Coupling of W-Doped SnO₂ and TiO₂ for Efficient Visible-Light Photocatalysis

Abstract

Five mol % tungsten-doped tin oxide ($W_{0.05}Sn_{0.95}O_2$, TTO5) was prepared by co-precipitation of $SnCl_4{\cdot}5H_2O$ and $WCl_4$, followed by calcination at $1000^{\circ}C$. The as-prepared TTO5 was in the pure cassiterite phase with a particle size of ~50 nm and optical bandgap of 2.51 eV. Herein it was applied for the formation of TTO5/$TiO_2$ heterojunctions by covering the TTO5 surface with $TiO_2$ by sol-gel method. Under visible-light irradiation (${\lambda}{\geq}420$ nm), TTO5/$TiO_2$ showed a significantly high photocatalytic activity in removing gaseous 2-propanol (IP) and evolving $CO_2$. It is deduced that its high visible-light activity is caused by inter-semiconductor holetransfer between the valence band (VB) of TTO5 and $TiO_2$, since the TTO5 nanoparticle (NP) exhibits the absorption edge at ~450 nm and its VB level is located more positive side than that of $TiO_2$. The evidence for the hole-transport mechanism between TTO5 and $TiO_2$ was also investigated by monitoring the holescavenging reaction with 1,4-terephthalic acid (TA).

Introduction

Photocatalytic removal of volatile organic compounds and environmental pollutants in air and aqueous solution has attracted extensive attentions for the last few decades.1-4 Even though TiO2 has been known as the most efficient photocatalyst, however, TiO2 can only utilize the photons in UV region (λ ＜ 380 nm) due to its wide bandgap (Eg = 3.2 eV), which limits its practical applications in sun light or indoor.5-9 Therefore, the design of new photocatalysts working under visible-light is necessary in order to utilize the major portion of the solar spectrum.

Thus far, several strategies have been attempted in designing visible-light photocatalysts, such as lowering the conduction band (CB) of TiO2 by doping transition metal ions10-13 and elevating the valence band (VB) by substituting anions to the oxygen site of TiO2.14-17 Another noticeable strategy in designing visible-light photocatalysts is coupling the narrow bandgap semiconductors to the TiO2.18-27 On the basis of relative energy band location between the sensitizers and TiO2, used as main photocatalyst, the heterojunction structure can be classified as two kinds. First, if sensitizer CB is positioned more negative side than that of TiO2 (denoted as Type-A heterojunction), the photo-induced electrons of the sensitizer can be transferred to TiO2 CB, and these electrons can be used for the chemical reactions. Metal calcogenide quantum dot-deposited or molecular dye-anchored TiO2 systems will be the typical examples of Type-A heterojunction.18-22 Second, if the VB level of sensitizer is located more positive side than that of TiO2, as shown in Scheme 1(a), the holes induced in the sensitizer VB under visible-light irradiation can be transferred to that of TiO2. This is classified as “Type- B heterojunction”,23-27 in which the photocatalytic reaction takes place on the basis of intersemiconductor hole transfer mechanism. Considering the powerful oxidative ability of the holes in the VB of TiO2, efficient and complete oxidation of organic compounds is expected with this heterojunction system under visible-light.

SnO2 (3.5 eV vs. NHE) is considered to be an appropriate candidate for the construction of Type-B heterojunction structure, since the VB of SnO2 (+3.5 V vs. NHE) has been reported to be located considerably more positive side than that of TiO2 (+2.7 V vs. NHE),28,29 but its wide bandgap (Eg) of 3.5 eV does not allow utilization of visible-light. Herein, we doped the W ions into the crystal lattice of SnO2, and successfully lowered the conduction band position of SnO2. Thus the preapred tungsten-doped tin oxide (WxSn1−xO2, TTO) can absorb the considerable portion of visible light. In this work, we prepared the several composites between W0.05Sn0.95O2 (TTO5) nanoparticle (NP) and TiO2 by sol-gel method. The visible-light phototocatalytic mechanism as well as the photocatalytic behavior of TTO5/TiO2 composites in decomposing gaseous IP was rigorously investigated.

Scheme 1.Energy band diagram describing the photo-generated charge transport in the TTO5/TiO2 heterojunction under visiblelight irradiation (a), and preparation strategy of TTO5/TiO2 (b).

