pubs.acs.org/Langmuir © 2009 American Chemical Society
Comparison of Dye Photodegradation and its Coupling with Light-to-Electricity Conversion over TiO2 and ZnO Yuanzhi Li,*,† Wei Xie,† Xuelei Hu,‡ Guofang Shen,† Xi Zhou,† Ye Xiang,† Xiujian Zhao,† and Pengfei Fang§ † Key Laboratory of Silicate Materials Science and Engineering (Wuhan University of Technology), Ministry of Education, 122 Luoshi Road, Wuhan 430070, China, ‡School of Chemical Engineering and Pharmacy, Hubei Key Laboratory of Novel Chemical Reactor & Green Chemistry Technology, Wuhan Institute of Technology, 693 Xiongchu Road, Wuhan 430073, China, and §Department of Physics, Wuhan University, Luojia Hill, Wuhan 430072, China
Received June 12, 2009. Revised Manuscript Received July 21, 2009 Through comparing the photocatalytic performance of microscale ZnO, nano ZnO, and Degussa P25 titania (P25), it was found that the microscale ZnO exhibited 2.6-35.7 times higher photocatalytic activity for the photodegradation of various dye pollutants than P25 under both UV-visible and visible irradiation and showed much better photostability than the nano ZnO. The photocatalysts were characterized with XRD, Raman, BET, DRUV-vis, adsorption of dye, photoelectrochemical measurement, and PL. The much higher photocataltyic activity of the microscale ZnO than P25 under UV-visible irradiation is attributed to the higher efficiency of generation, mobility, and separation of photoinduced electrons and holes. The much higher visible photocataltyic activity of the microscale ZnO than P25 is due to the higher photosensitization efficiency of electron transfer from an excited dye to the conduction band of the microscale ZnO than that of P25. The much better photostability of the microscale ZnO than the nano ZnO is due to its better crystallinity and lower defects. The photostability of the microscale ZnO is greatly improved by the surface modification of ZnO with a small amount of TiO2. On the basis of the excellent photocatalytic performance of the microscale ZnO and TiO2-modified ZnO, a novel device of coupling photodegradation with light-to-electricity conversion was developed, which is a promising candidate for the photocatalytic removal of dye pollutants and a renewable energy source.
Introduction Dye effluents from textile industries and photographic industries are becoming a serious environmental problem because of their toxicity, unacceptable color, high chemical oxygen demand content, and resistance to chemical, photochemical, and biological degradation.1 Heterogeneous photocatalysis based on nanostructured TiO2 has been extensively studied as an important destructive technology leading to the total mineralization of a wide range of organic dyes, which has been being an international hot topic for decades. ZnO is another semiconductor investigated as a potential photocatalyst in recent years. Various kinds of nanostructured ZnO, such as nanoparticle, nanorod, nanobelt, nanoplate, hollow sphere, and micro/nanostructure, have been used for the photodegradation of dye pollutants.2-7 In some cases, ZnO exhibits a better photocatalytic efficiency than TiO2. But few works reported that the photocatalytic activity of ZnO is better than Degussa P25 titania (P25),7,8 which has been proved to be one of most efficient photocatalysts, and widely used as a benchmarking photocatalyst. However, both TiO2 and ZnO are activated only under UV irradiation because of their large band gap, which greatly limits their application in environmental
decontamination as solar spectra only contain 5% of UV. Therefore, it is crucial to explore efficient methods to extend their photocatalytic response from UV to visible region. Photosensitization of TiO2 by dyes has been proved to be one of the most effective ways, which has been used for the visible photodegradation and mineralization of dye pollutants. In this system, the dye instead of the TiO2 photocatalyst is excited by visible light irradiation. The excited dye molecule transfers electrons into the conduction band of TiO2, while the dye itself is converted to its cationic radical. The injected electrons can react with dioxygen adsorbed on the surface of TiO2 to generate a series of active oxygen species such as O2-, •OH, and H2O2. The subsequent radical chain reaction can lead to the degradation of the dye.9,10 The question is: is it possible to efficiently extend the photoresponse of ZnO from UV to the visible region by photosensitization similar to TiO2? If the answer is yes, could the photosensitization be used in the visible photocatalytic removal of dye pollutants? One of the major drawbacks of ZnO photocatalyst is its photoinstability in aqueous solution due to its photocorrosion with UV irradiation, which significantly decreases the photocatalytic activity of ZnO and blocks its practical application in environment purification.11 It is anticipated that ZnO would
*Corresponding author. E-mail:
[email protected]. (1) Yanagisawa, K.; Ovenstone, J. J. Phys. Chem. B 1999, 103, 7781. (2) Jang, E. S.; Won, J. H.; Hwang, S. J.; Choy, J. H. Adv. Mater. 2006, 18, 3309. (3) Sun, T. J.; Qiu, J. S.; Liang, C. H. J. Phys. Chem. C 2008, 112, 715. (4) Ye, C. H.; Bando, Y.; Shen, G. Z.; Golberg, D. J. Phys. Chem. B 2006, 110, 15146. (5) Yu, J. G.; Yu, X. X. Environ. Sci. Technol. 2008, 42, 4902. (6) Deng, Z. W.; Chen, M.; Gu, G. X.; Wu, L. M. J. Phys. Chem. B 2008, 112, 16. (7) Lu, F.; Cai, W. P.; Zhang, Y. G. Adv. Funct. Mater. 2008, 18, 1047. (8) Sakthivel, S.; Neppolian, B.; Shankar, M. V.; Arabindoo, B.; Palanichamy, M.; Murugesan, V. Solar Energy Mater. Solar Cells 2003, 77, 65.
