the anatase crystallites, thus altering the organization of the structure and consequently obstructing crystallite growth [30,31]. Such defects are formed by partial substitution of Ti4+ sites by Fe3+ ions as observed in XRD analyses indeed promoting the anatase to brookite transformation. Additional information was obtained using SEM-EDX analyses. Fig. 3 shows the micrographs obtained for Fe(3)–TiO2 and Fe(10)–TiO2 . Inset shows the elemental EDX analysis of the sam-
Fig. 3. SEM micrographs of the: (a) Fe(3)TiO2 and (c) Fe(10)TiO2 samples, and its magnifications: (b and d), respectively. Insets to (b and d) show the EDX spectra for the 3% and 10% Fe loaded samples.
C.A. Castro et al. / Materials Chemistry and Physics 129 (2011) 1176–1183
1179
Fig. 4. XPS spectra of: (a) Ti 2p and (b) Fe 2p of the Fe–TiO2 and TiO2 samples.
ples. It is observed particles with irregular shape dimensions with sizes from 2 to 60 m and even smaller, such wide distribution is due to the manual grinding process. Magnifications (Fig. 3b and d) show a large quantity of small particles with polyhedral shapes with sizes of <1 m. The large particles with straight edges and sharp corners, as observed in Fig. 3a and c, show smooth surfaces with smaller deposits in the 10% Fe loaded sample (Fig. 3c). Such deposits seem to have a more homogeneous distribution of sizes than in the case of the 3% Fe loaded TiO2 . EDX analyses of both samples showed iron uniformly distributed between particles and within one particle, moreover, Fe concentration was increased due to the enrichment of surface layers with Fe. This is in agreement with XRD results suggests again the absence of Fe oxides in the samples and its insertion into the TiO2 structure. The surface composition analysis of the sample gives additional information. Fig. 4a shows the Ti 2p XPS spectra of the Fe(3)–TiO2 and the TiO2 samples. The Ti 2p spectrum is constituted by two peaks at: 458.9 eV and 464.55 eV, assigned to Ti 2p3/2 and Ti 2p1/2 , respectively, indicating a predominant state of Ti4+ in the surface of bare and Fe loaded samples [21]. No additional peaks for Ti species, such as: Ti3+ , Ti2+ or Ti0 , could be assigned following the information on constraints and restrictions for XPS deconvolution and curve fitting suggested by Biesinger et al. [21]. In addition, no shift was found for Ti in the Ti 2p signal for the 3 mol% Fe loaded TiO2 sample’s spectrum. Furthermore, Fig. 4b shows the Fe 2p XPS spectra of the samples: Fe(1)–TiO2 and Fe(3)–TiO2 . These spectra are constituted by Fe peaks at 711.1 and 724.4 eV of Fe 2p3/2 and Fe 2p1/2 , respectively, assigned to Fe3+ [22]. As expected, the increase in concentration leads to an increase in the signal intensity due to the enrichment of surface atomic layers with Fe as observed in EDX analyses. Probably, the low Fe concentrations did not allow to observe changes of the Ti binding energies due to the Fe substitution of Ti atoms as suggested in SEM-EDX and XRD characterizations. Fig. 5 shows the DRS spectra of the TiO2 and Fe–TiO2 samples with different Fe concentrations. The inset shows the decrease in the energy band gap due to the increased nominal concentration of Fe3+ in the TiO2 s network. These spectra show an effective increase in the visible light absorption capacity of TiO2 that increases with Fe3+ nominal concentration. As a consequence, calculated Eg values show a decrease with increasing Fe concentration. Such decrease
may arise from the overlapped TiO2 conduction band with the dorbitals of Fe3+ ions which will decrease the band gap. Furthermore, the spectra show an absorption shoulder-like peak at ∼480 nm for the Fe modified samples which is assigned to d–d transitions as suggested by He et al. [32]. The FTIR spectra of the Fe modified TiO2 and the TiO2 samples are shown in Fig. 6. The spectra show a broad band in the region 3600–2600 cm−1 attributed to the stretching vibrations of O–H with a shoulder at 3470 cm−1 . The band centered at 1660 cm−1 is assigned to H–O–H bending indicating that samples have hydroxylated surfaces. The wide band at 400–700 cm−1 which is attributed to Ti–O stretching and Ti–O–Ti bridging stretching modes [33] did not show significant changes with Fe loading. Furthermore, the peak at 1389 cm−1 is attributed to NO3 − [34] coming from the dissociation of HNO3 during the synthesis process. In fact, Fe loaded samples show an increase in intensity of this peak due to the contribution of NO3 − coming from the Fe precursor; the Fe(NO3 )3 . Moreover, no peaks were found for N–Ti interactions
Fig. 5. DRS spectra for the Fe(x)–TiO2 samples (x = 3, 2, 1 and 0 nominal mol% of Fe). In the inset, the effect of x on the Eg calculated value using the Kubelka–Munk theory.
