Effects of Surface Oxygen Vacancies on Photophysical and

This should be valuable for the practical application of nanosized TiO2 in fields ..... TiO2 samples, T550 (0% Zn), ZT550 (0.5% Zn), ZT550 (1% Zn), ZT...
0 downloads 0 Views 96KB Size
17860

J. Phys. Chem. B 2006, 110, 17860-17865

Effects of Surface Oxygen Vacancies on Photophysical and Photochemical Processes of Zn-Doped TiO2 Nanoparticles and Their Relationships Liqiang Jing, Baifu Xin, Fulong Yuan, Lianpeng Xue, Baiqi Wang, and Honggang Fu* The Laboratory of Physical Chemistry, School of Chemistry and Materials Science, Heilongjiang UniVersity, Harbin 150080, China ReceiVed: May 23, 2006; In Final Form: July 15, 2006

In this paper, TiO2 nanoparticles doped with different amounts of Zn were prepared by a sol-gel method and were mainly characterized by means of X-ray photoelectron spectroscopy (XPS), photoluminescence (PL), and surface photovoltage spectrum (SPS). The effects of surface oxygen vacancies (SOVs) of Zn-doped TiO2 nanoparticles on photophysical and photocatalytic processes were investigated along with their inherent relationships. The results show that the SOVs easily bind photoinduced electrons to further give rise to PL signals. The SOVs can result in an interesting sub-band SPS response near the band edge in the TiO2 sample consisting of much anatase and little rutile, except for an obvious band-to-band SPS response. Moreover, the intensities of PL and SPS signals of TiO2, as well as the photocatalytic activity for degrading phenol solution, can be enhanced by doping an appropriate amount of Zn. These improvements are mainly attributed to the increase in the SOV amount. It can be suggested that the SOVs should play an important role during the processes of PL, surface photovoltage, and photocatalytic reactions, and, for the as-prepared TiO2 samples doped with different amounts of Zn by thermal treatment at 550 °C, the larger the SOV amount, the stronger the PL and SPS signal, and the higher the photocatalytic activity.

1. Introduction Semiconductor functional materials are playing a more and more important role in science and technology for their unique optical, electrical, magnetic, catalytic, and chemical properties. They are widely used in fields such as solar energy conversion, photoluminescence, photocatalysis, and nanoelectronics.1-5 In recent years, much attention has been paid to TiO2 in the photocatalysis and photoelectric conversion because of its exceptional optical and electronic properties, strong oxidizing power, nontoxicity, low cost, and long-term stability against photocorrosion and chemical corrosion.6-9 The preparation, characterization, and modification of nanosized TiO2 have long been the central contents during the research processes of semiconductor photoelectric chemistry and photocatalysis.10-13 Photocatalytic reaction is a type of special photochemical process, and photoluminescence (PL) and surface photovoltage (SPV) are two kinds of main photophysical phenomena. Since the photochemical and photophysical processes are closely related to dynamic behaviors of photoinduced charges, there are certain intrinsic relationships among PL, SPV, and photocatalysis. Generally speaking, nanosized semiconductor materials usually contain a large amount of surface oxygen vacancies (SOVs) because of surface effects and size effects.14 These SOVs can strongly affect dynamic processes of photoinduced charges, further modifying photochemical and photophysical performances. Therefore, the influences of the SOVs on photophysical and photochemical processes should be considered as crucial questions during the designing and application courses of semiconductor nanomaterials. To further improve the performance of TiO2, doping metal ions and coupling other semiconductors are often adopted as * Authors to whom correspondence should be addressed. Tel: 086-45186608458; fax: 086-451-86673647; e-mail: [email protected].

effective modification methods.15-18 At present, some literature about doping Zn or coupling ZnO to enhance the photocatalytic and photoelectric performance of TiO2 has been reported.19-22 Among those papers, the performance improvement is often attributed to the increase in the separation rate of photoinduced charges, which is often supposed. However, few papers are involved with the SOV effects. To the best of our knowledge, papers devoted to a systematic study on the SOV effects on the photophysical and photochemical processes, mainly by means of X-ray photoelectron spectroscopy (XPS), PL, and surface photovoltage spectrum (SPS) methods, have seldom been reported so far. The XPS method is a kind of effective surface test technique about elemental composition and chemical states.11 The SPS and PL methods are two kinds of effective tools to investigate the photophysics, which can provide some important information, such as SOVs, charge separation or recombination, and transfer behavior.23-31 In this paper, the effects of the SOVs on the processes of PL, surface photovoltage (SPV), and photocatalysis were systematically investigated along with their relationships. Also, the Zn modification mechanisms related to the SOVs were discussed in detail. The results demonstrate that the SOVs should play an important role during the processes of photophysics and photocatalysis. This should be valuable for the practical application of nanosized TiO2 in fields such as photoelectric conversion, photoluminescence, and photocatalysis and to help understand photophysical and photochemical processes of nanosized semiconductors. 2. Materials and Methods 2.1. Chemicals and Reagents. All substances used in this study were analytical grade and were used without further purification from Shenyang Chemical Company in China. Deionized water was used in all experiments.

