催化学报  2014, Vol. 35 Issue (10): 1609-1618   PDF (806 KB)    
扩展功能
加入收藏夹
复制引文信息
加入引用管理器
Email Alert
RSS
本文作者相关文章
余长林
周晚琴
余济美
刘鸿
魏龙福
Design and fabrication of heterojunction photocatalysts for energy conversion and pollutant degradation
Changlin Yua , Wanqin Zhoua, Jimmy C. Yub, Hong Liuc , Longfu Weia    
a School of Metallurgy and Chemical Engineering, Jiangxi University of Science and Technology, Ganzhou 341000, Jiangxi, China;
b Department of Chemistry, The Chinese University of Hong Kong, Shatin, New Territories, Hong Kong, China;
c Chongqing Institute of Green and Intelligent Technology, Chinese Academy of Sciences, Chongqing 400714, China
Received: 2014-5-24. Accepted: 2014-6-5. Published: 2014-10-20.
* Corresponding author. Changlin Yu, Tel: +86-797-8312334; E-mail: yuchanglinjx@163.com; Hong Liu, Tel: +86-23-65935801; E-mail: liuhong@cigit.ac.cn
Foundation Item: This work was supported by the National Natural Science Foundation of China (21067004, 21263005), Young Science and Technology Project of Jiangxi Province Natural Science Foundation (20133BAB21003), Young Scientist Training Project of Jiangxi Province (20122BCB23015), the Implement Project of Science and Technology of Colleges and Universities in Jiangxi Province (KJLD14046), and One Hundred Talents Program of Chinese Academy of Sciences.
Abstract: Photocatalysis has attracted much attention for its promise in converting solar energy to chemical energy and in degrading various pollutants. Many recent investigations have demonstrated photocatalysts with well-defined junctions between two semiconductors with matched electronic band structures. Such structures effectively facilitate charge transfer and suppress recombination of photogenerated electrons and holes, leading to extremely high activity and stability. In this review, we focus on the influence of the heterojunction on the performance of semiconductor photocatalysts, including TiO2-based, ZnO-based, and Ag-based semiconductor photocatalysts. We also investigate fabrication methods for heterojunctions and attempt to understand the mechanisms behind photocatalysis. Finally, we propose challenges to design and clarify the mechanism for enhancing the effect of the heterojunction on photocatalyst performance.
Key words: Photocatalyst     Heterojunction     Semiconductor     Photocatalytic performance     Organic pollutant degradation     Hydrogen production    

1. Introduction

Photocatalysis has attracted much attention since Fujishima and Honda [1] demonstrated hydrogen evolution from water on an illuminated TiO2 electrode. Homogeneous and heterogeneous photocatalysts can convert solar energy into chemical energy as clean, renewable hydrogen fuel [2, 3, 4, 5, 6] or by reducing CO2 into solar fuels [7]. Photocatalysis is environmentally friendly especially because it can decompose pollutants into more benign end products, even CO2 and H2O [8, 9, 10, 11].

Heterogeneous photocatalysis begins with photons impinging on the photocatalyst. Photons with energy equal to or greater than the material’s band-gap energy will produce electron-hole pairs. These pairs can either recombine in the bulk to produce thermal energy or migrate to the surface and react with adsorbed species. The second process can produce reductive and oxidative reactions, which can extract hydrogen energy or nonselectively decompose pollutant molecules. Various photocatalytic materials have been fabricated that, irradiated by UV or visible light, can produce hydrogen and/or oxygen and decompose numerous organic substrates, including phenol, dyes, chlorinated compounds, aromatics, alkenes, and alkanes.

For wide-band-gap semiconductor photocatalysts such as TiO2 and ZnO, many approaches have been developed to improve their photocatalytic kinetics by either enhancing the separation efficiency of electron-hole pairs or by extending their range of excitation wavelengths from UV to visible. Common ways to extend this range for TiO2 include coupling a narrow-band-gap semiconductor [12, 13], adding transition metals [14, 15], and doping with nonmetals [16, 17]. Alternative narrow-band-gap photocatalysts have also been rapidly developed, such as Au nanoclusters (Au25(SR)18/TiO2) [18], TiO2-xNx [19], Rh(Pd,Pt)/BiOX (Cl, Br, I) [20], g-C3N4 [21], plasmonic noble metal (e.g., Au, Ag) nanoparticles [22], and Ag3PO4 [23]. These photocatalysts are very active under visible-light irradiation.

However, these photocatalysts still have energy conversion efficiencies too low for practical applications. This is mainly caused by the low efficiency of the three crucial steps of heterogeneous photocatalysis: light harvesting, charge separation and transportation, and the catalytic reduction and oxidation reactions. The optical properties and catalytic performance of semiconductor photocatalysts strongly depend on their nano- or microstructures. Forming a semiconductor interface or heterojunction can enhance the separation of photogenerated electron-hole pairs, increasing activity and stability [24]. Specifically, forming a phase junction between anatase and rutile NPs can greatly enhance their photocatalytic activity for H2 production [25]. Forming Ag2O/Ag2CO3 on the surface of Ag2CO3 increases its activity in decomposing methyl orange (MO) dye by 68 and 30 times over Ag2CO3 and Ag2O, respectively [26]. Therefore, tuning the physical properties (e.g., crystal structure, crystallinity, particle size) of photocatalysts can improve their charge separation and migration.

This review focuses on high-efficiency semiconductor- heterojunction photocatalysts (e.g., TiO2, ZnO, CdS, Ag-based semiconductors) and their applications in organic-pollutant degradation and H2 production. We describe the fabrication methods for these heterojunctions, the challenges in clarifying the mechanism by which the heterojunctions enhance photocatalytic performance, and the current understanding of photocatalysis.

2. The role of heterojunction in enhancing separation efficiency of electron-hole pairs

Photocatalyst performance mainly depends on how efficiently it separates photogenerated electrons and holes. Photocatalytic activity is traditionally improved by depositing noble metals (e.g., Pt, Ag, Pd, Au) or non-metal anions, or by doping with metal cations [27, 28, 29, 30, 31, 32, 33]. In a noble-metal composite system, photogenerated electrons accumulate on the metal, and holes remain on the photocatalyst surface, suppressing the recombination of charge carriers. Doping with non-metal anions (e.g., C, N, S, F) [34, 35, 36, 37] could also slow the recombination of photogenerated electrons and holes or extend the light absorption of titania into the visible region.

