Local structures, excited states, and photocatalytic reactivities of highly dispersed catalysts constructed within zeolites

Journal of Photochemistry and Photobiology C: Photochemistry Reviews 3 (2003) 225–252

Review

Local structures, excited states, and photocatalytic reactivities of highly dispersed catalysts constructed within zeolites Masaya Matsuoka, Masakazu Anpo∗ Department of Applied Chemistry, Graduate School of Engineering, Osaka Prefecture University, 1-1 Gakuen-cho, Sakai, Osaka 599-8531, Japan Accepted 29 October 2002

Abstract Transition metal oxides (Ti, V, Mo, Cr) incorporated within the framework of zeolites as well as transition metal ions (Cu+ , Ag+ , Pr3+ ) exchanged within the zeolite cavities were found to exhibit high and unique photocatalytic activities for various reactions such as the decomposition of NOx (NO, N2 O) into N2 and O2 or the reduction of CO2 with H2 O to produce CH4 and CH3 OH. Various in situ spectroscopic investigations of these catalytic systems using photoluminescence, X-ray absorption fine structure (XAFS) (X-ray absorption near edge structure (XANES) and Fourier transform of EXAFS (FT-EXAFS)), electron spin resonance (ESR), FT-IR, etc. revealed that the photo-excited states of these transition metal oxides or ions play a vital role in these photocatalytic reactions. The photocatalytic reactivity of these oxides and ions in their efficiency and selectivity were found to depend strongly on their local structures, which are controlled by the unique and restricted framework structures of zeolites. © 2002 Japanese Photochemistry Association. Published by Elsevier Science B.V. All rights reserved. Keywords: Photocatalysis; Transition metal oxides; Transition metal cations; Zeolites; Mesoporous molecular sieves; NOx decomposition; CO2 reduction

Contents 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Characterization of vanadium, molybdenum and chromium oxides photocatalysts incorporated into the framework of zeolites or mesoporous zeolites and their photocatalytic activities. . . . . . . . . 2.1. Vanadium oxide (V-silicalite) catalyst . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. Molybdenum oxide (Mo-MCM-41) catalyst . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3. Chromium oxide (Cr-HMS) catalyst . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. Characterization of titanium oxide photocatalysts anchored in the cavities of various zeolites and their photocatalytic activities. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Photocatalytic reduction of CO2 with H2 O on titanium oxide photocatalysts anchored in the cavities of various zeolites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1.1. Preparation of the Ti oxide/Y-zeolite by an ion-exchange method and their photocatalytic activity for the reduction of CO2 with H2 O . . . . . . . . . . . . . . . . . . . . . . . . . 3.1.2. Preparation of Ti-mesoporous zeolites by hydrothermal synthesis and their photocatalytic activity for the reduction of CO2 with H2 O . . . . . . . . . . . . . . . . . . . . . . . . . 3.1.3. Effect of Pt loading on the photocatalytic activity and selectivity of Ti-containing zeolite catalysts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. The photocatalytic decomposition of NO into N2 and O2 on the titanium oxide photocatalyst anchored in Y-zeolite cavities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4. Preparation of Cu+ /ZSM-5 catalyst and its photocatalytic reactivity for the decomposition of NO 4.1. Preparation of the Cu2+ /ZSM-5 sample and Cu+ /ZSM-5 catalyst and their local structures . . 4.2. Excited state of the Cu+ /ZSM-5 catalyst . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3. The interaction of NO with Cu+ /ZSM-5 catalyst . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4. The photocatalytic reactivity of Cu+ /ZSM-5 catalyst for the decomposition of NO . . . . . . . . . .

∗ Corresponding author. Tel.: +81-72-254-9282; fax: +81-72-254-9910. E-mail address: [email protected] (M. Anpo).

1389-5567/02/$ 20.00 © 2002 Japanese Photochemistry Association. Published by Elsevier Science B.V. All rights reserved. PII: S 1 3 8 9 - 5 5 6 7 ( 0 2 ) 0 0 0 4 0 - 0

226 227 227 230 233 234 235 235 237 237 238 239 240 241 243 243

226

M. Matsuoka, M. Anpo / Journal of Photochemistry and Photobiology C: Photochemistry Reviews 3 (2003) 225–252

5. Design of Ag+ /ZSM-5 catalyst and its photocatalytic reactivity for the decomposition of NO . . . . 5.1. Preparation of the Ag+ /ZSM-5 catalyst and determination of its local structure . . . . . . . . . . . . . 5.2. Excited state of the Ag+ /ZSM-5 catalyst and its interaction with NO . . . . . . . . . . . . . . . . . . . . . . 5.3. The photocatalytic reactivity of Ag+ /ZSM-5 catalyst for the decomposition of NO . . . . . . . . . . 5.4. The photocatalytic decomposition of N2 O on the Ag+ /ZSM-5 catalyst . . . . . . . . . . . . . . . . . . . . . 6. The photocatalytic decomposition of N2 O on lanthanoid ion-exchanged mordenite catalysts . . . . . . 7. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. Introduction The application of photocatalysis to reduce toxic agents in air and water by developing catalysts that can utilize clean and abundant solar energy and convert it into useful chemical energy is one of the most important challenges facing scientists today. Recently, investigations to address such concerns using semiconducting TiO2 catalysts have been carried out intensively, resulting in various practical applications such as the decomposition of NOx in the air [1,2], the degradation of toxic organic impurities diluted in water [3,4], and the decomposition of water into H2 and O2 [5,6]. Studies on the dynamics and mechanisms behind such photocatalysis have shown that the electrons and holes produced by UV-irradiation of the semiconducting catalysts play a significant role in the formation of an active intermediate species such as O2 − , OH• , H• , etc. depending on the reaction conditions [7,8]. On the other hand, the unique and fascinating properties of zeolites involving transition metal ions within the zeolite frameworks or cavities have opened new possibilities for many applications, not only in catalysis [9–11] but also for various photochemical processes [12–16]. The transition metal ions in metallosilicate catalysts are considered to be highly dispersed at the atomic level and well defined, existing in a specific structure of the zeolite framework. According to the Löwenstein rules, the Al atoms within the zeolite framework cannot connect with each other directly, a property observed in most zeolites. If this is also the case for metallosilicate zeolites, the well-prepared zeolite sample should contain only isolated metal ions in their framework structures. This phenomenon is of great significance in the design of highly dispersed transition metal oxides such as Ti, V, Cr, Mo, etc. which can be excited under UV-irradiation to form the corresponding charge-transfer excited state involving an electron transfer from O2 − (l) to Mn + (l) : hν

(n−1)+

n+ [M(l) –O2− (l) ]→[M(l)

∗ –O− (l) ]

(M : Ti, V, Cr, Mo, . . . )

The high reactivities of these charge-transfer excited states, i.e. electron–hole pair states, which are localized quite near to each other as compared to the electron and hole produced in semiconducting materials, induce various significant photocatalytic reactions such as the decomposition of NO into N2 and O2 [17–24], the degradation of

244 245 245 246 248 248 249 249

organic impurities in water [25–27], the photo-oxidation reaction of hydrocarbons [28,29], the photo-induced metathesis reaction of alkanes [30,31], and the reduction of CO2 with H2 O to produce CH4 and CH3 OH [32–35]. These photocatalytic reactions were found to proceed with a high efficiency and selectivity, displaying quite different reaction mechanisms from those observed on semiconducting photocatalysts, in which the photoelectrochemical reaction mechanism or charge separation plays an important role in determining the efficiency. These findings indicate that a fundamental understanding of the coordination structure and electronic state of the active species is important in the design and development of applicable photocatalysts having high reactivity and selectivity. Furthermore, the counter-cations in zeolites can very easily be exchanged for a variety of other cations by an ionexchange method. The exchangeable sites are separated from each other within the zeolite cavities under well-controlled conditions, so that by applying an ion-exchange method, metal ions having photocatalytic capabilities, such as Cu+ or Ag+ ions, can be prepared within zeolites as new and unique photocatalysts. The condensation effect attributed to the larger surface area and specific channel systems of zeolites make such materials the most promising candidates for useful and highly reactive photocatalysts. In fact, we have reported that Cu+ or Ag+ ions can be successively exchanged into the zeolites in an isolated state to produce catalysts effective for various photocatalytic reactions such as the decomposition of NO or N2 O into N2 and O2 under UV-irradiation [36–41]. In these reactions, the electronic excited state of the Cu+ or Ag+ ions, i.e. an “s” electron and a “d” hole, which are produced on the identical metal ion (d9 s1 electronic configuration) under UV-irradiation, play an important role as an active species. In other words, as shown in the following equations, the localized electron–hole pair states of these metal ions (d9 s1 ) can induce unique and new photocatalytic reactions that cannot be realized on either semiconducting photocatalysts nor on highly dispersed metal oxide photocatalysts. hν

Cu+ ([Ar]3d10 )→Cu+∗ ([Ar]3d9 4s1 ) hν

Ag+ ([Kr]4d10 )→Ag+∗ ([Kr]4d9 5s1 ) Beside acting as a support to control the dispersion and local structure of the active sites, zeolites have many other

M. Matsuoka, M. Anpo / Journal of Photochemistry and Photobiology C: Photochemistry Reviews 3 (2003) 225–252

227

Scheme 1. Applications of the various properties of zeolites to the design and development of photocatalysts utilized to reduce toxic agents in the air or water.

unique and useful properties such as the condensation effect for reactant gasses or shape selectivity due to their unique pore structure and restricted molecular-scale size. As shown in Scheme 1, these unique properties can be used to design highly efficient and selective photocatalysts that work to reduce toxic agents in air and water by utilizing the zeolite cavity as a molecular reaction field. A combination of well-defined highly dispersed catalysts and multifunctional materials will make it possible to develop novel photocatalytic systems, such as an artificial soil that can completely decompose toxic organic compounds such as dioxin into completely oxidized nontoxic compounds such as CO2 , H2 O and HCl, as shown in Scheme 2. In this review article, we summarize the photocatalytic reactivities of zeolite catalysts into which various transition metal oxides such as Ti, V, Mo, Cr, as well as transition metal ions such as Cu+ and Ag+ , and rare earth cations such as Pr3+ have been incorporated into their framework structures or cavities at ambient temperatures. The local structures of the transition metal oxides and ions, as well as rare earth

cation species incorporated are discussed based on results obtained by means of various in situ spectroscopic techniques. Special attention has been focused on the relationship between the local structures of the active sites and their reactivities for various photocatalytic reactions.

2. Characterization of vanadium, molybdenum and chromium oxides photocatalysts incorporated into the framework of zeolites or mesoporous zeolites and their photocatalytic activities 2.1. Vanadium oxide (V-silicalite) catalyst Zeolites having transition metal cations in their frameworks have been the focus of much attention for their interesting and distinctive properties. So far, several types of such vanadium silicalite have been developed [42–45], and, the true chemical nature and reactivities of the vanadium silicalites, especially their photochemical properties, are

Scheme 2. Complete detoxification of dioxin using artificial soil that contains photocatalysts working under visible light irradiation.

228

M. Matsuoka, M. Anpo / Journal of Photochemistry and Photobiology C: Photochemistry Reviews 3 (2003) 225–252

beginning to be understood. According to our previous works, vanadium oxides highly dispersed on silica show unique photocatalytic properties towards the partial oxidation of alkenes as well as other reactions [46–48]. In this case, the dispersion level of vanadium oxide species is the key factor in controlling the photocatalytic reactivity. A carefully and well-prepared vanadium silicalite contains uniform vanadium oxide moieties in its framework, allowing the catalyst to exhibit unique and high photocatalytic reactions. The distinct characteristics of the vanadium silicalite (VS-1) photocatalyst and its photocatalytic reactivity for the direct decomposition of NO into N2 and O2 as well as the photocatalytic reaction of NO with propane at 298 K under UV-irradiation are discussed here at the molecular level using various in situ spectroscopic techniques such as photoluminescence, X-ray absorption fine structure (XAFS), including X-ray absorption near edge structure (XANES) and extended X-ray absorption fine structure (EXAFS) and electron spin resonance (ESR) spectroscopies. The vanadium silicalite catalyst (Si/V = 120) was hydrothermally synthesized in accordance with previous literature [49] using tetraethyl-orthosilicate and VOSO4 ·3H2 O as the starting materials and tetrapropyl ammonium hydroxide as the template. Prior to in situ spectroscopic measurements and photocatalytic reactions, the catalyst was calcined in the presence of O2 at 723 K and then degassed at 473 K. The photocatalytic reaction of NO in the absence and presence of propane were both carried out at 298 K with a high-pressure mercury lamp with a color filter (λ > 270 nm). The XRD patterns indicated that the VS-1 catalyst has a silicalite-1 structure with high crystallinity. The IR spectrum exhibited significant bands at around 960–970 cm−1 , indicating successful incorporation of the vanadium ions into the zeolite framework [50]. The local structure of the vanadium oxide species included within the zeolite framework was investigated using an in situ ESR technique in which the V4+ ions produced by photoreduction in the presence of H2 at 77 K were monitored. Photoreduction of catalyst at 77 K under UV-irradiation was so mild that no local structural rearrangement around the vanadium ions occurred, allowing us to obtain the information on the local structures. As shown in Fig. 1, the VS-1 catalyst shows an ESR signal of eight equally spaced lines due to the I = 7/2 spin of the V atoms. The ESR parameters of the signal (g = 1.915, g⊥ = 1.962; A = 148, A⊥ = 16.3 G) are in good agreement with those of V4+ ions in a tetrahedral coordination environment, indicating that the vanadium oxide species incorporated in the zeolite framework involves a tetrahedrally coordinated V5+ [51,52]. Fig. 2 shows the V K-edge XAFS (XANES and Fourier transform of EXAFS (FT-EXAFS)) spectra of the VS-1 catalyst (a and a ) and VO(O-i-C3 H7 )3 reference compound (b and b ). The XANES spectrum of VS-1 (a) is quite similar to that of the VO(O-i-C3 H7 )3 compound (b) which has tetrahedral coordination, and exhibits an intense pre-edge peak