Langmuir 2010, 26(1), 591–597
(9) Cho, Y.; Choi, W.; Lee, C. H.; Hyeon, T.; Lee, H. I. Environ. Sci. Technol. 2001, 35, 966. (10) (a) Wang, Q.; Chen, C. C.; Zhao, D.; Ma, W. H.; Zhao, J. C. Langmuir 2008, 24, 7338. (b) Wu, T.; Lin, T.; Zhao, J.; Hidaka, H.; Serpone, N. Environ. Sci. Technol. 1999, 33, 1379–1387. (11) (a) Meulenkamp, E. A. J. Phys. Chem. B 1998, 102, 7764. (b) van Dijken, A.; Janssen, A. H.; Smitsmans, M. H. P.; Vanmaekelbergh, D.; Meijerink, A. Chem. Mater. 1998, 10, 3513. (c) Jongh, P. E.; Meulenkamp, E. A.; Vanmaekelbergh, D.; Kelly, J. J. J. Phys. Chem. B 2000, 104, 7686.
Published on Web 08/11/2009
DOI: 10.1021/la902117c
591
Article
Li et al.
become an excellent photocatalyst if the photocorrosion could be greatly suppressed. Recently, several methods have been developed to improve its photostability, including surface organic coating of ZnO,12 and surface hybridization of ZnO with carbon and fullerenes C60.13 These works provided possibility for the practical application of ZnO in the photocatalytic removal of dye pollutants. As the photogenerated active oxygen species have very strong oxidative ability, the organic compound or carbon used could be gradually oxidized during long photocatalytic process, thus probably resulting in the deterioration of their photostable role. Therefore, it is desirable to improve the photostability of ZnO by modification with inert and photostable inorganic compound. Currently, energy shortage is another main challenge in the world besides the environmental pollution. The dye-sensitized solar cells (DSSC) based on TiO2 and ZnO,14,15 especially for TiO2, have been extensively studied, and proved to be one of the most promising candidates for a new renewable energy source as they possess advantages of being flexible, inexpensive, and easier to manufacture than brittle silicon solar cells.14b,16 The reported researches on the photocatalysis and DSSC have independently proceeded during the last decades, and the cost of DSSC needs to be reduced as the dye used in DSSC (e.g., ruthenium polyridyl complex) is very expensive. It is highly desirable and greatly challenging to combine the advantages of photodegradation and light-to-electricity conversion, and develop a novel inexpensive device with two functions of DSSC and photodegradation of waste dye pollutants. In this article, it was found that microscale ZnO exhibited much higher photocatalytic activity for the photodegradation of various dye pollutants than P25 under both UV-visible and visible irradiation and showed much better photostability than nano ZnO. The photosensitization of ZnO by dye has been proved to be an effective way to extend its photoresponse into the visible region, and can be used in the visible photocatalytic removal of dye pollutants. The photostability of the microscale ZnO can be greatly improved by surface modification with a small amount of TiO2. On the basis of the high photocatalytic performance of the microscale ZnO and TiO2 modified ZnO, a novel concept of coupling photodegradation with DSSC was developed. The purification of waste dye pollutants and generation of renewable energy source were achieved at the same time.