 

Experimental

Preparation of Tungsten Doped Tin Oxides (TTOs). TTO NPs were synthesized by co-precipitation and subsequent calcination process.30 In typical synthesis of 5 mol ％ W-doped SnO2 (TTO5), 9.5 mmol of tin(IV) chloride pentahydrate (SnCl4·5H2O, Aldrich) and the stoichiometric amount of tungsten(IV) chloride (WCl4, Aldrich) were dissolved in 50 mL anhydrous ethanol. After vigorous stirring for 1 h, aqueous ammonia solution (NH4OH, 30 wt ％ aqueous solution) was slowly added to obtain pH 7.5. The solution was then precipitated, and the collected precipitate was washed with de-ionized water for several times. It was then dried overnight in air at 60 ℃, followed by calcination at 1000 ℃ for 3 h.

Preparation of TTO5/TiO2 Heterojunction. For the formation of TTO5/TiO2 heterojunction structures, 3.56 g titanium isopropoxide (97％, Aldrich) was stabilized in a mixed solution of 30 mL ethanol, 1 mL concentrated nitric acid, and 1 mL water. The mixed Ti-precursor solution was gently stirred for 1 h for the stabilization. The stoichiometric amount of TTO5 NP was then added to this solution. Typically, in preparing the composite consisting of 3 wt ％ TTO5 and 97 wt ％ TiO2 (3/97 TTO5/TiO2), 30 mg TTO5 NP pre-suspended in 10 mL ethanol was added to the Tiprecursor solution and gently stirred overnight. The sample was then dried at 90 ℃ for 24 h, and subsequently heattreated at 300 ℃ for 3 h.

Characterization. X-ray diffraction (XRD) patterns were obtained for the powder samples by using a Rigaku Multiflex diffractometer with monochromatic light-intensity Cu Kα radiation. XRD scanning was performed under ambient conditions over 2θ region of 20-700 at a rate of 2°/min (40 kV, 20 mA). UV-visible diffuse reflectance spectra were acquired by a Perkin-Elmer Lambda 40 spectrophotometer. BaSO4 was used as the reflectance standard. SEM image was observed by Field emission scanning electron microscope (FE-SEM, Hitachi S-4500). Transmission electron microscope (TEM) images were obtained by a JOEL JEM 2100F operated at 200 kV, 1 mg of powder samples were dispersed in 50 mL of ethanol, and a drop of the suspension was then spread on a holey amorphous carbon coated Ni grid (JEOL Ltd.).

Evaluation of Photocatalytic Activity. The visible-light photocatalytic efficiencies of the photocatalytic samples were estimated by monitoring the evolved amount of CO2 by decomposing 2-propanol (IP) in gas phase. An aqueous suspension containing 8.0 mg of photocatalytic sample was spread on a 2.5 × 2.5 cm2 Pyrex glass in a film form and subsequently dried at room temperature. The gas reactor system used for this photocatalytic activity has been described elsewhere.31 The net volume of the gas-tight reactor was 200 mL, and the photocatalytic film was located at the center of the reactor. The entire area of the photocatalytic film (2.5 cm × 2.5 cm) was irradiated by a 300 W Xe lamp through a UV cut-off filter (λ ＜ 422 nm, ZUL0422 Asahi Co.) and a water filter to cut-off IR. After evacuation of the reactor, 1.6 μL of the IP diluted in water (IP:H2O = 1:9 in volume) was injected into the reactor. The initial concentration of gaseous IP in the reactor was maintained at 117 ppm in volume (ppmv). Thus the ultimate concentration of CO2 evolved, with all of the IP decomposed, will be 351 ppmv, as shown in the following equation:

The total pressure of the reactor was then adjusted to 750 Torr by adding oxygen gas. Under this condition, the IP and H2O remained in the vapor phase. After a certain irradiation interval, 0.5 mL of the gas in the reactor was automatically picked up and sent to a gas chromatograph (Agilent Technologies, Model 6890N) using an auto sampling valve system. For CO2 detection, a methanizer was installed between the GC column outlet and the FID detector.

 

Results and Discussion

TTO particles in different W-doping levels were prepared by co-precipitation of SnCl4·5H2O and WCl4, followed by calcination at 1000 ℃ for 3 h. X-ray diffraction patterns in Figure 1(a) show that the pure cassiterite phase (JCPDS, No. 77-0450) can be retained up to 5 mol ％ of W ion incorporation, whereas further incorporation leads to formation of the impurity peak, inherent from the orthorhombic SnWO4 phase (JCPDS, No. 29-1354). It is suggested that only a limited amount of W ions can be doped into the SnO2 lattice and the maximum doping level is 5 mol %. Hence, 5 mol ％ W-doped SnO2 (W0.05Sn0.95O2, TTO5) was prepared and utilized in fabricating the coupled structure with TiO2 in this work. In TTO structure, the W6+ ions replace the Sn4+ in the SnO2 lattice.32 Thus in the energy band structure of TTO they form the interbands, originating from W 5d and 6s, below the CB of SnO2, as shown in Scheme 1(a).