10.1021/jp063148z CCC: $33.50 © 2006 American Chemical Society Published on Web 08/23/2006

Processes of Zn-Doped TiO2 Nanoparticles 2.2. Preparation. The synthesis of TiO2 samples in this work was similar to that described in previous papers,24,32 employing a sol-gel method with Ti(OBu)4 as raw material, see Supporting Information (SI) 1. 2.3. Characterization. The samples were analyzed by XRD using a Rigaku D/MAX-rA powder diffractometer with a nickelfiltered Cu KR radiation source at 30 kV and 20 mA. The UVvis DRS spectra of the samples were recorded with a PE Lambda20 spectrometry. The PL spectra of the samples were recorded with a PE LS 55 spectrofluorophotometer. The surface composition of the samples was examined by XPS using a VG ESCALAB MgKR X-ray source. The pressure was maintained at 6.3 10-6 Pa. The binding energies were calibrated with respect to the signal for adventitious carbon (binding energy ) 284.6 eV). Relative quantitative analysis was carried out using the sensitivity factors supplied by the instrument; the SPS measurement of the samples was carried out with a home-built apparatus that has been described elsewhere.26,33 Monochromatic light was obtained by passing light from a 500 W xenon lamp (CHF XQ500W, Global Xenon Lamp Power, made in China) through a double-prism monochromator (Hilger and Watts, D 300, made in England). A lock-in amplifier (SR540, made in U.S.A), synchronized with a light chopper (SR540, made in U.S.A.), was employed to amplify photovoltage signal. The SPS measurement was performed by a photovoltage cell, mainly consisting of two ITO quartz glass electrodes. For a powder experiment, the powder sample was sandwiched between the two ITO quartz glass electrodes. The raw SPS data were normalized using the illuminometer (Zolix UOM-1S, made in China). 2.4. Photocatalytic Activity Evaluation. Phenol (C6H5OH) is a common chemical material that is extensively used in a variety of industrial applications.34 Therefore, it is chosen to be a model pollutant. The degradation intermediates were not determined. The experiments were carried out in a 500 mL quartz photochemical reactor, open to air, having the shape of a vertical cylinder. The light was provided from a side of the reactor by a 350 W high pressure Hg lamp without filter, which was placed at about 20 cm from the reactor, with the stronger light of 365 nm. The total treated volume was 300 mL, and the initial concentration of phenol was equal to 0.10 mmol/L. The phenol solution was continuously stirred with a magnetic stirrer. The solution was first stirred for 20 min after 0.2 g of TiO2 samples was added into the reaction system, it has been shown that this period was sufficient to reach the adsorption equilibrium, and then the solution began to illuminate. The phenol concentrations at different times were measured using the colorimetric method of 4-aminoantipyrine with a model 721 spectrophotometer at the wavelength of 510 nm after centrifugation.32 3. Results and Discussion 3.1. Crystal Phase and Surface Composition. TiO2 exists in two main crystallographic forms, anatase (A) and rutile (R).8 The XRD peaks at 2θ ) 25.28° (A101) and 2θ ) 27.4 ° (R110) are often taken as the characteristic peaks of anatase and rutile crystal phase, respectively.10 Figure 1 showed the XRD patterns of TiO2 nanoparticles undoped by respective thermal treatment at 450 °C (T450), 550 °C (T550), and 650 °C (T650) and doped with 3 mol % Zn by respective thermal treatment at 450 °C (ZT450), 550 °C (ZT550), and 650 °C (ZT650). It can be seen that the ZT450 is anatase phase and the ZT650 mainly contains a rutile phase. However, the ZT550 consists of much anatase and little rutile. The XRD intensities of the anatase (A101) and rutile (R110) characteristic peaks are analyzed. The mass percentage of anatase in the samples can be estimated from the

J. Phys. Chem. B, Vol. 110, No. 36, 2006 17861

Figure 1. XRD patterns of TiO2 nanoparticles undoped and doped with Zn by respective thermal treatment at 450, 550, and 650 °C.