The formation of a well-defined heterojunction between two semiconductors with matching electronic band structures can also suppress the recombination of photogenerated electrons and holes [24, 25, 26, 38]. Fig. 1 illustrates how p-n heterojunctions enhance the separation efficiency of electron-hole pairs. The p-n heterojunction is a junction between two semiconductors, one doped with a donor (n-type) and one with an acceptor (p-type). A strong local electric field exists near the junction, pointing from n toward p, because of the juxtaposition of high concentrations of negatively and positively charged ions. The difference of the electric potential in the electric field can enhance the separation of photogenerated electrons and holes, increasing the quantum efficiency of the photocatalytic reactions.

Fig. 1. Schematic showing the role of the p-n heterojunction in enhancing the separation of electron-hole pairs.

Other types of heterojunctions can be used for photocatalysis. For example, heterojunctions between twosemiconductors with matching electronic band structures could similarly enhance the separation of photogenerated electron-hole pairs. Thus, engineering the junction between semiconductors is essential for improving photocatalytic activity.

3. Typical heterojunction photocatalysts
3.1. TiO2-based heterojunction photocatalysts

As a photocatalyst, TiO2 has some obvious advantages, including its high oxidative power for organic pollutants, photo-stability, low cost, and nontoxicity [39, 40, 41, 42, 43]. The photocatalytic performance of TiO2 depends much on its phase composition. The common phases of TiO2 are anatase (A) and rutile (R). Zhang et al. [25] reported that forming an A/R heterojunction over the surface of TiO2 can remarkably enhance its photocatalytic H2 production. To produce the A/R heterostructure, different amounts of A-TiO2 are loaded on the surface of R-TiO2 by the impregnation method. Fig. 2 shows high-resolution transmission electron microscopy (HR-TEM) images of a surface A/R junction formed on R-TiO2. The phase junction formed between the surface anatase and rutile particles enhances the photocatalytic activity for H2 production over that of pure R-TiO2.

Fig. 2. HRTEM image of an A-TiO2/R-TiO2 sample. Reproduced with permission from reference [25]. Copyright 2008, Wiley.

Xiang et al. [44] found that a TiO2/MoS2/graphene heterojunction composite with 0.5 wt% MoS2/graphene cocatalyst has much improved H2 production, caused by the synergy between the MoS2 and graphene components in this hybrid cocatalyst.

Very recently, Qiu et al. [45] reported a simple one-step hydrothermal method toward in situ growth of mesoporous TiO2 nanocrystals with (001) facets on 3D-graphene aerogels (GAs). The TiO2/GAs they obtained exhibited highly cyclable photocatalytic activity for degradation of methyl orange. The strong interaction between TiO2 and GAs produced highly active photocatalysis, high rate capability, and stable cycling.

A major drawback of TiO2 is its large band-gap of 3.2 eV; it is excited by only ~2%-3% of UV in the solar spectrum, making wavelengths under 400 nm indispensable for excitation. This fact considerably limits its use with natural solar light or artificial visible light. Forming a heterojunction of TiO2 with a narrow-gap semiconductor could improve its visible-light activity. Using an impregnating-hydroxylation method, Dai et al. [46] fabricated p-n junction BiOI/TiO2 nanotube arrays (NTs) by coating the walls of the TiO2 NTs with BiOI. Degrading MO under visible-light irradiation using photocatalysis, they found that BiOI/TiO2 exhibited a rate constant over three times that of BiOI. Fig. 3 shows the schematic of charge transfer at the BiOI/TiO2 interface for photoelectrocatalytic degradation of MO. BiOI is a p-type semiconductor with a narrow band gap (~2.1 eV), and anatase TiO2 is an n-type semiconductor with a large band gap (~3.2 eV). The formation of the p-n junction caused the position of the BiOI conduction band to be more negative than that of TiO2, letting electrons excited to the conduction band of p-type BiOI migrate to the conduction band of n-type TiO2. Under an external electrostatic field, these electrons travel along the TiO2 NTs and pass through the interface to the external circuit, leaving the photogenerated holes in the valence band of BiOI. This process separates the photogenerated electron-hole pairs. The separated electrons and holes then react with adsorbates on the electrode surface, enhancing photoelectrocatalytic activity.

Fig. 3. Energy band diagram of p-n junction at equilibrium and transfer of photoinduced electrons from p-BiOI to n-TiO2 under visible-light irradiation. Reproduced with permission from reference [46]. Copyright 2008, Wiley.

We have shown that forming an Ag2CO3/TiO2 heterojunction can greatly enhance UV and visible-light activity [47]. The degradation activity of 1% Ag2CO3/TiO2 irradiated by UV or visible light is 6 times that of TiO2 and 4 times that of Ag2CO3 or 3.4 times that of TiO2 and 1.7 times that of Ag2CO3. The Ag2CO3/TiO2 heterojunction promotes visible-light absorption and suppresses recombination electron-hole pairs, as shown in Fig. 4. Also, the additional surface ·OH groups in the Ag2CO3/TiO2 composite can react with photogenerated h+ and produce ·OH radicals, which decompose the dye and improve the performance of Ag2CO3/TiO2 compared with Ag2CO3.

Fig. 4. Proposed mechanism for the enhanced activity of the Ag2CO3/TiO2 heterojunction photocatalyst. Reproduced with permission from reference [47]. Copyright 2014, ACS.

Table 1 lists TiO2 heterojunction photocatalysts and their photocatalytic properties compared with reference photocatalysts for many applications, including degradation of aqueous and air pollutants. These comparisons show that a heterojunction of TiO2 and another photocatalyst has much better photocatalytic performance than either constituent alone.

Table 1
Photocatalytic properties of TiO2-based heterojunction photocatalysts.
3.2. Ag-based heterojunction photocatalysts

Ag-based semiconductor photocatalysts have recently become interesting because of their strong visible-light absorption. Yi et al. [53] reported an Ag3PO4 photocatalyst with extremely high photo-oxidative capability for water splitting and organic dye decomposition under visible-light irradiation. Unfortunately, this photocatalyst is photochemically unstable because it readily decomposes photochemically when no sacrificial reagent is present [54, 55]. Indeed, during photocatalysis, corrosion seems an insurmountable problem for Ag-based photocatalysts (e.g., Ag3PO4, AgI, Ag3AsO4), though they exhibit high activities [56, 57].