Fig. 1. ESR spectrum of the VS-1 catalyst photo-reduced in the presence of 20 Torr of H2 at 77 K (UV-irradiation was carried out with a high-pressure mercury lamp, λ > 280 nm at 77 K).

due to the 1s–3d transition, indicating that the VS-1 catalyst consists of vanadium oxide species having tetrahedral coordination [53]. As shown in Fig. 2a , the Fourier transform of EXAFS of VS-1 exhibits only a single peak due to the presence of the neighboring oxygen atoms (V–O) at around 1.2 Å (without phase-shift correction), suggesting that vanadium ions are highly dispersed in the catalyst. A curve-fitting analysis of the EXAFS spectrum was also performed, and the best fitting of the Fourier-filtered EXAFS spectrum was obtained with one oxygen at a shorter V–O distance of 1.68 Å and three oxygen atoms at a long V–O distance of 1.78 Å. The results of this curve-fitting analysis are shown in Table 1. These

Fig. 2. XANES (left) and Fourier transformation of EXAFS (right) spectra of (a and a ) VS-1 catalyst and (b and b ) VO(O-i-C3 H7 )3 compound as the reference sample.

M. Matsuoka, M. Anpo / Journal of Photochemistry and Photobiology C: Photochemistry Reviews 3 (2003) 225–252 Table 1 The results of the curve-fitting of the V K-edge EXAFS data for the VS-1 catalyst Catalyst

Edge position (eV)

Shell

R (Å)

CN

σ2

VS-1 catalyst

5480.6

V–O (1) V–O (2)

1.68 1.78

0.92 2.98

0.0030 0.0025

(Å2 )

R: bond distances; CN: coordination number; σ 2 : Debye–Waller factor.

results clearly indicate that a vanadium oxide species that has a distorted tetrahedal structure is incorporated within the zeolite framework in a highly dispersed state. As shown in Fig. 3, the VS-1 catalyst exhibits a photoluminescence spectrum having a vibrational fine structure at around 400–650 nm upon excitation at around 300 nm due to the highly dispersed tetrahedrally coordinated V–O moieties in C3v symmetry [46–48]. The excitation and photoluminescence spectra are attributed to the following charge-transfer processes on the V–O moieties of the tetrahedral vanadate ions (VO4 3− ), involving an electron transfer from the O2 − to V5+ ions and a reverse radiative decay from the

Fig. 3. Ordinary photoluminescence spectrum observed for (A and a) VS-1 catalyst at 77 K; (B) its second derivative; and (A, b–g) the effect of the addition of NO and propane on the photoluminescence at 77 K. Pressure of NO added: (a) 0 Torr; (b) 1.8 Torr; (d) 2.5 Torr; (f) 3.5 Torr; (g) degassed at 298 K after the measurement of (f). Pressure of propane added: (c) 0.265 Torr; (e) 0.4 Torr.

229

charge-transfer excited triplet state to its ground state: hν

(V5+ ==O2− )
(V4+ –O− )∗ hν

Furthermore, from the second derivative of the photoluminescence spectrum, the energy separation between the (0 → 0) and (0 → 1) vibrational transitions was determined to be about 960 cm−1 , and was attributed to the vibrational transition in the V=O bond. The energy separation of 960 cm−1 obtained from the photoluminescence spectrum of VS-1 was found to be slightly different from that of the vanadium oxide species highly dispersed on silica (1035 cm−1 ) [47,48,54]. The V=O bond length of the V–O moieties within the zeolite framework structure of VS-1 obtained from the curve-fitting results of FT-EXAFS was 1.68 Å. The V=O bond length (1.68 Å) was found to be longer than that of the V–O moieties anchored on the surface of Vycor glass or silica (1.62 Å), showing that the O=V–O(Si) bond angle was smaller for the V–O moieties within the zeolite framework structure [55,56]. These results indicate that the highly dispersed vanadium oxide species is present within the zeolite framework as a tetrahedrally coordinated species, and the charge-transfer excited state of this oxide species is well localized in the shorter V=O bond. As shown in Fig. 3, the addition of propane or NO onto VS-1 leads to the efficient quenching of the photoluminescence as well as a shortening of the photoluminescence lifetime, indicating that propane or NO interacts with the excited state of the V–O moieties so that this catalyst may act as a photocatalyst for reactions with NO or propane [52,57]. Fig. 4 shows the reaction time profile of the yields of N2 in the photocatalytic decomposition of NO in the absence and the presence of propane on the VS-1 catalyst. UV-irradiation of the catalyst in the presence of NO leads

Fig. 4. Reaction time profiles of the photocatalytic decomposition of NO (a) with and (b) without propane on VS-1. Propane added: 1.97 × 10−4 mol/g-cat; NO: 1.82 × 10−4 mol/g-cat.

230

M. Matsuoka, M. Anpo / Journal of Photochemistry and Photobiology C: Photochemistry Reviews 3 (2003) 225–252

to the photocatalytic decomposition of NO while the evolution of N2 is also observed together with O2 and N2 O as minor products. The reaction proceeds much more efficiently under a mixture of NO and propane while leading to the formation of propylene, ethylene and oxygen-containing compounds such as CH3 CHO and CO2 in addition to the evolution of N2 . The yield as well as the selectivity for the formation of N2 increased in proportion to the amount of added propane, while the yield of the formation of N2 O decreased and the yields of the oxygen-containing products increased [52,57]. Furthermore, the turnover frequency of the catalyst for the decomposition of NO exceeded unity after prolonged UV-irradiation. These results clearly indicate that this reaction proceeds photocatalytically and that the photocatalytic reduction of NO proceed efficiently in the presence of propane on the VS-1 catalyst. The reduction of NO with propane on VS-1 was about five times higher than on the V oxide photocatalyst anchored onto porous Vycor glass or silica. These results indicate that the difference in the coordination structure and the electronically excited states of the V species are the main factors controlling the reactivity for the photocatalytic reduction of NO. The efficiency of the photocatalytic reduction of NO was found to depend strongly on the kind of hydrocarbons used, such as methane and ethane, and among these hydrocarbons, propane showed the highest enhancement in the reaction rate, indicating that the abstraction of the H atom from hydrocarbons by the charge-transfer excited state, (V4+ –O− )∗ , plays a significant role in the enhancement of the reaction [29,58,59]. The importance of the abstraction of a H atom from the hydrocarbons is also confirmed by the fact that the presence of CO does not enhance the photocatalytic reduction of NO. The photo-induced adsorption of NO and propane were also investigated. Only small amounts of NO were adsorbed onto the catalyst under UV-irradiation, while the adsorption of NO was greatly enhanced in the co-existence of propane. These results suggest the importance of the intermediate species formed from NO and hydrocarbon radicals, which was subsequently followed by further reactions with NO to produce N2 as well as oxygen-containing products such as CH3 CHO and CO2 . In conclusion, in situ characterizations of the VS-1 catalyst at the molecular level clearly provides an insight into the local structure of the active sites and the photocatalytic reaction properties at ambient temperature. Vanadium oxides incorporated into the zeolite framework were found to exist as isolated tetrahedrally coordinated V–O moieties having a terminal vanadyl group (V=O), and the coordination structure was quite different from that of the isolated tetrahedrally coordinated V–O moieties anchored onto porous Vycor glass or silica. The charge-transfer excited state of these V–O moieties, (V4+ –O− ), shows a high photocatalytic reactivity for the decomposition of NO into N2 and O2 . In the presence of propane, the photocatalytic reaction was dramatically enhanced. The reaction rate for the

photocatalytic reduction of NO with propane on VS-1 was about five times higher than on the V oxide photocatalyst anchored onto porous Vycor glass or silica. From these results, it can be emphasized that the differences in the coordination structures and the electronically excited states of the V species are the main factors controlling the photocatalytic activity for the direct decomposition as well as reduction of NO. The incorporation of the V species into the zeolite framework makes it possible to control the local structure as well as the electronic state of the V species on a molecular scale. Thus, using zeolites as host materials for the active sites in various photocatalytic reactions can be considered as an especially promising way to design and develop unique and well-defined photocatalysts having high activity and selectivity. 2.2. Molybdenum oxide (Mo-MCM-41) catalyst Transition metal ions-containing zeolites such as MFI or MEL type zeolites have received attention as effective and efficient catalysts with high selectivities for various reactions. However, the concentration of the Mo oxide species that can be incorporated in the framework of zeolite is limited due to the great strain that occurs by the insertion of such large transition metal ions as Mo, W and Nb within the rigid zeolite framework. On the other hand, Mo ions can be easily incorporated within the framework of MCM-41 due to its flexibility in the structure. In fact, it is possible to incorporate highly dispersed Mo oxide species within the MCM-41 structure with high loading amounts, at least up to 1.0 wt.% Mo, while only 0.1 wt.% of Mo can be anchored onto the SiO2 surface in an highly dispersed state using a facile reaction of MoCl5 with the surface OH groups [60]. Furthermore, it has been reported that highly dispersed Mo oxide species anchored onto SiO2 exhibit high catalytic activity, not only for thermal reactions such as hydrogenation, oxidation and metathesis but also photocatalytic reactions such as the partial oxidation of hydrocarbons [28] and metathesis [30,31]. Thus, Mo-MCM-41, which incorporates highly dispersed Mo oxide with high concentration, can act as efficient and unique photocatalysts. In this chapter, special attention is focused on the photocatalytic decomposition reaction of NO in the presence of hydrocarbons or CO molecules on Mo-MCM-41 catalysts. The Mo-MCM-41 mesoporous molecular sieves were synthesized using tetraethylorthosilicate (TEOS) and (NH4 )6 Mo7 O24 ·4H2 O as the starting materials and cetyltrimethylammonium bromide (CTMABr) as the template, in accordance with previous literature [61]. After the products were recovered by filtration, washed with distilled water several times and dried at 373 K, calcination of the samples was carried out under a dry flow of air at 773 K. Prior to spectroscopic measurements and photocatalytic reactions, the catalysts were degassed at 773 K and calcined in O2 at 773 K and then degassed at 473 K.

M. Matsuoka, M. Anpo / Journal of Photochemistry and Photobiology C: Photochemistry Reviews 3 (2003) 225–252

231

Fig. 6. Effect of the addition of NO on the phosphorescence spectrum and corresponding excitation spectrum of the Mo-MCM-41 (1.0 wt.% Mo) catalyst. Pressure of added NO: (A and a) 0 Torr; (b) 0.07 Torr; (C and c) 0.4 Torr; (d) excess; (e) degassed after (d).

Fig. 5. XANES and FT-EXAFS spectra of the Mo-MCM-41 catalysts with various Mo loadings. Mo loadings were: (A and a) 1.0 wt.% Mo; (B and b) 2.0 wt.% Mo; (C and c) 4.0 wt.% Mo.

