Hydrogen Production by Photocatalytic Water Splitting

Transcription

1 280 Journal of the Japan Petroleum Institute, 56, (5), (2013) [Review Paper] Hydrogen Production by Photocatalytic Water Splitting Su Su Khine MA, Takashi HISATOMI, and Kazunari DOMEN Dept. of Chemical System Engineering, The University of Tokyo, Hongo, Bunkyo-ku, Tokyo , JAPAN (Received May 6, 2013) Hydrogen gas as a fuel has the potential to alleviate the threat of global climate change and help to avoid various undesirable effects caused by the mass consumption of fossil fuel. Photocatalytic water splitting has been widely studied as a potential method to produce H 2 from renewable solar energy. Photocatalysts that can operate under visible light irradiation (λ 400 nm), which forms the main part of sunlight, are highly desirable. Successful two-step water splitting systems (Z-scheme) that operate with or without a reversible redox couple have been reported for the application of visible light-driven photocatalysts. The Z-scheme system, which mimics photosynthesis in green plants, consists of two photocatalysts, one for H 2 evolution, and the other for O 2 evolution. The Z-scheme process can utilize a wider range of visible light than the conventional one-step excitation system because the energy required to drive each photocatalyst can be reduced. This review presents recent research progress in the development of visible light-driven photocatalytic materials with a focus on Z-scheme water splitting. Keywords Hydrogen production, Two-step water splitting, Z-scheme, Visible light, Photocatalyst 1. Introduction To whom correspondence should be addressed. domen@chemsys.t.u-tokyo.ac.jp Recently, natural sources of energy have become increasingly important to provide alternative fuels for environmentally friendly, sustainable development. Hydrogen, which has been proposed as the main energy source of the future, could offer a solution to the threat of global climate change and help avoid the undesirable effects caused by mass consumption of fossil fuels. Hydrogen is a clean energy carrier when used in fuel cells because only water is emitted as an oxidation product. Efficient, abundant and economical hydrogen production will have important implications for the hydrogen economy. The growing interest in environmentally benign and energy-saving technologies has intensified developments in the field of distributed hydrogen production. In particular, production of hydrogen using solar energy has attracted significant attention as a key issue in the utilization of the sun as the most abundant source of renewable energy. Photocatalytic hydrogen generation from water is one of the attractive, environmentally friendly methods for harvesting solar energy 1). The photocatalytic splitting of water into hydrogen and oxygen using a powdered photocatalyst has become the subject of research, mainly focusing on the visible light sensitization of photocatalysts in order to effectively utilize the available solar energy. Sunlight (AM1.5G) consists of three main components in terms of wavelengths: ultraviolet rays (λ 400 nm), visible light (400 nm λ 800 nm), and infrared rays (λ 800 nm), accounting for 4, 53, and 43 % of the solar energy, respectively. The theoretical maximum efficiency of solar energy conversion increases with the wavelength. The maximum solar energy conversion efficiencies are calculated to be 2, 16, and 32 %, at 400, 600, and 800 nm, respectively, given that the quantum efficiency of photocatalytic water splitting is unity 2). 2. Photocatalytic Water Splitting Basic Principles of Water Splitting on a Heterogeneous Photocatalyst Photocatalytic overall water splitting is a simple chemical process in which hydrogen (H2) and oxygen (O2) are produced from water by utilizing the energy of light. This process is also called artificial photosynthesis. The overall water splitting reaction is endothermic with an increase in Gibbs free energy of 238 kj mol 1. Theoretically, only solar energy (photons), water, and a photocatalyst are needed for this process. The chemical process of water splitting using a heterogeneous photocatalyst is illustrated in Fig. 1. The photocatalyst incorporates semiconductor material