1180
C.A. Castro et al. / Materials Chemistry and Physics 129 (2011) 1176–1183
Fig. 8. TOC evolution during the photocatalytic degradation of Or-II under UV–vis using the () TiO2 , and Fe(x)-TiO2 x = () 0.5, () 1, and (X) 3 Fe nominal mol%. Fig. 6. FTIR spectra of the Fe(x)-TiO2 samples, with x (mol%): of: (a) 10, (b) 3, and the (c) TiO2 as reference.
such as Ti–N–O–Ti at 1458 cm−1 , as proposed by Huang et al. [34], suggesting that N doping was not promoted with the presence of nitrate anion, which indeed remains in the structure, probably, as an interstitially located impurity. In fact, as observed in DRS analyses, vis light absorption of the TiO2 sample, containing NO3 − , was almost zero thus confirming Fe loading as the determinant parameter for the increase in light absorption. Therefore, characterization results show that the Fe–TiO2 photocatalyst is constituted of aggregates with anatase as the major phase component (Fig. 1) with surface-dispersed Fe3+ species (as observed on SEM-EDX analyses) and NO3 − ions in both Fe loaded and bare TiO2 samples as a local impurity (Fig. 6). In addition, Fe3+ was suggested to decrease crystallinity due to rearrangements in the TiO2 crystalline structure due to the replacing of Ti atoms inducing brookite formation and its coexistence with anatase. Indeed, such phase mixing has been observed to prop up TiO2 photoactivity in organic contaminants degradation [17,35]. Indeed, brookite XRD characteristic peak (Fig. 1) was observed to increase when increasing Fe loading. Such increase in Fe concentration also led to a decrease in crystallite size, which in fact suggests an increase in available area for photon absorption and activation. In addition, the increase in Fe loading turned into an increased light absorption capacity, thus suggesting that iron loading leads to an increase in photoactivity of the TiO2 sample. 3.2. Effect of Fe on the photooxidation activity of TiO2 under different light irradiation sources The Or-II degradation under simulated solar light using the TiO2 and the Fe–TiO2 samples is shown in Figs. 7 and 8. Fig. 7 shows
Fig. 7. Photocatalytic degradation of Or-II under UV–vis light irradiation by: () TiO2 , and Fe(x)-TiO2 with x = () 0.5, () 1, () 2, and (X) 3 nominal mol% of Fe. (♦) The photolysis blank. In the inset the Or-II conversion percentage achieved by the Fe–TiO2 samples at different Fe contents after 120 min of UV–vis irradiation.