Experimental Section Materials. Microscale ZnO was purchased from Wuhan Zhongbei Chemical Reagent Co. Titanium butoxide (Ti(OBu)4) was purchased from Shanghai Aisi Chemical Reagent Co. Titania P25 (TiO2, ca. 80% anatase, 20% rutile) was purchased from Degussa Co. Zn(NO3)2.6H2O, Na2CO3, nitric acid, poly(ethylene glycol), crystal violet(CV), methylene blue (MB), orange G (OG), methyl orange (MO), and ethanol were purchased from Shanghai Chemical Co. All of these chemicals were used without further purification. The chemical structures of the dyes are shown in Figure S1 (Supporting Information). (12) Comparelli, R.; Fanizza, E.; Curri, M. L.; Cozzi, P. D.; Mascolo, G.; Agostiano, A. Appl. Catal., B 2005, 60, 1. (13) (a) Zhang, L. W.; Cheng, H. Y.; Zong, R. L.; Zhu, Y. F. J. Phys. Chem. C 2009, 113, 2368. (b) Fu, H. B.; Xu, T. G.; Zhu, S. B.; Zhu, Y. F. Environ. Sci. Technol. 2008, 42, 8064. (14) (a) Gratzel, M. Nature 2001, 414, 338. (b) Nazeeruddin, M. K.; Deangelis, F.; Fantacci, S.; Selloni, A.; Viscardi, G.; Liska, P.; Ito, S.; Takeru, B.; Gratzel, M. J. Am. Chem. Soc. 2005, 127, 16835. (15) Cheng, H. M.; Chiu, W. H.; Lee, C. H.; Tsai, S. Y.; Hsieh, W. F. J. Phys. Chem. C 2008, 112, 16359. (16) Smestad, G.; Bignozzi, C.; Argazzi, R. Sol. Energy Mater. Sol. Cells 1994, 32, 259.
592 DOI: 10.1021/la902117c
Preparation of Photocatalysts and Photoelectrode. A 1.33 mol L-1 Na2CO3 aqueous solution was added to 100 mL of 0.2 mol L-1 Zn(NO3)2 aqueous solution under magnetic stirring until its pH value was 6-7. The formed precipitate was filtered, washed thoroughly with distilled water, and dried at 100 °C for 2 h. Nano ZnO was obtained by calcining the dried sample at 500 °C for 2 h. TiO2-modified ZnO was prepared according to the following procedure. A known amount of the Ti(OBu)4 ethanol solution with pH value of 2 adjusted by adding nitric acid was mixed with 1 g of the microscale ZnO. The mixture was dried under an infrared lamp and then calcined at 450 °C for 2 h. A mixture of the photocatalyst powder and poly(ethylene glycol) with a weight ratio of 5% was added to 15 mL of ethanol and ultrasonicated for 30 min. The obtained mixture was uniformly spread on an ITO glass substrate (1 cm 1.2 cm). After the evaporation of ethanol, the sample was heated to 480 °C in a muffle furnace at a rate of 2 °C min-1 and remained at this temperature for 2 h to completely remove poly(ethylene glycol). Characterization. X-ray diffraction (XRD) patterns were obtained on a Rigaku D/Max-IIIA X-ray diffractometer. Raman spectra were recorded on a Renishaw inVia Raman microscope with an excitation of 514.5 nm laser light. Diffusive reflectance UV-vis(DRUV-vis) absorption spectra were recorded on a UV2550 spectrophotometer. Photoluminescence (PL) spectra were recorded at room temperature on a Shimadzu RF-5301 PC spectrometer using 320 nm excitation light. The BrunauerEmmett-Teller (BET) surface area was measured on Autosorb-1 using N2 adsorption at -196 °C for the sample predegassed at 200 °C in a vacuum for 2 h. Photoelectrochemical measurements were carried out by using a homemade three-electrode quartz cell, Pt wire as the counter electrode, a saturated Ag/AgCl electrode as the reference electrode, and the thin film of the photocatalysts on ITO as the working electrode on an electrochemical analyzer (CHI750). The electrolyte used was a solution of 0.2 mol L-1 Na2SO4 and 4.0 10-6 mol L-1 dye. A 125 W high pressure Hg lamp (Shanghai Yaming Lighting Appliance Co. Ltd.) was used as light source. The intensity of UV light in the region of 320-400 nm and in the region of 400-1000 nm was measured with a UV-A Radiometer and a FZ-A Radiometer, respectively. Before the measurement, the electrolyte was purged by pure N2 to remove the dissolved oxygen. To measure the photocurrent under visible irradiation (λ > 420 nm), a UV cut-filter, which can filter out UV light with wavelengths below 420 nm, was placed between the lamp and cell. Photocatalytic Activity. The photocatalytic activity of the photocatalysts was evaluated by the photodegradation of several dyes. The light source was a 125 W high-pressure Hg lamp. The reaction was maintained at ambient temperature. In a typical experiment, aqueous suspensions of dye (50 mL, 1 10-4 mol L-1) and 0.2000 g of the photocatalyst powder were placed in the beaker. Prior to irradiation, the suspension was magnetically stirred in the dark to ensure the establishment of an adsorption/desorption equilibrium. The suspension was kept under constant air-equilibrated conditions. At the intervals of given irradiation time, 1 mL of the suspension was collected and centrifuged to remove the particles. The dye concentration was determined by measuring the UV-vis absorbance of the dye aqueous solution. To measure the photocatalytic activity under visible irradiation (λ > 420 nm), a UV cut-filter was used to filter out UV light with wavelengths below 420 nm.