2 281 Fig. 1 Process of Photocatalytic Water Splitting Reaction on a Heterogeneous Photocatalyst which has an electronic band structure. The highest occupied energy band is called the valence band (VB) and the lowest empty band is called the conduction band (CB). These bands are separated by a band gap on the order of a few electron volts 3). The process of photocatalytic water splitting using a semiconductor photocatalyst consists of four main steps: (i) absorption of photons with energies greater than the band gap, (ii) charge separation of photoexcited electron-hole pairs in the bulk catalyst, (iii) migration of charge carriers from the bulk to the surface of the catalyst, and (iv) simultaneous reduction of H to H2 by electrons (Eq. (1)) and oxidation of H2O to O2 by holes (Eq. (2)) at active sites on the catalyst surface. Overall, two water molecules are split into two H2 and one O2 molecules via the transfer of four electrons. As the numbers of electrons and holes involved in the surface redox reactions are the same, the photocatalyst remains unaltered. Reduction: 2H + + 2e H2 (1) Oxidation: + + 2HO+ 4h 4H + O (2) 2 2 Overall reaction: 2HO 2 2H2 + O2 (3) The feasibility of this optimum conversion reaction 4) has been limited by two main factors: the band structures of known photocatalysts and the reaction efficiency. The thermodynamic requirements are particularly stringent. The semiconductor photocatalyst should have a small band gap to absorb as much light as possible. However, the photogenerated electrons can reduce H to H2 only if the potential of the conduction band is more negative than the potential of H2 evolution (0 V versus HE at ph 0). Likewise, photogenerated holes can oxidize H2O to O2 only if the potential of the valence band is more positive than the potential of O2 evolution ( 1.23 V versus NHE at ph 0). Most oxides that act as active photocatalysts for water splitting have a band gap energy that is too large to absorb visible light, because the valence bands of the oxides mainly consist of O2p orbitals with potentials of about 3 V versus NHE (ph 0) 5). Consequently, the band gap energy of the oxides inevitably exceeds the 3 ev needed to satisfy the thermodynamic requirement for water dissociation that the band gap of the photocatalyst must include the potentials of water reduction and oxidation. For this reason, oxides have limited potential for photocatalytic solar energy conversion. The recombination of electron-hole pairs should also be suppressed to achieve high efficiency in the photocatalytic reactions, instead of energy losses in the generation of phonons or heat 6). Efficient charge separation and fast electron injection can be achieved by modifying the photocatalyst surfaces with cocatalysts that collect photoexcited carriers and catalyze the surface redox reactions. However, such surface active sites may favor undesirable water formation from H2 and O2 instead of water splitting, so strategies to suppress undesirable side reactions should be established. Note that the rate of water splitting also decreases in the presence of Ar gas, although Ar is not involved in the O2 reduction reaction. Apparently even an inert gas can suppress the formation and desorption of bubbles of the product gases and enhance O2 reduction during the overall water splitting reaction 7). Side reactions such as O2 reduction and H2O formation from H2 and O2 often drastically reduce the rates of photocatalytic overall water splitting. Therefore, it is important to develop a cocatalyst to avoid side reactions even in an ambient atmosphere. In addition, (photo)corrosion of photocatalysts and loss of catalytic capability gradually reduce the photocatalytic activities 8). Stabilization and regeneration of photocatalytic systems should be carefully considered for the use of photocatalytic systems in practical applications Half-reactions Using Sacrificial Electron Donors and Acceptors The endothermic requirements of water splitting with associated rapid recombination of photo-generated conduction band electrons and valence band holes are difficult to achieve. Consequently, sacrificial reagents are often used in the initial stage of photocatalyst development. Sacrificial reagents are employed in partial water splitting reactions (termed half-reactions/test reactions) to determine the photocatalytic activity of a particular photocatalyst and to evaluate whether the physicochemical properties satisfy the kinetic and thermodynamic potentials required for H2 or O2 formation. Such test reactions for H2 formation use methanol or sulfide ions as the electron donors. These additives, rather than water, are oxidized irreversibly by the photogenerated holes in the valence band. The test reaction of O2 formation usually uses silver ions or iodate as