the Or-II photodegradation followed by UV–vis spectrophotometry. The photolysis blank in Fig. 7 evidences almost zero degradation due to UV–vis light alone. What is more, photoactivity of TiO2 is increased due to the presence of Fe. The inset to Fig. 7 shows the effect on the Or-II degradation percentage, after 120 min of reaction, of the increase in Fe concentration. As observed, the TiO2 s photoactivity is increased by increasing Fe concentration up to 1 mol%, after which, there is a decrease in the photodegradation of Or-II. Higher concentrations of Fe result in the promotion of recombination centers for the photogenerated charges [36,37]. Thus, Fe concentration is a determinant parameter in the design of Fe modified TiO2 photocatalysts. The suggested increase in photoactivity due to: (i) decrease in crystallite size, (ii) the coexistence of anatase and brookite phases, and (iii) the increased Vis light absorption capacity, as previously suggested for Fe loaded TiO2 samples, were surpassed by the Fe negative effect after 1 mol% loading. Consequently, recombination of the photogenerated electron–hole pair (e− –h+ ) seems to be promoted using 2 and 3 mol% of Fe loading. However, before proposing a mechanism of interaction between e− –h+ and Fe3+ species it was necessary to evaluate the photooxidative action of the Fe(x)–TiO2 photocatalysts on direct mineralization of the Or-II target using TOC analysis. The evolution of TOC concentration during the photocatalytic degradation of Or-II under solar light irradiation is shown in Fig. 8. In accordance to this, Fe enhances the photooxidation activity of the TiO2 sample. When comparing velocities of Figs. 7 and 8 it is seen that discoloration is faster than mineralization as observed in TOC decay analyses. This is due to mineralization of Or-II by reactive oxygen species that undergoes through attack of the N N bond as first step, thus decreasing absorption at 486 nm. Proposed oxidation mechanisms reported in the literature show numerous intermediates formed during Or-II degradation, such as, HSO4 − , NH4 + , NO3 − , 4-hydroxybenzenesulfonic acid, nitrogen and sulfocontaining products, benzensulfonate, carboxylic and dicarboxylic acids and their anions [38]. However, the TOC evolution of the photocatalytic degradation of Or-II under UV irradiation shows contrary results as presented in Fig. 9. In this case, the performances of the Fe modified TiO2 samples were opposite to the obtained results under simulatedsolar light. This implies that under UV–vis light there possibly is a beneficial effect of vis irradiation for the photooxidation of the dye. This effect is not due to the promotion of e− to the TiO2 conduction band since this would increase the photoactivity which by the contrary is diminished. Moreover, Or-II photooxidation tests were done using a set of visible light emitting lamps; results (not presented) showed no significant activity for all of the photocatalysts during 6 h of visible light irradiation. In fact, nitrogen presence in Fe(x)–TiO2 photocatalysts could be related to an increase in visible light absorption and thus to an
C.A. Castro et al. / Materials Chemistry and Physics 129 (2011) 1176–1183
1181
Fig. 9. Or-II photocatalytic mineralization under UV followed by TOC evolution using the () TiO2 , and Fe(x)-TiO2 samples, x = Fe nominal mol% of: () 0.5, () 1, and (X) 3.
increase in photoactivity [34]. However, zero photoactivity under vis irradiation discarded any beneficial effect due to NO3 − impurities observed in FTIR analyses. In addition, it is possible to reject any given interaction of the dye with leached Fe3+ that could promote oxidation of Or-II. Such possibility arises from the Fe3+ complexing capacity with organic compounds in solution under UV irradiation [39]. Nevertheless, that possible leaching did not increase the dye oxidation under the different light irradiation systems used here. Therefore, the photooxidation pathway of the Fe–TiO2 photocatalyst is due to a photodynamic effect in which Or-II photosensitizes Fe–TiO2 particles; process promoted by the presence of Fe3+ . 3.3. Fe promoted electron caption mechanism on Or-II photooxidation by Fe–TiO2 under UV–vis irradiation To understand the photooxidation mechanism it is important to evaluate the recombination of the photogenerated charges and its interaction with Fe3+ . The photocatalytic oxidation of phenol using TiO2 has been proved to be due to the indirect oxidation with • OH radicals [24,25,40]. Therefore, it is possible to get indirect information of the recombination when degrading phenol since • OH is directly related to the availability of the photoproduced holes. Fig. 10 shows the photocatalytic degradation of phenol after 2 h of UV by the Fe–TiO2 and TiO2 samples. In this case the Fe modification of TiO2 decreases photoactivity, thus, under UV irradiations there is a negative main effect of the Fe3+ in the TiO2 matrix. Consequently, during the TiO2 excitation by UV light (Eq. (1)), Fe ions may act as traps for the photogenerated e− –h+ (Eqs. (1) and (2)) and interact promoting the recombination (Eq. (3)), thus decreasing the TiO2 photoactivity. Furthermore, the performance of the photocatalysts towards phenol photooxidation under 6 h of vis light irradiation (data
Fig. 10. Photocatalytic degradation of phenol during 2 h of UV irradiation by the () TiO2 , and, Fe(x)-TiO2 samples, x = Fe nominal mol% of: () 0.5, () 1, and (X) 3.
