* The preview only display some random pages of manuals. You can download
full content via the form below.
Applied Catalysis A: General 401 (2011) 98–105
Contents lists available at ScienceDirect
Applied Catalysis A: General journal homepage: www.elsevier.com/locate/apcata
Experimental and theoretical studies of Fe-doped TiO2 films prepared by peroxo sol–gel method Diana V. Wellia a , Qing Chi Xu a , Mahasin Alam Sk a , Kok Hwa Lim a , Tuti Mariana Lim b , Timothy Thatt Yang Tan a,∗ a b
School of Chemical and Biomedical Engineering, Nanyang Technological University, Singapore 637459, Singapore School of Life Sciences and Chemical Technology, Ngee Ann Polytechnic, Singapore 599489, Singapore
a r t i c l e
i n f o
Article history: Received 22 February 2011 Received in revised form 30 April 2011 Accepted 3 May 2011 Available online 11 May 2011 Keywords: Peroxo titanic acid (PTA) Visible light Fe-doped TiO2 Density functional theory (DFT)
a b s t r a c t Transparent, uniform, crack-free and visible light activated Fe-doped TiO2 thin films have been prepared by an organic-free approach using titanium tetrachloride (TiCl4 ), Fe3+ and ammonium hydroxide solutions to form yellow aqueous peroxo titanic acid (PTA) solution. Pure glass was made superhydrophilic after heating at 550 ◦ C for 1 h and was used as the coating substrate. The PTA solution dispersed uniformly on the superhydrophilic glass by forming a thin film, resulting in a crack-free Fe-doped TiO2 film as observed under FESEM. XRD results confirmed the presence of only anatase phase for all samples after calcination at 550 ◦ C. The Fe-doped TiO2 films exhibited slight red-shift in absorbance and enhanced absorbance in the visible-light region compared with undoped TiO2 , attributed to bandgap narrowing by successful Fe3+ doping into TiO2 , which is supported by DOS calculations. Photodegradation of stearic acid (SA) revealed that Fe3+ doping increased visible light photocatalytic activity four-fold compared to undoped TiO2 at optimal Fe-doped TiO2 (1 wt% Fe3+ ) film. At higher Fe ions concentrations, the existence of increased charge recombination sites and segregated iron oxide phase suppressed the photoactivity. © 2011 Elsevier B.V. All rights reserved.
1. Introduction Titanium dioxide (TiO2 ) as a photocatalyst offers an inexpensive, nontoxic, stable, highly oxidizing and reducing power in the degradation of organic pollutants [1,2]. The application of TiO2 in thin film (immobilized form) was superior to the suspension system as the latter required expensive and complex post-operation separation [3]. However, TiO2 can only be activated under UV irradiation due to its wide bandgap (3.2 eV) [4]. Numerous efforts have been dedicated to modify the photocatalyst so that it can be photoactivated under visible light. These include depressing the bandgap by doping TiO2 with metal ions (Fe [5,6], Cr [7,8], Co [9], Cu [10], V [11]). Iron-doped TiO2 has gained attention due to the fact that Fe3+ radius (78.5 pm) is similar to that of Ti4+ (74.5 pm) [12] resulting in easier insertion of Fe3+ into the crystal structure of TiO2 [13]. As a dopant, the role of Fe3+ is still controversial [13]. Some authors suggested detrimental effect of Fe3+ as a promoter that increases the rate of charge recombination [14] while other authors reported the beneficial effect of Fe3+ in enhancing electron/hole separation and thus increasing the photocatalytic activity [6,9,15,16]. The sol–gel method is one of the most widely used techniques to prepare TiO2 coatings as it presents many advantages such
∗ Corresponding author. Tel.: +65 63168829/64608526. E-mail addresses:
[email protected] (T.M. Lim),
[email protected] (T.T.Y. Tan). 0926-860X/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.apcata.2011.05.003