The results of the XRD patterns and the BET surface area of the Mo-MCM-41 catalyst indicated that they have a hexagonal lattice having mesopores larger than 20 Å and that they possess a high BET surface area (1000 m2 /g) as compared with amorphous silica (300 m2 /g), so that they can be considered as effective photocatalysts. Fig. 5 shows XAFS (XANES and FT-EXAFS) spectra of the Mo K-edge spectra of Mo-MCM-41. A characteristic feature of the XANES spectra of Mo-MCM-41 is the appearance of a pre-edge peak due to the 1s–4d transition of the Mo atoms, suggesting the presence of a terminal Mo–oxonium group (Mo=O) [62]. It was also found that the shape of the XANES spectra are quite similar to that of the K2 MoO4 compound having a tetrahedral coordination, indicating that Mo oxides with tetrahedral coordination are formed on MCM-41. The intensity of the pre-edge peak is the highest for Mo-MCM-41 (1.0 wt.% Mo); however, it decreases slightly and the shape of the XANE spectra changes when the Mo content is increased from 1.0 to 4.0 wt.% Mo. Furthermore, it was found that the FT-EXAFS of Mo-MCM-41 (1.0 wt.% Mo) exhibits only a well-resolved peak due to the neighboring oxygen atoms (Mo–O) at ca. 0.8–2.0 Å (without phase-shift correction) while an additional peak can be observed at ca. 3.0 Å due to the Mo–O–Mo bond or Mo–O–Si bond in the case of Mo-MCM-41 (4.0 wt.% Mo). The curve-fitting analysis showed that isolated tetrahedrally coordinated Mo oxides having two shorter Mo–O bonds (1.68 Å) and two longer ones (1.88 Å) are formed in Mo-MCM-41 (1.0 wt.% Mo), while oligomeric tetrahedral Mo oxides (MoO4 2− )n with an

additional Mo–O–Mo bonds (3.19 Å) and Mo–O–Si (3.15 Å) are formed in Mo-MCM-41 (4.0 wt.% Mo). As shown in Fig. 6, Mo-MCM-41 (1.0 wt.% Mo) exhibits photoluminescence at around 400–600 nm upon excitation at around 295 nm (defined as X), which coincides with the photoluminescence of the tetrahedrally coordinated Mo oxide species highly dispersed on SiO2 [31]. The excitation and emission spectra are attributed to the following charge-transfer processes on the Mo–O moieties of the isolated tetrahedral molybdate ions (MoO4 2− ), involving an electron transfer from the O2 − to Mo6+ ions and a reverse radiative decay from the charge-transfer excited triplet state. The width and the wavelength at the maximum intensity of the emission band do not change upon varying the excitation wavelength, indicating that there is only one luminescent moiety. hν

hν

[Mo6+ ==O2− ]→[Mo5+ ==O− ]→[Mo6+ ==O2− ] On the other hand, as shown in Fig. 7, there exist at least two luminescent species (the absorption spectrum can be deconvoluted into two components having wavelength regions of X and Y: 295 and 310 nm, respectively) on Mo-MCM-41 (4.0 wt.% Mo) [63]. The increase of Mo content leads to the formation of not only the emitted X site with a photoluminescence lifetime of 2.25 ms but also another emitting Y site with low photoluminescence yields and short photoluminescence lifetime (0.91 ms). Taking into account the results of XAFS measurements, only the isolated tetrahedrally coordinated Mo oxides are formed in lower Mo loadings, while isolated tetrahedrally coordinated Mo oxides as well as oligomeric tetrahedral (MoO4 2− )n species are formed in higher Mo loadings. As shown in Fig. 6, the addition of propane or NO onto Mo-MCM-41 (1.0 wt.% Mo) leads to the efficient quenching of the photoluminescence as well as a shortening of the photoluminescence lifetime. On the other hand, as shown in

232

M. Matsuoka, M. Anpo / Journal of Photochemistry and Photobiology C: Photochemistry Reviews 3 (2003) 225–252

Fig. 7. Effect of the addition of NO on the photoluminescence spectrum and corresponding excitation spectrum of the Mo-MCM-41 (4.0 wt.% Mo) catalyst. Pressure of added NO: (A and a) 0 Torr; (b) Torr 0.6; (C and c) 4 Torr; (d) excess; (e) degassed after (d). Spectrum (A) can be deconvoluted into (X) and (Y).

Fig. 7, in the case of Mo-MCM-41 (4.0 wt.% Mo), NO or propane interacts with the X site more efficiently than the Y site. It can, therefore, be expected that isolated tetrahedral Mo oxide species would show high photocatalytic reactivity as compared to (MoO4 2− )n species. Measurements on the photocatalytic reaction of NO in the presence of propane were performed on Mo-MCM-41. Under UV-irradiation, the photocatalytic reaction of NO with propane was found to proceed efficiently, leading to the formation of N2 and oxygen-containing compounds such as CH3 COCH3 and CO2 . Fig. 8 shows the relationship between the yields of N2 , or acetone formation and relative intensity of the absorption spectra observed in the region of 295 nm for Mo-MCM-41 (0.5, 1.0, 2.0 and 4.0 wt.% Mo). The in-

tensities of the absorption spectra at 295 nm have a good relationship with the yields of N2 or acetone formation, suggesting that the charge-transfer excited triplet state of the tetrahedrally coordinated Mo oxide species in a highly dispersed state plays a significant role in the reaction. After UV-irradiation of Mo-MCM-41 in the presence of propane, its subsequent evacuation at 295 K did not lead to the recovery of the original photoluminescence intensity but to the appearance of the ESR signals due to Mo5+ ions. This shows that the charge-transfer excited triplet state of the Mo oxide species abstracts the H atom from propane to form Mo5+ ion and the hydrocarbon radical [57]. Furthermore, only small amounts of NO or propane were photoadsorbed under UV-irradiation; however, a great enhancement of the photoadsorption occurred in the presence of a mixture of NO and propane. Taking these results into consideration, the following reaction mechanism can be proposed, that is, the intermediate species formed between NO and the hydrocarbon radicals, which is formed by the H abstraction of the photo-excited Mo oxide species from propane, subsequently reacts with NO to produce N2 as well as oxygen-containing compounds. It was also found that the photocatalytic decomposition reactions of NO were dramatically enhanced in the presence of CO, leading to the formation of N2 and CO2 . Fig. 9 shows the relationship between the yields of N2 formation for the decomposition of NO on Mo-MCM-41 (0.5–4.0 wt.% Mo) in the presence of CO and the relative intensity of the absorption spectra observed in the total region (X and Y) of the catalyst. The yields of N2 have a good relationship with the intensities of the absorption spectra in the total region of X and Y as well as with the amount of Mo4+ ions generated through the photoreduction of Mo6+ with CO (the number of Mo4+ ions are estimated by the number of photo-formed CO2 molecules) [62]. These results indicate

Fig. 8. Relationships between the yields of N2 (shaded bars), acetone (non-shaded bars) formation and the relative intensity of the absorption in the region of 295 nm (X) calculated by the deconvolution of the original excitation spectrum of the Mo-MCM-41 catalysts (0.5, 1.0, 2.0 and 4.0 wt.% Mo). Added propane or NO: 180 ␮mol/g-cat.

M. Matsuoka, M. Anpo / Journal of Photochemistry and Photobiology C: Photochemistry Reviews 3 (2003) 225–252

233

Fig. 9. Relationships between the yields of N2 formation for the photocatalytic decomposition reactions of NO in the presence of CO for 3 h (left), amount of generated Mo4+ ions in the photoreduction of Mo6+ with CO under UV-irradiation for 0.5 h, the relative intensity of the photoluminescence (dashed line; a) and the yields of N2 per 1 g of Mo (solid line; b) of Mo-MCM-41 catalysts (0.5, 1.0, 2.0 and 4.0 wt.% Mo). Added NO or CO: 180 ␮mol/g-cat.

that the charge-transfer excited triplet states of both isolated tetrahedral Mo oxides and (MoO4 2− )n play a significant role in the reaction. UV-irradiation of Mo-MCM-41 in the presence of CO alone and its subsequent evacuation at 293 K led to an efficient quenching of the photoluminescence. However, under identical conditions, no ESR signals due to the Mo5+ ions were detected, suggesting that the charge-transfer excited triplet state, i.e. the [Mo5+ –O− ]∗ complex react with CO to form Mo4+ ions and CO2 . It was found that the Mo4+ ions thus formed reacts efficiently with NO and N2 O under dark conditions to produce N2 O and N2 , respectively, and also to the reoxidation of Mo4+ to Mo6+ ions, which is indicated by the reappearance of the photoluminescence of Mo6+ ions after the addition of NO and N2 O. From these results, it can be concluded that the catalytic cycle shown in Scheme 3 plays a role in the photocatalytic decomposition of NO in the presence of CO, that is, Mo4+ ions that are formed through the reaction of the charge-transfer excited

Scheme 3. Catalytic cycle for the photocatalytic decomposition reaction of NO in the presence of CO.

triplet state of Mo6+ ==O− 2 with CO are reoxidized back to the original Mo6+ species in the presence of NO or N2 O. 2.3. Chromium oxide (Cr-HMS) catalyst Highly dispersed Mo or Cr oxides catalysts have been shown to exhibit high activities for various photocatalytic reactions such as the photo-oxidation reaction of hydrocarbons [28] or the photo-induced metathesis reaction of alkanes [30,31]. Recently, it was found that the Cr-containing mesoporous zeolite (Cr-HMS) shows photocatalytic activities for the photocatalytic decomposition of NO into N2 and O2 and the partial oxidation of propane with O2 under UV or even visible light irradiation. In this article, special attention is focused on local structure Cr oxide species on Cr-HMS and their photocatalytic activities for NO decomposition under UV and visible light irradiation [64,65]. Cr-HMS mesoporous molecular sieves (0.02, 0.2, 1.0, 2.0 wt.% as Cr) were synthesized using tetraethyl orthosilicate and Cr(NO3 )3 ·9H2 O as the starting materials and dodecylamine as a template [66,67]. Calcination of the sample was carried out in a flow of dry air at 773 K. Prior to spectroscopic measurements and photocatalytic reactions, the catalysts were degassed at 723 K, heated in O2 at the same temperature and then finally evacuated at 473 K. The results of the XRD analysis indicated that Cr-HMS have a mesopore structure and that the Cr oxide moieties are highly dispersed in the framework of HMS [66], while no other phases are formed. Cr-HMS exhibited a sharp, intense pre-edge peak in the XANES region which is characteristic of Cr oxide moieties in a tetrahedral coordination [67]. In the Fourier transforms of the EXAFS spectra, only a single peak due to the neighboring oxygen atoms was observed, showing that Cr ions are highly dispersed in the Cr-HMS. Analysis of EXAFS spectra of Cr-HMS indicated that tetrahedrally coordinated Cr oxide (chromate) moieties having two terminal

234

M. Matsuoka, M. Anpo / Journal of Photochemistry and Photobiology C: Photochemistry Reviews 3 (2003) 225–252

Fig. 10. UV-Vis spectra of Cr-HMS catalysts: (a) 2.0 wt.%; (b) 1.0 wt.%; (c) 0.2 wt.% as Cr.

Cr=O bonds existed in an isolated state [two oxygen atoms (Cr=O) at 1.57 Å and two oxygen atoms (Cr–O) at 1.82 Å. As shown in Fig. 10, the UV-Vis spectra of the Cr-HMS catalysts exhibit three distinct absorption bands at around 250, 360 and 480 nm, which can be assigned to charge-transfer from O2 − to Cr6+ of the tetrahedrally coordinated Cr oxide moieties [68]. Absorption bands above 550 nm assigned to the dichromate or Cr2 O3 cluster cannot be observed, indicating that the tetrahedrally coordinated Cr oxide species exists in an isolated state. Cr-HMS exhibited photoluminescence spectra at ca. 550–750 nm upon excitation of the absorption (excitation) bands at ca. 250, 360 and 480 nm. These absorption and photoluminescence spectra are similar to those obtained with well-defined, highly dispersed Cr oxides anchored onto Vycor glass or silica [69,70] and can be attributed to charge-transfer processes on tetrahedrally coordinated Cr oxide moieties involving an electron transfer from O2 − to Cr6+ and a reverse radiative decay, respectively. UV-irradiation (λ > 270 nm) of the Cr-HMS in the presence of NO in the gas phase at 275 K led to the photocatalytic decomposition of NO and the evolution of N2 , N2 O and O2 . The Cr-HMS also showed photocatalytic reactivity even under visible light irradiation (λ > 450 nm). As shown in Fig. 11, the N2 yields increase linearly with the irradiation time. The reaction stopped immediately when irradiation ceased. After prolonged irradiation, the amount of decomposed NO to form N2 per total number of Cr ions included within the catalyst exceeded unity. These results clearly indicate that the presence of both Cr oxide species (included within the HMS) as well as light irradiation are indispensable for the photocatalytic reaction and that the direct decomposition of NO occurs photocatalytically. Although the reaction rate under visible light irradiation is less than that under UV-irradiation, the selectivity for N2 formation (97%) under visible light irradiation is higher than that of UV-irradiation (45%). These results indicate that Cr-HMS can absorb visible light and act as an efficient photocatalyst not only under UV light but also under visible light irradi-

Fig. 11. Reaction time profile of N2 formation in the photocatalytic decomposition of NO on the Cr-HMS catalyst at 273 K (2.0 wt.% as Cr) under UV light irradiation (a: λ > 270 nm) and visible light irradiation (b: λ > 450 nm).

ation, and in particular, Cr-HMS can be useful to form N2 under visible light irradiation. The addition of NO to the Cr-HMS led to an efficient quenching of the photoluminescence spectrum of the catalyst, indicating that the charge-transfer excited state of the tetrahedrally coordinated isolated Cr oxide moieties, (Cr5+ –O− )∗ , easily interact with NO, and this photo-excited species plays an important role in the photocatalytic reaction under UV and visible light irradiation. These results indicate that the design of transition metal ion-containing zeolite catalysts is one of the most promising ways to develop unique photocatalytic systems that can convert abundant visible or solar light energy into useful chemical energy.

3. Characterization of titanium oxide photocatalysts anchored in the cavities of various zeolites and their photocatalytic activities The design of highly efficient and selective photocatalytic systems that work without any loss in the use of solar energy through chemical storage is of vital interest. In particular, the development of efficient photocatalytic systems that are able to decompose NO directly into N2 and O2 or to reduce CO2 with H2 O into chemically valuable compounds such as CH4 or CH3 OH is among the most desirable and challenging goals in the research on environmentally friendly catalysts [71–74]. Recently, we have reported that these photocatalytic reactions successfully proceed on powdered TiO2 at room temperature [75–79]. Using various types of well-characterized powdered TiO2 catalysts, the effect of the structure of the photocatalysts on the catalytic activity was investigated and it was found that extremely small TiO2 particles, having large band gaps, show the highest activity [77].