Figure 2. XPS spectra of O1s on TiO2 undoped and doped with Zn by thermal treatment at 550 °C.

respective integrated characteristic XRD peak intensities according to the mass coefficient ratio (1.265) of anatase to rutile. The crystallite size can also be determined from the broadening of the corresponding X-ray spectral peak by the Scherrer formula.10 The evaluated results show that the anatase crystallite size of the ZT450 and the rutile crystallite size of the ZT650 are about 6 and 26 nm, respectively. The anatase content of ZT550 is about 90%, and its anatase and rutile crystallite sizes are 10 and 13 nm, respectively. By comparison, it can be confirmed that doping Zn has a little inhibiting effect on the phase transformation from anatase to rutile. The inhibiting effect slightly increases with enhancing the Zn amount in our experiment, see SI 2. The ZT550 sample is made of little rutile and much anatase, which is a similar phase composition to the internationally commercial P-25 TiO2. This commercial TiO2 usually exhibits high photocatalytic activity because of its phase composition.6,8 In addition, the ZnTiO3 phase (PDF No. 150591) can be produced while the thermal treatment temperature is 650 °C. The XPS spectra of Ti2p, Zn2p, and O1s of TiO2 nanoparticles undoped (T550) and doped with 3 mol % Zn (ZT550) by thermal treatment at 550 °C were performed, see SI 3. The XPS peaks of Zn2p and Ti2p are sharp and symmetrical, and the XPS peak positions of Zn2p3 and Ti2p3 are at about 1021.8 and 458.6, respectively, demonstrating that the main chemical states of Zn and Ti in the samples are +2 and +4 valence, respectively.32,35 The O1s XPS spectra are wide and asymmetric, indicating that there are at least two kinds of chemical states according to the binding energy range from about 528 to 533 eV, including crystal lattice oxygen (OL) and chemiadsorbed oxygen (OH) with increasing binding energy.35 Thus, the O1s XPS spectrum is fitted to two kinds of chemical states by the Origin software with Gaussian rule, as shown in Figure 2, and the corresponding XPS data are listed in Table 1. The binding energy of OL and OH are 529.9 and 531.6 eV, respectively. The OL XPS is mainly

17862 J. Phys. Chem. B, Vol. 110, No. 36, 2006

Jing et al.

TABLE 1: Results of Curve Fitting of O1s XPS Spectra of T550 and ZT550 samples T550 ZT550

OL (Ti-O)

OH (O-H)

529.9 1.6 65.6 529.9 1.4 54.4

531.6 3.5 34.4 531.6 2.8 45.6

Eb/eV fwhm ri/% Eb/eV fwhm ri/%

TABLE 2: Atomic number ratio of OL, Ti, and Zn and the SOV relative content of T550 and ZT550 samples

atomic number ratio of OL, Ti, and Zn

SOV content (%)

T550 ZT550

100:59:0 100:62:3

15 21

attributed to the contribution of Ti-O in TiO2 crystal lattice, and the OH XPS is closely related to the hydroxyl groups resulting mainly from the chemisorbed water.11 For a careful XPS analysis, the evaluated data of atomic number ratio and relative SOV content are shown in Table 2. As seen from Table 2, the atomic number ratio of Ti to Zn is 62:3, lower than the desired atomic number ratio (100:3), illustrating that the Zn2+ content on the surfaces is larger than that in the inner. For ZT550, however, no Zn phase can be found in the XRD patterns. Also, the XRD peaks of TiO2 doped with Zn do not shift compared with TiO2, indicating that Zn does not come into the TiO2 lattices to substitute Ti. This is possibly because of the larger valence difference between Zn2+ and Ti4+. Thus, on the basis of the above XRD and XPS analysis, it can be concluded that Zn possibly exists as the main chemical form of lots of ZnO clusters and is mainly dispersed on TiO2 crystallite surfaces (or among TiO2 crystallites because of aggregation), further producing ZnO and TiO2 composite nanoparticles. Generally speaking, the phase transformation process from anatase to rutile easily takes place at the beginning of the surfaces of anatase crystallites because of higher surface energy. On the basis of XPS analysis, Zn2+ content on the surfaces is larger than that in the inner. Therefore, it can be deduced that the easy existence of the phase related to Zn on the anatase crystallite surfaces would inhibit slightly the production and growth of the rutile phase, further inhibiting phase transformation of TiO2. In addition, the percentages of SOV content of TiO2 nanoparticles undoped and doped with Zn are 15 and 21, respectively, demonstrating that the SOV content of TiO2 is enhanced by doping Zn. This corresponds with the XRD results. The increase in the SOV content is responsible for the increase in the OH amount. The surface hydroxyl groups can favor photocatalytic reactions.6-8 3.2. Optical Absorption and PL Characteristics. The UVVis-DRS spectra of different TiO2 nanoparticles are shown in Figure 3; the DRS spectrum gradually shifts to the red with increasing thermal treatment temperature, which is due to the increase in the crystallite size and crystallinity. According to the energy band structure of TiO2, the optical absorption at the wavelength range of shorter than 400 nm is mainly attributed to the electron transitions from the valence band to conduction band (band-to-band transition, O2pfTi3d), while the weak optical absorption at the wavelength range from 400 to 550 nm results from sub-band transitions closely related to the SOVs.33 By comparison, it can be found that the optical absorption range of TiO2 expands after doping Zn, which is in good agreement with XRD and XPS results about the SOVs. The absorption