3 282 electron acceptors or electron scavengers to consume the photogenerated electrons in the conduction band. The overall redox reactions in the presence of such sacrificial reagents can be expressed as follows: Water reduction (H2 formation): () i CH 3OH + HO 2 CO2 + 3H 2 (4) 2 ( ii ) SO 3 + HO 2 SO4 2 + H 2 (5) Water oxidation (O2 formation): + + () i 4Ag + 2HO 2 O2 + 4Ag+ 4H (6) ( ii ) 2IO3 2I + 3O2 (7) Note that the reduction of iodate and the oxidation of iodide are reversible reactions. This reversibility is utilized in redox shuttle mediators in Z-scheme water splitting, as described in a later section. Even if a photocatalyst can reduce and/or oxidize water separately, overall water splitting is not always possible to achieve, presumably because of rapid charge recombination and reverse water formation in the absence of sacrificial reagents. Accordingly, halfreactions using sacrificial reagents are regarded as suitable test reactions to assess whether a semiconductor of interest has the potential to produce H2 or O2 under light irradiation Heterogeneous Photocatalysts for Overall Water Splitting The discovery of the photocatalytic splitting of water on TiO2 electrodes by Fujishima and Honda in the early 1970s resulted in the beginning of a new era in photocatalysis and prompted extensive research concerning solar energy conversion 9),10). In their pioneering research, UV irradiation of a TiO2 single-crystal electrode resulted in O2 evolution under an electrochemical bias, associated with H2 evolution on the counter electrode. This important result prompted extensive work on photocatalytically active semiconductors, focusing solar energy conversion to produce H2 as a combustible fuel from water. By 2010, over 130 photocatalysts had been identified as active for the water splitting reaction (water oxidation and/or reduction) in the presence of sacrificial agents 11),12). These materials included ZrO2 13), titanates 14),15), tantalates 16),17), niobates 18), metal nitrides 19) 21), phosphides 22), and sulphides 23),24). On the basis of the previously established characteristics of photocatalysts, metal compounds with d 0 ions (Ti 4, Zr 4, Nb 5, and Ta 5 ) and d 10 ions (Ga 3, In 3, Ge 4, Sn 4, and Sb 5 ) are empirically accepted as having activity for photocatalytic water splitting. Photocatalysts based on transition metal cations with empty d-orbitals and typical metal cations with fully filled d-orbitals are called photocatalysts with d 0 - and d 10 -electronic configurations, respectively 21),25). The common feature of these d 0 /d 10 -type photocatalysts is that the metal ions are in the highest oxidation state. There have been no reported cases of highly stable water splitting achieved by materials consisting of metal ions with partially filled d-orbitals (d 1 to d 9 ), regardless of whether the catalyst is an oxide, nitride, or sulphide Non-oxide Semiconductors as Visible Lightdriven Photocatalysts Photocatalysts that function only under ultraviolet light will not allow photocatalytic water splitting as a practical means of solar energy conversion, because ultraviolet light accounts only for 4 % of the energy of solar irradiation. Therefore, considerable efforts have been invested in developing photocatalysts which function under the less energetic but more abundant visible light, which accounts for nearly half of the incoming solar irradiation received at the surface of the earth. As mentioned in section 2. 