M. Matsuoka, M. Anpo / Journal of Photochemistry and Photobiology C: Photochemistry Reviews 3 (2003) 225–252

Concurrently, the activity of a highly dispersed titanium oxide catalyst anchored onto silica glass or zeolite was also investigated [80–83] and it was found that the highly dispersed titanium oxide catalyst exhibits a unique, high photocatalytic reactivity as compared to bulk TiO2 powder, i.e. a selectivity for the formation of N2 and O2 in the NO decomposition reaction as well as the formation of CH3 OH in the reduction of CO2 with H2 O were increased as the extent of the dispersion of the titanium oxide became higher [84–87]. These results indicate that by using zeolites as supports, highly dispersed titanium oxides can be produced, leading to the development of environmentally friendly photocatalytic systems having high catalytic efficiency, selectivity and other fascinating properties such as shape selectivity and a reactant gas condensation effect, which can be derived from the physicochemical properties of zeolites. In the following section, the distinct characteristics of the titanium oxide species anchored onto the zeolite framework (Ti oxide/Y-zeolite, Ti-MCM-41 and Ti-MCM-48) and their photocatalytic activities for the decomposition reaction of NO as well as for the reduction of CO2 with H2 O will be discussed at the molecular level using various in situ spectroscopic techniques such as photoluminescence, UV-Vis, XAFS (XANES and EXAFS), and ESR spectroscopies. The relationship between the structure of the titanium oxide species and the reaction activity as well as the selectivity will be summarized. 3.1. Photocatalytic reduction of CO2 with H2 O on titanium oxide photocatalysts anchored in the cavities of various zeolites A Ti oxide/Y-zeolite (1.1 wt.% as TiO2 ) was prepared by ion-exchange with an aqueous titanium ammonium oxalate solution using Y-zeolite sample (SiO2 /Al2 O3 = 5.5) (ex-Ti oxide/Y-zeolite). The Pt-loaded ex-Ti oxide/Y-zeolite (1.0 wt.% as Pt) was prepared by impregnation with an aqueous solution of H2 PtCl6 . Ti oxide/Y-zeolites having different Ti contents (1.0 and 10 wt.% as TiO2 ) were prepared by impregnating the Y-zeolite with an aqueous solution of titanium ammonium oxalate (imp-Ti oxide/Y-zeolite). Ti-MCM-41 (Si/Ti = 100) and Ti-MCM-48 (Si/Ti = 80) were hydrothermally synthesized according to procedures reported previously [86]. A TiO2 powdered catalyst, JRC-TIO-4 (anatase 92%, rutile 8%), supplied by the Catalysis Society of Japan was used. UV-irradiation of the catalysts in the presence of CO2 and gaseous H2 O was carried out using a high-pressure Hg lamp (λ > 280 nm) at 328 K. The reaction products were analyzed by gas chromatography. 3.1.1. Preparation of the Ti oxide/Y-zeolite by an ion-exchange method and their photocatalytic activity for the reduction of CO2 with H2 O UV-irradiation of powdered TiO2 and Ti oxide/Y-zeolite catalysts prepared by ion-exchange or impregnation methods in the presence of a mixture of CO2 and H2 O led to the

235

Fig. 12. Reaction time profiles of the photocatalytic reduction of CO2 with H2 O to produce (a) CH4 and (b) CH3 OH on the ex-Ti oxide/Y-zeolite catalyst.

evolution of CH4 and CH3 OH in the gas phase at 328 K, as well as trace amounts of CO, C2 H4 and C2 H6 . The evolution of small amount of O2 was also observed. As shown in Fig. 12, the yields of these photo-formed products increase linearly as a function of the UV-irradiation time and the reaction immediately ceases when irradiation is discontinued, indicating the photocatalytic reduction of CO2 with H2 O on the catalysts. The specific photocatalytic reactivities for the formation of CH4 and CH3 OH are shown in Fig. 13. It is clear that the photocatalytic reaction rate and selectivity for the formation of CH3 OH depend strongly on the type of catalyst. It can be seen that the specific photocatalytic

Fig. 13. Product distribution of the photocatalytic reduction of CO2 with H2 O on (a) anatase TiO2 powder; (b) imp-Ti oxide/Y-zeolite (10 wt.% as TiO2 ); (c) imp-Ti oxide/Y-zeolite (1.0 wt.% as TiO2 ); (d) ex-Ti oxide/Y-zeolite (1.1 wt.% as TiO2 ); and (e) Pt-loaded ex-Ti oxide/Y-zeolite catalysts.

236

M. Matsuoka, M. Anpo / Journal of Photochemistry and Photobiology C: Photochemistry Reviews 3 (2003) 225–252

Fig. 15. Ordinary photoluminescence spectrum observed for: (a) ex-Ti oxide/Y-zeolite catalyst; (EX) its corresponding excitation spectrum; (b and c) the effects of the addition of CO2 and H2 O, respectively; (d) the effect of the loading of Pt on the photoluminescence spectrum at 77 K. Excitation at 290 nm; emission monitored at 490 nm. Amounts of added CO2 : (b) 8.5 mmol/g; H2 O: (c) 2.9 mmol/g.

Fig. 14. XANES (left) and FT-EXAFS (right) spectra of (a and a ) anatase TiO2 powder; (b and b ) imp-Ti oxide/Y-zeolite (10 wt.% as TiO2 ); (c and c ) imp-Ti oxide/Y-zeolite (1.0 wt.% as TiO2 ); (d and d ) the ex-Ti oxide/Y-zeolite; (e and e ) the Pt-loaded ex-Ti oxide/Y-zeolite catalysts.

reactivities of the Ti oxide/Y-zeolite catalysts, which have been normalized by unit gram of Ti in the catalysts, are much higher than bulk TiO2 . The ex-Ti oxide/Y-zeolite exhibits a high reactivity and a high selectivity for the formation of CH3 OH, while the formation of CH4 was found to be the major reaction on bulk TiO2 as well as on the imp-Ti oxide/Y-zeolite. Fig. 14 shows the XANES spectra of the Ti oxide/Y-zeolite catalysts. The XANES spectra of the Ti oxide catalysts at the Ti K-edge show several well-defined pre-edge peaks, which are related to the local structures surrounding the Ti atom. The ex-Ti oxide/Y-zeolite exhibits an intense single pre-edge peak, indicating that the Ti oxide species in this catalyst has a tetrahedral coordination [18,86]. On the other hand, the imp-Ti oxide/Y-zeolite exhibits three characteristic weak pre-edge peaks attributed to crystalline anatase TiO2 . Fig. 14 also shows the FT-EXAFS spectra of the catalysts, distances uncorrected for phase shifts. The ex-Ti oxide/Y-zeolite exhibits a peak only at around 1.6 Å assigned to the neighboring oxygen atoms (Ti–O), indicating the presence of isolated Ti oxide species in this catalyst. From the curve-fitting analysis of the EXAFS spectra, it

was found that the ex-Ti oxide/Y-zeolite catalyst consists of four-coordinate titanium ions with a coordination number (N) of 3.7 and an atomic distance (R) of 1.78 Å. On the other hand, the imp-Ti oxide/Y-zeolite catalysts exhibit an intense peak at around 2.7 Å assigned to the neighboring titanium atoms (Ti–O–Ti), indicating the aggregation of the Ti oxide species in these catalysts. Fig. 15 shows that the ex-Ti oxide/Y-zeolite catalyst exhibits a photoluminescence spectrum at around 490 nm by excitation at around 290 nm at 77 K. The observed photoluminescence and absorption bands are in good agreement with those previously observed with highly dispersed tetrahedrally coordinated Ti oxides prepared in silica matrices [80–83]. We can, therefore, conclude that the observed photoluminescence spectrum can be attributed to the radiative decay process from the charge-transfer excited state to the ground state of the highly dispersed Ti oxide species in tetrahedral coordination, as shown below. hν

(Ti4+ –O2− )
(Ti3+ –O− )∗ hν

On the other hand, the imp-Ti oxide/Y-zeolite catalysts did not exhibit any photoluminescence. Thus, these results clearly indicate that the ex-Ti oxide/Y-zeolite catalyst consists of highly dispersed isolated tetrahedral Ti oxide species, while the imp-Ti oxide/Y-zeolite catalysts involve the aggregated octahedral Ti oxide species, which do not exhibit any photoluminescence spectrum. As shown in Fig. 15, the addition of H2 O or CO2 molecules to the ex-Ti oxide/Y-zeolite catalyst leads to an efficient quenching of the photoluminescence. The lifetime of the charge-transfer excited state was also found to be

M. Matsuoka, M. Anpo / Journal of Photochemistry and Photobiology C: Photochemistry Reviews 3 (2003) 225–252

shortened by the addition of CO2 or H2 O, its extent depending on the amount of added gasses. Such an efficient quenching of the photoluminescence with CO2 or H2 O suggests not only that the tetrahedrally coordinated Ti oxide species is located at positions accessible to the added CO2 or H2 O but also that added CO2 or H2 O interacts and/or reacts with the Ti oxide species in both its ground and excited states. UV-irradiation of the anchored Ti oxide catalyst in the presence of CO2 and H2 O at 77 K led to the appearance of ESR signals due to the Ti3+ ions, H atoms, and carbon radicals [75,87]. From these results, the reaction mechanism in the photocatalytic reduction of CO2 with H2 O on the highly dispersed Ti oxide catalyst can be proposed in the following way: CO2 and H2 O molecules interact with the excited state of the photo-induced (Ti3+ –O− )∗ species, and the reduction of CO2 and the decomposition of H2 O proceed competitively. Furthermore, H atoms and OH• radicals are formed from H2 O and these radicals react with the carbon species formed from CO2 to produce CH4 and CH3 OH. 3.1.2. Preparation of Ti-mesoporous zeolites by hydrothermal synthesis and their photocatalytic activity for the reduction of CO2 with H2 O In situ photoluminescence, ESR, UV-Vis and XAFS investigations indicated that the Ti oxide species in the Ti-mesoporous zeolites (Ti-MCM-41 and Ti-MCM-48) and in the TS-1 zeolite are highly dispersed within the zeolite framework and exist in a tetrahedral coordination. Upon excitation with UV light at around 260–290 nm, these zeolites exhibit photoluminescence spectra at around 450–650 nm. The addition of CO2 or H2 O to these zeolites results in a significant quenching of these photoluminescence spectra, suggesting the excellent accessibility of the Ti oxide species to CO2 and H2 O. In addition, quenching with H2 O is much more effective than with CO2 , reflecting the stronger interaction of H2 O with the Ti oxide species [85–87]. UV-irradiation of the Ti-mesoporous zeolites and the TS-1 zeolite in the presence of CO2 and H2 O also led to the formation of CH3 OH and CH4 as the main products. The yields of CH4 and CH3 OH per unit weight of the Ti-based catalysts are shown in Fig. 16. It can be seen that Ti-MCM-48 exhibits much higher reactivity than either TS-1 or Ti-MCM-41. Besides the higher dispersion state of the Ti oxide species, other distinguishing features of these zeolite catalysts are: TS-1 has a smaller pore size (ca. 5.7 Å) and a three-dimensional channel structure; Ti-MCM-41 has a large pore size (>20 Å) but one-dimensional channel structure; and Ti-MCM-48 has both a large pore size (>20 Å) and three-dimensional channels. Thus, the higher reactivity and higher selectivity for the formation of CH3 OH observed with the Ti-MCM-48 zeolite than with the other catalysts may be due to the combined contribution of the high dispersion state of the Ti oxide species and the large pore size with a three-dimensional channel structure. The results observed strongly indicate that mesoporous zeolites with highly dispersed Ti oxide species in their framework are

237

Fig. 16. Product distribution of the photocatalytic reduction of CO2 with H2 O on (a) anatase TiO2 powder; (b) TS-1; (c) Ti-MCM-41; (d) Ti-MCM-48; and (e) the Pt-loaded Ti-MCM-48 catalysts.