Figure 3. DRS spectra of different TiO2 nanoparticles.

Figure 4. PL spectra of different amounts of Zn-doped TiO2 with the excitation wavelength of 300 nm.

increase at the wavelength range of shorter than 400 nm is possibly attributed to fine ZnO particles, while that at the wavelength range of larger than 400 nm is mainly because of the increase in the SOV amount. In general, bulk and single-crystal TiO2 cannot exhibit PL signals at room temperature, however, nanosized TiO2 has PL signals.28-31,36-39 Figure 4 shows the PL spectra of TiO2 nanoparticles doped with different amounts of Zn by thermal treatment at 550 °C using the excitation light of 300 nm. The energy of the used excitation light is enough to promote electronic transitions from the valence band to the conduction band of TiO2 according to the above DRS spectra. All the samples can exhibit an obvious PL signal with a similar curve shape, demonstrating that the Zn species do not result in a new PL phenomenon. All PL spectra are wide at the wavelength range from 400 to 550 nm, with two obvious PL peaks at about 420 and 480 nm, respectively. Considering the wavelength of the above PL signals and DRS results, it can be confirmed that these PL signals mainly result from the SOVs.30,31,38 Although the excitation wavelength is different, the PL peak position does not change, indicating that the sub-band levels related to the SOV are fairly stable, see SI 4. To illustrate PL process mechanisms and SOV effects, an energy level schematic of the valence band and the conduction band of TiO2 and ZnO as well as the sub-bands related to the SOVs is shown in Figure 5. The potential position of the valence band and the conduction band, versus the normal hydrogen electrode (NHE) at pH ) 1, can be determined.7 The energy band gaps of bulk ZnO and anatase TiO2 are 3.2 eV and that of bulk rutile TiO2 is 3.0 eV.6-9 According to DRS and PL spectra, the sub-band levels related to the SOVs possibly exist near the bottom of the conduction band or the top of the valence band. If it is near the bottom of the conduction band (proved by means of SPS method later), there are at least three main steps during the PL processes. First, the electrons are excited to the con-

Processes of Zn-Doped TiO2 Nanoparticles

Figure 5. An energy level schematic of the valence band and the conduction band of TiO2 and ZnO as well as sub-bands related to the SOVs.