Therefore, a challenge in Ag-based visible-light photocatalysts is promoting charge separation and photocatalytic stability. Forming a heterojunction between an Ag-based semiconductor and another semiconductor can enhance separation of photogenerated electrons and holes, as well as improve stability [58]. For example, Wang et al. [59] reported that an Ag3PO4/AgBr/Ag photocatalyst exhibited enhanced photocatalytic activity and stability. Following is a typical example for designing a heterojunction to enhance the photocatalytic performance in degradation dyes or phenol [26].

Ag2O/Ag2CO3 heterostructures have been successfully synthetized via facile phase transformation. Fig. 5(a) illustrates the formation of an Ag2O/Ag2CO3 heterostructure during calcination. In this system, Ag2CO3 particles are covered by a surface layer of Ag2O, producing a core-shell-like Ag2O/Ag2CO3 heterostructure with a well-contacted interface. The valence band (VB) edge positions of Ag2CO3 and Ag2Oare calculated using electronegativity, and the band gaps of Ag2CO3 and Ag2O are 2.5 and 1.3 eV, respectively. The calculated VB and CB edges of Ag2CO3 are 2.77 and 0.27 eV, respectively, and the estimated VB and CB of Ag2O are 1.5 and 0.2 eV, respectively. Fig. 5(b) shows detailed electronic band structures of Ag2CO3 and Ag2O.

Fig. 5. (a) Schematic of the phase transformation Ag2CO3 → Ag2O/Ag2CO3 → Ag2O; (b) Electronic band structures of Ag2CO3 and Ag2O.

In this Ag2O/Ag2CO3 heterostructure, Ag2O has a more negative potential of the conduction band (CB: 0.2 eV) and valence band (VB: 1.5 eV) than those of Ag2CO3 (CB: 0.27 eV, VB: 2.77 eV). Under visible-light irradiation, the photogenerated electrons in the Ag2O can quickly transfer to the Ag2CO3 surface. Simultaneously, the photoinduced holes on the Ag2CO3 surface migrate to the Ag2O surface, increasing the separation of photo-excited electrons and holes, suppressing electron-hole recombination. Ag2O/Ag2CO3 exhibited an MO photocatalytic decomposition rate of 67 and 31 times those of Ag2CO3 and Ag2O, respectively (see Table 2). Moreover, during their second cycle Ag2CO3 and Ag2O almost completely lose their activity from photocorrosion caused by the formation of metallic silver. Even after six successive cycles, Ag2O/Ag2CO3 still exhibited a 94% MO degradation rate after 40 min of visible-light irradiation, indicating its high stability and great promise in practical applications. Table 3 summarizes the catalytic properties of other typical Ag-based heterojunction photocatalysts.

Table 2
Textural properties, band gaps, and photocatalysis rate constants in photocatalysts for degradation of MO and phenol.

Table 3
Photocatalytic properties of Ag-based heterojunction photocatalysts.
3.3. ZnO-based heterojunction photocatalysts

ZnO is advantageous because it is abundant, non-toxic, and inexpensive [65, 66, 67, 68, 69, 70, 71]. However, ZnO has lower photocatalytic activity than commercial TiO2 photocatalysts. ZnO exhibits a high recombination rate of photogenerated electron-hole pairs and has lower stability, caused by photochemistry corrosion. This photocorrosion quickly decreases the photocatalytic activity of the catalyst, making it difficult to recycle. Moreover, photocorrosion of ZnO releases Zn2+, a pollutant, greatly limiting its use in environmental purification.

Forming a ZnO heterojunction such as WO3/ZnO [71], Bi2WO6/ZnO [72], BiOCl/ZnO [73], or MO3/ZnO [74] can improve activity and stability. Fig. 6 shows how forming the SnO2/ZnO heterojunction typically enhances these properties [75]. When SnO2 and ZnO form a heterojunction, the difference in their work functions transfers negatively charged carriers from SnO2 to ZnO until their Fermi levels align, creating an interfacial electrostatic field. At thermal equilibrium, the CB and VB of SnO2 and ZnO bend, forming a depletion layer. Under UV-light irradiation, electrons in the VB can be excited to the CB while holes are generated in the VB. These photogenerated electrons and holes are separated by the electrostatic field induced by the differing work functions. The photoinduced electrons can be easily trapped by electronic acceptors such as adsorbed O2, producing super-oxide anion radicals (·O2-). The photoinduced holes can be easily trapped by OH, producing ·OH species. Both ·O2- and ·OH can partially or completely mineralize organic molecules. Under 40 min of UV-light irradiation, SnO2, ZnO, and SnO2/ZnO exhibited MO degradation rates of 20%, 65%, and 100%, respectively.

Fig. 6. Energy band diagram and photocatalytic mechanism of an as-synthesized SnO2/ZnO heterojunction nanocatalyst. vac, vacuum level; Ef, Fermi level; CB, conduction band; VB, valence band. Reproduced with permission from reference [75]. Copyright 2009, ACS.
3.4. Other heterojunction photocatalysts

In other heterojunction systems, one semiconductor acts as the main photocatalyst, and the other acts as a cocatalyst [76, 77]. For example, in intimately joined MoS2/CdS heterojunctions, MoS2 acts as the cocatalyst and shows much higher photocatalytic H2 evolution activity than Pt/CdS in lactic acid. Even though Pt performs better than MoS2 for activating H2 in electrochemical systems, the lactic acid acts as the sacrificial electron donor. Forming atomically well-defined junctions between CdS and PdS might facilitate charge transfer. For example, chemically deposited MoS2/CdS exhibits much higher activity than mixed CdS and MoS2. Results from other heterojunction systems, such as Bi5O7I/BiOI [78], Bi24O31Br/BiOBr [79], and CuO/BiVO4 [80], further confirmed this conclusion. Table 4 summarizes other typical heterojunction photocatalysts.