Results and Discussion Nanosized materials are generally believed to perform much better than their bulk counterparts in photocatalytic properties due to the higher surface-to-volume ratio and shorter transfer distance of photogenerated charge carriers (e.g., electron and hole) from bulk interior to surface. This is especially important for designing TiO2-based photocatalysts of high efficiency because Langmuir 2010, 26(1), 591–597
Li et al.
Article
Figure 2. Time course of the decrease in the concentration (A) and ln(C0/C) (B) for the photodegradation of CV under both UVvisible and visible irradiation.
Figure 1. XRD patterns (A), Raman spectra (B) and DRUV-vis spectra (C) of the photocatalysts.
the mobility of charge carriers in TiO2 (0.1-1.0 cm2 V-1 s-1)17 is quite low, and only in nanosized TiO2 do the photogenerated electron and hole in bulk have more chance to move to its surface, and contribute to photocatalytic reaction. On the other hand, nanosized materials usually have abundant defects (especially for ZnO) due to their poor crystallinity. These defects are acted as recombination centers of photogenerated holes and electrons, which result in the decrease of their photocatalytic efficiency. Most seriously, in contrast to the photostability of TiO2, the abundant defects in nanosized ZnO induce photocorrosion,18 which results in its photoinstability. As the mobility (100205 cm2 V-1 s-1) in ZnO19 is much higher than that in TiO2, much more photogenerated electrons and holes in ZnO than in TiO2 could transfer to its surface, and contribute to photocatalytic reaction, suggesting that ZnO may exhibit higher photocatalytic activity than in TiO2. Therefore, the beneficial nanosized effect is not a main consideration for designing highly efficient and photostable ZnO photocatalyst. With this idea in mind, microscale ZnO with good crystallinity and low defects was chosen for the photodegradation of dye pollutants. XRD, Raman, and DRUV-Vis. Figure 1A shows XRD pattern of the microscale ZnO. The observed diffraction peaks can be indexed to those of hexagonal wurtzite ZnO (JCPDS 89-0511). No impurity phases were detected. Its average crystal size is determined to be 353 nm according to Scherrer formula (17) Frederikse, H. P. R. J. Appl. Phys. 1961, 32, 2211. (18) Kislov, N.; Lahiri, J.; Verma, H.; Goswami, D. Y.; Stefanakos, E.; Batzill, M. Langmuir 2009, 25, 3310. (19) Look, D. C.; Reynolds, D. C.; Sizelove, J. R.; Jones, R. L.; Litton, C. W.; Cantwell, G.; Harsch, W. C. Solid State Commun. 1998, 105, 399.
Langmuir 2010, 26(1), 591–597
(L = 0.89λ/β[cos θ]). Its BET surface area is 3.0 m2 g-1. Figure 1B shows the Raman spectra of microscale ZnO. There are observed exhibits strong bands at 583.1, 537.8, 435.3, 382.0, and 331.6 cm-1, which are attributed to the E1(LO), A1(LO), E2, A1(TO), and A1 modes of wurtzite ZnO, respectively.20 As shown in Figure 1C, the microscale ZnO possesses higher absorption than P25 in UV region, indicating a higher efficiency for the generation of electron-hole pairs than P25. The more photogenerated electronhole pairs together with their much higher mobility in the microscale ZnO than in P25 suggest that the microscale ZnO would show higher photocatalytic activity than P25. Photocatlytic Activity. Figure 2A shows the time course of the decrease in the concentration of crystal violet (CV). Under visible irradiation (λ > 420 nm), CV adsorbed on P25 is photodegraded very slowly due to the photosensitization of titania by the dye.9,10 In contrast, CV is almost completely photodegraded on the microscale ZnO in 440 min. The photodegradation of dye follows first-order kinetics as shown in Figure 1B. The rate constant on the microscale ZnO is 35.7 times higher than that on P25. The visible photocatalytic activity of the microscale ZnO is attributed to the photosensitization of ZnO by the dye similar to TiO2-dye system. Our finding of the highly effective visible photocatalytic activity of the microscale ZnO exhibits that the photosensitization of ZnO by dyes is an effective way to extend its photoresponse into the visible region, and can be used in the visible photocatalytic removal of dye pollutants. It is well-documented that P25 shows high activity for the photodegradation of various kinds of dyes under UV irradiation. We tested the photocatalytic activity of P25 and ZnO under UV-visible irradiation. Similarly, ZnO reveals much higher photocatalytic activity than P25, and its rate constant is 2.6 times higher than hat on P25. The time course of the decrease in the absorption of CV on microscale ZnO under UV-visible and visible irradiation was shown in Figures S2 and S3, respectively. Under UV-visible (20) Wang, R. P.; Xu, G.; Jin, P. Phys Rev. B 2004, 69, 13303.