duction band from the top of the valence band. Next, the excited electrons are captured or bound by different levels of sub-bands related to oxygen vacancies. Finally, the electrons can recombine with the holes in the valence to give rise to PL signals. The SOVs of nanosized semiconductor easily capture or bind photoinduced electrons, which has been proved by the electronic spin resonance method.35 Thus, the larger the SOV content is, the stronger the PL signal possibly is. As seen from Figure 4, the Zn dopant has a great effect on the PL intensity of TiO2 nanoparticles. The PL intensity increases with increasing the Zn content and arrives at the higher degree while the Zn content is 3 mol %. If the Zn content continues to increase, namely, more than 3 mol %, the PL intensity begins to go inversely down. On the basis of the above PL mechanisms and the results of XRD and XPS, the increase in the PL intensity is mainly due to the increase in the SOV amount. Zn2+ content on the surfaces is larger than that in the inner by means of the XPS method. If the Zn dopant amount is too large, TiO2 would be covered with excess ZnO, even with a shell ZnO, which can make both the effective area for absorbing light and the surface oxygen vacancy amount decrease. This will result in the decrease of PL intensity of TiO 2 nanoparticles. Therefore, it can be predicted that the separation situation of photoinduced chargers of TiO2 nanoparticles can be improved by doping an appropriate amount of Zn since the SOVs easily capture or bind photoinduced electrons. 3.3. SPS and EFISPS Analysis. The SPV method, with a very high sensitivity, is a well-established contactless and nondestructive technique for semiconductor characterization that relies on analyzing illumination-induced changes in the surface voltage.23-27,33 The photovoltage generation mainly arises from the creation of electron-hole pairs, followed by the separation under a built-in electric field (also called space-charge layer). The SPS is a kind of action spectrum on the basis of optical absorption. Thus, it can reflect photogenerated charge separation and transfer behavior as well as optical absorption characteristics of semiconductor samples, especially for the EFISPS method, in which the SPS is combined with the electric-field-modified technique.40 The SPS responses of 3 mol % Zn-doped TiO2 nanoparticles calcined at different temperatures are shown in Figure 6. For the three samples, an obvious SPS response, with an SPS peak (P1) at about 345 nm, can be found at the wavelength range from 300 to 370 nm, which can be mainly attributed to the electron transitions from the valence band to the conduction band (O2pfTi3d) on the basis of DRS spectra and TiO2 band structure.24,27,33 During the process of SPV measurement, the ITO glass can strongly absorb the light with the wavelength

J. Phys. Chem. B, Vol. 110, No. 36, 2006 17863

Figure 6. SPS responses of Zn-doped TiO2 nanoparticles calcined at different temperatures (inset: a schematic diagram of a photovoltage measurement).

lower than 330 nm, which results in a corresponding weak or even no SPV signal.26 The SPS response assigned to ZnO cannot be distinguished in that TiO2 and ZnO have similar band structures and energy band gaps. The ZT550 sample has a very strong SPS response, indicating that its photoinduced electronhole pairs are easily separated on the basis of the SPS principle.23 This sample is a coexistence of much anatase and little rutile. A little rutile cannot affect the optical adsorption of TiO2. The recombination of photoinduced charges can be inhibited because the electrons can easily transfer from the anatase conduction band to rutile because of the energy match (see Figure 5). These factors are responsible for a very stronger SPS response of ZT550 than that of ZT450 and ZT650. The ZT650 mainly consists of rutile phase, and its optical absorption decreases compared with ZT450 and ZT550, resulting in a very weak SPS signal. Moreover, its narrow band gap is not favorable for the increase in the separation situation of photoinduced charges. This weak SPS response of rutile is in good agreement with its low photocatalytic activity.6-8 Interestingly, except for an SPS P1 response, a new SPS response, with an SPS peak (P2) at about 405 nm situated near the energy band edge, can be seen at the wavelength range from 370 to 500 nm in ZT550 and ZT650 and is especially very remarkable in ZT550. On the basis of the above analysis, this SPS P2 response is reasonably ascribed to the sub-band transitions and is closely related to the rutile phase. Although the PL peak signal and the SPS P2 response both result from sub-band transitions related to the SOVs, their corresponding wavelengths are different, which can be attributed to their different process mechanisms. The sub-band transitions have great effects on the SPV signal of an n-type semiconductor,23 see SI 5. If the subbands related to the SOVs are near the bottom of the conduction band, an SPS response cannot appear near the band edge. On the contrary, a corresponding SPS response should appear if the sub-bands are located near the top of the valence band. In fact, an SPS P2 response cannot appear in an anatase sample like ZT450 with a lot of SOVs. Therefore, it can be confirmed that the sub-bands related to the SOVs are near the bottom of the conduction band. This is in accordance with the literature in theoretical calculation.41 For a sample with a coexistence of anatase and rutile like ZT550, however, the excited electrons in the anatase sub-bands between the conduction bands of anatase and rutile possibly transfer to the conduction band of rutile because of the energy match, which is responsible for a remarkable SPS P2 response. Thus, it can be deduced that, if the TiO2 sample contains a certain amount of rutile phase, the larger its anatase SOV amount is, the stronger its SPS P2 response is. The ZT650 mainly made

17864 J. Phys. Chem. B, Vol. 110, No. 36, 2006

Jing et al.

