Catalysis Science & Technology

0 downloads 0 Views 3MB Size Report
May 2, 2013 - systems such as FeTiO3/TiO2, Ag3PO4/TiO2, W18O49/TiO2, and Sb-doped SnO2 (ATO)/TiO2 demonstrated noticeably higher photocatalytic ...

Catalysis Science & Technology View Article Online

Published on 02 May 2013. Downloaded by University of Michigan Library on 20/01/2014 13:56:15.

PAPER

Cite this: Catal. Sci. Technol., 2013, 3, 1822

View Journal | View Issue

Design of visible-light photocatalysts by coupling of narrow bandgap semiconductors and TiO2: effect of their relative energy band positions on the photocatalytic efficiency† Sher Bahadur Rawal,a Sandipan Bera,a Daeki Lee,b Du-Jeon Jangb and Wan In Lee*a According to relative energy band positions between TiO2 and visible-light-absorbing semiconductors, three different types of heterojunction were designed, and their visible-light photocatalytic efficiencies were analyzed. In Type-A heterojunction, the conduction band (CB) level of sensitizer is positioned at a more negative side than that of TiO2, whereas in Type-B system its valence band (VB) level is more positive than that of TiO2 and in Type-C system the sensitizer energy level is located between the CB and VB of TiO2. In evolving CO2 from gaseous 2-propanol (IP) under visible-light irradiation, the Type-B systems such as FeTiO3/TiO2, Ag3PO4/TiO2, W18O49/TiO2, and Sb-doped SnO2 (ATO)/TiO2 demonstrated noticeably higher photocatalytic efficiency than the Type-A such as CdS/TiO2 and CdSe/TiO2, while the

Received 4th January 2013, Accepted 28th April 2013

Type-C such as NiTiO3/TiO2, CoTiO3/TiO2, and Fe2O3/TiO2 did not show any appreciable improvement.

DOI: 10.1039/c3cy00004d

explained by inter-semiconductor hole-transfer mechanism between the VB of the sensitizer and that of TiO2. The evidence for the hole-transport between sensitizer and TiO2 was also obtained by monitoring

www.rsc.org/catalysis

the hole-scavenging reactions with iodide (I ) and 1,4-terephthalic acid (TA).

Remarkably high visible-light photocatalytic activity of Type-B heterojunction structures could be

1. Introduction Remediation of environmental pollutants by photocatalytic reaction has attracted extensive attention in the last few decades.1,2 TiO2 has been known as the most efficient photocatalyst among various semiconductors,3–6 but it cannot utilize the photons in the visible-light range, occupying a major portion of solar spectrum due to its wide bandgap (Eg = 3.2 eV).7–11 Thus, the development of photocatalysts functioning under visible-light will be a crucial issue. Thus far, several strategies, including the substitution of various transition elements to the Ti site,12–14 several anions such as N, C, B, and S to the oxygen site,15–18 and incorporation of carbon nanomaterials such as carbon nanotubes and graphene,19,20 have been attempted to extend the band edge of TiO2 up to the visible-light range. Another promising strategy will be coupling of TiO2 with

other narrow bandgap semiconductors capable of harvesting the photons in the visible range.21–24 Conceptually, according to the relative energy band location between the sensitizer and TiO2, the heterojunction structures can be classified by the following three different types. First, as described in Scheme 1a, the conduction band (CB) of the sensitizer is positioned to a more negative side than that of TiO2 (denoted as Type-A heterojunction). For example, several metal chalcogenide quantum dots or molecular dyes are loaded on the TiO2 surface to form Type-A heterojunction.25–27 With visible-light irradiation to this system, the sensitizer is excited, and the electrons are then transported to the CB of TiO2, since the CB level of TiO2 is lower than that of the sensitizer. These electrons can induce various reduction reactions or participate in decoloration reactions of organic dyes.28–30 Complete oxidation of organic pollutants is also possible by forming the  O2 and HO2 , as shown in eqn (1)–(3).30–32 O2 + e -  O2 ,

E0 =

0.284 V (vs. NHE)

(1)

a

Department of Chemistry, Inha University, Incheon 402-751, Republic of Korea. E-mail: [email protected]; Fax: +82-32-867-5604; Tel: +82-32-863-1026 b School of Chemistry, Seoul National University, Seoul 151-747, Republic of Korea † Electronic supplementary information (ESI) available: Detailed photocatalytic activities, bandgaps, and PL spectra. See DOI: 10.1039/c3cy00004d

1822

Catal. Sci. Technol., 2013, 3, 1822--1830



O2 + H+ - HO2 ,

0.046 V (vs. NHE)

(2)

HO2 + organic compounds - - - CO2 + H2O

(3)

This journal is

E0 =

c

The Royal Society of Chemistry 2013

View Article Online

Published on 02 May 2013. Downloaded by University of Michigan Library on 20/01/2014 13:56:15.

Paper

Catalysis Science & Technology Hence, no synergetic effect enhancing visible-light photocatalytic activity is expected for this system.37 Thus far various coupled systems were investigated to design efficient visible-light photocatalysts, but most of them were limited to the Type-A systems, and the other types have been scarcely studied. In the present study, various narrow bandgap semiconductors such as CdS, CdSe, Sb-doped SnO2 (ATO), Ag3PO4, W18O49, FeTiO3, NiTiO3, CoTiO3 and Fe2O3 were prepared, and they were coupled with TiO2 to form the Type-A, Type-B and Type-C heterojunction structures, respectively. Correlation of the relative energy band positions between the sensitizer and TiO2 and the resultant visible-light photocatalytic activities have been systematically investigated. We found that the relative energy band positions are highly important in determining the visible-light photocatalytic activity. The obtained results will provide insight in designing highly efficient visible-light photocatalysts based on heterojunction structures as well as understanding of the photocatalytic reaction mechanism.