as the use of very simple equipment, low cost, the possibility of using different substrates, and the ability to control the microstructure, homogeneity and density of thin films [17–20]. However, this technique suffers from several disadvantages such as the use of organometallic precursors that are expensive and easily hydrated in air [21]. Besides, this method also requires acid or base to stabilize the prepared sol, rendering it difficulty in the application on corrosive substrate [22]. The peroxo sol–gel method is a promising approach to overcome these issues in the preparation of TiO2 thin film [23]. The peroxo sol–gel method also offers other advantages such as neutral pH condition, low material cost and is environmentally friendly as it uses water instead of organic solvents [22–24]. Moreover, peroxo titanic acid (PTA) is stable in air [23,25] and its preparation is both simple and cost-efficient [25]. PTA solution can be prepared from titanium alkoxide [22,26] or inorganic salt such as TiCl4 [26] or TiCl3 [27]. Ge and Xu added polyethylene glycol (PEG) to a PTA sol to develop porous TiO2 structure [21] while Murakami et al. controlled the shape of TiO2 by adding polyvinyl alcohol [28]. Sonawane et al. used PEG as a stabilizer in the preparation of Fedoped TiO2 thin films [29]. Due to a significant amount of work on Fe-doped TiO2 , theoretical work using density functional theory (DFT) has become important to investigate the electronic properties of Fe-doped TiO2 . Recent study on Fe-doped TiO2 by generalized gradient approximation Perdew–Burke–Eznerhof (GGA-PBE) method showed that the Fe doping “split” the bandgap of pure anatase TiO2 by the introduction of mid-gap state. The predicted
D.V. Wellia et al. / Applied Catalysis A: General 401 (2011) 98–105
99
Fig. 1. Scheme for the preparation of Fe-doped TiO2 samples from peroxo titanic acid (PTA) solution.
bandgap difference (Egap ) of Fe-doped TiO2 is 0.9 eV [30], while another study reported the Egap of Fe-doped TiO2 to be 0.69 eV using GGA-FP-LAPW method [31]. In our previous work, we have demonstrated the preparation of N-doped [32] and C–N-codoped [33] TiO2 thin films by a peroxo sol–gel method. In the present work, we use a similar approach to prepare transparent, uniform, crack-free and visible-light activated Fe-doped TiO2 thin films without the addition of a stabilizer. We also performed DFT (GGA-PBE) calculations to study the structural and electronic properties of pure anatase TiO2 , Fe-doped TiO2 , and Fe–TiO2 with elemental Fe in the lattice cage. The density of state (DOS) was determined to investigate the effect of Fe3+ dopant concentration on the bandgap. The visible light photocatalytic activity of the Fe-doped TiO2 thin films is also investigated using stearic acid (SA) as the model organic compound and the photodegradation efficiency is compared with undoped TiO2 . 2. Experimental 2.1. Chemicals All chemicals were used directly without further purification. Titanium (IV) chloride (TiCl4 ), purchased from Merck was used as TiO2 source; iron (III) chloride hexahydrate (FeCl3 ·6H2 O, Alfa Aesar) was used as the source of Fe dopant; ammonia solution (NH3 ·H2 O, 25%) and hydrogen peroxide (H2 O2 , 30%) were obtained from Sigma and VWR BDH Prolabo, respectively.
2.2. Preparation of Fe-doped TiO2 Coatings The schematic of Fe-doped TiO2 preparation is depicted in Fig. 1. Typically, 3.6 mL of TiCl4 was added drop wise into 300 mL Fe3+ solution containing 0.5 wt% of Fe3+ in an ice-water bath and stirred with a magnetic stirrer. Ammonia solution was then added into the solution, resulting in the formation of white precipitate [Ti(OH)4 ]. After stirring for 24 h, the obtained precipitates were washed thoroughly with DI water repeatedly until no Cl− ions were detected. 86 mL distilled water was added to disperse the precipitate and then 20 mL H2 O2 was added drop-wise into the mixture under continuous magnetic stirring until a transparent yellow solution was obtained. This was used as the coating solution. The same procedure was applied for 1.0, 1.5 and 5.0 wt% of Fe3+ . The glass substrates were first cleaned by ultrasonication in a bath with distilled water for 30 min and then washed with isopropanol, ethanol and acetone sequentially. The glass slides were then heated in a furnace at 550 ◦ C for 1 h with a heating rate of 10 ◦ C min−1 . The glass slides were then cooled to room temperature and the contact angle measured was 2.9◦ indicating the glass slides were superhydrophilic. The prepared glass slide was dip coated with the PTA solution using the KSV Dip Coater at a speed of 0.2 cm/s. The coated glass slide was then heated in the furnace at 550 ◦ C for 1 h. To prepare the TiO2 powder, 30 mL of the yellow coating solution was put into a flask in water evaporator for 1 h followed by oven