1., the valence band of oxide semiconductors is more positive than the O2 evolution potential. Therefore, the development of photocatalysts based on non-oxide semiconductors with narrower band gaps is required for efficient solar energy utilization. One approach to decrease the band gap is to incorporate N atoms into metal oxide structures through nitridation 21),25) 27), which is accomplished by heat treatment of the oxide precursors under NH3 flow at high temperature. NH3 thermally decomposes at high temperature to form free radicals (H and N 3 ) 29). As nitridation progresses, these free radicals extract O from the oxide precursors and replace it with N. This process forms metal (oxy)nitrides. The incorporated N atoms form a new valence band consisting of hybridized N2p and O2p orbitals. Because the N2p atomic orbital has a more negative potential than the O2p atomic orbitals, the top of the valence band is shifted to a more negative potential. On the other hand, the potential of the conduction band edge, which consists of the d or sp orbitals of metals, is largely unchanged by the substitution of N for O. Consequently, the band gap energy of the corresponding oxide is decreased whereas the conduction band level remains unaffected, resulting in a visible light-driven photocatalyst with band edge potentials suitable for overall water splitting. Note that the band structures of (oxy)nitrides are different from those of nitrogen-doped oxides, in that the N2p orbitals are hybridized with O2p orbitals to form a thick and continuous band. Generally, N atoms doped in an oxide form discontinuous impurity levels within the band gap because of the limited amount of doped nitrogen 28),29). Therefore, the visible light absorption of nitrogendoped oxides is relatively weak and is often observed only as shoulder absorption. Schematic band structures of Ta2O5, TaON, and Ta3N5 are shown in Fig. 2 30). During the nitridation process, the three O 2 anions in the precursor Ta2O5 are

4 283 Fig. 2 Schematic Band Structures of Ta 2 O 5, TaON, and Ta 3 N 5 (Modified from 30) ) Fig. 3 Energy Diagram of Two Approaches for the Water Splitting Reaction: one-step, and two-step photoexcited water splitting (Z-scheme) replaced by two N 3 anions, whereas the valence state of Ta 5 is maintained. The top valence band of Ta2O5 consists of a O2p orbital which is located at ca. 3.4 V versus NHE. In contrast, the N2p orbitals of TaON and Ta3N5 are incorporated in the valence bands, modifying the top of the valence bands to ca. 2.0 V versus NHE for TaON and 1.5 V versus NHE for Ta3N5. The lower conduction bands of the three Ta-compounds mainly consist of empty Ta5d orbitals and are located at similar potentials, 0.3 to 0.5 V versus NHE. Consequently, the light absorption spectra of TaON and Ta3N5 are expanded up to ca. 500 nm and 600 nm, respectively, although light absorption by Ta2O5 is limited in the UV region. Therefore, the nitrided products TaON and Ta3N5 have suitable band gap positions for the photocatalysis of water splitting. In the last decade, substantial efforts have been made to develop (oxy)nitrides and (oxy)sulphides as photocatalysts for water splitting under visible light irradiation. One excellent candidate is (Ga1 xznx)(n1 xox) which is a visible light-driven d 10 -type (oxy)nitride photocatalyst and has been successfully used in water splitting through one-step photoexcitation 31),32). Using Rh2 ycryo3 loaded (Ga1 xznx)(n1 xox), a high apparent quantum efficiency of 5.1 % at 410 nm was obtained for overall water splitting, one of the highest efficiencies yet reported for photocatalytic water splitting under visible light irradiation. However, the absorption edge wavelength of (Ga1 xznx)(n1 xox) is located at 500 nm, whereas highly efficient photocatalytic systems must also absorb longer wavelength photons. This drawback has motivated further research to identify potential candidates that could function under a wider spectrum. Other (oxy)nitrides, including the d 0 -type (oxy) nitrides TaON, Ta3N5, LaTiO2N, and ATaO2N (A Ca, Sr, Ba) with absorption edges at nm (band gap energies of ev), cannot catalyze water splitting, but can evolve H2 and O2 separately from solutions containing appropriate sacrificial reagents 21). This effect possibly results from charge recombination due to insufficient reaction efficiencies. Although overall water splitting has not been achieved in the one-step excitation scheme despite various modifications, presumably because of rapid charge recombination, correctly modified TaON can split water into H2 and O2 under visible light irradiation in the two-step excitation (Z-scheme) system. These favorable results have opened up a range of possibilities for the modification of d 0 -type (oxy)nitrides for use in the Z-scheme water splitting system. 