DOI: 10.1021/la902117c
593
Article
Li et al.
Figure 3. Time course of the decrease in the concentration of MB on the photocatalysts under both UV-visible and visible irradiation.
Figure 4. Time course of the decrease in the concentration of OG on the photocatalysts under both UV-visible and visible irradiation.
Table 1. Photocatalytic Rate Constant and Saturation Adsorption Amount of Different Dyes rate constant 10-3 (min-1) vis
UV-vis
adsorption amount (μmol g-1)
dye
ZnO
TiO2
ZnO
TiO2
ZnO
TiO2
CV MB OG MO
10 5.9 2.9 2.2
0.28 0.49 0.38 no
29 75 90 64
11 15 20 11
0.46 0.51 0.05 0.02
0.19 0.09 0.69 0.15
irradiation, both the maximum absorption peak at 584 nm and the peaks in the range of 200-320 nm decrease quickly with the increase of irradiation time. After irradiation for 160 min, all these peaks disappears, indicating cleavage of the whole conjugated chromophore structure of CV and the single aromatic ring of intermediates, and further mineralization. Under visible irradiation, both the maximum absorption peak at 584 nm and the peaks in the range of 200-320 nm gradually decrease with the increase of irradiation time. In contrast, after irradiation for 440 min, the former peak disappears, but small peaks in the range of 200320 nm still exist. This result suggests that some intermediates with the single aromatic ring could not be further photodegraded due to lack of photosensitization. The photocatalytic activity of MB, another kind of cationic dye like CV, was tested on the microscale ZnO and P25. Their photocatalytic activity was shown in Figure 3, and the corresponding rate constants are summarized in Table 1. As in case of photodegradation of CV, the microscale ZnO shows much higher photocatalytic activity than P25 under both UV and visible irradiation. Its rate constants under visible irradiation (7.5 10-2 min-1) and under UV-visible irradiation are 12 and 5 times higher than those on P25 (1.5 10-2 min-1), respectively. The photodegradation of anionic dyes such as OG and MO on the microscale ZnO and P25 was also examined. Their photocatalytic activity was shown in Figure 4 and 5, and the corresponding rate constants are summarized in Table 1. As can be seen from Table 1, the microscale ZnO has much higher photocatalytic activity than P25 for the photodegradation of OG and MO under both UV-visible and visible irradiation. For the photodegradation of OG, the rate constants on the microscale ZnO under visible irradiation and under UV-visible irradiation are 7.6 and 4.5 times higher than those on P25, respectively. In the case of the photodegradation of MO, no detectable photodegradation on P25 was observed under visible irradiation, indicating that the photodegradation of the dye on P25 through photosensitization pathway is negligible. In contrast, the microscale ZnO shows considerable photocatalytic activity with a rate 594 DOI: 10.1021/la902117c
Figure 5. Time course of the decrease in the concentration of MO on the photocatalysts under both UV-visible and visible irradiation.