Table 4
Photocatalytic properties of heterojunction photocatalysts.
4. Fabricating heterojunction photocatalysts

The performance of a photocatalyst depends much on its fabrication method. Heterojunction photocatalysts can be fabricated by many methods, including phase transformation [26, 78, 79], impregnation-deposition[25, 86], dip-coating [48], liquid-phase deposition [89], high-temperature solid-state reaction [81], hydrothermal or solvothermal methods [84, 85], electro-deposition [90], cation exchange [91], and sol-gel methods [49]. A simple method is calcination-induced phase transformation, which has been used to fabricate photocatalysts including Ag2O/Ag2CO3 [26], α-Bi2O3/Bi2O2CO3, Bi5O7I/BiOI [78], Bi5O7I/Bi2O3 [78], and Bi24O31Br/BiOBr [79]. In this process, a thermally unstable semiconductor precursor is decomposed by heat treatment. The relative content of the two phases in the heterojunction can be controlled by changing the calcination temperature and time. Here, we give an example of forming Bi5O7I/BiOI and Bi5O7I/Bi2O3 heterojunctions using this method. Using thermogravimetric analysis, we can understand the stability and transformations of BiOI under heat treatment in air, as shown in Fig. 7. From 350 to 520 °C, BiOI began to decompose corresponding to the reaction: 5BiOI + O2 → Bi5O7I + 2I2. The total weight decrease was ~27.00%. From 600 to 850 °C, the total decrease in weight was ~9.16%, corresponding to 2Bi5O7I + 1/2O2 → 5Bi2O3 + I2. Thus, at calcination temperatures of 350-520 °C, a Bi5O7I/BiOI heterojunction can be produced. Similarly, a Bi5O7I/Bi2O3 heterojunction can be produced at a calcination temperature of ~620°C.

Fig. 7. TGA profile of BiOI powders in air flow rate of 30 mL/min.
5. Conclusions

To use photocatalysts for large-scale solar conversion and environmental purification, their photocatalytic efficiency must be improved. Engineering junctions between light-harvesting semiconductors will surely improve catalytic activity. Forming well-defined junctions between semiconductors with matching electronic band structures can effectively facilitate charge transfer and suppress recombination of photogenerated electrons and holes, resulting in extremely high activity and stability. Further work should clarify the mechanism by which heterojunctions enhance performance, as it remains unclear. New methods are expected to quantify the individual semiconductor phase over the surface of photocatalysts.