3. Two-step Photoexcitation Processes (Z-scheme) Basic Principle There are two basic approaches for water splitting: the one-step and two-step photoexcitation schemes (Fig. 3). The Z-scheme system incorporates H2 and O2 evolution photocatalysts and usually a redox mediator. The two semiconductors with small band gaps are connected by reversible redox reagents. This photocatalytic system is inspired by green plant photosynthesis, and has been termed the Z-scheme 33), based on the similarities of the excitation and transfer processes of the photoexcited electrons. The idea of Z-scheme water splitting was originally proposed in the late 1970s and early 1980s by Bard and associates, who developed a dual n-type semiconductor model for biological photosynthesis and a possible means of application to artificial systems 33),34). Later, Fujihara et al. 35) also constructed a Z-scheme water splitting system using a TiO2-rutile photocatalyst and two redox mediators (Br2/ Br and Fe 3 /Fe 2 ). A detailed schematic of Z-scheme water splitting is presented in Fig. 3 2),36), in which H2 and O2 evolution systems are connected by a shuttle redox couple (termed Red/Ox) in the solution. In the ideal scenario, both reduction of H to H2 and oxidation of redox mediators occur on the H2 evolution photocatalyst, while the reduction of redox mediators and oxidation of H2O to O2 occur simultaneously on the O2 evolution photocatalyst 37) 41). The advantage of Z-scheme water

5 284 Fig. 4 Schematic Illustration of Various Z-scheme Water Splitting Systems Driven by Visible Light in the Presence of (a) IO 3 / I and (b) Fe 3 /Fe 2 Redox Mediators Respectively (Modified from 38) ) splitting over one-step photoexcitation water splitting is that a wider range of visible light can be used, as semiconductors with either water reduction or oxidation potentials can be employed. Both H2 and O2 generation from water can be independently studied by using sacrificial electron donors (for H2 generation) and electron acceptors (for O2 formation). Indeed, Z-scheme water splitting using optimized photocatalysts with high efficiency in half-reactions can be feasible and beneficial. On the other hand, the Z-scheme requires balanced photocatalytic activity of the H2 and O2 evolution photocatalysts. In addition, the number of photons required to generate a certain amount of H2 is double that required for one-step splitting. In addition to the inherent quality of the photocatalysts, several other factors affect the photocatalytic activity of Z-scheme water splitting systems: ph, cocatalysts, and redox mediators 40),42). First, the ph of the reactant solution is related to the redox processes of both water reduction and oxidation, as well as the zeta-potential of the H2 and O2 evolution photocatalysts 40),42). Optimizing the ph can improve the photocatalytic activity of Z-scheme systems by a factor of two or more 38),41),42). ph also has a direct influence on the stability of the photocatalytic materials. Second, the cocatalysts provide the sites for redox reactions on the photocatalyst surface. Cocatalysts in a Z-scheme process could facilitate charge transfer via redox couples and, to some extent, prevent the thermodynamically favorable reverse reaction of the redox mediator 25). Third, the redox mediator shuttles photogenerated carriers between the H2 and O2 evolution photocatalysts. The most commonly employed redox mediators are the Fe 3 /Fe 2 or IO3 /I redox couples. The presence of a redox mediator is critical for driving the electron relay efficiently. The efficiency of a Z-scheme water splitting system depends strongly on the choice of suitable redox mediators Z-scheme Water Splitting Using Redox Mediators (Fe 3 /Fe 2 or IO3 /I ) Since the conception of the Z-scheme, both photosystems have been studied separately using different photocatalytic materials and effective redox mediators 2),40),43) 51). Z-scheme photocatalytic water splitting was first reported by Sayama et al. in 1997 using WO3 powder