TABLE 3: Photocatalytic Degradation Rate Data of Phenol on Different TiO2 Samples TiO2 samples

T550 (0% Zn)

ZT550 (0.5% Zn)

ZT550 (1% Zn)

ZT450 (3% Zn)

ZT550 (3% Zn)

ZT650 (3% Zn)

ZT550 (5% Zn)

degradation rate of phenol (%)

45

51

58

65

76

37

33

of rutile exhibits a weak SPS P2 response, attributed to a small amount of anatase SOVs. The SPS responses of TiO2 nanoparticles doped with different amounts of Zn by thermal treatment at 550 °C are shown in Figure 7. It can be found that the Zn amount has a great effect on the SPS intensity of TiO2. The SPS response becomes gradually strong as the Zn amount increases. When the Zn amount is 3 mol %, the SPS intensity reaches a higher degree, indicating that the corresponding TiO2 sample has a high separation rate of photoexcited carriers. However, the SPS response can become weak if the Zn amount is too much. Surprisingly, the effects of the Zn amount on SPS and PL signals are very similar, namely, the stronger the PL signal, the stronger the SPS response. On the basis of the above PL analysis, the separation situation of photogenerated charges of band-to-band transitions can be improved because the SOVs can easily capture or bind electrons. Thus, it can be deduced that the increase in the SPS intensity can be attributed to the increase in the SOV amount. Moreover, the photoexcited electrons can easily transfer from the ZnO to the TiO2 conduction band because of their energy match, which also is favorable for the SPS intensity increment. However, if TiO2 is covered with excess ZnO, this possibly makes the SPS response become weak. Therefore, it can be concluded that the separation situation of photoinduced chargers of TiO2 nanoparticles can be improved by doping an appropriate amount of Zn, demonstrating that the SOVs should play an important role during the PL and SPV photophysical processes. The EFISPS responses of TiO2 nanoparticles doped with 3 mol % Zn by thermal treatment at 450 (A), 550 (B), and 650 (C) °C were measured, see SI 6. For the three samples, the SPS responses greatly change under an outer electrical field, demonstrating that the SPS responses related not only to bandto-band transitions but also to the SOVs are impressionable to an outer field.26 When a positive electric field is employed, the SPS intensity markedly rises with increasing field strength, which is attributed to the same direction of added-outer as builtin field.33 Moreover, the Stark effect also can explain the intensity increase of the SPS response.42 The electronic density at the top of the valence band is enhanced with increasing electric field strength. The increment of state density will increase the probability of excited transitions. On the contrary, the SPS response decreases while a negative electrical field is added. For ZT450, the EFISPS response range extends to the longer wavelength direction to 500 nm, which is assigned to

sub-band transitions related to the SOVs. Under no external field, although the electronic transitions from the valence band to the sub-bands can occur, the SPS response is weak or even undetectable. If an external field is applied, the probability of electronic transitions related to sub-bands can increase, and the corresponding separation rate of photoinduced charges can also be improved, resulting in a detectable SPS response. Moreover, the SPS P2 position can change after an outer field is added, demonstrating that an outer field can affect the electronic injections of different levels of sub-bands to the rutile phase. This possibly is a characteristic of the EFISPS response related to sub-band transitions.26 In addition, it can be found that the SPS response of ZT450 cannot reverse under a negative field, while that of ZT550 and ZT650 can. By comparison, it can be concluded that the phase composition and energy band gap are responsible for the results. 3.4. Photocatalytic Activity and Its Enhancement Mechanisms. The photocatalytic activity of TiO2 samples was evaluated by photocatalytic degradation of phenol solution. The degradation rate of phenol was calculated, see SI 7. The larger the degradation rate of phenol is, the higher the photocatalytic activity is. The experiment results demonstrate that the direct photolysis of phenol, about 3%, can be ignored compared with the photocatalytic degradation. The photocatalytic degradation rate data of phenol on different TiO2 samples for 2 h are shown in Table 3. As seen from Table 3, among TiO2 samples doped with 3 mol % Zn by respective thermal treatment at 450, 550, and 650 °C, the ZT550 sample exhibits a higher photocatalytic activity. Generally speaking, the stronger the SPS response is, the higher the separation rate of photoinduced charges is, which is responsible for the above activity order.24 Very interestingly, the activity orders of TiO2 samples doped with different amounts of Zn by thermal treatment at 550 °C are in good agreement with their corresponding PL and SPS intensity orders, in other words, the stronger the PL and SPS signal, the higher the photocatalytic activity. These results are principally attributed to the SOVs on the basis of the above analysis of PL and SPS responses. The increase in the SOV amount is favorable for the increase in the PL and SPS intensity, indicating that the SOVs are highly advantageous to the separation of photoinduced charges so as to enhance the activity. Moreover, the increase in the SOV amount also results in the increment in the surface hydroxyl amount, which is good to photocatalytic reactions.6-8 This demonstrates that the SOVs should play an important role in photocatalytic reactions. 4. Conclusions

Figure 7. SPS responses of TiO2 doped with different amounts of Zn by thermal treatment at 550 °C.