References
[1] Fujishima A, Honda K. Nature, 1972, 238: 37
[2] Cui W Q, Liu Y F, Liu L, Hu J S, Liang Y H. Appl Catal A, 2012, 417-418: 111
[3] Jing D W, Jing L, Liu H, Yao S, Guo L J. Ind Eng Chem Res, 2013, 52: 1992
[4] Ahmed A Y, Kandiel T A, Oekermann T, Bahnemann D. J Phys Chem Lett, 2011, 2: 2461
[5] Zhang J Y, Wang Y H, Zhang J, Lin Z, Huang F, Yu J G. ACS Appl Mater Interfaces, 2013, 5: 1031
[6] Cui W Q, Ma S S, Liu L, Liang Y H. Chem Eng J, 2012, 204-206: 1
[7] Wang Y, Yu J G, Xiao W, Li Q. J Mater Chem A, 2014, 2: 3847
[8] Yu C L, Yang K, Xie Y, Fan Q Z, Yu J C, Shu Q, Wang C Y. Nanoscale, 2013, 5: 2142
[9] Yu C L, Cao F F, Li X, Li G, Xie Y, Yu J C, Shu Q, Fan Q Z, Chen J C. Chem Eng J, 2013, 219: 86
[10] Zhou W Q, Yu C L, Fan Q Z, Wei L F, Chen J C, Yu J C. Chin J Catal (周晚琴, 余长林, 樊启哲, 魏龙福, 陈建钗, Yu J C. 催化学报), 2013, 34: 1250
[11] Yu C L, Chen J C, Cao F F, Li X, Fan Q Z, Yu J C, Wei L F. Chin J Catal (余长林, 陈建钗, 操芳芳, 李鑫, 樊启哲, Yu J C, 魏龙福. 催化学报), 2013, 34: 385
[12] Li K, Chai B, Peng T Y, Mao J, Zan L. ACS Catal, 2013, 3: 170
[13] Grötzel M. J Photochem Photobiol C, 2003, 4: 145
[14] Gong X Q, Selloni A, Dulub O, Jacobson P, Diebold U. J Am Chem Soc, 2008, 130: 370
[15] Rodrigues S, Ranjit K T, Uma S, Martyanov I N, Klabunde K J. Adv Mater, 2005, 17: 2467
[16] Liu G, Zhao Y N, Sun C H, Li F, Lu G Q, Cheng H M. Angew Chem Int Ed, 2008, 47: 4516
[17] Chen X F, Wang X C, Hou Y D, Huang J H, Wu L, Fu X Z. J Catal, 2008, 255: 59
[18] Yu C L, Li G, Kumar S, Kawasaki H, Jin R C. J Phys Chem Lett, 2013, 4: 2847
[19] Asahi R, Morikawa T, Ohwaki T, Aoki K, Taga Y. Science, 2001, 293: 269
[20] Yu C L, Cao F F, Li G, Wei R F, Yu J C, Jin R C, Fan Q Z, Wang C Y. Sep Purif Technol, 2013, 120: 110
[21] Wang X C, Maeda K, Thomas A, Takanabe K, Xin G, Carlsson J M, Domen K, Antonietti M. Nat Mater, 2009, 8: 76
[22] Zhou X M, Liu G, Yu J G, Fan W H. J Mater Chem, 2012, 22: 21337
[23] Bi Y P, Hu H Y, Ouyang S X, Lu G X, Cao J Y, Ye J H. Chem Commun, 2012, 48: 3748
[24] Heremans P, Cheyns D, Rand B P. Acc Chem Res, 2009, 42: 1740
[25] Zhang J, Xu Q, Feng Z C, Li M J, Li C. Angew Chem Int Ed, 2008, 47: 1766
[26] Yu C L, Li G, Kumar S, Yang K, Jin R C. Adv Mater, 2014, 26: 892
[27] Su R, Tiruvalam R, Logsdail A J, He Q, Downing C A, Jensen M T, Dimitratos N, Kesavan L, Wells P P, Bechstein R, Jensen H H, Wendt S, Catlow C R A, Kiely C J, Hutchings G J, Besenbacher F. ACS Nano, 2014, 8: 3490
[28] Yu C L, Wei L F, Li X, Chen J C, Fan Q Z, Yu J C. Mater Sci Eng B, 2013, 178: 344
[29] Yu C L, Yang K, Zhou W Q, Fan Q Z, Wei L F, Yu J C. J Phys Chem Solids, 2013, 74: 1714
[30] Yu C L, Fan C F, Meng X J, Yang K, Cao F F, Li X. React Kinet Catal Lett, 2011, 103: 141
[31] Yu C L, Yu J C, Fan C F, Wen H R, Hu S J. Mater Sci Eng B, 2010, 166: 213
[32] Xing M Y, Yang B Y, Yu H, Tian B Z, Bagwasi S, Zhang J L, Gong X Q. J Phys Chem Lett, 2013, 4: 3910
[33] Yu J G, Xiong J F, Cheng B, Liu S W. Appl Catal B, 2005, 60: 211
[34] Yu C L, Fan Q Z, Xie Y, Chen J C, Shu Q, Yu J C. J Hazard Mater, 2012, 237-238: 38
[35] Barolo G, Livraghi S, Chiesa M, Paganini M C, Giamello E. J Phys Chem C, 2012, 116: 20887
[36] Yu C L, Cai D J, Yang K, Yu J C, Zhou Y, Fan C F. J Phys Chem Solids, 2010, 71: 1337
[37] Yu C L, Yu J C. Catal Lett, 2009, 129: 462
[38] Yu J G, Wang Y, Xiao W. J Mater Chem A, 2013, 1: 10727
[39] Kumar N, Maitra U, Hegde V I, Waghmare U V, Sundaresan A, Rao C N R. Inorg Chem, 2013, 52: 10512
[40] Boppana V B R, Lobo R F. J Catal, 2011, 281: 156
[41] Yu C L, Yu J C, Zhou W Q, Yang K. Catal Lett, 2010, 140: 172
[42] Xing M Y, Qi D Y, Zhang J L, Chen F, Tian B Z, Bagwas S, Anpo M. J Catal, 2012, 294: 37
[43] Xing M Y, Zhang J L, Chen F, Tian B Z. Chem Commun, 2011, 47: 4947
[44] Xiang Q J, Yu J G, Jaroniec M. J Am Chem Soc, 2012, 134: 6575
[45] Qiu B C, Xing M Y, Zhang J L. J Am Chem Soc, 2014, 136: 5852
[46] Dai G P, Yu J G, Liu G. J Phys Chem C, 2011, 115: 7339
[47] Yu C L, Wei L F, Chen J C, Xie Y, Zhou W Q, Fan Q Z. Ind Eng Chem Res, 2014, 53: 5759
[48] Xu Q C, Wellia D V, Ng Y H, Amal R, Tan T T Y. J Phys Chem C, 2011, 115: 7419
[49] Huang H J, Li D Z, Lin Q, Zhang W J, Shao Y, Chen Y B, Sun M, Fu X Z. Environ Sci Technol, 2009, 43: 4164
[50] Zhang J Y, Zhu H L, Zheng S K, Pan F, Wang T M. ACS Appl Mater Interfaces, 2009, 1: 2111
[51] Kim Y J, Gao B F, Han S Y, Jung M H, Chakraborty A K, Ko T, Lee C, Lee W I. J Phys Chem C, 2009, 113: 19179
[52] Sun M, Chen G D, Zhang Y K, Wei Q, Ma Z M, Du B. Ind Eng Chem Res, 2012, 51: 2897
[53] Yi Z G, Ye J H, Kikugawa N, Kako T, Ouyang S X, Stuart-Williams H, Yang H, Cao J Y, Luo W J, Li Z S, Liu Y, Wither R L. Nat Mater, 2010, 9: 559
[54] Bi Y P, Ouyang S X, Umezawa N, Cao J Y, Ye J H. J Am Chem Soc, 2011, 133: 6490
[55] Bi Y P, Ouyang S X, Cao J Y, Ye J H. Phys Chem Chem Phys, 2011, 13: 10071
[56] Yang X F, Cui H Y, Li Y, Qin J L, Zhang R X, Tang H. ACS Catal, 2013, 3: 363
[57] Tang J T, Liu Y H, Li H Z, Tan Z, Li D T. Chem Commun, 2013, 49: 5498
[58] Yang J H, Wang D E, Han H X, Li C. Acc Chem Res, 2013, 46: 1900
[59] Wang W S, Du H, Wang R X, Wen T, Xu A W. Nanoscale, 2013, 5: 3315
[60] Zhu L, Wei B, Xu L L, Lü Z, Zhang H L, Gao H, Che J X. CrystEngComm, 2012, 14: 5705
[61] Xu H, Xu Y G, Li H M, Xia J X, Xiong J, Yin S, Huang C J, Wan H L. Dalton Trans, 2012, 41: 3387
[62] Zhou W J, Liu H, Wang J Y, Liu D, Du G J, Cui J J. ACS Appl Mater Interfaces, 2010, 2: 2385
[63] Yao W F, Zhang B, Huang C P, Ma C, Song X L, Xu Q J. J Mater Chem, 2012, 22: 4050
[64] Shen K, Gondal M A, Siddique R G, Shi S, Wang S Q, Sun J B, Xu Q Y. Chin J Catal (沈凯, Gondal M A, Siddigue R G, 施珊, 王斯琦, 孙江波, 徐庆宇. 催化学报), 2014, 35: 78
[65] He W W, Kim H K, Wamer W G, Melka D, Callahan J H, Yin J J. J Am Chem Soc, 2014, 136: 750
[66] Mclaren A, Valdes-Solis T, Li G Q, Tsang S C. J Am Chem Soc, 2009, 131: 12540
[67] Li P, Wei Z, Wu T, Peng Q, Li Y D. J Am Chem Soc, 2011, 133: 5660
[68] Chu D, Masuda Y, Ohji T, Kato K. Langmuir, 2010, 26: 2811
[69] Yu C L, Yang K, Yu J C, Peng P, Cao F F, Li X, Zhou X C. Acta Phys-Chim Sin (物理化学学报), 2011, 27: 505
[70] Lai Y L, Meng M, Yu Y F, Wang X T, Ding T. Appl Catal B, 2011, 105: 335
[71] Yu C L, Yang K, Shu Q, Yu J C, Cao F F, Li X. Chin J Catal (余长林, 杨凯, 舒庆, Yu J C, 操芳芳, 李鑫. 催化学报), 2011, 32: 555
[72] Yu L L, Yang K, Yu J C, Cao F F, Li X, Zhou X Z. J Inorg Mater (余长林, 杨凯, Yu J C, 操芳芳, 李鑫, 周晓春. 无机材料学报), 2011, 26: 1157
[73] Yang K, Yu C L, Zhang L N, Yu J C. J Synth Cryst (杨凯, 余长林, 张丽娜, 余济美. 人工晶体学报), 2012, 41: 171
[74] Yu C L, Yang K, Shu Q, Yu J C, Cao F F, Li X, Zhou X C. Sci China Chem, 2012, 55: 1802
[75] Zheng L R, Zheng Y H, Chen C Q, Zhan Y Y, Lin X Y, Zheng Q, Wei K M, Zhu J F. Inorg Chem, 2009, 48: 1819
[76] Zong X, Yan H J, Wu G P, Ma G J, Wen F Y, Wang L, Li C. J Am Chem Soc, 2008, 130: 7176
[77] Yang J H, Wang D E, Han H X, Li C. Acc Chem Res, 2013, 46: 1900
[78] Yu C L, Fan C F, Yu J C, Zhou W Q, Yang K. Mater Res Bull, 2011, 46: 140
[79] Yu C L, Zhou W Q, Yu J C, Cao F F, Li X. Chin J Chem, 2012, 30: 721
[80] Yu C L, Yang K, Yu J C, Cao F F, Li X, Zhou X Z. J Alloys Compd, 2011, 509: 4547
[81] Lin X P, Xing J C, Wang W D, Shan Z C, Xu F F, Huang F Q. J Phys Chem C, 2007, 111: 18288
[82] Xie T P, Liu C L, Xu L J, Yang J, Zhou W. J Phys Chem C, 2013, 117: 24601
[83] Shenawi-Khalil S, Uvarov V, Fronton S, Popov I, Sasson Y. J Phys Chem C, 2012, 116: 11004
[84] He Z Q, Shi Y Q, Gao C, Wen L N, Chen J M, Song S A. J Phys Chem C, 2014, 118: 389
[85] Chang C, Zhu L Y, Wang S F, Chu X L, Yue L F. ACS Appl Mater Interfaces, 2014, 6: 5083
[86] Reddy K H, Martha S, Parida K M. Inorg Chem, 2013, 52: 6390
[87] Xu L L, Ni L, Shi W D, Guan J G. Chin J Catal (许蕾蕾,倪磊,施伟东,官建国.催化学报), 2012, 33: 1101
[88] Yu J G, Jin J, Cheng B, Jaroniec M. J Mater Chem A, 2014, 2: 3407
[89] Chai S N, Zhao G H, Zhang Y N, Wang Y J, Nong F Q, Li M F, Li D M. Environ Sci Technol, 2012, 46: 10182
[90] Yang L X, Luo S L, Li Y, Xiao Y, Kang Q, Cai Q Y. Environ Sci Technol, 2010, 44: 7641
[91] Zhang J, Yu J G, Zhang Y M, Li Q, Gong J R. Nano Lett, 2011, 11: 4774
设计和制备能量转换和环境净化的高效异质结光催化剂
余长林a , 周晚琴a, 余济美b, 刘鸿c , 魏龙福a    
a 江西理工大学冶金与化学工程学院, 江西 赣州 341000;
b 香港中文大学化学系, 香港 新界 沙田;
c 中国科学院重庆绿色智能技术研究院, 重庆 400714
收稿日期: 2014-5-24. 接受日期: 2014-6-5. 出版日期: 2014-10-20.
基金来源：国家自然科学基金（21067004，21263005）；江西省自然科学基金青年科学基金计划（20133BAB21003）；江西省青年科学家培养项目（20122BCB23015）；江西省高等学校科技落地计划（KJLD14046）；中国科学院百人计划.
*通讯联系人.余长林, 电话: (0797)8312334; 电子信箱: yuchanglinjx@163.com; 刘鸿, 电话: (023)65935801; 电子信箱: liuhong@cigit.ac.cn
摘要：在过去的几十年中，光催化由于具有将太阳能转化为清洁氢化学能和降解各种污染物的广泛应用前景，因而引起了人们广泛关注. 近期，很多研究表明，两个具有相匹配电子能级结构的半导体形成接触良好的异质结，可以有效地促进电荷转移和抑制光生电子（e-）和空穴（h+）的复合，从而显著提高光催化剂的活性和稳定性. 本文主要讨论了异质结对半导体光催化剂的促进作用；分析了异质结对一些典型光催化剂如TiO2，ZnO和Ag基半导体等光催化性能的影响；讨论了异质结光催化剂的制备方法和对光催化过程影响的基本机理；最后，提出了设计和理解异质结促进光催化反应机理所面临的挑战.
关键词： 光催化剂     异质结     半导体     光催化性能     有机污染物降解     制氢    