Figure 1.XRD patterns of the bare SnO2 and several WxSn1−xO2 (TTO) in different doping levels (a), and those of TiO2, TTO5, and several TTO5/TiO2 composites (b).

UV-visible diffuse reflectance spectrum of TTO5 in Figure 2(a) shows that the absorption band edge is ~450 nm. The bandgap of the prepared TTO5 was determined by the following equation.

where α, hν, A and Eg are the optical absorption coefficient, the photon energy, proportionality constant, and band gap, respectively.33 In this equation, n decides the type of the transition in a semiconductor (n = 1, direct absorption; n = 4, indirect absorption). By applying n = 1, the direct bandgap of the prepared TTO5 was determined from the plot of (αhν)2 vs. hν, as shown in the inset of Figure 2. By extrapolating the straight line to the x-axis in this plot, Eg of the TTO5 was estimated to 2.51 eV.

Figure 2.UV-visible diffuse reflectance spectrum of TTO5 obtained at absorbance mode (a), and those spectra for TiO2, SnO2, 1/99 TTO5/TiO2, 3/99 TTO5/TiO2, 5/95 TTO5/TiO2, TTO5 obtained at reflectance mode (b). The plot of (αhν)2 vs. (hν) for TTO5 is shown in the inset of a.

The SEM image in Figure 3(a) indicates that the asprepared TTO5 is moderately monodispersed with a size of 50-60 nm. Figure 3(b) exhibits the TEM image of a single TTO5 NP and its dotted area was further magnified, as shown in Figure 3(c). The uniform fringe patterns observed over the whole area suggest that the individual particle is a single crystal. The spacing of the fringe patterns was measured to be 0.335 nm, corresponding to d110 of the cassiterite SnO2 phase.

Figure 3.SEM images of TTO5 NPs (a). TEM image of a TTO5 NP (b), and high resolution image taken over the dotted area in b (c).

Scheme 1(b) describes the preparation strategy of TTO5/ TiO2 coupled structure derived from sol-gel method. Figure 1(b) shows the XRD patterns of the pure TTO5, TiO2, and TTO5/TiO2 composites in different compositions. The diffraction peaks at 25.36°, 37.90°, 48.15°, 54.05°, 55.20°, 62.86° and 75.27° corresponding to (101), (004), (200), (105), (211), (204) and (215) peaks, respectively, of the anatase TiO2 (JCPDS, No. 73-1764), clearly indicating that the solgel processed TiO2 has been sufficiently crystallized. XRD patterns of the several TTO5/TiO2 composites also indicate no presence of other impurity peaks, suggesting that there was no chemical reaction occurred between TTO5 and TiO2 during the heat-treatment at 300 ℃. UV-visible diffuse reflectance spectra of the TTO5/TiO2 in several compositions are shown in Figure 2(b). Since TTO5 reveals significant absorption toward the photons shorter than 450 nm, TTO5/ TiO2 composites can also show appreciable absorption in the visible region, and its absorbance gradually increases with increase of TTO5 content. The TEM image of the 3/97 TTO5/TiO2 is shown in Figure 4(a). The TTO5 NP with a size of ~50 nm was fully covered with TiO2. Figure 4(b) is a high-magnification image focusing the interface between TiO2 and TTO5. The sharp interface without any gap or interface clearly indicates that the TiO2 is tightly bound to the TTO5.

Figure 4.TEM images of 3/97 TTO5/TiO2 heterojunction (a) and its high resolution image monitoring the interface of TiO2 and TTO5 (b).

Photocatalytic activities of the bare TTO5, TiO2, and several TTO5/TiO2 composites in decomposing gaseous IP were evaluated under visible-light irradiation (λ ≥ 420 nm). It is well known that the photocatalytic reaction proceeds through the oxidation of IP to acetone and it is then fully mineralized to CO2. Figure 5(a) describes the photocatalytic removal of IP with several photocatalytic samples as a function of visible-light irradiation time. The TTO5/TiO2 composites exhibit remarkably higher catalytic activity than the bare TTO5 or TiO2. Especially, the 3/97 TTO5/TiO2 exhibited the highest photocatalytic activity: About 72％ of IP was removed in 2 h irradiation, whereas 5.7％ and 3.5％ was removed with Degussa P25 and bare TTO5, respectively.