suspended in an aqueous FeSO4 solution under UV irradiation (λ 200 nm) 43). In this system, Fe 2 is excited by UV light followed by photochemical reduction of water to form H2 and Fe 3, while WO3 photocatalyzes water oxidation using Fe 3 as an electron acceptor. Another Z-scheme system consists of Pt-modified anatase TiO2 as the H2 evolution photocatalyst and bare rutile TiO2 as the O2 evolution photocatalyst under UV irradiation (λ 300 nm) in the presence of an iodate/ iodide (IO3 /I ) shuttle redox mediator 44). This pioneering Z-scheme for water splitting employed photocatalysts for both H2 and O2 evolution. Since the band edge potentials and the band gap energies of semiconductors can be modified by changing the compositions, various photocatalysts can be active under visible light. Z-scheme water splitting was achieved under visible light (λ 420 nm) using suspended particles of Pt-loaded SrTiO3 doped with Cr and Ta (H2 evolution photocatalyst) and Pt-loaded WO3 (O2 evolution photocatalyst) with an IO3 /I redox pair 37). These findings established the principles of Z-scheme water splitting under visible light using various H2 and O2 evolution photocatalysts in the presence of IO3 /I or Fe 3 /Fe 2 redox mediators. Figures 4(a) and 4(b) show an overview of various Z-scheme photocatalysis systems with IO3 /I and Fe 3 / Fe 2 redox couples, respectively. The most effective system is (Pt/ZrO2/TaON) _ (Pt/WO3) using the IO3 /I redox couple with an apparent quantum yield of 6.3 % at 420 nm 49), which is currently the highest apparent quantum yield reported for a non-sacrificial visible

6 285 Fig. 5 Forward and Reverse Reactions in the Two-step Photoexcitation System (Z-scheme) (Modified from 2),38) ) light-driven water splitting system. ZrO2/TaON loaded with nanoparticulate Pt functions efficiently as the H2 evolution photocatalyst in combination with Pt/WO3 48), RuO2/TaON 49), or Ir/TiO2/Ta3N5 50) as the O2 evolution photocatalyst in the presence of an IO3 /I redox mediator, far outperforming the analogous Pt/TaON system 48). The range of wavelengths available for O2 evolution has also been extended, resulting in water splitting with visible light up to 520 nm wavelength 39),40). These Z-scheme systems consisted of Pt- or Ru-modified SrTiO3 doped with Rh (H2 evolution photocatalyst) and BiVO4 (O2 evolution photocatalyst), with electron transport between the two photocatalysts by an Fe 3 /Fe 2 redox mediator Z - s c h e m e Wa t e r S p l i t t i n g t h r o u g h Interparticle Electron Transfer without Redox Mediator The thermodynamically favorable reverse reaction is a major obstacle to the use of redox mediators, despite the vital effect of the redox mediators on Z-scheme water splitting. As illustrated in Fig. 5, the reverse reactions (indicated by the dotted lines) of the redox mediator proceed readily over both photocatalysts and suppress the forward reactions (H2 and O2 evolution). Although certain combinations of photocatalyst and redox mediator resulted in evolution of either H2 or O2 under light irradiation, gas evolution was terminated during the reaction because of the reverse reaction in most cases. Therefore, high-efficiency Z-scheme water splitting requires a photocatalytic system with high selectivity for the forward reactions (indicated by the solid lines). The use of redox couples can also cause additional problems. Part of the visible light spectrum will be absorbed by a redox mediator with color, which absorbs some of the irradiated incident light. In addition, the differences in favorable redox concentrations for the individual photocatalysts makes the optimum performance of the respective photocatalysts difficult to identify. Thus, the construction of Z-scheme systems consisting of only H2 and O2 evolution photocatalysts in the Fig. 6 Mechanism of Two-step Photoexcited Z-scheme Water Splitting Driven by Electron Transfer between H 2 and O 2 Evolution Photocatalysts Using Ru/SrTiO 3 :Rh and Co/Ir/ Ta 3 N 5 in the Absence of Redox Mediator 51) absence of electron relay systems is highly desirable, and can be achieved through interparticle electron transfer from one side to the other, mediated by physical contact between the H2 and O2 evolution photo catalysts, respectively. Recently, Ru-loaded