Fig. 2. 2 × 2 × 1 Supercell model of (a) anatase TiO2 with the dopant sites, (b) with elemental Fe (blue sphere) in the lattice cage. Grey spheres, Ti atoms; red spheres, O atoms; blue spheres, doping sites. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)
100
D.V. Wellia et al. / Applied Catalysis A: General 401 (2011) 98–105
drying at 70 ◦ C for 12 h. The powder was then calcined at 550 ◦ C for 1 h. 2.3. Characterization The crystalline structures of the synthesized Fe-doped TiO2 were analyzed by X-ray diffraction (XRD). XRD pattern was collected using Rigaku X-ray diffractometer with Cu K␣ radiation ˚ with a scan step of 0.01 min−1 in the range from 20◦ ( = 1.54060 A) ◦ to 70 . Bruker AXS TOPAS v.3 was used to analyze the composition of the crystalline phase. The chemical structures of the samples were determined by FTIR spectrophotometer (Digilab FTS3100). The thermal property of the samples was studied by TG/DTA (Perkin Elmer) in the temperature ranges from room temperature to 800 ◦ C with air flow rate of 20 mL/min and heating rate of 10 ◦ C min−1 . UV–Vis spectra of the thin films were obtained using a UV-Visible spectrophotometer (Shimadzu). The morphology of Fe-doped TiO2 films were evaluated by FESEM (JEOL JSM-6700F SEM). EDX supported in the FESEM was used to determine the amount of Fe present in the Fe-doped TiO2 samples. Water contact angles of coated glasses were measured on a commercial contact angle meter (FTA200 Dynamic Contact Angle Analyzer) at ambient temperature using a sessile drop method. 2.4. Photocatalytic activity test The visible-light photocatalytic activity of the prepared Fedoped TiO2 coating was investigated using stearic acid as a model organic compound [32,33]. Stearic acid was dissolved in methanol (0.035 mol in 100 mL methanol). 300 L of this solution was spuncoated by WS-400B-6NPPL/LITE for 2 min at 2000 rpm onto the glass substrate and dried in an oven at 70 ◦ C. Prior to stearic acid coating, the samples were UV irradiated for 2 h to clean the surface from any organic contaminants that might have been adsorbed from atmosphere. The visible light source used was a 300 W halogen lamp held at 15 cm from the sample with 420 nm UV filter. The degradation of stearic acid was determined by monitoring the concentration of stearic acid using FTIR (Digilab FTS3100), where the absorbance at 2917 cm−1 was converted to a thickness on the basis of an earlier observation that an absorbance of 0.01 corresponded to a thickness of 12.5 nm [32,33]. 2.5. Theoretical calculation
Fig. 3. FTIR spectra (i) before calcinations, (ii) after calcinations, of Fe-doped TiO2 samples with different concentrations of Fe3+ after calcination at 550 ◦ C for 1 h: (a) 0 wt%, (b) 0.5 wt%, (c) 1.0 wt%, (d) 1.5 wt%, and (e) 5.0 wt%.
The structural and electronic properties of pure anatase, and iron doped-TiO2 , TiO2 with elemental iron in the lattice cage (see Fig. 2) was studied using density functional theory (DFT) method. The interaction of elemental Fe with TiO2 crystal was also studied by placing one Fe atom in TiO2 cage. All the calculations were performed using generalized gradient approximation Perdew–Burke–Eznerhof (GGA-PBE) method [34] as implemented in Vienna ab initio simulation package (VASP) [35]. For TiO2 model and computational parameters used in our present study, we would like to refer to our previous work [36]. 3. Results and discussion The isoelectric point of TiO2 is in the range of pH 4.5–6.8. At pH values far from the isoelectric point, the TiO2 particles would bear an electric charge and hence need to be stabilized in the form of sol. Sasirekha et al. reported that at neutral and basic pH, the TiO2 particles are stabilized by the mutual repulsion of the negative charges present at the surface of the TiO2 particle and maintain their particle size [23]. Xu et al. reported N-doped TiO2 was formed at alkaline pH [32]. In this work, pH 7 was chosen to avoid the presence of nitrogen in the Fe-doped TiO2 photocatalyst.