promising candidates to serve as effective photocatalysts for the photoreduction of CO2 with H2 O [84–87]. 3.1.3. Effect of Pt loading on the photocatalytic activity and selectivity of Ti-containing zeolite catalysts The effect of Pt-loading on the photocatalytic reactivity of Ti-containing zeolite has also been investigated and the changes in the yields of CH4 and CH3 OH formation are shown in Figs. 13 and 16. Although the addition of Pt to the Ti-containing zeolites is effective for an increase in the photocatalytic reactivity, only the formation of CH4 is promoted, accompanied by a decrease in the CH3 OH yields. The absorption spectra of the Pt-loaded catalysts were the same as those observed with the original Ti-containing zeolite without Pt-loading. As shown in Fig. 14, the Pt-loaded catalyst also exhibits the same pre-edge peak in the XANES spectra and the same Ti–O-bonding peak in the FT-EXAFS spectra as those of the original Ti-containing zeolite. Furthermore, as shown in Fig. 15, Pt-loading on the Ti-containing zeolite catalyst leads to an efficient quenching of the photoluminescence, accompanied by the shortening of its lifetime. Because the results obtained by XAFS and absorption measurements indicate that the local structure of the Ti oxide species dispersed in the zeolite was not altered by the Pt loading, the effective quenching of the photoluminescence can be attributed to the electron transfer from the photo-excited Ti oxide species to metallic Pt that exists in the neighborhood of the Ti oxide species. The electrons are easily transferred from the charge-transfer excited state of the Ti oxide species, the electron–hole pair state of (Ti3+ –O− )∗ , to the Pt moieties while the holes remain in the Ti oxide species, resulting in the charge separation of electrons and holes from the photo-formed electron–hole pair states. As result, on the Pt-loaded Ti-containing zeolite catalyst, photocatalytic reactions that proceed in the same

238

M. Matsuoka, M. Anpo / Journal of Photochemistry and Photobiology C: Photochemistry Reviews 3 (2003) 225–252

ence of titanium oxides included within the zeolites as well as the UV-irradiation light are indispensable for the photocatalytic reaction to take place, while the direct decomposition of NO to produce N2 , O2 and N2 O occurs photocatalytically on the surface of the titanium oxide catalyst. The photocatalytic reactivities of various titanium oxide catalysts for the direct decomposition of NO are shown in Table 2. Of special interest is the comparison of the photocatalytic activities of the Ti oxide/Y-zeolite catalysts with that of the widely used bulk TiO2 powdered catalyst. It can be seen that the specific photocatalytic reactivities of the Ti oxide/Y-zeolite catalysts, which have been normalized for the unit amount of TiO2 in the catalysts, are much higher than that for the bulk TiO2 catalysts [18,19]. Table 2 also shows the yields of the photo-formed N2 and N2 O (efficiency) and their distribution (selectivity) in the photocatalytic decomposition of NO on various types of titanium oxide catalysts. From Table 2, it is clear that the efficiency and selectivity for the formation of N2 strongly depend on the type of catalyst. The ex-Ti oxide/Y-zeolite catalyst exhibits a high reactivity and a high selectivity for the formation of N2 , while the formation of N2 O was found to be the major reaction on the bulk TiO2 catalyst as well as on the imp-Ti oxide/Y-zeolite catalyst. Thus, the results obtained with the ex-Ti oxide/Y-zeolite clearly show the large difference in selectivity as well as efficiency with the imp-Ti oxide/Y-zeolite and the bulk TiO2 catalyst [18,19,88]. XAFS (XANES and EXAFS) investigations of Ti oxide catalysts at the Ti K-edge were performed, and the results revealed that the titanium oxide species has a tetrahedral coordination in the case of the ex-Ti oxide/Y-zeolite catalyst, while the titanium oxide species has a octahedral coordination in the case of the imp-Ti oxide/Y-zeolite catalyst, as shown in Fig. 14. Fig. 18 shows the relationship between the coordination number of titanium oxide species and the selectivity for N2 formation in the photocatalytic decomposition of NO on various titanium photocatalysts. A clear dependence of the N2 selectivity on the coordination number of the titanium oxide species can be observed, i.e. the lower the coordination number of the titanium oxide species, the higher the N2 selectivity. From these results, it is proposed that a highly efficient, highly selective photocatalytic reduction of NO into N2 and O2 can be achieved using the ex-Ti oxide/Y-zeolite, which includes the highly dispersed isolated tetrahedral titanium oxide as the active species, while

Fig. 17. Reaction time profiles of the photocatalytic decomposition of NO into N2 ( ) and N2 O ( ) on the ex-Ti oxide/Y-zeolite.

manner as on bulk TiO2 catalysts become predominant, and the reduction reaction by electrons and the oxidation reaction by holes occur separately from each other on different sites, leading to the selective formation of CH4 . 3.2. The photocatalytic decomposition of NO into N2 and O2 on the titanium oxide photocatalyst anchored in Y-zeolite cavities The Ti oxide/Y-zeolite (1.1 wt.% as TiO2 ) was prepared by ion-exchange with an aqueous titanium ammonium oxalate solution using Y-zeolite samples (SiO2 /Al2 O3 = 5.5) (ex-Ti oxide/Y-zeolite). Ti oxide/Y-zeolites having different Ti contents (1.0 and 10 wt.% as TiO2 ) were prepared by impregnating the Y-zeolite with an aqueous solution of titanium ammonium oxalate (imp-Ti oxide/Y-zeolite). UV-irradiation of the powdered TiO2 and the Ti oxide/Y-zeolite catalysts prepared by ion-exchange or impregnation methods in the presence of NO were found to lead to the evolution of N2 , O2 and N2 O in the gas phase at 275 K with different yields and different product selectivities [18,19]. As shown in Fig. 17, the yields of these photo-formed N2 , O2 and N2 O increased linearly with the UV-irradiation time, and the reaction immediately ceased when irradiation was discontinued, indicating that the pres-

Table 2 The yields of the photo-formed products, N2 and N2 O in the photocatalytic decomposition of NO at 275 K and their distribution on various Ti-based photocatalysts Catalysts

Ex-Ti oxide/Y-zeolite Imp-Ti oxide/Y-zeolite Imp-Ti oxide/Y-zeolite TiO2 powder

Ti content (wt.% as TiO2 )

1.1 1.0 10

Yields (␮mol/g of TiO2 h)

Selectivity (%)

N2

N2 O

Total

N2

N2 O

14 7 5 2

1 10 22 6

15 17 27 8

91 41 19 25

9 59 81 75

M. Matsuoka, M. Anpo / Journal of Photochemistry and Photobiology C: Photochemistry Reviews 3 (2003) 225–252

Fig. 18. Relationship between the coordination number of the Ti oxide species and the selectivity for N2 formation in the photocatalytic decomposition of NO on various titanium oxide photocatalysts.

the formation of N2 O as the major product was observed for the bulk TiO2 catalysts and on the imp-Ti oxide/Y-zeolite catalysts, which include the aggregated octahedrally coordinated titanium oxide species. Photoluminescence investigations of the ex-Ti oxide/ Y-zeolite catalyst were also carried out, and the catalyst was shown to exhibit a photoluminescence spectrum at around 490 nm by excitation at around 290 nm at 77 K due to the highly dispersed tetrahedrally coordinated titanium oxides, as shown in Fig. 15, while the imp-Ti oxide/Y-zeolite catalysts did not exhibit any photoluminescence spectrum. The addition of NO onto the ex-Ti oxide/Y-zeolite catalyst leads to an efficient quenching of the photoluminescence spectrum. The lifetime of the charge-transfer excited state was also found to be shortened by the addition of NO, its extent depending on the amount of NO added. These results indicate not only that the tetrahedrally coordinated titanium oxide species maybe located at positions accessible to the added NO but also that the added NO easily interacts with the charge-transfer excited state of the species [18,19]. From the above results, the reaction mechanism for the photocatalytic decomposition of NO on the isolated tetrahedral titanium oxide species can be proposed, as shown in Scheme 4. The NO species are able to adsorb onto these oxide species as weak ligands to form the reaction precursors. Under UV-irradiation, the charge-transfer excited complexes of the oxides, (Ti3+ –O− )∗ , are formed. Within their lifetimes, the electron transfer from the trapped electron center, Ti3+ , into the ␲-anti-bonding orbital of NO takes place, and simultaneously, the electron transfer from the ␲-bonding orbital of another NO into the trapped hole center, O− , occurs. These electron transfers lead to the direct decomposition of two sets of NO on (Ti3+ –O− )∗ into N2 and O2 under UV-irradiation in the presence of NO even at 275 K. On the other hand, with the aggregated or

239

Scheme 4. Reaction scheme of the photocatalytic decomposition of NO into N2 and O2 on the Ti oxide/Y-zeolite catalyst at 275 K.

bulk TiO2 catalysts, the photo-formed holes and electrons rapidly separate from each other with large space distances between the holes and electrons, thus preventing the simultaneous activation of two NO on the same active sites and resulting in the formation of N2 O and NO2 in place of N2 and O2 . The decomposed N and O species react with NO on different sites to form N2 O and NO2 , respectively. These results clearly demonstrate that zeolites used as supports enable the anchoring of the titanium oxide species in a highly dispersed state within the zeolite cavities, and thus, such tetrahedrally coordinated titanium oxide photocatalysts are promising candidates for unique and applicable photocatalysts for the reduction of toxic NOx elements. 4. Preparation of Cu+ /ZSM-5 catalyst and its photocatalytic reactivity for the decomposition of NO The protection and recovery of our environment from global air pollution caused by NOx (NO, NO2 , and N2 O) as well as SOx is currently an urgent, serious problem. Ammoxidiation using the V2 O5 /TiO2 catalyst has been developed as a de-NOx -ing process [89]. However, it operates only at high temperatures and requires NH3 . Strongly desired are catalytic systems for the direct decomposition of NOx , operating at ambient temperatures and ambient pressure conditions without NH3 . To address these concerns, ion-exchanged Cu+ /ZSM-5 zeolite catalysts have been the focus of much attention as potential catalysts for the direct decomposition of NO into N2 and O2 at around 723 K [90]. Utilization of photocatalytic processes in gas–solid systems also seems to be one of the most promising approaches to dissolve and reduce environmental toxins. Along these lines, the photocatalytic activity of the Cu+ /ZSM-5 catalyst for the decomposition of NO has been investigated and it was

240

M. Matsuoka, M. Anpo / Journal of Photochemistry and Photobiology C: Photochemistry Reviews 3 (2003) 225–252 Table 3 The results of the curve-fitting of the Cu K-edge EXAFS data for the Cu2+ /ZSM-5 sample and Cu+ /ZSM-5 catalysts evacuated at various temperatures Samples

Tm. (K)

Shell

R (Å)

CN

Cu2+ /ZSM-5

373

Cu–O

2.00

4.2

Cu+ /ZSM-5

573 773 973

Cu–O Cu–O Cu–O

1.94 1.94 1.94

2.5 2.4 2.3

Tm.: evacuation temperature; R: bond distances; CN: coordination number.

Fig. 19. XANES (left) and FT-EXAFS (right) spectra of (a and a ) Cu2+ /ZSM-5 sample and (b and b ) Cu+ /ZSM-5 catalyst. The catalyst was prepared by evacuation of the original Cu2+ /ZSM-5 sample at 973 K.

found that this catalyst acts as a photocatalyst to directly decompose NO into N2 and O2 and that the photo-excited state of the isolated Cu+ ion plays an significant role in the reaction [36]. In the following section, the characteristics of the Cu+ species anchored in the nano-pores of the ZSM-5 zeolite by means of in situ photoluminescence, ESR, XAFS and UV-Vis measurements and their reactions with gaseous NO under UV-irradiation will be discussed in detail on a molecular scale. 4.1. Preparation of the Cu2+ /ZSM-5 sample and Cu+ /ZSM-5 catalyst and their local structures The copper(II) ion-exchanged ZSM-5, Cu2+ /ZSM-5 sample was prepared by ion-exchange with an aqueous Cu(NH3 )4 2+ solution, followed by a drying process in air at 373 K. Fig. 19 shows the Cu K-edge XANES and Fourier transform of EXAFS spectra of the Cu2+ /ZSM-5 sample (a and a ) and the Cu+ /ZSM-5 catalyst prepared by the evacuation of the original Cu2+ /ZSM-5 sample at 973 K (b and b ), respectively. XANES and FT-EXAFS spectra provide useful and detailed information on the coordination structure and electronic states of elements which absorb X-ray photons. As shown in Fig. 19a, the Cu2+ /ZSM-5 sample dried at 373 K exhibits a well-separated weak pre-edge band (A) due to the 1s–3d transition as well as an intense band due to the 1s–4p transition. Band (B), attributed to the 1s–4pz (1s–4p␲∗ ) transition, can be observed as a shoulder of the intense band (C), attributed to the 1s–4px,y (1s–4p␴∗ ) transition, accompanied by their weak shake-down bands (B and C ). The presence of band (A), attributed to the 1s–3d transition, which is forbidden by the selection rule in the case of perfect octahedral symmetry, as well as the presence of the shake-down bands (B and C ), arising from a 1s–4p transition coupled with a simultaneous

ligand-to-metal electron transfer, which is characteristic for a metal ion having a d9 electronic configuration, indicate that the Cu2+ /ZSM-5 sample contains Cu2+ ions having slightly distorted symmetries as the major species [91–94]. These findings coincide with results obtained by ESR studies (shape, g-tensors and A-factors) indicating the presence of hydrated Cu2+ ions with elongated octahedral symmetry in the Cu2+ /ZSM-5 sample [36,95]. The ESR spectrum of Cu2+ /ZSM-5 sample measured at ambient temperature gives a broad isotropic signal, suggesting that the hydrated Cu2+ ions are not anchored onto the zeolite surface but are rotating freely in the pore structure of the ZSM-5 zeolite. As shown in Fig. 19b, the Cu+ /ZSM-5 catalyst that is prepared by the evacuation of the original Cu2+ /ZSM-5 sample at 973 K exhibits a very strong, sharp band (B) due to the 1s–4pz transition. In the case of a Cu+ ion having a d10 electronic configuration, a shake-down transition coupled with a ligand-to-metal electron transfer does not occur. The absence of the shake-down peak related to the sharp band (B) indicates that Cu+ is the main component in the Cu+ /ZSM-5 catalyst. It is known that in a planar or a linear geometry, the 1s–4pz transition is not affected by the ligands, and therefore, the copper compounds having these geometries exhibit a strong, sharp band (B) attributed to the 1s–4pz transition in the pre-edge region. Band (B) is intense enough to identify the copper species as the isolated Cu+ ions with a planar three-coordinate or linear two-coordinate geometry [91–93]. Fig. 19 also shows the corresponding FT-EXAFS spectra of the copper cation sample and catalyst, distances uncorrected for phase shifts. Table 3 shows the results obtained by the curve-fitting analyses of the Cu K-edge EXAFS spectra employing the iterative nonlinear least-squares method of Levernberg [53] and the empirical backscattering parameter sets extracted from the shell features of copper compounds. As can be seen in Fig. 19b , the Cu+ /ZSM-5 catalyst exhibits only the Cu–O peak at around 1.5 Å due to the presence of a neighboring O atom adjacent to the Cu atom, and neither the Cu–O–Cu peak that is observed for the aggregated Cu oxide species nor the Cu–Cu peak that is observed for the Cu metal species is observed. These results clearly indicate the presence of isolated Cu+ ions on the Cu+ /ZSM-5 catalyst. The curve-fitting analysis of the Cu–O peak indicates that the isolated Cu+ ions are present