2. Experimental section 2.1. Preparation of Type-A heterojunction structures (CdS/TiO2 or CdSe/TiO2)

Scheme 1 Schematic diagrams of the photo-induced charge flow under visiblelight irradiation for Type-A (a), Type-B (b), and Type-C (c) heterojunction structures.

Second, as described in Scheme 1b, the valence band (VB) of the sensitizer is located at a more positive side than that of the TiO2 (Type-B heterojunction). With irradiation of visible-light to this coupled system, the electrons in the sensitizer VB are excited to its CB. Thereby the holes in the sensitizer VB can be transferred to that of TiO2. As a result, holes are generated in the VB of TiO2 by inter-semiconductor hole-transfer mechanism,33–35 and in turn they initiate various oxidation reactions by generating the  OH radical on the TiO2 surface, as described in eqn (4) and (5).36 (H2O)ads + h+ - H+ +  OH

(4)

(OH )ads + h+ -  OH

(5)

Considering the powerful oxidative ability of the holes generated in the VB of TiO2, efficient and complete decomposition of organic compounds could be achieved. Third, as described in Scheme 1c, the CB and VB of the sensitizer are located between those of the TiO2 (Type-C heterojunction). Under visible light irradiation the electrons in the VB of the sensitizer are excited to its CB, but neither electrons in CB nor holes in VB of the sensitizer can be transferred to the CB or VB of TiO2, due to unfavourable energy band matching.

This journal is

c

The Royal Society of Chemistry 2013

CdS and CdSe quantum dots (QDs) were synthesized by the procedures reported in the literature.38,39 In a typical syntheses, 0.1 mmol CdO (Aldrich), 1.2 mmol oleic acid (OA, Aldrich) (or 2.4 mmol trioctylphosphine oxide (TOPO, Aldrich) for CdSe synthesis), and 3.0 ml 1-octadecene (ODE, Aldrich) were mixed in a three-neck flask, and heated to 300 1C under Ar flow. In a separate flask, a solution of sulfur (0.05 mmol, Daejung Chem. Co.) in 1.91 ml ODE or 0.24 mmol selenium (Aldrich) with 0.96 mmol trioctylphosphine (TOP, Aldrich) in ODE was prepared, and injected swiftly to the hot solution. The temperature of the mixture was then adjusted to B260 1C for the growth of CdS or CdSe QDs. After 5 min, the solution was immediately cooled down by adding the 30 ml cold toluene solution to obtain OA-capped CdS or TOPO-capped CdSe QDs. The OA and TOPO groups on the CdS and CdSe surfaces, respectively, were exchanged to mercapto propionate (MPA) group,33,40 in order to anchor the CdS and CdSe QDs onto the TiO2 surface. Typically, 0.2 mmol MPA (Aldrich) was dissolved in 10 mL anhydrous methanol, and the pH was adjusted to 11.4 by adding tetramethylammonium hydroxide (TMAH, Aldrich). 40 mg OA-capped CdS or TOPO-capped CdSe was then suspended in this solution, followed by heating at 63 1C under dry Ar atmosphere for B24 h. The formed MPA-capped CdS or CdSe QD was precipitated by adding a mixture of ethyl acetate and diethyl ether (1 : 1 in volume). The collected precipitate was washed several times with ethyl acetate to remove residual MPA, OA or TOPO. To anchor the CdS or CdSe QDs onto the TiO2 surface, 50 ml ethanol solution containing 0.5 g TiO2 and the stoichiometric amount of the MPA-capped CdS (or CdSe) was magnetically stirred at 60 1C for 6 h. The prepared CdS/TiO2, or CdSe/TiO2 (Type-A heterojunctions) suspended in the solution were then

Catal. Sci. Technol., 2013, 3, 1822--1830

1823

View Article Online

Catalysis Science & Technology precipitated by centrifugation, and the collected powders were subsequently annealed at 250 1C for 3 h.

Published on 02 May 2013. Downloaded by University of Michigan Library on 20/01/2014 13:56:15.

2.2. Preparation of Type-B and Type-C heterojunction structures Antimony doped tin oxide (ATO), Ag3PO4, W18O49, FeTiO3, NiTiO3, CoTiO3, and Fe2O3 particles were prepared by the procedures reported in the literatures.33–44 10 mol% Sb-doped SnO2 was prepared by co-precipitation of SnCl45H2O and SbCl3, followed by post heat treatment method.33 FeTiO3 nanodisc was synthesized by the hydrothermal reaction of FeSO4 7H2O, KOH, titanium(IV) isopropoxide (TTIP, Aldrich) stabilized in aqueous solution of tetrabutylammonium hydroxide (TBAH, Aldrich) at 220 1C.34 Ag3PO4 nanoparticle (NP) was synthesized by ion-exchange reaction between AgNO3 and Na3PO4 in solid phase.41 W18O49 nanorods were synthesized by heating the reaction mixture of WCl4, oleic acid and oleylamine at 350 1C under argon environment.42 NiTiO3 (or CoTiO3) particle was prepared by co-precipitation of nickel acetate (or cobalt acetate), titanium(IV) butoxide (Aldrich) and citric acid, followed by subsequent heat treatment method.43 Fe2O3 NP was synthesized by a hydrothermal reaction of FeCl36H2O stabilized in 25/75 ammonia/water solution at 180 1C.44 For the formation of Type-B or Type-C heterojunction, 3.67 g titanium isopropoxide (97%, Aldrich) was stabilized in the mixed solution of 40 ml ethanol, 1 ml concentrated nitric acid, and 1 ml water, and the mixture was then gently stirred for 1 h. A stoichiometric amount of each sensitizer particle was added to this solution. For example, to obtain 5/95 ATO/TiO2 (in wt% ratio), 53 mg ATO particle was added and gently stirred overnight. The amorphous titania-coated samples were then dried at 80 1C for 24 h, and subsequently heat treated at 300 1C for 3 h to crystallize the TiO2. As a blank sample, bare TiO2 was prepared by the same procedure without adding the sensitizer particle. 2.3.