Fig. 4. TG-DTA of Fe-doped TiO2 with 1.0 wt% Fe3+ .
D.V. Wellia et al. / Applied Catalysis A: General 401 (2011) 98–105
101
Fig. 5. (a) Image of transparent Fe-doped TiO2 coated glass; (b) FESEM image of Fe-doped TiO2 thin films (1.0 wt% Fe3+ ); (c) EDX pattern of the undoped TiO2 and (d) Fe-doped TiO2 (1.0 wt% Fe3+ ).
In this work, Fe ion was added as the dopant together with other precursors at the first stage to obtain stable PTA coating solution. It was noted that if the Fe ions were added into the PTA at a final stage, the viscosity of solution immediately increased with Fe ion being precipitated out. The precipitate never reverted to the original transparent PTA solution even if
ammonia solution or hydrogen peroxide solution was added. This was probably due to the formation of TiO(OH)(OOH) immediate species when Fe3+ polyvalent ion was added, resulting in the precipitation [26]. The FTIR spectra of the yellow peroxo titanic acid (PTA) before and after heat treatment are presented in Fig. 3(i) and
102
D.V. Wellia et al. / Applied Catalysis A: General 401 (2011) 98–105
Fig. 6. XRD patterns of Fe-doped TiO2 samples with different concentrations of Fe3+ after calcination at 550 ◦ C for 1 h: (a) 0 wt%, (b) 0.5 wt%, (c) 1.0 wt%, (d) 1.5 wt%, and (e) 5.0 wt%.
(ii) respectively. In Fig. 3(i), a peak at 900 cm−1 corresponding to the stretching vibration of the O–O bond (peroxo group) in the Ti–O–O–H bond of the peroxo titanic acid was observed. The wide bands at 3100–3700 cm−1 are attributed to the stretching vibration of the hydrogen-bonded OH groups of the adsorbed water. The absorption around 1630 cm−1 is assigned to the bending vibration of adsorbed H2 O molecules. The peaks around 1400 cm−1 are assigned to the stretching vibrations of the N–H bonds in NH4 + . PTA has unstable chained atoms of hydrogen, which will decompose and release oxygen molecules to form stable Fe-doped TiO2 during heating. In this process, NH4 + also decomposes. The disappearance of 900 cm−1 and 1400 cm−1 bands in Fig. 3(ii) evidences the decomposition of PTA and NH4+ , respectively [21,26]. TG-DTA profile (Fig. 4) shows an endothermic minimum at 100 ◦ C and exothermic peak maximum at 260 ◦ C. The first stage of weight loss (endothermic) is attributed to the removal of physically adsorbed water. The second stage (exothermic) is attributed to the decomposition of peroxo group [37]. Other broad exothermic peak at about 320 ◦ C is probably due to the decomposition of remaining NH4 Cl [26]. There is no further weight loss in the range of 400–420 ◦ C, but a broad exothermic peak was observed which is attributed to the slow conversion of amorphous phase to anatase [37]. Our observation is consistent with the 3 distinct stages reported during the annealing process: (1) removal of water, (2) decomposition of a peroxo group, and (3) amorphous-anatase phase transformation [37]. A prepared Fe-doped TiO2 coated glass is shown in Fig. 5a. The coated glass is visually transparent and uniform with a yellow tint due to Fe doping. FESEM image (Fig. 5b) reveals that the Fe-doped TiO2 thin films have no cracks, attributed to uniform dispersion of PTA on the superhydrophilic pure glass. EDX measurement (Fig. 5c and d) further confirmed the presence and amount of iron present in the Fe-doped TiO2 samples. Fe element was observed in the EDX pattern in the Fe-doped TiO2 (Fig. 5d), while no trace of Fe element was found in the EDX pattern of the undoped TiO2 (Fig. 5c). The amounts of Fe element in the undoped TiO2 , Fe-doped TiO2 (0.5 wt% Fe3+ ), Fe-doped TiO2 (1.0 wt% Fe3+ ), Fe-doped TiO2 (1.5 wt% Fe3+ ) and Fe-doped TiO2 (5.0 wt% Fe3+ ) were determined to be 0, 0.60, 0.96, 1.74 and 4.96 wt%, respectively. The results are consistent with the original amounts of Fe in the samples. Fig. 6 shows the XRD patterns of Fe-doped TiO2 samples with different concentrations of Fe3+ after calcination at 550 ◦ C for 1 h. The XRD spectra show the presence of only anatase phase. No peaks due to haematite (Fe2 O3 ) are observed in any of the samples. This
Fig. 7. XRD patterns of samples after calcination at 800 ◦ C for 1 h with different concentrations of Fe3+ : (a) 0.5 wt%, (b) 1.0 wt%, (c) 1.5 wt%, and (d) 5.0 wt% (r = rutile, o = iron titanium oxide peak).