M. Matsuoka, M. Anpo / Journal of Photochemistry and Photobiology C: Photochemistry Reviews 3 (2003) 225–252

241

perature dependence of the ESR, UV-Vis and XANES spectra of Cu2+ /ZSM-5 and Cu+ /ZSM-5 also clearly indicate that the isolated Cu2+ species are reduced to the isolated Cu+ ions by the evacuation of the Cu2+ /ZSM-5 samples at temperatures higher than 473 K. Temperature-programmed desorption profiles with the Cu2+ /ZSM-5 sample show the desorption of large amounts of H2 O, NH3 , N2 and O2 in the temperature range above 473 K [36]. From these results, the reduction mechanism of Cu2+ ions can be proposed in the following equations, where H2 O and NH3 coordinated to the Cu2+ ions act as reductants [96]: 2Cu2+ + H2 O → 2Cu+ + 2H+ (Br o¨ nsted acid) + 21 O2 3Cu2+ + NH3 → 3Cu+ + 3H+ + 21 N2 Fig. 20. Effects of the evacuation temperature of the Cu2+ /ZSM-5 sample on (a) the relative intensity of the ESR signal due to Cu2+ ; (b) the relative intensities of the UV absorption at 650 nm due to Cu2+ ; (c) at 300 nm due to Cu+ ; (d) the relative intensity of the XANES band due to the 1s–4pz transition of Cu+ .

in the Cu+ /ZSM-5 catalyst with a two- or three-coordinate geometry [94]. The presence of the isolated two- or three-coordinate Cu+ ions directly suggests the formation of Cu+ ions with a planar three or linear two-coordinate geometry, as suggested by the XANES studies. Fig. 20 shows the effect of the evacuation temperature of the original Cu2+ /ZSM-5 sample on the intensity of the ESR signals attributed to Cu2+ ions (a), the intensity of the UV-Vis spectra of the Cu2+ (b) and Cu+ ions (c), and the intensity of the XANES band due to the 1s–4pz transition of the Cu+ ions (d). As can be seen in this figure, the tem-

4.2. Excited state of the Cu+ /ZSM-5 catalyst As shown in Fig. 20, after the evacuation of the Cu2+ /ZSM-5 sample at 973 K, the ESR signal assigned to the Cu2+ species became very weak in its intensity. It is clear that at this temperature most of the Cu2+ ions were reduced to Cu+ ions. Fig. 21 shows that the Cu+ /ZSM-5 catalyst prepared in this way exhibits a photoluminescence spectrum with its maxima at around 440 nm upon excitation at around 280–300 nm. It has been reported that the emission spectrum of the free Cu+ ion in the gas phase is observed at around 450 nm, which is attributed to the radiative deactivation (3d9 4s1 → 3d10 + hν ) of the free Cu+ ion, while the presence of a crystal field due to the zeolite structure makes the emission peak shift to higher energy by raising the energy of the 4s orbital with respect to that of the 3d orbitals [97]. On the other hand, as the

Fig. 21. Ordinary photoluminescence spectrum observed for (a) Cu+ /ZSM-5 catalyst; (b) corresponding excitation spectrum; (curves 1–5) the effect of the addition of NO on the photoluminescence measured at 77 K. The addition of NO was carried out at 298 K. Pressure of NO added: (1) 0.1 Torr; (2) 0.3 Torr; (3) 0.5 Torr; (4) 1 Torr; (5) 20 Torr. The excitation spectrum was monitored at 460 nm emission. The catalyst was prepared by evacuation of the original Cu2+ /zeolite sample at 1173 K.

242

M. Matsuoka, M. Anpo / Journal of Photochemistry and Photobiology C: Photochemistry Reviews 3 (2003) 225–252

distance between the Cu+ ions is decreased, the overlap of the 4s atomic orbitals, which leads to the formation of the ␴(4s) molecular orbital, is increased, pushing the ␴(4s) bond energy down closer to that of the ␴*(3d)-anti-bonding molecular orbital derived from the 3d atomic orbitals. Thus, Cu+ –Cu+ pairs are expected to give photoluminescence (␴(4s) → ␴∗ (3d)) in the wavelength region longer than the position of the photoluminescence band due to the free Cu+ ion in the gas phase (450 nm) [98]. Along these arguments, the excitation (absorption) and photoluminescence spectra shown in Fig. 21 are attributed to the presence of isolated Cu+ ions coordinated by the oxygen of the zeolite framework, i.e. the electronic excitation of the isolated Cu+ ion (3d10 + hν → 3d9 4s1 ) and its reverse radiative deactivation (3d9 4s1 → 3d10 + hν ), respectively [98–101]. This assignment of the photoluminescence is well supported by the results obtained by FT-EXAFS which indicate that, in the ZSM-5 zeolite, most of the copper cations are included as isolated Cu+ monomer species. On the other hand, in the case of using Y-zeolite as a support instead of ZSM-5 zeolite, the Cu+ /Y-zeolite exhibits two different intense bands, at around 420 nm due to the Cu+ monomer, and one at around 515 nm, which is attributed to the radiative deactivation (␴(4s) → ␴∗ (3d)) of the Cu+ dimer species [93,98]. Considering the differences in the void space volumes of the supercages and types of the channel connections in these two types of zeolites, it can be expected that the copper(I) cations in the Y-zeolite diffuse more easily to form the Cu+ –Cu+ dimer species, while in the narrow channels of ZSM-5, the mobility of the copper(I) cation is low, so that it remains intact as an isolated Cu+ monomer species. In addition to these differences, the density or number of the ion-exchangeable sites

in the zeolites is remarkably different. In the case of the Y-zeolite, the SiO2 /Al2 O3 ratio (5.5) is lower than that for ZSM-5 (23.3), and has a large density of ion-exchangeable sites, thus allowing the copper cations to exist close to each other. Such a situation easily causes aggregation of copper cations during thermal treatments at higher temperatures [93]. In accordance with these results, the FT-EXAFS spectrum of the Cu+ /Y-zeolite exhibits an intense peak at 2.6 Å due to the neighboring copper atoms (Cu–O–Cu), in addition to a peak at 1.5 Å due to the Cu–O bond, indicating that the aggregation of the Cu+ ions easily occurred on the Cu+ /Y-zeolite [93]. Fig. 22 shows that the intensity of the photoluminescence spectrum of the Cu+ /ZSM-5 catalyst increases when the degassing temperature of the original Cu2+ /ZSM-5 sample is increased, passing through a maximum at 1173 K, and then decreasing in the region of much higher evacuation temperatures. In the highest temperature region, where the intensity of the photoluminescence becomes weaker, the color of the catalysts changed from white to light red, and a peak due to the Cu–Cu bond was also observed in the FT-EXAFS spectrum of the catalyst, suggesting that the evacuation of the samples at temperatures higher than 1173 K leads to further reduction of copper(I), Cu+ to Cu0 . It can be seen in Fig. 22 that there is a gap between the temperature where the decrease in the intensity of the ESR signal starts and that where the increase in the intensity of the photoluminescence begins. This can be explained by quenching of the photoluminescence of the Cu+ ions by the existence of the surface OH groups and/or adsorbed O2 near the Cu+ ions in the temperature regions of 473–673 K, and these quenchers can be desorbed only by evacuation treatment at temperature higher than 673 K.

Fig. 22. Effects of the evacuation temperature of the Cu2+ /ZSM-5 sample on (a) the relative intensity of the ESR signal due to Cu2+ ; (b) the relative yields of the photoluminescence due to Cu+ ; (c) the relative conversions (yields) of the photocatalytic decomposition of NO (NO pressure: 2 Torr) at 298 K.

M. Matsuoka, M. Anpo / Journal of Photochemistry and Photobiology C: Photochemistry Reviews 3 (2003) 225–252

Fig. 23. ESR signal obtained after the addition of 2 Torr of NO to the Cu+ /ZSM-5 catalyst at 77 K.

4.3. The interaction of NO with Cu+ /ZSM-5 catalyst Fig. 23 shows the ESR signal due to the NO species adsorbed on the Cu+ /ZSM-5 catalyst. The four equally spaced g⊥ components and the spin Hamiltonian parameter of the spectrum indicate that NO molecules are adsorbed on the Cu+ ion (I = 3/2) to form nitrosylic adducts on Cu+ having a slight electron transfer from Cu+ to NO (it can be written as Cu+ –NOδ− ) [95,102–104]. However, the fact that the ESR signal due to the Cu2+ cannot be observed clearly indicates that the complete one-electron transfer from Cu+ to NO has not occurred. Cu+ –NOδ− adduct species are stable in the presence of low pressure NO (<1 Torr), while after the evacuation of the system at ambient temperature, the ESR signal disappears. On the other hand, in the presence of higher pressure NO (>1 Torr) the ESR signal of Cu2+ becomes observable, suggesting that the oxidation of the Cu+ ion with NO to form Cu2+ ion proceeds. As shown in Fig. 24, after the addition of NO to the Cu+ /ZSM-5 catalyst, an IR spectrum due to the Cu+ –NO species is observed at 1813 cm−1 . The wavenumber of this species (1813 cm−1 ) is lower than that of NO gas (1876 cm−1 ) and higher than that of NO− (1100 cm−1 ). These results indicate that the Cu+ –NOδ− adduct species is formed on the Cu+ /ZSM-5 catalyst in the presence of low-pressure NO, in good agreement with the ESR measurements [95]. With increasing NO pressure, the intensity of the IR band at 1911 cm−1 which is identified as an NO species adsorbed on the Cu2+ , increases, and new bands attributed to a dimeric NO species with a cis configuration also become observable at 1733 and 1827 cm−1 [95]. These results clearly indicate that the major NO adsorption species on the Cu+ /ZSM-5 catalyst is a nitrosylic adducts Cu+ –NOδ− .

243

Fig. 24. IR spectra of the NO species adsorbed on the Cu+ /ZSM-5 catalyst prepared by the evacuation of the Cu2+ /ZSM-5 sample at 1173 K. The adsorption of NO was carried out at 298 K. Pressure of NO added: (a) 0.1 Torr; (b) 2.0 Torr; (c) 3.0 Torr; (d) 5.0 Torr; (e) 10 Torr; (f) 20 Torr. The spectra were recorded at 298 K.

4.4. The photocatalytic reactivity of Cu+ /ZSM-5 catalyst for the decomposition of NO UV-irradiation of the Cu+ ion catalyst even at 275 K in the presence of NO was found to lead to the formation of N2 and O2 , with a good linear relationship between the UV-irradiation time and the NO conversion. Fig. 25 shows

Fig. 25. Reaction time profiles of the photocatalytic decomposition of NO (NO pressure: 2 Torr) into N2 and O2 at 298 K on the (a) Cu+ /ZSM-5 and (b) Cu+ /Y catalysts.