1. 前言

自从Fujishima等[1]发现TiO2电极在紫外光照射下可以分解水产氢以来, 光催化引起了人们的广泛关注.  这一发现使人们可以利用均相或多相光催化反应将太阳能以清洁氢能方式转化为化学能[2, 3, 4, 5, 6], 同时可以将CO2还原成太阳燃料[7].  光催化还可以将有机污染物完全降解矿化成CO2和H2O[8, 9, 10, 11].  因此, 光催化也被认为是未来在处理环境问题上可以取得新突破的环境友好方法.  

多相光催化反应的第一步是, 当照射光的光能大于或者等于光催化剂的带隙能时, 光催化剂价带中的电子被激发到导带, 产生光生电子(e-)与空穴(h+)对.  光生电子和空穴可以在体相中复合以热能形式消耗掉, 也可以转移至催化剂表面, 与表面吸附的物质发生一系列还原和氧化反应.  这些反应可以用来产氢, 也可以无选择性地分解有机污染物分子.  目前, 光催化已经取得了很大的进展, 已经研制出各种光催化材料.  在紫外光或者可见光照射下, 这些光催化材料能催化产氢或氧, 也能降解各种有机反应底物, 如苯酚、染料、氯化物、芳烃、烯烃和烷烃等.  

对于TiO2和ZnO等典型宽带隙半导体光催化材料的研究, 主要考虑的是如何提高光生e--h+对的分离效率, 提高光催化反应的量子效率, 或者延长其光谱响应范围(从紫外光至可见光).  通常采用的方法有窄带隙半导体耦合[12, 13]、过渡金属的添加[14, 15]和非金属掺杂[16, 17]等.  与此同时, 各种新型窄带隙光催化剂也被开发出来, 如Au纳米簇(Au25(SR)18/TiO2)[18], TiO2-xNx[19], Rh(Pd,Pt)/ BiOX(Cl, Br, I)[20], g-C3N4[21], 等离子贵金属(Au, Ag)纳米颗粒[22]和Ag3PO4[23]等.  这些 窄带隙的光催化剂在可见光照射下表现出明显的光催化活性.  