SrTiO3 doped with Rh (Ru/SrTiO3:Rh) achieved the functionality of a H2 evolution photocatalyst for Z-scheme water splitting even without a shuttle redox mediator 25),40). In this system, many metal oxide photocatalysts, active for O2 evolution from aqueous AgNO3 solution but inactive in the presence of a redox couple, can be applied for the O2 evolution function. When the H2 and O2 evolution photocatalysts are photoexcited, H2 and O2 evolution reactions occur at the respective photo catalysts, and electron transfer from one side to the other is mediated by physical contact. More specifically, interparticle electron transfer occurs from the conduction band of the O2 evolution photocatalyst to the donor function provided by Rh dopants in the forbidden band of SrTiO3:Rh, thus allowing stoichiometric H2 and O2 evolution without requiring an electron relay redox mediator system. Among the metal oxide photocatalysts examined, BiVO4 was the most active component for O2 evolution in the redox-free Z-scheme system with SrTiO3:Rh, achieving a solar energy conversion efficiency of 0.12 % under optimal conditions 25),40). However, the absorption band edge of the O2 evolution photocatalyst, BiVO4, is located at ca. 520 nm. Efficient utilization of solar energy will require the replacement of BiVO4 with another component with a wider absorption band in the visible light region. Recently, a redox mediatorfree Z-scheme system was successfully constructed using Ir/CoOx/Ta3N5 (λ 600 nm) in combination with Ru/SrTiO3:Rh, which obtained a solar energy conversion efficiency of % under simulated sunlight, as shown in Fig. 6 51). Conclusions Photocatalytic water splitting using semiconductor catalysts under visible light may become an important

7 286 method for supplying clean H2 fuel. This review described the construction of Z-scheme photocatalytic systems using powdered photocatalysts. Such systems offer significant capability to use various photocatalysts that are active for either H2 or O2. In particular, Z-scheme systems that do not rely on redox mediator systems have been the target of active numerous studies because the undesirable reverse reactions and negative effects due to redox mediators can be excluded. The present review also described the challenges in application of the Z-scheme systems, development of suitable photocatalysts with visible light activity, and the rate determining factors of Z-scheme water splitting on hetero geneous photocatalysts. Efficient and stable use of solar energy using twostep photoexcited Z-scheme water splitting is possible by employing visible light-driven photocatalysts absorbing high wavelength light and suitable catalyst modification methods to construct reaction sites and promote interparticle electron transfer. Such modifications have expanded the possibility of using various active (oxy)nitrides and nitrides in Z-scheme water splitting systems. Although photon energy conversion using powdered photocatalysts has not reached the stage of practical implementation, the potential of photocatalytic water splitting has been advanced with these visible light-driven photocatalysts in Z-scheme systems. More efficient photocatalytic materials with band gaps of longer wavelength must be developed for application to hydrogen production. A mechanistic understanding of photocatalytic reactions would also be useful for the development of such photocatalysts. Acknowledgment This work was financially supported by the Grant-in- Aid for Specially Promoted Research (No ) of the Ministry of Education, Culture, Sports, Science, and Technology (MEXT) of Japan, the Advanced Low Carbon Technology Research and Development Program (ALCA) of the Japan Science and Technology Agency (JST), the Global Center of Excellence (GCOE) Program for Chemistry Innovation through Cooperation of Science and Engineering, and the A3 Foresight Program of the Japan Society for the Promotion of Science (JSPS). References 1) Youngblood, W. J., Lee, S.-H. A., Maeda K., Mallouk, T. E., Acc. Chem. Res., 42, (12), 1966 (2009). 2) Abe, R., J. Photochem. Photobiol. C, 11, (4), 179 (2010). 3) Litter, M. I., Appl. Catal. B: Environmental, 23, (2), 89 (1999). 4) Bard, A. J., Fox, M. A., Acc. Chem. Res., 28, (3), 141 (1999). 