Characterizations

X-ray diffraction (XRD) patterns were obtained for the heterojunction composite powder samples by using a Rigaku Multiflex diffractometer with monochromatic light-intensity Cu Ka radiation. XRD scanning was performed under ambient conditions over 2y region of 20–701 at a rate of 21 min 1 (40 kV, 20 mA). UV-visible diffuse reflectance spectra were acquired by a PerkinElmer Lambda 40 spectrophotometer. BaSO4 was used as the reflectance standard. Field emission transmission electron microscope (FE-TEM) images were obtained by a JEOL JEM2100F operated at 200 kV. One milligram of the synthesized particles was dispersed in 50 mL of ethanol, and a drop of the suspension was then spread on a holey amorphous carbon film deposited on the copper grid.45 2.4.

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 the 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

1824

Catal. Sci. Technol., 2013, 3, 1822--1830

Paper dried at room temperature. The gas reactor system used for this photocatalytic activity has been described elsewhere.45 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 (l o 422 nm, ZUL0422 Asahi Co.) and a water filter to cut-off IR. After evacuation of the reactor, 1.6 mL 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: 2 (CH3)2CHOH(g) + 9 O2(g) - 6 CO2(g) + 8 H2O(g)

(6)

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.

3. Results and discussion The X-ray diffraction patterns in Fig. 1a indicate that the prepared CdS and CdSe QDs are in the cubic (JCPDS, No. 75-0581) and hexagonal (JCPDS, No. 88-2346) phases, respectively, with no impurity peaks. The average crystallite sizes of CdS and CdSe QDs, as determined from the corresponding (111) peaks by applying the Scherrer equation, were 3.8 and 5.5 nm, respectively. Fig. 2a shows the UV-vis diffuse absorbance spectra of CdS and CdSe QDs loaded on the TiO2 particles. The absorption edge of CdS and CdSe QDs appeared at B500 and B660 nm, respectively, exhibiting the capability of sensitizing the visible-light. Antimony doped tin oxide (Sn0.9Sb0.1O2, ATO), Ag3PO4, W18O49, and FeTiO3 particles were prepared by the procedures reported in the literature.33,34,41,42 X-ray diffraction patterns in Fig. 1b confirm that the prepared sensitizers are in the pure phase. The average crystallite size of ATO, Ag3PO4, W18O49, and FeTiO3 particles, as determined from the (110) peak of ATO, (210) peak of Ag3PO4, (010) peak of W18O49 and (104) peak of FeTiO3 by applying the Scherrer equation, were 49 nm, 62 nm, 43 nm and 46 nm, respectively, clearly indicating their high crystallinity. UV-vis diffuse absorbance spectra of the prepared sensitizers are shown in Fig. 2b. Ag3PO4 reveals an absorption band edge at B560 nm, whereas ATO, W18O49, and FeTiO3 exhibited profound absorbance in the entire visible region. The NiTiO3, CoTiO3, and Fe2O3 particles were also prepared by the reported methods in the literature.43,44 The prepared samples were in the pure phase with high crystallinity, as shown in the XRD patterns in Fig. 1c. UV-vis diffuse absorbance spectra in Fig. 2c indicate that the prepared NiTiO3, CoTiO3, and Fe2O3 particles also exhibit profound absorption in the visible region.

This journal is

c

The Royal Society of Chemistry 2013

View Article Online

Published on 02 May 2013. Downloaded by University of Michigan Library on 20/01/2014 13:56:15.

Paper

Catalysis Science & Technology

Fig. 2 UV-vis diffuse absorbance spectra for the sensitizers belonging to Sen-A (a), Sen-B (b), and Sen-C (c).

Table 1 CB and VB potential levels (vs. NHE) and bandgaps of various visiblelight-absorbing semiconductors measured at pH 744–48 Fig. 1 X-ray diffraction patterns for the sensitizers belonging to Sen-A (a), Sen-B (b), and Sen-C (c).

NiTiO3 showed the absorption edges at B550 nm and B410 nm, caused by the transition of Ni2+ - Ti4+ and O2 - Ti4+,43 respectively. CoTiO3 revealed a broad absorption peak at B610 nm with the absorption band edges at B670 nm and B490 nm, respectively, inherent from the transition of Co2+ Ti4+ and O2 - Ti4+.43 Fe2O3 powder with a bandgap of 2.20 eV displays an absorption band edge at B610 nm. Table 1 illustrates the CB and VB energy levels and the bandgaps of the synthesized sensitizers reported in the literature.46–50 Herein we also calculated the bandgaps of these sensitizers from the Kubelka Munk (KM) or Tauc plots versus wavelength. The determined values, shown in Fig. S1 and S2 (ESI†), were quite close to the reported ones. The sensitizers were categorized as Sensitizer-A (Sen-A), Sensitizer-B (Sen-B), and Sensitizer-C (Sen-C), according to their energy band positions relative to those of TiO2. Considering that the CB and VB levels of TiO2 at pH 7 are 0.5 and +2.7 V (vs. NHE), respectively, CdS and CdSe are classified to Sen-A, whereas FeTiO3,