Figure 5.Visible-light photocatalytic activities of TiO2, TTO5, and several TTO5/TiO2 composites in removing IP (a) and evolving CO2 (b).

The photocatalytic activity was also evaluated, according to the amount of CO2 evolved under visible-light irradiation. The 3/97 TTO5/TiO2 composite also demonstrated the highest photocatalytic activity in evolving CO2 among the several composites in different compositions, as shown in Figure 5(b). The amount of CO2 evolved in 2 h is 6.9 ppmv, which is 6.2 times that of Degussa P25 TiO2 and 15.1 times that of the bare TTO5. Noticeably, its photocatalytic efficiency reaches to 80.2％ that of the typical N-doped TiO2, as illustrated in Figure 6.

Figure 6.Comparison of photocatalytic activities of TiO2, TTO5, 3/97 TTO5/TiO2 and N-doped TiO2 in evolving CO2 in 2 h visiblelight irradiation.

It is deduced that the high visible-light photocatalytic activity of TTO/TiO2 is caused by the relative energy band location between these two semiconductors. As shown in Scheme 1(a), the VB position of TTO5 is located more positive side, compared with that of TiO2. Under visiblelight irradiation, the electrons in the VB of TTO5 are excited to its CB. Thereby, the holes in the TTO5 VB can be transferred to the VB of TiO2. By this inter-semiconductor hole-transport mechanism, holes are generated on the TiO2 VB, resultantly producing the •OH radicals by the Eqs. (3) and (4).

As a result, the generated •OH radicals can completely decompose the organic compounds to CO2 and H2O.

In order to support the hole transfer mechanism, the presence of •OH radicals in the TiO2 side of the TTO5/TiO2 during the visible-light irradiation was monitored. That is, 20 mg TTO5, TiO2 or 3/97 TTO5/TiO2 was suspended in 50 mL aqueous solution containing 0.01 M NaOH and 3 mM 1,4-terephthalic acid (TA). Before exposure to the visiblelight, the suspension was stirred in dark for 30 min. Five mL of the solution was then taken after 2 h irradiation for the fluorescence measurement. It is well known that •OH radical reacts with TA in basic solution and generates 2-hydroxy terephthalic acid (TAOH), which emits the unique fluorescence peak at 426 nm.34,35

Bare TiO2 suspended in TA solution did not show no fluorescence peak under visible-light (λ ≥ 420 nm) irradiation, as shown in Figure 7. Only a small amount of •OH radical was formed with the bare TTO5, even though TTO5 can be sensitized by visible-light. This will be caused by the fast recombination of electron-hole pairs in the TTO5 NP. Contrarily, the TTO5/TiO2 shows the intense characteristic fluorescence peak. This clearly indicates that lots of holes are formed at the TiO2 side and that they have been transported from the TTO5 VB by the inter-semiconductor holetransport mechanism.

Figure 7.Fluorescence spectra measured after 2 h visible-light (λ ≥ 420 nm) irradiation for each TiO2, TTO5, and 3/97 TTO5/ TiO2 suspension in 3 mM TA. Wavelength of the excitation light for obtaining the fluorescence spectra is 312 nm.

 

Conclusion

About 50 nm-sized TTO5 NP was successfully prepared by co-precipitation and subsequent heat-treatment at 1000 ℃. The 5 mol ％ doping of W ion in the lattice site of SnO2 reduced the band gap of SnO2 from 3.5 to 2.51 eV. The coupled structure of TTO5 and TiO2 exhibited a significantly enhanced visible-light photocatalytic activity. In evolving CO2 from the gaseous IP with 3/97 TTO5/TiO2, the amount of CO2 evolved in 2 h was 6.9 ppmv, which is 6.2 times that of Degussa P25 TiO2 and 80.2％ that of the typical N-doped TiO2. The significantly high activity of TTO5/TiO2 is considered to be caused by the inter-semiconductor holetransport mechanism. The evidence for the hole-transport between TTO5 and TiO2 was also investigated by monitoring the OH radicals in TiO2 side during the visible-light irradiation using 3 mM TA. It was monitored that the 3/97 TTO5/TiO2 composite converted TA to TAOH with high rate, indicative of hole-transport between TTO5 and TiO2.