may due to Fe3+ ions substituting Ti4+ ions in the TiO2 lattice since the radii of Ti4+ and Fe3+ ions are similar (78.5 pm and 74.5 pm for Ti4+ and Fe3+ radii [12] respectively) and the electronegativity of both ions are reasonably close (Fe3+ : 1.96 [12], Ti4+ : 1.5 [38]). Based on Hume–Rothery rule, if the % difference of atomic radii is less than 15% and the electronegativity of the elements are similar, a substitutional solid solution is most likely to be formed. The phenomenon of Fe3+ ion doping to TiO2 crystal satisfies the third of Hume–Rothery rule as well which states that a lower-valent metal will be soluble in a higher-valent host [39]. Hence the Fe ions may be inserted into the lattice site of Ti4+ [13]. To detect possible segregated iron oxide phase in the Fe-doped TiO2 , the samples were calcined at 800 ◦ C for 1 h. The XRD spectra (Fig. 7) show the presence of rutile for all samples. For Fe-doped TiO2 samples with 0.5, 1.0 and 1.5 wt% Fe3+ , no other peak was observed. However, for the sample with 5.0 wt% Fe3+ , two peaks at 2 = 33◦ and 60.5◦ were identified as perovskite structure of iron titanium oxide (Fe2 TiO5 ) [13]. Analysis using Bruker AXS TOPAS v.3 showed that 0.5, 1.0 and 1.5 wt% Fe3+ samples contain 100% rutile, while 5.0 wt% Fe3+ sample contain 0.84% Fe2 TiO5 in addition to TiO2 rutile phase. These data suggest that almost all of the Fe3+ in 0.5, 1.0
Fig. 8. Absorbance spectra of Fe-doped TiO2 with different concentrations of Fe3+ : (a) 0 wt% (undoped TiO2 ), (b) 0.5 wt%, (c) 1.0 wt%, (d) 1.5 wt% and (e) 5.0 wt%.
D.V. Wellia et al. / Applied Catalysis A: General 401 (2011) 98–105
103
Fig. 9. Density of state (DOS) of pure and Fe-doped TiO2 , TiO2 with elemental Fe in the lattice cage. The dotted line is Fermi Energy level.
and 1.5 wt% Fe3+ samples were trapped in the crystal lattice of TiO2 , whereas the 5.0 wt% Fe3+ sample contains segregated iron oxide in addition to Fe3+ doped TiO2 . DFT calculation was used to prove the existence of doping and segregated oxide or other phase. In our present study, we have considered 4.3 and 8.6 wt% Fe doping in TiO2 . To investigate the stability of pure and Fe-doped TiO2 , the formation energy (Ef ) is calculated according to Eq. (1) as follows: Ef = Ecrystal − ni i
(1)
where, Ecrystal is the total energy of the crystal; ni and i are the number of atoms and chemical potential of various elements (Ti, Fe, O, etc.), respectively. The chemical potential of Ti and Fe is derived Table 1 Calculated formation energies difference Ef of Fe-doped TiO2 and TiO2 with elemental Fe in the lattice cage. Please refer to Fig. 2 for the definition of doping. No
Systems
Ef a
1 2 3
4.3 wt% Fe–TiO2 b 8.6 wt% Fe–TiO2 c Fe–TiO2 –Ed
0.13 0.25 0.10
a b c d
Pure TiO2 is taken as reference formation energy. 4.3 wt% Fe–TiO2 : Ti1 is replaced by Fe. 8.6 wt% Fe–TiO2 : both Ti1 and Ti2 are replaced by Fe. TiO2 with atomic Fe in the lattice cage.