Fig. 11. Energy band diagram for the suggested mechanism of oxidation of Or-II using the Fe–TiO2 photocatalyst under solar light irradiation.
not shown) was almost zero, suggesting no activation of the photocatalyst by this type of irradiation. Fe3+ + h+ → Fe4+ −
3+
+ e → Fe
2+
4+
Fe Fe
+ Fe
(1)
2+ −
(2) +
→ (e –h ) + Fe Fe3+
3+
(3)
leaching, which was ∼28% of the Fe content in In addition, the Fe(1)–TiO2 sample, did not promote the phenol degradation in Fenton like reactions promoted by H2 O2 production during TiO2 irradiation. Therefore, this result also discards the effect of leached Fe in the photocatalytic process. As observed above, in contrast to the diminished activity of Fe–TiO2 towards the phenol degradation, there was an increase in photoactivity of the Fe modified TiO2 photocatalyst towards the photodegradation of Or-II under UV–vis light irradiation, thus suggesting that photoactivity is due to the simultaneous irradiation of the system under vis and UV irradiation. This assumption implies a possible photodynamic mechanism of photooxidation in which there is an interaction between the Or-II molecule and the Fe–TiO2 surface. Therefore, it is possible to suggest that Fe3+ increase the TiO2 s electron caption capacity from a photosensitizer. Hence, vis-light-photoexcited Or-II (Or-II*; LUMO state) injects an e− to the Fe–TiO2 particles promoting the formation of reactive oxygen species, such as, superoxide radical (O2 •− ) to efficiently oxidize the Or-II molecule. Thus, the photooxidation of Or-II is due to a double oxidation pathway promoted by Fe3+ in the TiO2 matrix: (i) the auto-photodegradation of the molecule due to the lost energy in the injection of the e− to the TiO2 , and (ii) the oxidation caused by the photoproduced oxidative species in the Or-II photosensitization. Fig. 11 shows a scheme of the double oxidation pathway proposed for the photodegradation of the Or-II due to UV–vis irradiation of the Fe–TiO2 photocatalyst. Thermodynamics dictate that the couple Fe3+ /Fe2+ is located just below the conduction band. Electrons in this state cannot reduce O2 to form O2 •− since the O2 /O2 •− redox potential, at ∼−0.51 V vs. the saturated calomel electrode (SCE), lies above the conduction band of TiO2 at ∼0.3 V vs. SCE [41]. Moreover, the Or-II’s LUMO state is positioned at ∼−1.5 V vs. SCE [42], thus, it can inject one electron to the Fe3+ forming Fe2+ . This process is not related to the formation of oxidative radicals. Subsequently, this e− may stabilize the positive vacancy (h+ ) in the TiO2 s valence band. This process leaves free the UV-photogenerated e− in the TiO2 s conduction band to effectively reduce O2 to form O2 •− , thus also involving a photocatalytic oxidation pathway. This process could explain that simultaneous irradiation of UV and vis is strictly necessary to promote the photoactivity of Fe–TiO2 since the Or-II excitation occurs under visible
1182
C.A. Castro et al. / Materials Chemistry and Physics 129 (2011) 1176–1183