This journal is

c

The Royal Society of Chemistry 2013

Sensitizers

Type

CB (V)

VB (V)

Eg (eV)

CdS CdSe Ag3PO4 ATO W18O49 FeTiO3 CoTiO3 NiTiO3 Fe2O3 TiO2

A A B B B B C C C —

0.85 0.60 +0.45 +0.95 +0.61 +0.20 +0.14 +0.20 +0.28 0.50

+1.55 +1.10 +2.90 +3.60 +3.21 +3.00 +2.39 +2.38 +2.48 +2.70

2.40 1.70 2.45 2.55 2.60 2.80 2.25 2.18 2.20 3.20

Ag3PO4, W18O49, and ATO belong to Sen-B, and Fe2O3, CoTiO3 and NiTiO3 belong to Sen-C. Each sensitizer was then coupled with TiO2 to form the three different types of heterojunctions, categorized as Type-A, Type-B, and Type-C. As shown in Scheme 2a, a few nanometer-sized Sen-A QDs were loaded on the surface TiO2 (Degussa P25) to form Type-A heterojunction. In order to form Type-B or Type-C heterojunction, relatively large Sen-B or Sen-C particles were fully covered with TiO2, as described in Scheme 2b. Fig. 3 shows the TEM images of the Type-A heterojunction structures prepared by anchoring the MPA capped CdS or CdSe QDs onto the surface of Degussa P25. The images in Fig. 3a and

Catal. Sci. Technol., 2013, 3, 1822--1830

1825

View Article Online

Published on 02 May 2013. Downloaded by University of Michigan Library on 20/01/2014 13:56:15.

Catalysis Science & Technology

Paper

Scheme 2 Schematic diagrams describing the preparation method for Type-A (a) and Type-B or Type-C (b) heterojunction structures.

Fig. 3 TEM images of CdS/TiO2 (a) and CdSe/TiO2 (b). High resolution (HR) image is shown in the inset of (a), obtained from the dotted rectangular parts in (a).

b shows the 2/98 CdS/TiO2 and 1/99 CdSe/TiO2 (both are in molar ratio), respectively, clearly indicating that the individual CdS and CdSe QDs with a size of B4 and B5 nm, respectively, are attached on the TiO2 surface. The dotted rectangular part in Fig. 3a was further magnified, as shown in the inset. The d-spacing of 0.335 nm was identified to be the (111) plane of CdS. Fig. 4 shows the TEM images of the Type-B heterojunction structures prepared by sol–gel method in depositing the TiO2 on the surface of large sensitizer particles. TEM images in Fig. 4a, c, e and f show the 5/95 ATO/TiO2, 7/93 W18O49/TiO2, 3/97 Ag3PO4/TiO2, and 5/95 FeTiO3/TiO2 (all are in wt% ratio), respectively. It is clear that large sensitizer particles are located in the core and their surfaces are fully covered by small TiO2 grains. The high resolution TEM images were also obtained for the dotted rectangular parts of Fig. 4a and c. Their characteristic fringe patterns, as shown in Fig. 4b and d, certify the presence of crystallized ATO and W18O49, respectively, in the 5/95 ATO/TiO2 and 7/93 W18O49/TiO2 composites. Type-C heterojunction structures were also prepared by the same sol–gel method. TEM images of the 3/97 NiTiO3/TiO2, 3/97 CoTiO3/TiO2, and 3/97 Fe2O3/TiO2 (all are in wt% ratio) are shown in Fig. 5a–c, revealing that the large Sen-C particles are fully covered with TiO2. High resolution TEM image of 3/97 Fe2O3/ TiO2 in Fig. 5d showed the fringe patterns, identified to be the (110) plane of Fe2O3 and (101) of TiO2.

1826

Catal. Sci. Technol., 2013, 3, 1822--1830

Fig. 4 TEM images of ATO/TiO2 (a), W18O49/TiO2 (c), Ag3PO4/TiO2 (e), and FeTiO3/TiO2 (f). HR-TEM images of ATO/TiO2 (b) and W18O49/TiO2 (d) obtained from the dotted rectangular parts in (a) and (c), respectively.

Photocatalytic efficiencies of three different heterojunction structures, that is, Type-A (CdS/TiO2, and CdSe/TiO2), Type-B (ATO/TiO2, Ag3PO4/TiO2, W18O49/TiO2, and FeTiO3/TiO2), and Type-C (NiTiO3/TiO2, CoTiO3/TiO2, and Fe2O3/TiO2) were evaluated by monitoring the decomposition of 2-propanol (IP) in gas phase under visible-light irradiation (l Z 422 nm). The amount of CO2 evolved in 2 h was monitored to evaluate the photocatalytic activity of each system. More detailed photocatalytic data for the individual heterojunction structures are shown in Fig. S3 (ESI†). First, as shown in Fig. 6a, the Type-A systems exhibited considerably higher catalytic activity than the bare TiO2. The amounts of evolved CO2 with 2/98 CdS/TiO2 and 1/99 CdSe/TiO2 (both are in molar ratio) were 4.3 and 3.1 ppm, respectively, whereas that with the bare TiO2 was only 0.95 ppm and those with the pure CdS and CdSe were 0.65 and 0.80 ppm, respectively. Second, visible-light photocatalytic activities of the Type-B systems were illustrated in Fig. 6b. The amount of CO2 evolved in 2 h was remarkably enhanced for all Type-B systems. The optimum ratios between the sensitizer and TiO2, with respect to photocatalytic efficiency, were determined to be 5/95 for ATO/ TiO2, 3/97 for Ag3PO4/TiO2, 7/93 for W18O49/TiO2, and 5/95 for FeTiO3/TiO2 (all are in wt% ratio). As shown in Fig. 6b, the

This journal is

c

The Royal Society of Chemistry 2013

View Article Online

Published on 02 May 2013. Downloaded by University of Michigan Library on 20/01/2014 13:56:15.