constant of 2.2 10-3 min-1. Under UV-visible irradiation, the rate constant on the microscale ZnO is 5.8 times higher than that on P25. These results show that no matter what the photodegraded dye is cationic or anionic, the microscale ZnO shows much higher photocatalytic efficiency than P25 under both UVvisible and visible irradiation. The specific surface area of microscale sized ZnO (3.0 m2 g-1) is far less than that of P25(59.5 m2 g-1). The much higher efficiency of microscale ZnO photocatalyst implies that surface area is not the determining factor for the photodegradation of organic dyes. The photocatalytic difference between ZnO and P25 is attributed to other factors, which need to be further probed. Adsorption of Dye. As the difference in the adsorption amount of dye on the microscale ZnO and P25 may influence their photocatalytic activity, by monitoring the concentration of the dyes in the bulk solution after the adsorption/desorption equilibrium was reached from various initial concentrations of different dyes, the adsorption isotherms of dyes in aqueous TiO2 and ZnO dispersions were examined. The adsorption isotherms in all the cases showed the typical Langmuir adsorption/desorption behaviors.10 Figure 6 presents the adsorption isotherms of CV on ZnO as an example. The good linear relationship (with a standard deviation R value of 0.9966 for ZnO systems) of the Langmuir isotherms(Ceq/Γ versus Ceq) suggests that CV tends to adsorb on the surface of ZnO in monolayer and single-site modes.10 The saturation amount of adsorption for the dyes on the microscale ZnO and P25 is summarized in Table 1. As the photocatalytic reaction takes place on the surface of photocatalyst, the much higher adsorption amount of CV and MB on the microscale ZnO than that on TiO2 seems to contribute to the much higher photocatalytic activity of the microscale ZnO than TiO2. However, as can be seen from Table 1, although the adsorption amount of OG and MO on the microscale ZnO is much lower Langmuir 2010, 26(1), 591–597
Li et al.
Article
Figure 8. Durability of the photocatalysts for the photodegradation of CV under UV-visible irradiation for 2 h.
Figure 6. Adsorption isotherms of CV on the microscale ZnO (A) and the corresponding plot of Ceq/Γ versus Ceq (B).
Figure 7. Photoelectrochemical response of ZnO and P25 electrode vs bias potential under visible (A) and UV-visible irradiation (B).
than that on TiO2, the microscale ZnO exhibits much higher photocatalytic activity than P25 under both UV-visible and visible irradiation. These results suggest that there is no direct relationship between photocatalytic activity and adsorption amount of the dyes. Photoelectrochemical Property. The photoelectrochemical performance of the microscale ZnO and P25 was studied to evaluate the separation efficiency of photogenerated electrons and holes. Under visible irradiation, the photocurrent of P25 in the solution of different dyes is quite low as shown in Figure 7A. Langmuir 2010, 26(1), 591–597
In contrast, the microscale ZnO exhibits much larger photocurrent than P25 in the bias potential range of -0.15 to þ0.7 V. As ZnO and TiO2 could not be excited by visible irradiation due to their large band gap, the observed photocurrent results from the electron transfer from visible-light-excited dye to the conduction band of ZnO or TiO2.14,15 The much larger photocurrent of the microscale ZnO than P25 indicates the higher efficiency of the electron transfer in ZnO than in TiO2, which results in the much higher visible photocataltyic activity of the microscale ZnO than P25. Under UV-visible irradiation, the electron in the valence band of ZnO and TiO2 can be excited to their corresponding conduction band, thus, the efficient separation of photoinduced electrons and holes leads to the generation of photocurrent. As can be seen from Figure 7B, the photocurrent of the microscale ZnO is much larger than that of P25, indicating a much higher separation efficiency of photoinduced electrons and holes in the microscale ZnO than in P25. Therefore, the higher efficiency of generation, mobility, and separation of photoinduced electrons and holes in the microscale ZnO than in P25 contributes to the much higher photocataltyic activity of the microscale ZnO than P25 under UV-visible irradiation. Photostability. To compare the photocatalytic performance between microscale and nano ZnO, nano ZnO was prepared by sol-gel method followed by calcination at 500 °C. The nano ZnO has a pure hexagonal wurtzite structure (Figure 1A). Its average crystal size and surface area are 33 nm and 32.5 m2 g-1, respectively. It shows absorption similar to the microscale ZnO in UV region as shown in Figure 1C. The nano ZnO exhibits photocatalytic activity similar to the microscale ZnO under UVvisible irradiation (Figure 2). However, its photocataltyic activity under visible irradiation is much lower than that of the microscale ZnO. The rate constant on the microscale ZnO is 3 times higher than that on the nano ZnO. The recycled experiments for the photodegradation of CV under UV-visible irradiation were performed to evaluate the photostability of the photocatalysts. It was reported that nano ZnO suffers from serious photocorrosion with a significant decrease of its photocatalytic activity.13 Similar phenomenon was observed in our case. As shown in Figure 8, 97.5% of CV is photodegraded when the nano ZnO is used for the first time. After four cycles, its photocatalytic conversion rapidly decreases to 7.0%. In contrast, when the microscale ZnO is recycled, it is observed that photocatalytic conversion decreases slightly from 99.8% to 86.3% after four cycles. Even after eight cycles, its photocatalytic conversion is as high as 46.7%. These results show that microscale ZnO exhibits much better photocorrosion inhibition and photostability than the nano ZnO. It is well-known that TiO2 is one of most photostable photocatalysts. It is expected that surface modification of ZnO by TiO2 DOI: 10.1021/la902117c
595
Article
Li et al.