light, while the promotion of electrons in the TiO2 s CB is possible under UV irradiation. As a result, these materials have a potential as heterogeneous photocatalysts in the degradation of azo dyes, or to promote an electron flux in dye photosensitized TiO2 for artificial photosynthesis due to its increased electron caption capacity by embedded Fe3+ in TiO2 . 4. Conclusions The chosen hydrothermal synthesis route leads to the formation of TiO2 particles constituted of anatase with embedded Fe3+ . This photocatalyst has an enhanced photooxidation activity towards the degradation of azo dyes. Mineralization tests of Orange II (Or-II) dye by the Fe–TiO2 photocatalysts, under different illumination set-ups, suggests that Fe3+ promotes the electron caption in TiO2 photosensitization by Or-II under solar-simulated light irradiation. Indeed, the photocatalytic oxidation of Or-II was observed under strict simultaneous irradiation of ultraviolet (UV) and visible (vis) of the Fe–TiO2 photocatalyst using a solar-simulated light source. Moreover, Fe3+ effectively promotes visible light absorption of TiO2 , even so, enhancing the recombination of the photogenerated electron–hole pair, as observed in the photocatalytic degradation of phenol under UV irradiation. Therefore, there is a remarked effect of the type of light irradiation source on the coupled photodynamic and photocatalytic degradation of Or-II. Such interaction leads to a double oxidation pathway of the OrII molecule: (i) the auto-degradation due to the charge injection to the TiO2 from its excited state and (ii) the oxidation caused by oxidative species produced in the photosensitization of the Fe–TiO2 photocatalyst.
Acknowledgements Authors wish to present their gratitude to COLCIENCIAS and SENA for the financial support of the project with code: 110234119419. C. Castro thanks also to the named Colombian government entities and to the UIS in Bucaramanga, Colombia, for their financial support to his PhD studies. Special thanks to the Group of Electrochemical Engineering in the EPFL, Switzerland, for the HPLC and TOC measurements.
References [1] T. Lazar, Color chemistry: synthesis, properties, and applications of organic dyes and pigments, Color. Res. Appl. 30 (2005) 313–314. [2] P. Pitter, J. Chudoba, Biodegradability of Organic Substances in the Aquatic Environment, CRC Press, Boca Raton, 1990. [3] O. Carp, C.L. Huisman, A. Reller, Photoinduced reactivity of titanium dioxide, Prog. Solid State Chem. 32 (2004) 33–177. [4] I.K. Konstantinou, T.A. Albanis, TiO2 -assisted photocatalytic degradation of azo dyes in aqueous solution: kinetic and mechanistic investigations. A review, Appl. Catal. B: Environ. 49 (2004) 1–14. [5] Z. Li, W. Shen, W. He, X. Zu, Effect of Fe-doped TiO2 nanoparticle derived from modified hydrothermal process on the photocatalytic degradation performance on methylene blue, J. Hazard. Mater. 155 (2008) 590–594. [6] C. Hu, Y. Lan, J. Qu, X. Hu, A. Wang, Ag/AgBr/TiO2 visible light photocatalyst for the destruction of azodyes and bacteria, J. Phys. Chem. B 110 (2006) 4066–4072. [7] D. Gumy, C. Morais, P. Bowen, C. Pulgarín, S.A. Giraldo, R. Hajdu, J. Kiwi, Catalytic activity of commercial TiO2 powders for the abatement of the bacteria (E. coli) under solar simulated light: influence of the isoelectric point, Appl. Catal. B: Environ. 63 (2006) 76–84. [8] A.G. Rincón, C. Pulgarin, Photocatalytic inactivation of E. coli: effect of (continous-intermittent) light intensity and of (suspended-fixed) TiO2 concentration, Appl. Catal. B: Environ. 44 (2003) 263–2849. [9] L. Zan, W. Fa, T. Peng, Z. Gong, Photocatalysis effect of nanometer TiO2 and TiO2 -coated ceramic plate on Hepatitis B, J. Photochem. Photobiol. B: Biol. 86 (2007) 165–169. [10] M.I. Litter, J.A. Navío, Photocatalytic properties of iron doped-titania semiconductors, J. Photochem. Photobiol. A: Chem. 98 (1996) 171–181.