244

M. Matsuoka, M. Anpo / Journal of Photochemistry and Photobiology C: Photochemistry Reviews 3 (2003) 225–252

reaction time profiles of the photocatalytic decomposition of NO at 275 K on the Cu+ /ZSM-5 and Cu+ /Y-zeolites. The formation of by-products such as N2 O and NO2 was negligible. As also shown in Fig. 22, the yields of the photocatalytic decomposition reaction of NO greatly depend on the pretreatment degassing temperature of the Cu2+ /ZSM-5 samples, showing a good parallel relation with the dependency of the intensity of the photoluminescence due to the Cu+ ion. These results indicate that the photo-excited states of Cu+ ion play a significant role in the photocatalytic decomposition of NO. As shown in Fig. 25, the Cu+ /ZSM-5 catalyst exhibits a higher apparent photocatalytic activity as compared to Cu+ /Y-zeolite catalyst. As mentioned above, in the Cu+ /ZSM-5 catalyst, isolated Cu+ ions are the predominant species, whereas, with the Cu+ /Y-zeolite, a considerable amount of Cu+ dimer species exists together with the Cu+ monomer species. Therefore, it can be concluded that the photocatalytic reactivity of the Cu+ monomer species is higher than that of the Cu+ dimer species. In fact, the photoluminescence spectrum of the Cu+ /ZSM-5 catalyst at around 440 nm is quenched more efficiently than with the Cu+ /Y-zeolite, indicating that the photo-excited state of the Cu+ monomer species interacts with NO more efficiently than with the Cu+ dimer species [36,93]. It should be also noted that Cu2 O powder, which is the model oxide compound of the aggregated Cu+ , did not show any photocatalytic activity for the reaction, suggesting that the presence of the highly dispersed Cu+ species is necessary for the photocatalytic reaction. UV-irradiation of the Cu+ /ZSM-5 catalyst having the Cu+ –NOδ− adduct species leads to a decrease in the intensity of the ESR signal assigned to the Cu+ –NOδ− species with UV-irradiation time, without the appearance of any new signal. After UV-irradiation was stopped, the intensity of the signal returned to its original level. These reversible changes in the ESR signal assigned to the Cu+ –NOδ− adduct species suggest not only that the Cu+ –NOδ− species acts as a reaction precursor but also that the photo-induced decomposition of NO proceeds catalytically. The addition of NO to the Cu+ /ZSM-5 catalyst leads to the efficient quenching of the photoluminescence due to Cu+ . The lifetime of the photoluminescence is shortened by the addition of NO, its value becoming shorter with increasing NO pressure, i.e. from about 85 ␮s in vacuum to 70 ␮s in the presence of 1.0 Torr NO [36]. These results clearly suggest that the added NO easily interacts with the Cu+ species in both its ground and excited states. From these results, together with the results obtained by in situ photoluminescence, ESR, and FT-IR measurements, the mechanism of the photocatalytic decomposition of NO into N2 and O2 on the Cu+ /ZSM-5 catalyst at 275 K under UV-irradiation is proposed as follows (Scheme 5): electron transfer from the excited state of the Cu+ (3d9 4s1 state) to an ␲-anti-bonding orbital of NO and simultaneous electron transfer from the ␲-bonding orbital of another

Scheme 5. Reaction scheme of the photocatalytic decomposition of NO into N2 and O2 on the Cu+ /ZSM-5 catalyst at 298 K (䊐 denotes an electron vacancy).

NO to the vacant electronic state of the Cu+ ion (3d9 4s0 state) occur, causing local charge separation and a weakening of the N–O bond of two NO molecules, initiating the decomposition of NO into N2 and O2 [14,36,105]. Two different types of NO, i.e. one type that constitutes the Cu+ –NOδ− adduct, and a second type, which is supplied from gas phase, are simultaneously activated on the photo-excited Cu+ site in the local electron transfer, resulting in the selective formation of N2 and O2 without any formation of N2 O and/or NO2 on the Cu+ /ZSM-5 catalyst at 275 K. 5. Design of Ag+ /ZSM-5 catalyst and its photocatalytic reactivity for the decomposition of NO As mentioned above, the Cu+ /ZSM-5 catalyst decomposes NO into N2 and O2 . However, preparation of Cu+ ion catalysts requires the pretreatment of the original Cu2+ /ZSM-5 samples at temperature higher than 973 K to reduce Cu2+ to Cu+ ions. On the other hand, Ag+ ion has the same electronic configuration as the Cu+ ion (d10 ). Therefore, it is expected that the Ag+ ion may exhibit similar catalytic and photocatalytic reactivity to that of the Cu+ ion. Furthermore, the Ag+ ion has the advantage of being chemically stable even in an oxidative atmosphere. In fact, ion-exchanged silver/zeolite catalysts have been reported to show very high activity for the disproportionation of ethylbenzene [106,107], photochemical/thermal cleavage of water to produce H2 and O2 [108], photo-oxygen production from water [109], photo-dimerization of alkanes [110], and the selective reduction of NO by ethylene at around 823 K [111] or by ethanol at around 723 K [112]. The following section shows that the Ag+ /ZSM-5 catalyst is of remarkably high photocatalytic reactivity for the decomposition of NO compared with the Cu+ /ZSM-5 catalyst, even in the presence of H2 O or O2 .

M. Matsuoka, M. Anpo / Journal of Photochemistry and Photobiology C: Photochemistry Reviews 3 (2003) 225–252

245

Fig. 27. UV-Vis spectra of the (a) Ag+ /ZSM-5; (b) H+ /ZSM-5; and (c) Ag0 /ZSM-5 catalysts.

Fig. 26. XANES (left) and FT-EXAFS (right) spectra of (a and a ) Ag+ /ZSM-5 catalyst; (b and b ) Ag2 O powder; and (c and c ) Ag foil.

5.1. Preparation of the Ag+ /ZSM-5 catalyst and determination of its local structure Silver(I) ion-exchanged ZSM-5, Ag+ /zeolite samples were prepared by ion-exchange with an aqueous Ag(NH3 )2 + solution, followed by evacuation at 298 K, calcined at 673 K in the presence of O2 , and then degassed at 473 K. The Ag0 /ZSM-5 catalyst was prepared by heating the Ag+ /ZSM-5 at 673 K in the presence of an H2 /H2 O mixture at a total pressure of 20 Torr (H2 /H2 O = 1). Fig. 26 shows the FT-EXAFS spectra of the Ag+ /ZSM-5 catalyst (a ), together with bulk Ag2 O (b ) and Ag foil (c ) as references, respectively. The FT-EXAFS spectrum of Ag2 O (b ) exhibits a peak at around 3.5 Å (distance is uncorrected for phase shift) which can be attributed to the Ag–O–Ag bonding, and the FT-EXAFS of the Ag foil (c ) exhibits a peak at around 2.5 Å due to Ag–Ag bonding. However, the FT-EXAFS of the Ag+ /ZSM-5 catalyst exhibits only a well-defined peak due to the neighboring oxygen atoms (Ag–O) at around 1.8 Å. This result suggests that silver is anchored within the micropores of the ZSM-5 zeolite in an isolated states forming neither clusters nor Ag metal or oxide crystals. Fig. 27 shows the UV-Vis spectra of the Ag+ /ZSM-5 (a), H+ /ZSM-5 (b), and Ag0 /ZSM-5 (c) catalysts. The H+ /ZSM-5 exhibits no intense absorption band in 200– 250 nm wavelength ranges. On the other hand, as shown in Fig. 27a, the Ag+ /ZSM-5 catalyst exhibits an intense absorption band at around 220 nm, which is attributed to

the 4d10 → 4d9 5s1 electronic transition of the Ag+ ions [113–115]. Although the Ag0 atoms, and Agn 0 and Agm n+ clusters are known to exhibit absorption bands at wavelength above 250 nm [116,117], no absorption band of the Ag+ /ZSM-5 appears in this region. Furthermore, no ESR signals assigned to the Ag0 atoms or Ag2+ species were observed with the Ag+ /ZSM-5 catalyst. These results firmly support the conclusion that silver ions are included within the pore structure of the ZSM-5 zeolite as highly dispersed Ag+ ions [38,39]. As shown in Fig. 27c, after H2 treatment of the Ag+ /ZSM-5 catalyst at 673 K, the intensity of the absorption band of Ag+ ions at around 220 nm drastically decreases, and broad absorption bands due to Agn 0 or Agm n+ clusters appear at wavelengths longer than 250 nm, indicating that reduction and aggregation of the Ag+ ions have occurred. Since Ag0 /ZSM-5 did not show any photocatalytic reactivity for the decomposition of NO, it is concluded that the Agn 0 or Agm n+ clusters are not associated with the reaction. On the other hand, in the case of the Ag+ /Y-zeolite, UV absorption bands above 250 nm appear, and with the same catalyst, a peak due to the neighboring Ag atom was observed in the FT-EXAFS spectrum of the catalyst at around 2.5 Å, these results suggesting the presence of Agn 0 or Agm n+ clusters on Ag+ /Y-zeolite [118]. 5.2. Excited state of the Ag+ /ZSM-5 catalyst and its interaction with NO Fig. 28 shows the ESR signal obtained by the addition of 7 Torr of NO to the Ag+ /ZSM-5 catalyst at 77 K. The hyperfine splitting of the signal due to the interaction of the electron spin of NO with a nucleus of Ag+ (I = 1/2) clearly indicates that NO molecules are adsorbed on the Ag+ ions to form a nitrosylic adduct species, i.e. (Ag–NO)+ [119]. The evacuation of the system after the adsorption of NO on Ag+ /ZSM-5 led to the disappearance of the signal, suggesting that the interaction of NO with the Ag+ ion is weak.

246

M. Matsuoka, M. Anpo / Journal of Photochemistry and Photobiology C: Photochemistry Reviews 3 (2003) 225–252

Fig. 28. ESR signal obtained after the addition of NO to the Ag+ /ZSM-5 catalyst at 77 K.

On evacuation, no new ESR signals appeared, indicating that the oxidation of Ag+ to Ag2+ by the addition of NO at high pressures (>1 Torr) did not occur, unlike the case of Cu+ /ZSM-5. Fig. 29 shows the effect of UV-irradiation upon the intensity of the ESR signal under the same conditions shown in Fig. 28. UV-irradiation of the Ag+ /ZSM-5 catalyst having the (Ag–NO)+ adduct species leads to a decrease in the intensity of the ESR signal with UV-irradiation time, without the appearance of any new signals. After UV-irradiation was stopped, the signal was found to recover its original intensity. These reversible changes suggest that the (Ag–NO)+ adduct species acts as a reaction precursor for the photocatalytic decomposition of NO. Fig. 30 shows the photoluminescence spectrum (a) and its corresponding excitation spectrum (b) of the Ag+ /ZSM-5 catalyst. A good coincidence of the excitation band position (220 nm) with that of the absorption band due to the isolated Ag+ ion (220 nm) indicates that these excitation and photoluminescence features can be attributed to the absorption due to 4d10 → 4d9 5s1 transition and its reverse radiative deactivation process 4d9 5s1 → 4d10 , respectively [38,39]. Fig. 30 also shows the effect of the addition of NO on the photoluminescence of the Ag+ /ZSM-5 catalyst. The

Fig. 29. Time profile of the intensity of the ESR signal due to (Ag–NO)+ species under UV-irradiation.

Fig. 30. Ordinary photoluminescence spectrum observed for (a, curve 1) Ag+ /ZSM-5 catalyst; (b) the corresponding excitation spectrum; and (curves 2–4) the effect of the addition of NO on the photoluminescence. The addition of NO was carried out at 298 K. NO pressure: 1, 0.0 Torr; 2, 0.2 Torr; 3, 4.0 Torr; 4, after the evacuation of NO at 298 K. The excitation spectra were monitored at 340 nm emission.

addition of NO leads to an efficient quenching of the photoluminescence, and the evacuation of the system after the complete quenching of photoluminescence leads to the recovery of the photoluminescence to its original intensity. These results also suggest that the interaction of NO molecules with Ag+ ions is weak, in an agreement with the result obtained by ESR measurements, and added NO easily interacts with the Ag+ species in its ground and excited states. The photoluminescence of the Ag+ /ZSM-5 catalyst was found to be more efficiently quenched by NO than with the Cu+ /ZSM-5 catalyst, suggesting that the photo-excited states of Ag+ ions interact with NO more efficiently than Cu+ ions. 5.3. The photocatalytic reactivity of Ag+ /ZSM-5 catalyst for the decomposition of NO UV-irradiation of the Ag+ /ZSM-5 catalyst in the presence of 10 Torr of NO at 298 K was found to lead to the formation of N2 , N2 O and NO2 . The reaction time profiles of the formation of N2 and N2 O are shown in Fig. 31. The formation of N2 and N2 O is found only under UV-irradiation and their yields increase with a good linearity versus the irradiation time, suggesting that the reaction proceeds photocatalytically. In fact, after prolonged irradiation, the turnover frequency (number of N2 molecules per number of Ag+ ions in the catalyst) exceeded 1.0, showing that the reaction proceeds catalytically. The most effective wavelengths of UV light for the photocatalytic decomposition of NO was examined by using various UV-cut filters. Under UV-irradiation of the catalyst through a UV-25 filter (λ > 250 nm), the photocatalytic decomposition of NO proceeded at 15% of the rate obtained under irradiation without any filters, i.e. under the full arc irradiation of the high-pressure mercury lamp. This shows

M. Matsuoka, M. Anpo / Journal of Photochemistry and Photobiology C: Photochemistry Reviews 3 (2003) 225–252

Fig. 31. Time profiles of the photocatalytic decomposition of NO into N2 and N2 O on the (a) Ag+ /ZSM-5 and (b) Cu+ /ZSM-5 catalysts at 298 K.