但是, 光催化反应总的能量转化效率很低, 还远未达到实际应用的要求.  这主要是由于光催化反应的三个关键步骤(光捕获、电荷分离和催化还原氧化反应)效率不高或不够协同.  半导体光催化剂的光吸收和催化性能强烈依赖于它们的纳米/微米结构.  研究表明, 两个半导体之间形成界面或者异质结可以明显提高光生电子与空穴的分离, 从而提高光催化反应活性和稳定性[24].  例如锐钛矿相TiO2和金红石相纳米粒子之间形成的界面可大幅度提高其光催化产氢活性[25].  在可见光光分解甲基橙(MO)中, Ag2CO3表面上形成Ag2O/Ag2CO3异质结使其活性提高了68倍(相对于Ag2CO3)和30倍(相对于Ag2O)[26].&# 8197; 因此, 可以通过调控光催化剂的物理结构(相组成、相界面和颗粒大小等)提高光生电子与空穴的分离效率.  

本文综述了近几年所报道的典型异质结型光催化剂在降解有机污染物、光催化产氢和还原CO2方面的应用研究进展, 讨论了异质结对光催化剂的促进作用和TiO2, ZnO, CdS和Ag基等半导体异质结催化剂的制备方法, 并提出了目前设计高效异质结光催化剂和理解异质结提高光催化反应机理所面临的一些挑战.  

2. 异质结提高光生e--h+分离作用

光催化剂的催化性能依赖于光生电子与空穴的分离效率.  为了提高光催化剂活性, 传统制备方法采用贵金属沉积、非金属阴离子或者金属阳离子掺杂等.  Pt, Ag, Pd和Au等贵金属通常用来沉积在光催化剂上, 减小光生电子与空穴的复合机率[27, 28, 29, 30, 31, 32, 33].  在贵金属/半导体复合体中, 贵金属颗粒可以捕获光生电子, 抑制光生电子与空穴的复合.  掺杂非金属阴离子(例如C, N, S, F)[34, 35, 36, 37]也可以达到有效抑制光生电子和空穴复合的目的.  

两个具有相匹配电子能级结构的半导体结合在一起形成异质结, 可以有效地抑制光生电子与空穴的复合[24, 25, 26, 38].  图1以p-n异质结为例说明了异质结在提高e--h+分离效率中的作用.  如图1所示, 在一个半导体中, 当一部分掺杂施主杂质形成n型半导体, 另外一部分与受主杂质掺杂形成p型半导体.  p型半导体与n型半导体相互结合形成p-n异质结.  在p-n异质结附近, 在p型部分存在丰富的负电荷离子, 在n型部分存在带正电的离子, 从而产生电场.  在p-n异质结中,& #8197;电场方向为n指向p.  电场中的电势差可以成为光生电子与空穴分离的驱动力, 从而提高光催化反应的量子效率.  

在光催化中, 异质结的作用不仅局限于p-n结.  其它两个具有相匹配电子能级结构的半导体形成异质结也可以提高光生电子/空穴对的分离效率.  因此, 通过半导体之间形成界面/异质结是提高光催化剂活性的有效方法之一.  

3. 典型异质结光催化剂
3.1. TiO2基异质结光催化剂

作为光催化剂, TiO2具有对有机污染物氧化能力强、光稳定性好、成本低和无毒等优势[39, 40, 41, 42, 43].  TiO2的光催化性能与它的相组成紧密相关.  TiO2常见的相为锐钛矿相(A)和金红石相(R).  Zhang等[25]报道在TiO2表面形成锐钛矿相(A)/金红石相(R)的异质结可明显提高其光催化产氢活性.  将不同含量的锐钛矿相TiO2通过浸渍法负载在金红石相TiO2表面, 可以获得A/R异质结.  图2的高分辨率透射电镜照片显示出在金红石表面形成了A/R相异质结.  

Xiang等[44]还发现, 形成异质结的TiO2/MoS2/石墨烯复合物具有很高的产氢性能, 且MoS2/石墨烯含量为0.5 wt%时活性最高.  其活性提高源于MoS2和石墨烯的协同作用.  最近Qiu等[45]利用一步水热法在具有介孔结构的TiO2的(001)晶面原位生长3D石墨烯气溶胶(Gas), 获得的TiO2/Gas同样具有很高的催化降解甲基橙的活性.  

TiO2的带隙能为3.2 eV, 只能被紫外光(λ < 400 nm)激发, 而紫外光在太阳光谱中的能量只占2%-3%, 因此TiO2无法利用大部分自然光或人工可见光.  与窄带隙半导体复合形成异质结还可以给TiO2带来意想不到的可见光活性.  Dai等[46]采用浸渍羟基化法制备出p-n结BiOI/TiO2纳米管阵列, 可见光下进行光催化降解MO的活性测试表明, BiOI/TiO2的速率常数是BiOI的3倍.  图3为光催化降解MO过程中BiOI/TiO2上的电荷转移过程示意图.  BiOI是一个窄带隙能(~2.1&# 8197;eV)的p型半导体, TiO2是一个宽带隙能(~3.2 eV)的n型半导体.  由于p-n结的形成, BiOI的导带位置比TiO2的更负.  因此, 在p型BiOI导带上的激发电子可以迁移至n型TiO2的导带上, 在BiOI的价带上留下光生空穴, 从而使光生电子-空穴得到了有效分离.  

我们的研究表明, 在TiO2表面形成Ag2CO3/TiO2异质结能使其光催化活性大幅度提高[47].  在紫外光下, 1%Ag2CO3/TiO2对MO的降解率是TiO2的6倍, Ag2CO3的4倍;  在可见光下, 是TiO2的3.4倍和Ag2CO3的1.7倍.  如图4所示, Ag2CO3/TiO2异质结有助于可见光的吸收和抑制e-/h+对的复合.  

表1列出了一系列TiO2异质结光催化剂和光催化性能比较.  由表可见, 异质结的形成可以明显提高TiO2的光催化性能.  

3.2. Ag基半导体异质结光催化剂

Ag基半导体通常具有很强的可见光吸收能力, 作为光催化剂已经引起广泛关注.  最近, Yi等[53]报道了Ag3PO4光催化剂在可见光照射下可以分解水和高效降解有机染料.  但是, Ag3PO4光催化剂的光化学性质不稳定性, 在没有添加其它牺牲试剂的情况下极易发生光化学分解[54, 55].  在光催化过程中, Ag基半导体光催化剂(Ag3PO4, AgI, Ag3AsO4等)通常不可避免地发生光化学腐蚀[56, 57].  因此, 如何提高Ag基半导体的光催化稳定性是其面临的一个挑战.  