On the basis of the above analysis, the following conclusions can be drawn: (1) Different levels of sub-bands resulting from the SOVs are located near the bottom of the conduction band, and the SOVs easily bind photoinduced electrons to further give rise to PL signals; (2) except for an SPS response attributed to band-to-band transitions, an interesting SPS response related to sub-band transitions appears in the TiO2 with coexistence of anatase and rutile, demonstrating that the excited electrons on appropriate levels of anatase sub-bands can transfer to the conduction band of rutile. The larger the anatase SOV amount is, the stronger the corresponding SPS response is; (3) an appropri-

Processes of Zn-Doped TiO2 Nanoparticles ate amount of Zn dopant can improve the photocatalytic activity as well as the intensities of PL and SPS response of TiO2, which is attributed to the increase in the SOV amount. Moreover, it can be suggested that the SOVs should play an important role during the photophysical and photochemical processes, and the larger the SOV amount, the stronger the PL and SPS signals, and the higher the photocatalytic activity. Acknowledgment. This project was supported from the Key Program Projects of National Nature Science Foundation of China (No. 20431030), the National Nature Science Foundation of China (No. 20301006 and 20501007), the Program for New Century Excellent Talents in University (NCET), the Key Nature Science Foundation of Heilongjiang Province of China (No. ZJG04-04), the Foundation for Excellent Youth of Heilongjiang Province of China, and the Science Foundation of Harbin City of China (No. 2005AFQXJ060), for which we were very grateful. Supporting Information Available: Description of the synthesis of the TiO2 samples, figures showing XRD patterns, XPS spectra, PL spectra, the effects of the sub-band transitions on the SPV signal, EFISPS responses, and description of the calculation for the degradation rate of phenol. This material is available free of charge via the Internet at http://pubs.acs.org. References and Notes (1) Fujishima, A.; Honda, K. Nature 1972, 238, 37-38. (2) O’Regan, B.; Gra¨tzel, M. Nature 1991, 353, 737-740. (3) Ma, W.; Li, J.; Tao, X.; He, J.; Xu, Y. M.; Yu, J. C.; Zhao, J. C. Angew. Chem., Int. Ed. 2003, 42, 1029-1032. (4) Nitzan, A.; Ratner, M. A. Science 2003, 300, 1384-1389. (5) Jose, D. A.; Shukla, A. D.; Kumar, D. K.; Ganguly, B.; Das, A.; Ramakrishna, G.; Palit, D. K.; Ghosh, H. N. Inorg. Chem. 2005, 44, 24142425. (6) Hoffmann, M. R.; Martin, S. T.; Choi, W.; Bahnemann, D. W. Chem. ReV. 1995, 95, 69-96. (7) Linsebigler, A. L.; Lu, G.; Yates, J. T. Chem. ReV. 1995, 95, 735758. (8) Litter, M. I. Appl. Catal. B 1999, 23, 89-114. (9) Hagfeldt, A.; Gratzel, M. Chem. ReV. 1995, 95, 49-68. (10) Zhang, Q.; Gao, L.; Guo, J. Appl. Catal. B 2000, 26, 207-215. (11) Yu, J. G.; Yu, H. G.; Cheng, B.; Zhao, X. J.; Yu, J. C.; Ho, W. K. J. Phys. Chem. B 2003, 107, 13871-13879. (12) Fujishima, A.; Rao, T. N.; Tryk, D. A. J. Photochem. Photobiol. C 2000, 1, 1-21. (13) Diebold, U. Surf. Sci. Rep. 2003, 48, 53-229.