[11] M.A. Khan, S.I. Woo, O. Yang, Hydrothermally stabilized Fe(III) doped titania active under visible light for water splitting reaction, Int. J. Hydrogen Energy 33 (2008) 5345–5351. [12] S. Liu, Y. Chen, Enhanced photocatalytic activity of TiO2 powders doped by Fe unevenly, Catal. Commun. 10 (2009) 894–899. [13] E. Piera, M.I. Tejedor, M.E. Zorn, M.A. Anderson, Relationship concerning the nature and concentration of Fe(III) species on the surface of TiO2 particles and photocatalytic activity of the catalyst, Appl. Catal. B: Environ. 46 (2003) 671–685. [14] Y. Wu, J. Zhang, L. Xiao, F. Chen, Preparation and characterization of TiO2 photocatalyst by Fe3+ doping together with Au deposition for the degradation of organic pollutants, Appl. Catal. B: Environ. 88 (2008) 525–532. [15] K. Mizushima, M. Tanaka, S. Lida, Energy levels of iron group impurities in TiO2 , J. Phys. Soc. Jpn. 32 (1972) 1519–1524. [16] M. Asiltürk, F. Sayilkan, E. Arpac¸, Effect of Fe3+ ion doping to TiO2 on the photocatalytic degradation of Malachite Green dy under UV and vis-irradiation, J. Photochem. Photobiol. A. Chem. 203 (2009) 64–71. [17] C.A. Castro-López, A. Centeno, S.A. Giraldo, Fe modified TiO2 photocatalysts for the oxidative degradation of recalcitrant water contaminants, Catal. Today 157 (2010) 119–124. [18] Z. Ambrus, N. Balázs, T. Alapi, G. Wittmann, P. Sipos, A. Dombi, K. Mogyorósi, Synthesis, structure and photocatalytic properties of Fe(III)-doped TiO2 prepared from TiCl3 , Appl. Catal. B: Environ. 81 (2008) 27–37. [19] J. Maier, Physics and Chemistry of Ionic Materials: Ions and Electrons in Solids, John Wiley & Sons Ltd., London, 2004. [20] K. Naeem, F. Ouyang, Preparation of Fe3+ -doped TiO2 nanoparticles and its photocatalytic activity under UV light, Physica B 405 (2010) 221–226. [21] M.C. Biesinger, L.W. Lau, A.R. Gerson, R.S. Smart, Resolving surface chemical states in XPS analysis of first row transition metals oxides and hydroxides: Sc, Ti, V, Cu and Zn, Appl. Surf. Sci. 257 (2010) 887–898. [22] K.M. Parida, G.K. Pradhan, Sol-gel synthesis and characterization of mesoporous iron-titanium mixed oxide for catalytic application, Mater. Chem. Phys. 123 (2010) 427–433. [23] N. Todorova, T. Giannakopoulou, G. Romanos, T. Vaimakis, J. Yu, C. Trapalis, Preparation of fluorine-doped TiO2 photocatalysts with controlled crystalline structure, Int. J. Photoenergy 2008 (2008) 1–9. ´ ´ [24] A. Sobczynski, L. Duczman, W. Zmudzinsky, Phenol destruction by photocatalysis on TiO2 : an attempt to solve the reaction mechanism, J. Mol. Catal. A 213 (2004) 225–230. [25] P. Górska, A. Zaleska, J. Hupka, Photodegradation of phenol by UV/TiO2 and vis/N,C–TiO2 processes: comparative mechanistic and kinetic studies, Sep. Purif. Technol. 