from bulk crystal and the chemical potential of O is derived from molecular O2 . The calculated formation energy difference (Ef ) is listed in Table 1 with the formation energy of TiO2 (−3.39 eV/atom) as a reference. We also observed that with increasing dopant concentration, the stability of Fe-doped TiO2 decreases compared to TiO2 . Thus, our results suggest that at higher dopant concentration, segregated Fe2 O3 phase may be formed, which is consistent with our experimental observation. The formation energy of TiO2 with elemental Fe in the lattice cage is similar to Fe-doped TiO2 , which suggests the possibility of entrapment of elemental Fe in TiO2 cage. In the doping process of Fe3+ into TiO2 , wherein Fe3+ ions replace Ti4+ , Fe3+ will act as an electron donor and form donor level close to the conduction band, resulting in a smaller energy transition, which may lead to visible light photoactivation [39,40]. In Fig. 8, the
Table 2 Calculated bandgap differences (Egap ) of Fe-doped TiO2 and TiO2 with elemental Fe in the lattice cage. No
Systems
Egap a
1 2 3
4.3 wt% Fe–TiO2 8.6 wt% Fe–TiO2 Fe–TiO2 –E
−0.53 −0.61 −1.34
a
Pure TiO2 is taken as the reference bandgap.
104
D.V. Wellia et al. / Applied Catalysis A: General 401 (2011) 98–105
is listed in Table 2. The calculated bandgap difference (Egap ) of 4.3 and 8.6 wt% Fe-doped TiO2 is −0.53 and −0.61 eV, respectively. The experimental bandgap difference (Egap = −0.69) of Fe-doped TiO2 [44] is comparable with our calculated bandgap difference of 8.6 wt% Fe-doped TiO2 . Our results indicate that Fe doping decreases the bandgap of TiO2 . Here, we observe that bandgap decreases with increasing Fe doping which is consistent with the current experimental observation of absorbance red shift (Fig. 8). Our results also show that the elemental Fe has strong effect on the bandgap of TiO2 as compared to doped Fe (Egap = −1.34). Although our theoretical study shows that the entrapment of elemental Fe in the cage of TiO2 is possible, the large bandgap difference, which will result in a very small bandgap for elemental Fe contained in the cage of TiO2 , is not supported by current experimental data. Therefore, we suggest that all the iron in the current Fe-doped TiO2 films were doped in the crystal lattice with a small amount of segregated phase for the case of 5.0 wt% Fe3+ . The photocatalytic activity of the Fe-doped TiO2 coated glass was evaluated by photodegradation study of stearic acid (SA) under visible light irradiation (Fig. 10). 1.0 wt% Fe3+ was found to be the optimal dopant concentration and the associated photocatalytic activity was 4 times higher than that of undoped TiO2 coated glass. Fe ions doping extends light absorption into the visible light region, supported by other studies [13,16] and our current results (Fig. 8). However, higher Fe ions concentration at 1.5 wt% may increase the number of trap sites [7,45], resulting in higher recombination rate and lower photocatalytic activity. At 5.0 wt% Fe ions concentration, the existence of segregated iron oxide phase further decreased the photoactivity [45–48]. 4. Conclusions
Fig. 10. (a) Photocatalytic activities of undoped TiO2 and Fe-doped TiO2 films under visible light illumination for 24 h; (b) evolution of the IR absorbance spectra of Fedoped TiO2 (1.0 wt%).