Figure 10. Room temperature photoluminescence spectra with excitation at 320 nm for the photocatalysts.
Figure 9. Time course of the decrease in the concentration of CV with UV-visible irradiation (A) and visible (B) on the microscale ZnO modified with a different amount of TiO2.
probably will block the surface defects of ZnO, and thus improve the photostability of ZnO. With this idea in our mind, the effect of TiO2 modification on the photocatalytic performance of the microscale ZnO was investigated. The results are shown in Figure 9. Although the modification of ZnO with a small amount of TiO2 (0.02-0.1 wt %) leads to the decrease of its photocatalytic activity under both UV-visible and visible irradiation, the TiO2modified ZnO photocatalysts show much better photocatalytic activity than P25 under visible irradiation. The optimal modification amount of TiO2 is 0.05%, and its rate constant is 13.6 times higher than that of P25. As shown in Figure 8, 83.6% of CV is photodegraded when the photocatalyst is used for the first time within 120 min. After eight cycles, its photocatalytic activity remains unchanged, indicating that the photostability of the microscale ZnO could be greatly improved by the TiO2 surface modification. One of the major drawbacks of nanosized photocatalysts for their application in treating wastewater is the special difficulty in separation from a suspension due to their nanosized particle dimension, which in turn significantly increases the running cost, and sometimes even produces a secondary pollution. To solve the problem, much effort has been focused on forming a composite with magnetite,21 which usually leads to the decrease of their photocatalytic efficiency. The very good photocatalytic performance of the microscale ZnO and TiO2-modified ZnO together with their easy separation from a suspension by conventional separation technology will facilitate their application in wastewater treatment. Photoluminescence. It is well-known that the photoluminescent property of ZnO is sensitive to its defects.22 A strong UV emission at 390 nm and several relatively weak visible emissions in the range of 400-520 nm are observed for the microscale ZnO (Figure 10). The UV emission is attributed to free excitonic (21) Beydoun, D.; Amal, R.; Low, G. K. C.; Mcevoy, S. J. Phys. Chem. B 2000, 104, 4387. (22) (a) Sagar, P.; Shishodia, P. K.; Mehra, R. M.; Okada, H.; Wakahara, A.; Yoshida, A. J. Lumin. 2007, 126, 800. (b) Vanheusden, K.; Seager, C. H.; Warren, W. L.; Tallant, D. R.; Voigt, J. A. Appl. Phys. Lett. 1996, 68, 403. (c) Fujihara, S.; Ogawa, Y.; Kasai, A. Chem. Mater. 2004, 16, 2965.
596 DOI: 10.1021/la902117c
emission near band edge. The visible emissions are due to transition in defect states, in particular the oxygen vacancies.22 Surface modification of ZnO by 0.05%TiO2 leads to a significant decrease of both the UV and visible emissions. The results of XRD (Figure 1A) and Raman (Figure 1B) show that no detectable structural change is observed for the microscale ZnO, and no crystalline TiO2 (anatase or rutile) is detected after the surface modification, suggesting that the surface modification does not change the bulk intrinsic property of the microscale ZnO, and TiO2 is highly dispersed on the surface of the microscale ZnO. Therefore, the great decrease of the UV emission suggests that considerable photogenerated electrons and holes move to the surface of the microscale ZnO due to the high mobility of charge carriers in ZnO, and further transfer to the highly dispersed TiO2. The considerable decrease of the visible emissions indicates that the surface defects in the microscale ZnO are greatly reduced after the surface modification. The photocorrosion of ZnO consists of two slow steps where two holes are trapped on the surface of ZnO, followed by the fast formation of an oxygen molecule and the fast expulsion of Zn2þ from the surface, and the overall reaction may be represented as: ZnO þ 2hþ ;Zn2þ þ 0.5O2.23 The charge transfer from ZnO to the highly dispersed TiO2 on the surface of the TiO2-modified ZnO could inhibit the photocorrosion reaction. The great decrease of defects after the surface modification further improve the photostability of the microscale ZnO.18 For the nano ZnO, there are only observed strong broad green emissions in the range 400-520 nm, indicating a lot of defects in the nano ZnO. This is why the nano ZnO is less photoactive and photostable than the microscale ZnO. It should be acknowledged that higher photocatalytic efficiency would be expected if nanostructured ZnO with good crystallinity and low defects could be obtained. Device of Coupling Photodegradation with Light-to-Electricity Conversion. On the basis of the very good photocatalytic performance of the microscale ZnO and TiO2-modified ZnO, a novel device with two functions of light-to-electricity conversion and photodegradation was developed as schematically illustrated in Figure 11A. The device is comprised of a film of the photocatalyst on a conducting glass plate of ITO as working electrode (2.0 2.0 cm), and a platinum wire as counter electrode, which were put in 25 mL of 0.1 mol L-1 Na2SO4 as electrolyte and 4.0 10-5 mol L-1 CV in a quartz cell. This quartz cell was set on a magnetic stirrer. A 125 W high pressure Hg Lamp was used as light source. Under magnetic stirring and UV-visible irradiation, the evolution of photocurrent and dye concentration was recorded under a certain operation voltage. As shown in Figure 11, parts B and C, under dark, no detectable photodgradation is (23) (a) Rudd, A. L.; Bresli, C. B. Electrochim. Acta 2000, 45, 1571. (b) Spathis, P.; Poulio, I. Corros. Sci. 1995, 51, 673.