5) Scaife, D. E., Sol. Ener., 25, (1), 41 (1980). 6) Ni, M., Leung, M. K. H., Leung, D. Y. C., Sumathy, K., Renew. Sust. Energy Rev., 11, (4), 401 (2010). 7) Maeda, K., Teramura, K., Masuda, H., Takata, T., Saito, N., Inoue, Y., Domen, K., J. Phys. Chem. B, 110, (26), (2006). 8) Yan, H. J., Yang, J. H., Ma, G. J., Wu, G. P., Zong, X., Lei, Z. B., Shia, J. Y., Li, C., J. Catal., 266, (2), 165 (2009). 9) Fujishima, A., Honda, K., Bull. Chem. Soc. Jpn., 44, (4), 1148 (1971). 10) Fujishima, A., Honda, K., Nature, 238, (5358), 238 (1972). 11) Inoue, Y., Energy Environ. Sci., 2, (4), 364 (2009). 12) Osterloh, F. E., Chem. Mater., 20, (1), 35 (2008). 13) Sayama, K., Arakawa, H., J. Phys. Chem., 97, (3), 531 (1993). 14) Inoue, Y., Kaneko, M., Okura, I., Eds.; Springer: New York, 12, (3), 249 (2002). 15) Sayama, K., Mukasa, K., Abe, R., Abe, Y., Arakawa, H., Chem. Commun., 23, 2416 (2001). 16) Kato, H., Kudo, A., Chem. Phys. Lett., 295, (5-6), 487 (1998). 17) Kato, H., Kudo, A., Catal. Lett., 58, (2-3), 153 (1999). 18) Domen, K., Kudo, A., Shibata, M., Tanaka, A., Maruya, K., Onishi, T., J. Chem. Soc. Chem. Commun., 23, 1706 (1986). 19) Yamasita, D., Takata, T., Hara, M., Kondo, J. N., Domen, K., Solid State Ionics, 172, (1-4), 591 (2004). 20) Maeda, K., Teramura, K., Saito, N., Inoue, Y., Kobayashi, H., Domen, K., Pure Appl. Chem., 78, (12), 2267 (2006). 21) Maeda, K., Domen, K., J. Phys. Chem. C, 111, (22), 7851 (2007). 22) Ohmori, T., Mametsuka, H., Suzuki, E., Int. J. Hydrogen Energy, 25, (10), 953 (2000). 23) Reber, J. F., Meier, K., J. Phys. Chem., 88, (24), 5903 (1984). 24) Reber, J. F., Rusek, M., J. Phys. Chem., 90, (5), 824 (1986). 25) Kudo, A., Miseki, Y., Chem. Soc. Rev., 38, (1), 253 (2009). 26) Hitoki, G., Takata, T., Kondo, J. N., Hara, M., Kobayashi, H., Domen, K., Electrochemistry, 70, (6), 463 (2002). 27) Maeda, K., Domen, K., J. Phys. Chem. Lett., 1, (18), 2655 (2010). 28) Asahi, R., Morikawa, T., Ohwaki, T., Aoki, K., Taga, Y., Science, 293, (5528), 269 (2010). 29) Rahinov, I., Ditzian, N., Goldman, A., Cheskis, S., Appl. Phys. B, 77, (79), 541 (2003). 30) Chun, W.-J., Ishikawa, A., Fujisawa, H., Takata, T., Kondo, J. N., Hara, M., Kawai, M., Matsumoto, Y., Domen, K., J. Phys. Chem. B, 107, (8), 1798 (2003). 31) Maeda, K., Takata, T., Hara, M., Saito, N., Inoue, Y., Kobayashi, H., Domen, K., J. Am. Chem. Soc., 127, (23), 8286 (2005). 32) Maeda, K., Xiong, A., Yoshinaga, T., Ikeda, T., Sakamoto, N., Hisatomi, T., Takashima, M., Lu, D., Kanehara, M., Setoyama, T., Teranishi, T., Domen, K., Angew. Chem., Int. Ed., 49, (24), 4096 (2010). 33) Bard, A. J., J. Photochem., 10, 59 (1979). 34) Mau, A. W.-H., Huang, C. B., Kakuta, N., Bard, A. J., J. Am. Chem. Soc., 106, (3), 6537 (1984). 35) Fujihara, K., Ohno, T., Matsumura, M., J. Chem. Soc., Faraday Trans., 94, (93), 3705 (1998). 36) Maeda, K., J. Photochem. Photobiol. C, 12, (2), 237 (2011). 37) Sayama, K., Mukasa, K., Abe, R., Abe, Y., Arakawa, H., Chem. Commun., (23), 2416 (2001). 38) Abe, R., Sayama, K., Sugihara, H., J. Phys. Chem. B, 109, (33), (2005). 39) Kato, H., Hori, M., Konta, R., Shimodaira, Y., Kudo, A., Chem. Lett., 33, (10), 1348 (2004). 40) Sasaki, Y., Nemoto, H., Saito, K., Kudo, A., J. Phys. Chem. C, 113, (40), (2009). 41) Kato, H., Sasaki, Y., Iwase, A., Kudo, A., Bull. Chem. Soc. Jpn., 80, (12), 2457 (2007). 42) Kudo, A., MRS Bull., 36, (17), 32 (2011). 43) Sayama, K., Yoshida, R., Kusama, H., Okabe, K., Abe, Y., Arakawa, H., Chem. Phys. Lett., 277, (4), 387 (1997). 44) Abe, R., Sayama, K., Domen, K., Arakawa, H., Chem. Phys.

8 287 Lett., 344, (3-4), 339 (2001). 45) Abe, R., Higashi, M., Zou, Z., Sayama, K., Abe, Y., Arakawa, H., J. Phys. Chem. B, 108, (3), 811 (2004). 46) Arakawa, H., Sayama, K., Catal. Surv. Jpn., 4, (1), 75 (2000). 47) Maeda, K., Terashima, H., Kase, K., Higashi, M., Tabata, M., Domen, K., Bull. Chem. Soc. Jpn., 81, (8), 927 (2008). 48) Maeda, K., Higashi, M., Lu, D., Abe, R., Domen, K., J. Am. Chem. Soc., 132, (16), 5858 (2010). 49) Maeda, K., Abe, R., Domen, K., J. Phys. Chem. C, 115, (30), 3057 (2011). 50) Tabata, M., Maeda, K., Higashi, M., Lu, D., Takata, T., Abe, R., Domen, K., Langmuir, 26, (12), 9161 (2010). 51) Ma, S. S. K., Maeda, K., Lu, D., Domen, K., Chem. Eur. J., 19, (23), 7480 (2013). Su Su Khine MA, Z 2 Z