UV–Vis absorption spectra of the Fe-doped TiO2 films show obvious enhanced absorption in the visible light region (400–600 nm) compared to that of the undoped TiO2 film, indicating their potential to absorb visible light and improve photocatalytic activities under visible light illumination, even though no significant shift of the absorption edge to visible light region is observed [41]. The absorption edges of undoped TiO2 , Fe-doped TiO2 (1.0 wt%) and Fedoped TiO2 (5.0 wt%) films are 340, 350 and 362 nm, respectively. According to the empirical formula, Eg = 1239/edge (edge = the wavelength of the optical absorption edge) [42], the bandgap energies of the undoped TiO2 , Fe-doped TiO2 (1.0 wt%) and Fe-doped TiO2 (5.0 wt%) films are 3.64, 3.54 and 3.42 eV, respectively. The bandgap energies of the undoped TiO2 , Fe-doped TiO2 (1.0 wt%) and Fe-doped TiO2 (5.0 wt%) films are higher than the usually reported in literature (3.2 eV for anatase TiO2 ), which may be attributed to quantum size effect of TiO2 film and the thermal stress in the films due to the difference in the thermal expansion coefficients between the fused substrate and coating material [43]. The density of states (DOS) were determined to study the effect of doping on bandgap variation. We have also investigated the effect of elemental Fe doping on the changes of bandgap structure of TiO2 . The DOS plots of TiO2 , Fe-doped TiO2 and TiO2 with elemental Fe in the lattice cage are shown in Fig. 9. The calculated bandgap difference (Egap ) of Fe-doped TiO2 with respect to TiO2
Visible light active Fe-doped TiO2 coated glass slides have been successfully prepared using a “green” aqueous PTA solution approach. The prepared thin films are transparent, uniform and crack free. The visible light photocatalytic activity of the Fe-doped TiO2 films is attributed to bandgap narrowing as observed from absorption red-shift and DOS results. The photocatalytic activity of optimal 1.0 wt% Fe3+ doped TiO2 was about 4 times higher than that of undoped TiO2 coated glass. DFT results indicated that Fe3+ was doped into the TiO2 crystal lattice but at higher dopant concentration, its stability may be lowered due the presence of segregated iron oxide phase. Acknowledgments Financial support from Nanyang Technological University AcRF Tier 1 RG29/07 is gratefully acknowledged. D.V.W. acknowledges a research scholarship from Nanyang Technological University. References [1] Y. Kuroda, T. Mori, K. Yagi, N. Makihata, Y. Kawahara, M. Nagao, S. Kittaka, Langmuir 21 (2005) 8026–8034. [2] R. Asahi, T. Morikawa, T. Ohwaki, K. Aoki, Y. Taga, Science 13 (2001) 269–271. [3] T.E. Doll, F.H. Frimmel, Acta Hydroch. Hydrob. 32 (2004) 201–213. [4] M. Ksibi, A. Zemzemi, R. Boukchina, J. Photochem. Photobiol. A 159 (2003) 61. [5] Y. Yalc¸ın, M. Kılıc¸, Z. C¸ınar, Appl. Catal. B: Environ. 99 (2010) 469–477. [6] C. Wang, C. Böttcher, D.W. Bahnemann, J.K. Dohrmann, J. Mater. Chem. 13 (2003) 2322–2329. [7] L. Kernazhitsky, V. Shymanovska, V. Naumov, V. Chernyak, T. Khalyavka, V. Kshnyakin, Ukr. J. Phys. Opt. 9 (3) (2008) 197–207. ˛ G. Stochel, Z. Stasicka, W. Macyk, Catal. Today 161 (2011) [8] J. Kuncewicz, P. Zabek, 78–83. [9] J. Choi, H. Park, M.R. Hoffmann, J. Phys. Chem. C 114 (2010) 783–792. [10] T. Morikawa, Y. Irokawa, T. Ohwaki, Appl. Catal. A: Gen. 314 (2006) 123–127. [11] S.M. Chang, W.S. Liu, Appl. Catal. B: Environ. 101 (2011) 333–342. [12] J. Chen, M. Yao, X. Wang, J. Nanopart. Res. 10 (2008) 163–171. [13] M. Asiltürk, F. Sayilkan, E. Arpac, J. Photochem. Photobiol. A: Chem. 203 (2009) 64–71.