Langmuir 2010, 26(1), 591–597
Li et al.
Article
than that in the photocatalytic test (2.0 10-4 mmol/g-catal. min). In the case of 0.05%TiO2-ZnO, similar results are obtained with 31.3% photodegradation of CV and 0.19% of light-toelectricity conversion rate. Although the light-to-electricity is low, it is anticipated that a higher light-to-electricity conversion rate comparable to TiO2 or ZnO-based DSSC could be achieved by optimizing the preparation condition of ZnO-based working electrode and other factors, which are currently under progress. Our approach provided a novel strategy of combining the advantages of photocatalysis and DSSC, which is an ideal and promising candidate for the photocatalytic removal of dye pollutants and a renewable energy source.
Figure 11. Schematic illustration of a device coupling photodegradation with DSSC (A) and the time course of the photocurrent (B) and the decrease in the concentration of dye (C) with UVvisible irradiation under an operation voltage of 0.5 V.
observed, and the dark current is negligible. Under UV-visible irradiation, photocurrent is generated with the photodegradation of the dye. For the device with P25/ITO as working electrode, although 24.3% of CV is photodegraded after UV-visible irradiation for 240 min, the light-to-electricity conversion rate is quite low (0.0024%). In contrast, when the microscale ZnO/ITO is used as working electrode, 34.8% of CV is photodegraded, and a light-to-electricity conversion rate of 0.23% is achieved. The light-to-electricity conversion rate for the microscale ZnO is 95.8 times higher than that for P25. The lower photodegradation conversion in the device than in the above-mentioned photocatalytic test is due to the very low amount of the photocatalyst on the working electrode(0.0040 g). The specific photodegradation rate in the device is 3.6 10-4 mmol/g-catal.min, which is larger
Langmuir 2010, 26(1), 591–597
Conclusion In summary, the microscale ZnO exhibits much higher efficiency for the photodegradation of various dye pollutants and much higher light-to-electricity conversion rate than TiO2 under UV-vis irradiation due to the higher efficiency of generation, mobility, and separation of photoinduced electrons and holes in the microscale ZnO than in TiO2. The highly efficient photosensitization of ZnO by dye is an effective way to extend its photoresponse into the visible region, and can be used in the visible photocatalytic removal of dye pollutants. The much higher visible photocataltyic activity of the microscale ZnO than P25 is attributed to the higher photosensitization efficiency of the electron transfer from excited dye to the conduction band of the semiconductor in the microscale ZnO than in P25. The microscale ZnO exhibits much better photostability than nano ZnO due to its better crystallinity and lower defects. The photostability of the microscale ZnO can be greatly improved by surface modification with a small amount of TiO2. On the basis of the excellent photocatalytic performance of the microscale ZnO and TiO2modified ZnO, a novel device of coupling photodegradation with light-to-electricity conversion was developed. Currently, environmental pollution and energy shortage are main challenge in the world. We demonstrated that the purification of waste dye pollutants and generation of renewable energy source could be realized at the same time. Acknowledgment. This work was supported by National Basic Research Program of China (2009CB939704), Important Project of Ministry of Education of China (309021), Scientific Research Foundation for the Returned Overseas Chinese Scholars ([2008] 890), the Program for New Teacher from Ministry of Education (20070497003), and Nippon Sheet Glass Foundation. Supporting Information Available: Figures showing the chemical structures of the dyes and the time course of the decrease in the absorption of CV. This material is available free of charge via the Internet at http://pubs.acs.org.
DOI: 10.1021/la902117c
597