Paper

Catalysis Science & Technology

Fig. 5 TEM images of NiTiO3/TiO2 (a), CoTiO3/TiO2 (b), and Fe2O3/TiO2 (c). HRTEM image of Fe2O3/TiO2 (d) obtained from the dotted rectangular part in (c).

amounts of CO2 evolved in 2 h with ATO/TiO2, Ag3PO4/TiO2, W18O49/TiO2, and FeTiO3/TiO2 were 8.3, 11.8, 6.8, and 8.1 ppm, respectively. Third, visible-light photocatalytic activities of the 3/97 NiTiO3/TiO2, 3/97 CoTiO3/TiO2, and 3/97 Fe2O3/TiO2 (all are in wt% ratio), classified as Type-C heterojunction, were also examined. As shown in Fig. 6c, the amounts of CO2 evolved in 2 h for the NiTiO3/TiO2, CoTiO3/TiO2, and Fe2O3/TiO2 were 0.85, 0.88, 0.96 ppm, respectively, indicating that the coupling of Type-C sensitizer with TiO2 did not induce appreciable enhancement in photocatalytic activity. Recently, Kubacka et al. reported that the photocatalytic activities of the coupled systems are influenced by the size of sensitizer, due to the energy band modification by the quantum size effect.51 However, the particle sizes of the metal oxides prepared in this work were in the range of 40–70 nm, which does not offer any quantum size effect. Thus the sizes of sensitizers in these ranges would not be critical in determining the catalytic activity. From the observed trends in the visiblelight photocatalytic activities of the Type-B and Type-C heterojunctions, it is obvious that relative energy band position between the sensitizer and TiO2 is a crucial factor. Hence, in designing efficient catalytic system based on heterojunction structure, the transport of electrons or holes from the sensitizer to TiO2 is regarded to be the most important issue, since the electron–hole pairs are generated in the sensitizer and the catalytic sites are formed on the TiO2 surface. In the Type-B heterojunction, the electrons in the sensitizer VB are excited to its CB under visible-light irradiation. The generated holes in the sensitizer VB will be transported voluntarily to that of TiO2. By this inter-semiconductor hole-transport mechanism, holes are generated on the TiO2 VB, followed by

This journal is

c

The Royal Society of Chemistry 2013

Fig. 6 Visible-light photocatalytic activities of several Type-A (a), Type-B (b), and Type-C (c) heterojunction structures. The amounts of CO2 evolved under visiblelight irradiation in 2 h were monitored.

formation of  OH radicals through the eqn (4) and (5). Hence, the generated  OH radicals can completely decompose the organic compounds to CO2 and H2O. For the case of Type-C heterojunction, the electrons in the sensitizer VB are excited to its CB under visible-light irradiation, but the photogenerated electrons and holes cannot be transported to TiO2 CB and VB, respectively, since the sensitizer’s energy levels lies in between the CB and VB of TiO2. Thus the coupling of these two semiconductors will not bring any synergetic effect in charge separations. As a result, NiTiO3/TiO2, CoTiO3/TiO2, and Fe2O3/ TiO2, belonging to Type-C heterojunction, did not show appreciable visible-light photocatalytic activity. In the Type-A heterojunction system, sensitizers are excited under visible-light irradiation, followed by the transportation of the photogenerated electrons to TiO2 CB. As described in eqn (1)–(3), the electrons in the TiO2 CB are transferred to O2 to form  O2 radicals, and further converted to HO2 by reacting with proton. Even though HO2 is not as strong as  OH in oxidation ability, it is able to induce the complete mineralization of organic compounds.30,46 This explains why CdS/TiO2 and

Catal. Sci. Technol., 2013, 3, 1822--1830

1827

View Article Online

Published on 02 May 2013. Downloaded by University of Michigan Library on 20/01/2014 13:56:15.

Catalysis Science & Technology

Paper

CdSe/TiO2 systems showed considerably enhanced visible-light photocatalytic activities, compared with the bare TiO2. Herein it was found that Type-B heterojunction systems exhibited relatively higher catalytic activity than the Type-A systems. Moreover, some of the Type-B systems showed comparable or even higher activity than the typical N-doped TiO2. Basically, direct comparison of activity among the different photocatalytic systems is not simple, since the several factors such as particle size of sensitizer, contact and charge transport between sensitizer and TiO2, and others, are involved in determining the photocatalytic activity. Nonetheless, Type-B systems seem to have a significant advantage in photocatalytic oxidation reactions due to the availability of  OH radicals in the TiO2 VB. The produced  OH radicals, known as the most powerful oxidant, can induce fast and complete decomposition of organic pollutants, rationalizing the enhanced photocatalytic efficiency of Type-B systems. In order to confirm the hole-transfer mechanism between the VB of sensitizer and TiO2 in the Type-B heterojunction, the evidence for the generation of holes in TiO2 VB was investigated by monitoring the chemical reaction of the iodide ion (I ), known as a hole scavenger. As a Type-B heterojunction system, ATO/TiO2 was used in this experiment. Generally, I /I3 redox couple has been used as electrolyte mediating the charges in the dye-sensitized solar cells, and the role of I ions is accepting the holes from the HOMO of dye.52,53 Therefore, it is deduced that the I ions can be oxidized to triiodide (I3 ) by reacting with the generated holes in the ATO or TiO2, as shown in eqn (7), since the redox potential of I /I3 is +0.536 V,54 which lies much higher than the VB position of ATO (+3.6 V)33 or TiO2 (+2.7 V).46 2h+(VB) + 3I (aq) - I3 (aq)