D.V. Wellia et al. / Applied Catalysis A: General 401 (2011) 98–105 [14] J.A. Navio, J.J. Testa, P. Djedjeian, J.R. Padron, D. Rodriguez, M.I. Litter, Appl. Catal. A: Gen. 178 (2) (1999) 191–203. [15] S.W. Lam, K. Chiang, T.M. Lim, R. Amal, G.K.C. Low, J. Catal. 234 (2005) 292–299. [16] J. Yu, Q. Xiang, M. Zhou, Appl. Catal. B: Environ. 90 (2009) 595–602. [17] W.A. Douod, J.H. Xin, Chem. Commun. (2005) 2110–2112. [18] B. Guo, Z. Liu, L. Hong, H. Jiang, Surf. Coat. Technol. 198 (2005) 24–29. [19] T. Ivanova, A. Harizanova, M. Surtchev, Z. Nenova, Sol. Energy Mater. Sol. Cells 76 (2003) 591–598. [20] W. Zhang, W. Liu, C. Wang, Wear 253 (2002) 377–384. [21] L. Ge, M. Xu, J. Sol–Gel Sci. Technol. 43 (2007) 1–7. [22] Y.J. Liu, M. Aizawa, Z.M. Wang, H. Hatori, N. Uekawa, H. Kanoh, J. Colloid Interface Sci. 322 (2008) 497–504. [23] Y.F. Gao, Y. Masuda, K. Koumoto, Chem. Mater. 16 (2004) 1062. [24] L. Ge, M. Xiu, M. Sun, H. Fang, Mater. Res. Bull. 41 (2006) 1560–1603. [25] Y.F. Gao, Y. Masuda, K. Koumoto, Langmuir 20 (2004) 3188. [26] T.P. Niesen, J. Bill, F. Aldinger, Chem. Mater. 13 (2001) 1552–1559. [27] C.K. Lee, D.K. Kim, J.H. Lee, J.H. Sung, I. Kim, J. Sol–Gel Sci. Technol. 31 (2004) 67–72. [28] N. Murakami, Y. Kurihara, T. Tsuboto, T. Ohno, J. Phys. Chem. C 113 (2009) 3062–3069. [29] R.S. Sonawane, B. Kale, M.K. Dongare, Mater. Chem. Phys. 85 (2004) 52–57. [30] E. Thimsen, S. Biswas, C.S. Lo, P. Biswas, J. Phys. Chem. C 113 (2009) 2014– 2021. [31] C.C. Yen, D.Y. Wang, M.H. Shih, L.S. Chang, H.C. Shih, Appl. Surf. Sci. 256 (2010) 6865–6870. [32] Q.C. Xu, D.V. Wellia, R. Amal, D.W. Liao, J.S.C. Loo, T.T.Y. Tan, Nanoscale 2 (7) (2010) 1122–1127.
105
[33] Q.C. Xu, D.V. Wellia, S. Yan, D.W. Liao, T.M. Lim, T.T.Y. Tan, J. Hazard. Mater. 188 (2011) 172–180. [34] J.P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 77 (1996) 3865–3868. [35] G. Kresse, J. Joubert, Phys. Rev. B 59 (1999) 1758–1775. [36] Q.C. Xu, D.V. Wellia, M.A. Sk, K.H. Lim, J.S.C. Loo, D.W. Liao, R. Amal, T.T.Y. Tan, J. Photochem. Photobiol. A 210 (2010) 181–187. [37] G.H. Kim, C.G. Lee, I. Kim, Met. Mater. Int. 10 (5) (2004) 423–427. [38] Y.C. Cao, T. He, Y. Chen, Y. Cao, J. Phys. Chem. C 114 (2010) 3627–3633. [39] H. Yamashita, M. Harada, J. Misaka, M. Takeuchi, B. Neppolian, M. Anpo, Catal. Today 84 (2003) 191–196. [40] C.Y. Wang, D.W. Bahnemann, J.K. Dohrmann, Chem. Commun. (2000) 1539–1540. [41] Q. Wang, Z. Jiang, Y. Wang, D. Chen, D. Yang, J. Nanopart. Res. 11 (2009) 375–384. [42] W. Weng, M. Ma, P. Du, G. Zhao, G. Shen, J. Wang, G. Han, Surf. Coat. Technol. 198 (2005) 340–344. [43] Q.C. Xu, D.V. Wellia, Y.H. Ng, R. Amal, T.T.Y. Tan, J. Phys. Chem. C (2011), doi:10.1021/jp1090137. [44] E.D. Jeong, P.H. Borse, J.S. Jang, J.S. Lee, O.S. Jung, H. Chang, J.S. Jin, M.S. Won, H.G. Kim, J. Ceram. Process. Res. 9 (2008) 250–253. [45] J.A. Navìo, G. Colón, M. Macías, C. Real, M.I. Litter, Appl. Catal. A: Gen. 177 (1999) 111–120. [46] B. Pal, M. Sharon, G. Nogami, Mater. Chem. Phys. 59 (1999) 254–261. [47] J.A. Wang, R. Limas-Barresteros, T. López, A. Moreno, R. Gómez, O. Navaro, X. Bokhimi, J. Phys. Chem. B 105 (40) (2001) 9692–9698. [48] Z.M. Wang, G. Yang, P. Biswas, W. Bresser, P. Boolchand, Powder Technol. 114 (2001) 197–204.