研究表明, Ag基半导体与其它半导体形成异质结在提高光生电子和空穴分离的同时可以提高其稳定性[58, 59].  下面是一个设计Ag2O/Ag2CO3异质光催化剂的例子[26].  Ag2O/Ag2CO3异质结可通过简单的焙烧-相转变路径合成, 如图5(a)所示.  在焙烧过程中, Ag2CO3表面首先开始分解产生Ag2O, 后者覆盖在Ag2CO3表面, 产生类似核-壳结构的Ag2O/Ag2CO3异质结.  能级结构计算表明, Ag2CO3和Ag2O的带隙能分别为2.5和1.3 eV;  Ag2CO3的价带(VB)顶和导带(CB)底的电势分别为2.77和0.27 eV;  Ag2O的VB和CB分别为1.5和0.2 eV.  Ag2CO3和Ag2O的电子能带结构如图5(b)所示.  

在这个Ag2O/Ag2CO3体系中, 与Ag2CO3 (CB: 0.27 eV, VB: 2.77 eV)相比, Ag2O有更负的导带(0.2 eV)和价带(1.5 eV).  在可见光照射下, Ag2O层上的光生电子可迅速转移到Ag2CO3表面.  与此同时, 在Ag2CO3表面上产生的光生空穴可迁移至Ag2O表面, 这样增加了彼此光生电子和空穴的分离效率.  Ag2O/Ag2CO3光催化降解MO的活性分别是Ag2CO3和Ag2O的67和31倍(见表2).  另外, 稳定性测试表明, 在第二次循环反应中, Ag2CO3和Ag2O由于光化学腐蚀导致Ag的生成而几乎失去了活性.  但是Ag2O/Ag2CO3即使经过6次循环反应之后, 在光照40 min内MO的降解率仍达到94%. 表3总结了一些典型的Ag基半导体异质结光催化剂的催化性能.  

3.3. ZnO基异质结光催化剂

作为光催化剂, ZnO具有来源丰富、无毒和成本低等优点[65, 66, 67, 68, 69, 70, 71].  但是, 与TiO2光催化剂相比, ZnO光催化活性低, 同时易发生光化学腐蚀而导致稳定性下降.  此外, ZnO发生光化学腐蚀释放的Zn2+对环境产生新的二次污染.  这些缺陷限制了ZnO在环境净化中的利用.  

形成WO3/ZnO[71], Bi2WO6/ZnO[72], BiOCl/ ZnO[73]和MO3/ZnO[74]等异质结有助于提高其活性和稳定性.  图6说明了SnO2/ZnO异质结提高活性的机理[75].  当SnO2和ZnO形成异质结时, 它们不同的功函导致负电荷从SnO2转移到ZnO上直到它们的费米能级能够匹配, 因此在界面产生了一个静电场.  在热平衡中, SnO2和ZnO的CB和VB弯曲形成一个过渡的界面.  在紫外光照射下, VB上的电子(e-)可以被激发到CB上;  同时, 空穴(h+)在VB产生.  在静电场的影响下, 光生电子和空穴容易得到分离.  电子受体如吸附的O2可以容易地诱捕光电子产生超氧自由基(·O2-).  光生空穴可轻易地诱捕OH进一步产生·OH自由基.  ·O2-和·OH可以部分或者全部矿化有机分子.  在紫外光照射下, SnO2, ZnO和SnO2/ZnO对MO的降解率分别为20%, 65%和100%.  

3.4. 其它异质结光催化剂

在异质结体系中, 一个半导体作为主体光催化剂, 另一个半导体称为共催化剂[76, 77].  例如, CdS和少量的MoS2形成紧密接触的MoS2/CdS, MoS2称为共催化剂.  MoS2/CdS的产氢活性比Pt/CdS高.  研究发现, 利用化学沉积合成的MoS2/CdS异质结表现出比CdS和MoS2简单混合更高的催化活性, 这进一步证实了形成异质结的作用.  在其它异质结体系如Bi5O7I/BiOI[78], Bi24O31Br/ BiOBr[79]和CuO/BiVO4[80]中也进一步证实了这一结论.  表4总结了其它典型异质结光催化剂.  

4. 异质结光催化剂的制备方法

光催化剂的性能强烈依赖于它们的制备方法.  目前报道合成异质结光催化剂的方法包括:  相转变法[26, 78, 79]、浸渍-沉积法[25, 86]、浸涂法[48]、液相沉积法[89]、高温固相反应法[81]、水热/溶剂热法[84, 85]、电沉积法[90]、阳离子交换法[91]和溶胶凝胶法[49]等.  相转变法可以简单制备出各种新型异质结光催化剂, 如Ag2O/Ag2CO3[26], α-Bi2O3/Bi2O2CO3, Bi5O7I/BiOI[78], Bi5O7I/Bi2O3[78]和Bi24O31Br/BiOBr[79]等.  相转变合成是通过精确控制热力学不稳定半导体的焙烧温度、时间和气氛, 使半导体发生部分相转变, 形成异质相结.  下面是采用相转变法制备Bi5O7I/BiOI和Bi5O7I/Bi2O3异质结催化剂的一个例子.  首先通过热重(TGA)判断BiOI在空气中发生相转变的温度.  如图7所示, 从350到520 ºC, BiOI开始分解失去I, 失重为27.00%, 相应的反应为5BiOI + O2 → Bi5O7I + 2I2.  从600到850 ºC, 总失重为9.16%, 对应反应为2Bi5O7I + 1/2O2 → 5Bi2O3 + I2.  如果控制BiOI的焙烧温度在350- 520 ºC, 同时控制焙烧时间, 可以获得Bi5O7I/BiOI异质结.  同样地, 控制BiOI焙烧温度在620 ºC左右可以获得Bi5O7I/Bi2O3.  

5. 结论

提高光催化效率是推进光催化在太阳能转化和环境净化领域大规模应用的关键.  在主体半导体上设计能级相互匹配的异质结, 形成紧密接触的相界面, 可以有效地促进光生电子和空穴分离, 大幅度地提高光催化剂的活性和稳定性.  目前存在的主要问题是, 难于准确表征分析催化剂表面的不同晶相和进行不同晶相的定量分析, 同时关于异质结含量和催化性能的关系及异质结增强光催化活性机理还不是非常清楚.  这是今后需要解决的问题.