(7)

40 mL KI (0.01 M) solution containing ATO, ATO/TiO2 or TiO2 (20 mg each) was irradiated for 2 h under visible-light (l Z 422 nm). Then the I3 formed in the solution could be identified from its characteristic absorption peak at 286 nm and 345 nm. Fig. 7 showed the UV-vis absorption spectra of KI solution, after the photocatalytic reaction with ATO, ATO/TiO2 and TiO2. First, pure 0.01 M KI solution, after visible-light irradiation for 2 h, did not show any characteristic absorption peak in 250–500 nm range. Second, when KI solution was

Fig. 7 UV-vis spectra of 0.01 M KI aqueous solution in the presence of several catalytic systems after visible-light (l Z 422 nm) or UV light irradiation for 2 h.

1828

Catal. Sci. Technol., 2013, 3, 1822--1830

irradiated in presence of the bare ATO, interestingly, there was no characteristic absorption peak of I3 , indicating that the bare ATO cannot oxidize I under visible-light irradiation. In case of pure ATO, electron and hole pairs will be generated in its CB and VB, respectively, but the holes generated in ATO did not seem to be consumed for the formation of I3 , presumably due to the faster electron–hole recombination than the reaction between the hole and I . Third, when KI solution was irradiated in presence of the bare TiO2 under visible-light, there was no characteristic absorption peak of I3 , but under UV light strong absorption peaks appeared at 286 nm and 345 nm, indicating the presence of I3 . It is deduced that the TiO2 itself cannot be excited by visible-light due to its wide band gap but that the generated holes in the VB by UV-light can induce the formation of I3 , as described in eqn (7). Fourth, when the KI solution was irradiated in presence of the ATO/TiO2, noticeably, the characteristic absorption peak of I3 was observed (Fig. 7), indicating that the holes are generated in the TiO2 VB. Therefore, the obtained result strongly supports that the visible-light photocatalytic activity of Type-B heterojunction systems originates from the inter-semiconductor hole-transport. The presence of  OH radicals on the ATO/TiO2 surface during the visible-light irradiation was also monitored in order to support the hole transfer mechanism.55,56 That is, 20 mg of ATO, TiO2 or ATO/TiO2 was suspended in 60 mL aqueous solution containing 0.01 M NaOH and 3 mM 1,4-terephthalic acid (TA). Before exposure to visible-light, the suspension was stirred in dark for 30 min. Then, 5 mL of the solution was taken after every 1 h for the fluorescence measurements. It is 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.56 Bare ATO as well as TiO2, suspended in TA solution, did not show any appreciable fluorescence peaks upon visible-light (l Z 422 nm) irradiation, as shown in Fig. 8a and c, suggesting that bare TiO2 or ATO cannot produce holes nor utilize them in producing  OH radicals. Contrarily, the ATO/TiO2 shows the characteristic fluorescence peak, and its intensity was increased with elapse of irradiation time, as shown in Fig. 8b. This clearly indicates that the holes are formed at the TiO2 side and they were transported from the ATO by the inter-semiconductor holetransport mechanism. For the Fe2O3/TiO2 system, we also performed the  OH radical test using TA in order to check the possibility of holetransfer between Fe2O3 and TiO2. Fig. 9 shows the fluorescence spectra of TAOH for the suspensions with TiO2, Fe2O3/TiO2, and Fe2O3, after 2 h irradiation of visible-light. It was found that the fluorescence peaks of the TAOH with the bare Fe2O3 and Fe2O3/TiO2 were not appreciably different, suggesting that the hole transport from Fe2O3 to TiO2 was blocked in the Fe2O3/ TiO2 system. The obtained result rationalize the reason for the low catalytic efficiency of Type-C systems including Fe2O3/TiO2. In order for the Type-B system to be an efficient catalyst, the excited electrons in the CB of sensitizer have to be scavenged. Considering the CB level of the sensitizer, direct electron transfer to the oxygen molecules, requiring 0.284 V (vs. NHE),

This journal is

c

The Royal Society of Chemistry 2013

View Article Online

Published on 02 May 2013. Downloaded by University of Michigan Library on 20/01/2014 13:56:15.

Paper

Catalysis Science & Technology The visible-light catalytic activity of Type-B systems, achieved in the present work, is comparable or even higher than that of N-doped TiO2, which is a well-known visible-light photocatalyst, but we believe that the catalytic activity can be enhanced much more, if the appropriate sensitizers are developed and the interface control between the sensitizer and TiO2 are optimized for the efficient hole-transport. In this regards, several properties will be required to the sensitizers for the design of efficient Type-B systems. The VB level of sensitizer has to be sufficiently lower than that of TiO2, and the inherent holetransport from sensitizer to TiO2 has to be efficient and fast with less charge recombination. Small band gap will be favorable for utilization of visible-light, but its CB level needs to be high enough for effective scavenging of the electrons in its CB.

4. Conclusions

Fig. 8 Fluorescence spectra measured for visible-light (310–540 nm) irradiated bare TiO2 (a), ATO/TiO2 (b), and ATO (c) suspensions in 3 mM TA. Wavelength of the excitation light for obtaining the fluorescence spectra was 320 nm.