28 (2009) 90–96. [26] M. Popa, L. Diamandescu, F. Vasiliu, C.M. Teodorescu, V. Cosoveanu, M. Baia, M. Feder, L. Baia, V. Danciu, Synthesis, structural characterization, and photocatalytic properties of iron-doped TiO2 aerogels, J. Mater. Sci. 44 (2009) 358–364. [27] D. Gazzoli, G. Minelli, M. Valigi, The FeOx –TiO2 system. An X-ray diffraction, thermogravimetric and magnetic susceptibility study, Mater. Chem. Phys. 21 (1989) 93–106. [28] G.K. Li, X.Q. Zhang, H.Y. Wu, W.G. Huang, J.L. Jin, Y. Sun, Z.H. Cheng, Electron spin resonance investigation of the subsitution of Fe3+ for Ti4+ in rutile TiO2 single crystal, Chin. Phys. Soc. 18 (2009) 3551–3554. [29] K. Wantala, L. Laokiat, P. Khemthong, N. Grisdanurak, K. Fukaya, Calcination temperature effect on solvothermal Fe–TiO2 and its performance under visible light irradiaiton, J. Taiwan Inst. Chem. Eng. 41 (2010) 612–616. [30] N.A. Jamalluddin, A.Z. Abdullah, Reactive dye degradation by combined Fe(III)/TiO2 catalyst and ultrasonic irradiation: effect of Fe(III) loading and calcination temperature, Ultrason. Sonochem. 18 (2010) 669–678. [31] A.R. Liu, S.M. Wang, Y.R. Zhao, Z. Zheng, Low-temperature preparation of nanocrystalline TiO2 photocatalyst with a very large specific surface area, Mater. Chem. Phys. 99 (2006) 131–134. [32] Y.P. He, Y.M. Miao, C.R. Li, S.Q. Wang, L. Cao, S.S. Xie, G.Z. Yang, B.S. Zou, Size and structure effect on optical transitions of iron oxide nanocrystals, Phys. Rev. B 71 (2005) 125411–125419. [33] J.A. Navio, G. Colon, M. Macias, C. Real, M.I. Litter, Iron-doped titania semiconductor powders prepared by a sol–gel method. Part I. Sythesis and characterization, Appl. Catal. A: Chem. 177 (1999) 111–120. [34] D. Huang, S. Liao, S. Quan, L. Liu, Z. He, J. Wan, W. Zhou, Synthesis and characterization of visible light responsive N-TiO2 mixed crystal by a modified hydrothermal process, J. Non-Cryst. Solids 354 (2008) 3965–3972. [35] S. Ardizzone, C. Bianchi, G. Cappelletti, S. Gialanella, C. Pirola, V. Ragaini, Tailored anatase/Brookite nanocrsytalline TiO2 . The optimal particle features for liquid and gas-phase photocatalytic reactions, J. Phys. Chem. C 111 (2007) 13222–13231. [36] J. Li, J. Xu, W.L. Dai, K. Fan, Direct hydro-alcohol thermal synthesis of special core–shell structured Fe-doped titania microspheres with extended visible light response and enhanced photoactivity, Appl. Catal. B: Environ. 85 (2009) 162–170. [37] B. Xin, Z. Xin, H. Ren, Hu, X. Zhang, C. Dong, K. Shi, L. Jing, H. Fu, Photocatalytic activity and interfacial carrier transfer of Ag–TiO2 nanoparticle films, Appl. Surf. Sci. 252 (2005) 2050–2055. [38] J. Bandara, J. Kiwi, Fast kinetic spectroscopy, decoloration and production of H2 O2 induced by visible light in oxygenated solutions of the azo dye Orange II, New J. Chem. 23 (1999) 717–724. [39] W. Du, Y. Xu, Y. Wang, Photoinduced degradation of Orange II on different Iron (Hydr)oxides in aqueous suspension: rate enhancement on addition of
C.A. Castro et al. / Materials Chemistry and Physics 129 (2011) 1176–1183 hydrogen peroxide, silver nitrate and sodium fluoride, Langmuir 24 (2008) 175–181. [40] C.M. Miller, R.L. Valentine, Mechanistic studies of surface catalyzed H2 O2 decomposition and contaminant degradation in the presence of sand, Water Res. 33 (1999) 2805–2816.
1183
[41] M. Gratzel, Energy Resources Through Photochemistry and Catalysis, Academic Press Inc., New York, 1983. [42] N. Helaïli, Y. Bessekhouad, A. Bouguelia, M. Trari, Visible light degradation of Orange II using xCuy Oz /TiO2 heterojuncions, J. Hazard. Mater. 168 (2009) 484–492.