J. Phys. Chem. B, Vol. 110, No. 36, 2006 17865 (14) Zhang, L.; Mou, J. Nanomaterials and nanostructure; Science Press: Beijing, 2001; p 12. (15) Vinodgopal, K.; Kamat, P. V. EnViron. Sci. Technol. 1995, 29, 841-845. (16) Kawahara, T.; Konishi, Y.; Tada, H.; Tohge, N.; Ito, S. Langmuir 2001, 17, 7442-7445. (17) Chu, S. Z.; Inoue, S.; Wada, K.; Li, D.; Suzuki, J. Langmuir 2005, 21, 8035-8041. (18) Wilke, K.; Breue, H. D. J. Photochem. Photobiol. A 1999, 121, 49-53. (19) Wang, Z. S.; Huang, C. H.; Huang, Y. Y.; Hou, Y. J.; Xie, P. H.; Zhang, B. W.; Cheng, H. M. Chem. Mater. 2001, 13, 678-682. (20) Willis, R. L.; Olson, C.; O’Regan, B.; Lutz, T.; Nelson, J.; Durrant, J. R. J. Phys. Chem. B 2002, 106, 7605-7613. (21) Sun, R. D.; Nakajima, A.; Fujishima, A.; Watanabe, T.; Hashimoto, K. J. Phys. Chem. B 2001, 105, 1984-1990. (22) Kandavelu, V.; Kastien, H.; Thampi, K.; Ravindranathan, K. Appl. Catal. B 2004, 48, 101-111. (23) Kronik, L.; Shapira, Y. Surf. Sci. Rep. 1999, 37, 1-206. (24) Jing, L. Q.; Fu, H. G.; Wang, B. Q.; Wang, D. J.; Xin, B. F.; Li, S. D.; Sun, J. Z. Appl. Catal. B 2006, 62, 282-291. (25) Lenzmann, F.; Krueger, J.; Burnside, S.; Brooks, K.; Gratzel, M.; Gal, D.; Ruhle, S.; Cahen, D. J. Phys. Chem. B 2001, 105, 6347-6452. (26) Lin, Y. H.; Wang, D. J.; Zhao, Q. D.; Yang, M.; Zhang, Q. L. J. Phys. Chem. B 2004, 108, 3202-3206. (27) Xin, B. F.; Jing, L. Q.; Ren, Z. Y.; Wang, B. Q.; Fu, H. G. J. Phys. Chem. B 2005, 109, 2805-2809. (28) Li, X. Z.; Li, F. B. EnViron. Sci. Technol. 2001, 35, 2381-2387. (29) Yu, J. C.; Yu, J. G.; Ho, W. K.; Jiang, Z. T.; Zhang, L. Z. Chem. Mater. 2002, 14, 3808-3816. (30) Zhang, W. F.; Zhang, M. S.; Yin, Z.; Chen, Q. Appl. Phys. B 2000, 70, 261-265. (31) Zhang, L. D.; Mo, C. M. Nanostruct. Mater. 1995, 6, 831-834. (32) Jing, L. Q.; Sun, X. J.; Xin, B. F.; Wang, B. Q.; Cai, W. M.; Fu, H. G. J. Solid State Chem. 2004, 177, 3375-3382. (33) Jing, L. Q.; Sun, X. J.; Shang, J.; Cai, W. M.; Xu, Z. L.; Du, Y. G.; Fu, H. G. Sol. Energy Mater. Sol. Cells 2003, 79, 133-151. (34) Dohnal, V.; Fenclova, D. J. Chem. Eng. Data 1995, 40, 478-483. (35) Jing, L. Q.; Xu, Z. L.; Sun, X. J.; Shang, J.; Cai, W. M. Appl. Surf. Sci. 2001, 180, 308-314. (36) Cohen, R.; Kronik, L.; Shanzer, A.; Cahen, D.; Liu, A.; Rosenwaks, Y.; Lorenz, J. K.; Ellis, A. B. J. Am. Chem. Soc. 1999, 121, 10545-10553. (37) Zhang, Y. X.; Li, G. H.; Jin, Y. X.; Zhang, Y.; Zhang, J.; Zhang, L. D. Chem. Phys. Lett. 2002, 365, 300-304. (38) Hong, J.; Cao, J.; Sun, J.; Li, H.; Chen, H.; Wang, M. Chem. Phys. Lett. 2003, 380, 366-371. (39) Serpone, N.; Lawless, D.; Khairutdinov, R. J. Phys. Chem. 1995, 99, 16646-16654. (40) Wang, D. J.; Zhang, J.; Shi, T. S.; Wang, B. H.; Cao, X. Z.; Li, T. J. J. Photochem. Photobiol. A 1996, 93, 21-25. (41) Lindan, P. J. D.; Harrison, N. M.; Gillan, M. J.; White, J. A. Phys. ReV. B 1997, 55, 15919-15926. (42) Szczepankiewicz, S. H.; Moss, J. A.; Hoffmann, M. R. J. Phys. Chem. B 2002, 106, 7654-7658.