Nine heterojunction systems (CdS/TiO2, CdSe/TiO2, FeTiO3/ TiO2, Ag3PO4/TiO2, W18O49/TiO2, ATO/TiO2, NiTiO3/TiO2, CoTiO3/TiO2, and Fe2O3/TiO2) were fabricated, and they were classified into three types, according to the relative energy band positions between sensitizer and TiO2. Among them Type-B systems such as FeTiO3/TiO2, Ag3PO4/TiO2, W18O49/TiO2, and ATO/TiO2 exhibited relatively higher visible-light photocatalytic activity in evolving CO2 from the gaseous IP under visible-light irradiation. Especially, Ag3PO4/TiO2 showed significantly higher catalytic efficiency than the typical N-doped TiO2. It is deduced that the higher photocatalytic activity of Type-B systems originates from the inter-semiconductor hole-transport. That is, the generated holes in the sensitizer VB is transported to that of TiO2, hence inducing the formation of  OH radicals. The evidence for the hole-transport between sensitizer and TiO2 was also investigated by monitoring the reaction with iodide (I ). By irradiating visible-light in the presence of ATO/TiO2, it was found that I was converted to I3 , clearly indicating that the holes are generated in the TiO2 VB. It was also found that the ATO/TiO2 system can convert the TA to TAOH, which is evident for the formation of  OH radicals in TiO2 side under visible-light irradiation.

Acknowledgements

Fig. 9 Fluorescence spectra of TAOH for the 3 mM TA suspensions with TiO2, Fe2O3/TiO2, and Fe2O3 after 2 h irradiation of visible-light. Wavelength of the excitation light for obtaining the fluorescence spectra was 320 nm.

will be difficult in general (eqn (1)).34 Thus it is considered that the electrons in the sensitizer CB are transported to the oxygen species through the processes described in eqn (8) and (9).54,57,58 O2 + H+ + e - HO2 , O2 + 2H + 2e - H2O2, +

This journal is

c

E0 =

0.046 V

E0 = +0.682 V

The Royal Society of Chemistry 2013

(8) (9)

This work has been supported by the National Research Foundation of Korea (Project No. 2011-0002995), and Korean Center for Artificial Photosynthesis (KCAP) funded by the Ministry of Education, Science, and Technology (NRF-2011-C1AAA0012011-0030278).

Notes and references 1 T. Inoue, A. Fujishima, S. Konishi and K. Honda, Nature, 1979, 277, 637. 2 M. R. Hoffmann, Chem. Rev., 1995, 95, 69. 3 E. S. Jang, J. H. Won, S. J. Hwang and J. H. Choy, Adv. Mater., 2006, 18, 3309.

Catal. Sci. Technol., 2013, 3, 1822--1830

1829

View Article Online

Published on 02 May 2013. Downloaded by University of Michigan Library on 20/01/2014 13:56:15.

Catalysis Science & Technology 4 Z. Ding, G. Q. Lu and P. F. Greenfield, J. Phys. Chem. B, 2000, 104, 4815. 5 Y. Ou, J. Lin, S. Fang and D. Liao, Catal. Commun., 2007, 8, 936. 6 Y. Huang, Z. Zheng, Z. Ai, L. Zhang, X. Fan and Z. Zou, J. Phys. Chem. B, 2006, 110, 19323. 7 C. Burda, Y. Lou, X. Chen, A. C. S. Samia, J. Stout and J. L. Gole, Nano Lett., 2003, 3, 1049. 8 V. Stengl, V. Houskova, S. Bakardjieva and N. Murafa, ACS Appl. Mater. Interfaces, 2010, 2, 575. 9 S. Sakthivel and H. Kisch, ChemPhysChem, 2003, 4, 487. 10 K. Y. Song, M. K. Park, Y. T. Kwon, H. W. Lee, W. J. Chung and W. I. Lee, Chem. Mater., 2001, 13, 2349. 11 L.-L. Tan, S.-P. Chai and A. R. Mohamed, ChemSusChem, 2012, 5, 1868. 12 M. Anpo and M. Takeuchi, J. Catal., 2003, 216, 505. 13 B. K. Vijayan, N. M. Dimitrijevic, J. Wu and K. A. Gray, J. Phys. Chem. C, 2010, 114, 21262. 14 P. Bouras, E. Stathatos and P. Lianos, Appl. Catal., B, 2007, 73, 51. 15 R. Asahi, T. Morikawa, T. Ohwaki, K. Aoki and Y. Taga, Science, 2001, 293, 269. 16 S. U. M. Khan, M. Al-Shahry and W. B. Ingler, Science, 2002, 297, 2243. 17 W. Zhao, W. Ma, C. Chen, J. Zhao and Z. Shuai, J. Am. Chem. Soc., 2004, 126, 4782. 18 J. C. Yu, W. Ho, J. Yu, H. Yip, P. K. Wong and J. Zhao, Environ. Sci. Technol., 2005, 39, 1175. 19 W. J. Ong, M. M. Gui, S. P. Chai and A. R. Mohamed, RSC Adv., 2013, 3, 4505. 20 Y. Zhang, N. Zhang, Z. R. Tang and Y. J. Xu, Phys. Chem. Chem. Phys., 2012, 14, 9167. 21 D. Liu and P. V. Kamat, J. Electroanal. Chem., 1993, 347, 451. 22 L. Spanhel, H. Weller and A. Henglein, J. Am. Chem. Soc., 1987, 109, 6632. 23 S. Y. Chai, Y. J. Kim, M. H. Jung, A. K. Chakraborty, D. Jung and W. I. Lee, J. Catal., 2009, 262, 144. 24 S. B. Rawal, S. D. Sung and W. I. Lee, Catal. Commun., 2012, 17, 131. 25 S. N. Frank and A. J. Bard, J. Phys. Chem., 1977, 81, 1484. 26 W. Ho and J. C. Yu, J. Mol. Catal. A: Chem., 2006, 247, 268. 27 E. Bae, W. Choi, J. Park, H. S. Shin, S. B. Kim and J. S. Lee, J. Phys. Chem. B, 2004, 108, 14093. 28 X. Yu, Q. Wu, S. Jiang and Y. Guo, Mater. Charact., 2006, 57, 333. 29 J. C. Kim, J. K. Choi, Y. B. Lee, J. H. Hong, J. I. Lee, J. W. Yang, W. I. Lee and N. H. Hur, Chem. Commun., 2006, 5024. 30 Y. Bessekhouad, N. Chaoui, M. Trzpit, N. Ghazzal, D. Robert and J. V. Weber, J. Photochem. Photobiol., A, 2006, 183, 218. 31 A. G. Agrios and P. Pichat, J. Appl. Electrochem., 2005, 35, 655. 32 U. I. Gaya and A. H. Abdullah, J. Photochem. Photobiol., C, 2008, 9, 1.

1830

Catal. Sci. Technol., 2013, 3, 1822--1830

Paper 33 S. B. Rawal, A. K. Chakraborty, Y. J. Kim, H. J. Kim and W. I. Lee, RSC Adv., 2012, 2, 622. 34 Y. J. Kim, B. Gao, S. Y. Han, M. H. Jung, A. K. Chakraborty, T. Ko, C. Lee and W. I. Lee, J. Phys. Chem. C, 2009, 113, 19179. 35 S. Shamaila, A. K. L. Sajjad, F. Chen and J. Zhang, J. Colloid Interface Sci., 2011, 356, 465. 36 H. Gnaser, M. R. Savina, W. F. Chalaway, C. E. Tripa, I. V. Veryovkin and M. J. Pellin, Int. J. Mass Spectrom., 2005, 245, 61. 37 B. Gao, Y. J. Kim, A. K. Chakraborty and W. I. Lee, Appl. Catal., B, 2008, 83, 202. 38 W. W. Yu and X. Peng, Angew. Chem., Int. Ed., 2002, 41, 2368. 39 Q. Dai, D. Li, H. Chen, S. Kan, H. Li, S. Gao, Y. Hou, B. Liu and G. Zou, J. Phys. Chem. B, 2006, 110, 16509. 40 K. S. Leschkies, R. Divakar, J. Basu, E. E. Pommer, J. E. Boercker, C. B. Carter, U. R. Kortshagen, D. J. Norris and E. S. Aydil, Nano Lett., 2007, 7, 1793. 41 Z. Yi, J. Ye, N. Kikugawa, T. Kako, S. Ouyang, H. S. Williams, H. Yang, J. Cao, W. Luo, Z. Li, Y. Liu and R. L. Withers, Nat. Mater., 2010, 9, 559. 42 J. W. Seo, Y. W. Jun, S. J. Ko and J. W. Cheon, J. Phys. Chem. B, 2005, 109, 5389. 43 Y. J. Lin, Y. H. Chang, W. D. Yang and B. S. Tsai, J. NonCryst. Solids, 2006, 352, 789. 44 J. Ma, J. Lian, X. Duan, X. Liu and W. Zheng, J. Phys. Chem. C, 2010, 114, 10671. 45 Y. T. Kwon, K. Y. Song, W. I. Lee, G. J. Choi and Y. R. Do, J. Catal., 2000, 191, 192. 46 N. Serpone, P. Maruthamuthu, P. Pichat, E. Pelizzetti and H. Hidaka, J. Photochem.Photobiol., A, 1995, 85, 247. 47 P. V. Kamat, Chem. Rev., 1993, 93, 267. 48 Y. Xu and A. A. Schoonen, Am. Mineral., 2000, 85, 543. 49 S. B. Rawal, S. Bera and W. I. Lee, Catal. Lett., 2012, 142, 1482. 50 Z. Q. Li, Y. L. Yin, X. D. Liu, L. Y. Li, H. Liu and Q. G. Song, J. Appl. Phys., 2009, 106, 083701. ´ndez-Garcı´a and G. Colo ´n, Chem. Rev., 51 A. Kubacka, M. Ferna 2012, 112, 1555. 52 M. Gratzel, Nature, 2001, 414, 338. 53 M. Wang, N. Chamberland, L. Breau, J. E. Moser, R. H. Baker, B. Marsan, S. M. Zakeeruddin and M. Gratzel, Nat. Chem., 2010, 2, 385. 54 Handbook of Chemistry and Physics, ed. C. R. Weast, CRC Press, Boca Raton, FL, 77th edn 1996. 55 T. Hirakawa and Y. Nosaka, Langmuir, 2002, 18, 3247. 56 G. Liu, C. Sun, L. Cheng, Y. Jin, H. Lu, L. Wang, S. C. Smith, G. Q. Lu and H. M. Cheng, J. Phys. Chem. C, 2009, 113, 12317. 57 A. B. Anderson and T. V. Albu, J. Am. Chem. Soc., 1999, 121, 11855. 58 H. Irie, S. Miura, K. Kamiya and K. Hashimoto, Chem. Phys. Lett., 2008, 457, 202.

This journal is

c

The Royal Society of Chemistry 2013