that the effective UV light for the reaction lies in the wavelength region from 200 to 250 nm, where the absorption (excitation) of the Ag+ ions (220 nm) is observed. These results clearly indicate that the photo-excited state of the Ag+ ions plays a significant role in the photocatalytic decomposition of NO. Under UV-irradiation of the Ag0 /ZSM-5, H+ /ZSM-5 catalysts and Ag2 O powder, the decomposition of NO scarcely proceeded, and the formation of N2 and N2 O was negligible, while on the Ag+ /Y-zeolite catalyst, the photocatalytic reaction proceeded at about 15% of the rate obtained on the Ag+ /ZSM-5 catalyst. These results suggest that the isolated Ag+ ion is responsible for the photocatalytic reaction, while Agn 0 or Agm n+ clusters are not responsible for the reaction. As also shown in Fig. 31, the rate of the N2 formation in the photocatalytic decomposition of NO on the Ag+ /ZSM-5 catalyst is 10 times faster than that on the Cu+ /ZSM-5 catalyst [36]. Such high efficiency of the Ag+ /ZSM-5 catalyst can be explained by the efficient interaction of the photo-excited Ag+ ion with NO as compared to the Cu+ ion. The effect of the addition of O2 on the photocatalytic decomposition of NO on the Ag+ /ZSM-5 catalyst was investigated. As suggested by the ESR measurements, the addition of O2 or NO at high pressures (>1 Torr) did not lead to the oxidation of Ag+ to Ag2+ in the Ag+ /ZSM-5 catalyst, in clear contrast to the easy oxidation of Cu+ to Cu2+ in the Cu+ /ZSM-5 catalyst [95]. Such a chemical stability of Ag+ , even in an oxidative atmosphere, is one advantage in the utilization of the Ag+ /ZSM-5 catalyst as a photocatalyst for the elimination of NOx in the atmosphere. Fig. 32 shows the effect of the addition of O2 on the yield of the photocatalytic decomposition reaction of NO on the Ag+ /ZSM-5 catalyst and Cu+ /ZSM-5. In the presence of O2 in the system, the reactivity of the Ag+ /ZSM-5 catalyst is maintained in spite of the decrease in the reaction yield, while the photocatalytic reactivity of Cu+ /ZSM-5 decreases dramatically.

247

Fig. 32. The effect of the addition of O2 on the formation of N2 in the photocatalytic decomposition of NO on the (a) Ag+ /ZSM-5 and (b) Cu+ /ZSM-5 catalysts. The effect of the addition of H2 O on the photocatalytic reactivity of the Ag+ /ZSM-5 catalyst for the direct decomposition of NO at 298 K (c).

Fig. 32 also shows the effect of the addition of H2 O on the photocatalytic reactivity of the Ag+ /ZSM-5 catalyst. In the presence of 20% H2 O, the photocatalytic reactivity of the catalyst is found to maintain 90% of the rate obtained in the absence of H2 O, showing that Ag+ /ZSM-5 catalyst acts as an efficient photocatalyst for the decomposition of NO even in the presence of H2 O. As mentioned above, in addition to these advantages, unlike the preparation Cu+ /ZSM-5 from Cu2+ /ZSM-5 samples, the evacuation pretreatment is not needed for the preparation of Ag+ /ZSM-5 catalysts [36–41]. From these various findings, as shown in Scheme 6, it is concluded that the photo-exited electronic state of highly dispersed Ag+ ions (4d9 5s1 ) plays a significant role in the photocatalytic decomposition of NO, while an electron transfer from the photo-excited Ag+ into the ␲-anti-bonding molecular orbital of NO initiates the weakening of the N–O bond. At the same time, an electron transfer from a

Scheme 6. Reaction scheme of the photocatalytic decomposition of NO on the Ag+ /ZSM-5 catalyst at 298 K.

248

M. Matsuoka, M. Anpo / Journal of Photochemistry and Photobiology C: Photochemistry Reviews 3 (2003) 225–252

␲-bonding orbital of another NO to the vacant orbital leads to the weakening of the N–O bond resulting in the decomposition of NO into N2 , N2 O, and NO2 . The remarkably high photocatalytic reactivity of the Ag+ /ZSM-5 catalyst is attributed to the high chemical stability of the Ag+ ion and the efficient interaction of the excited electronic state of the Ag+ ion with NO, as compared with those properties of the Cu+ ion on the Cu+ /ZSM-5 catalyst [38,39]. 5.4. The photocatalytic decomposition of N2 O on the Ag+ /ZSM-5 catalyst The direct decomposition or selective reduction of NO using a thermal catalytic process has been extensively investigated; however, relatively little attention has been paid to the decomposition reaction of the N2 O, which plays a role in causing the greenhouse effect and in the destruction of the ozonosphere. As shown in Fig. 33, UV-irradiation of Ag+ /ZSM-5 (2.9 wt.%) in the presence of 1 Torr of N2 O at 298 K leads to the efficient formation of N2 and O2 (N2 /O2 = 3). The yields of N2 and O2 increase with good linearity versus the UV-irradiation time, while under dark conditions these products could not be detected [120]. The value of the yield of the photo-formed N2 molecules per total number of Ag+ ions included in the catalyst exceeded 2.0 after prolonged UV-irradiation, and even after this time, the decomposition of N2 O proceeded linearly with the UV-irradiation time, indicating that the reaction proceeds photocatalytically. On the other hand, only a small amount of N2 was observed on Ag0 /ZSM-5 (2.9 wt.%). These results clearly indicate that Ag+ ions play a significant role in the photocatalytic decomposition of N2 O. Under UV-irradiation of the catalyst through UV-25 filter (λ > 250 nm), the photocatalytic decomposition of N2 O proceeded at 4% of the rate under the full spectrum of the high-pressure mercury lamp.

This shows that the UV light effective for the reaction lies in wavelength regions of 200–250 nm, where the UV absorption band of the Ag+ ion exists, suggesting that the electron transfer from the photo-excited state of Ag+ to the anti-bonding molecular orbital of N2 O is responsible for the reaction [36,120]. As shown in Fig. 33, Ag+ /ZSM-5 (2.9 wt.%) shows higher photocatalytic reactivity as compared with Cu+ /ZSM-5 (1.6 wt.%), and the N2 /O2 ratios of the reaction products are lower for Ag+ /ZSM-5. These results can be attributed to the fact that the Ag+ ions more easily desorb the oxygen atoms that are formed during the photocatalytic reaction and are in equilibrium with O2 in the gas phase, as compared to the Cu+ ions [121,122].

6. The photocatalytic decomposition of N2 O on lanthanoid ion-exchanged mordenite catalysts It has been reported by Ebitani et al. [123] that lanthanoid ions (La, Pr, Sm, Eu, and Gd) exchanged onto the mordenite zeolite also act as photocatalysts for the decomposition of N2 O into N2 and O2 , as shown in Table 4. Among these lanthanoid catalysts, a Pr/Mordenite catalyst that was evacuated at 873 K was found to exhibit the highest reactivity for the decomposition of N2 O; it also exhibited a good stoichiometric value (2:1) for the N2 :O2 ratio, while that observed for the other lanthanoid ions exchanged onto mordenite were less than the stoichiometric value, i.e. around 2.6–3.3:1. In the case of the Pr/Mordenite catalyst, the effective wavelength of the incident light for the reaction is shorter than 250 nm, where the strong absorption due to the f–f transition (3H4 level to 1S0 level) or 4f–5d transition of the Pr3+ (4f2 ) cation appears (at around 240 nm) [124,125]. These results suggest that the photocatalytic decomposition of N2 O on the Pr/Mordenite is initiated by the absorption of photons by Pr3+ cations anchored on the mordenite zeolite. The photocatalytic reaction rate of the decomposition of N2 O on the Pr6 O11 powder was found to be six times lower Table 4 Results of the photocatalytic decomposition of N2 O on various praseodymium-loaded catalysts and lanthanoid cation-exchanged mordenites

Fig. 33. Reaction time profiles of the photocatalytic decomposition of N2 O into N2 and O2 on the Ag+ /ZSM-5 (2.9 wt.%), Ag0 ZSM-5 (2.9 wt.%) and Cu+ /ZSM-5 (1.6 wt.%) catalysts.

Catalyst

N2 formation rate (␮mol/min)

N2 :O2 ratio

Total turnover for Ln ion

Pr-Y zeolite Pr-mordenite La-mordenite Sm-mordenite Eu-mordenite Gd-mordenite Na-mordenite Pr/Al2 O3 Pr/Al2 O3 Pr/SiO2 -Al2 O3 Pr/SiO2 -Al2 O3 Pr/SiO2 Pr6 O11

0.0075 0.1173 0.0128 0.0135 0.1059 0.0230 0.0147 0.0585 0.0582 0.0486 0.0327 0.0021 0.0189

2.3 2.1 3.3 2.8 2.6 2.8 3.1 2.8 2.8 2.5 2.5 12.0 3.9

0.3 5.0 0.5 0.6 4.6 1.0 – 0.5 2.5 0.4 0.3 0.02 0.02

M. Matsuoka, M. Anpo / Journal of Photochemistry and Photobiology C: Photochemistry Reviews 3 (2003) 225–252

than Pr/Mordenite, indicating that the high dispersion of the Pr ions is necessary to achieve the high photocatalytic reactivity [126,127]. It was found that the amount of N2 O irreversibly adsorbed on the Pr/Mordenite catalyst is 2.5 N2 O molecules per one Pr3+ cation. The initial reaction rate of the photocatalytic decomposition of N2 O on the Pr/Mordenite in the presence of sufficient N2 O in the gas phase is the same as the rate obtained after the evacuation of gaseous N2 O from the reaction system, indicating that the N2 O species that was irreversibly adsorbed on the Pr3+ cation plays a significant role in the reaction [126,127]. On the other hand, it was found that the adsorption of NH3 or H2 O at ambient temperatures leads to a remarkable decrease in the photocatalytic reaction of N2 O and that in such systems a disappearance of the absorption band due to the Pr3+ cation (240 nm) and a remarkable decrease in the amount of N2 O species irreversibly adsorbed on the Pr3+ cation can be observed. These results indicate that the adsorption of NH3 or H2 O leads to a critical modification of the local coordination structure of the Pr3+ cation, resulting in a significant suppression of the irreversible adsorption of N2 O, which is directly responsible for the photocatalytic decomposition of N2 O [126,127].

7. Conclusions In this paper, the local structures of the transition metal oxides (Ti, V, Mo, Cr) incorporated into the zeolite framework structures as well as local structures of the transition metal or rare earth ions (Cu+ , Ag+ , Pr3+ ) exchanged into the zeolite cavities were discussed, based on results of various in situ spectroscopic techniques such as ESR, UV-Vis, photoluminescence and XAFS (XANES and EXAFS). The interactions of these active species with gaseous NOx (NO and N2 O) and CO2 were investigated, and the photocatalytic reactivity of the catalysts for the decomposition of NOx as well as the reduction of CO2 with H2 O have been summarized. The metal oxide catalysts (Ti, V, Mo, Cr), which are incorporated into the frameworks or cavities of various zeolites by means of ion-exchange or hydrothermal synthesis, exist in highly dispersed tetrahedral coordination states and act as efficient photocatalysts for the decomposition of NO into N2 and O2 as well as the reduction of CO2 with H2 O to produce CH4 and CH3 OH. Photoluminescence investigations revealed that the efficient interaction of the charge-transfer photo-excited complexes of these oxides, (Me(n −1)+ –O− )∗ , i.e. electron–hole pair state with reactant molecules, such as NO, CO2 and H2 O, plays a significant role in the photocatalytic reactions. On the other hand, highly dispersed Cu+ , Ag+ and 3+ Pr ions anchored onto the surfaces of zeolite cavities by ion-exchange act as efficient and unique photocatalysts for the decomposition of NOx (NO and N2 O) into N2 and

249

O2 . In the case of the Cu+ /ZSM-5 and Ag+ /ZSM-5 zeolite catalysts, the interaction of Cu+ or Ag+ ions with the reactant NOx molecules was observed by ESR, IR and photoluminescence spectroscopic measurements. Based on these results, the reaction mechanisms involving a local electron transfer from the excited Cu+ or Ag+ ions to the ␲-anti-bonding orbital of NOx and simultaneous electron transfer from the ␲-bonding orbital of another NOx to the vacant orbital of the Cu+ or Ag+ ions have been proposed for the decomposition of NOx into N2 and O2 at 275 K. In the case of the Pr3+ /Mordenite catalyst, it was found that the N2 O species irreversibly adsorbed on the Pr3+ cation plays a significant role in the photocatalytic reaction of N2 O. The localized excitation of the highly dispersed metal oxides, as well as of the Cu+ , Ag+ or Pr3+ ions, leads to unique photocatalytic properties that are quite different from semiconductor type photocatalysts (e.g. TiO2 , ZnO, etc.), in which the photo-formed electrons and holes rapidly separate from each other at relatively large distance, leading to reduction and oxidation reactions, respectively, at the different surface sites. It was also found that by using zeolites as a support, the local environment of the introduced cations can be significantly modified, and it becomes possible to achieve the mono-atomic dispersion of metal oxides or ions, resulting in a remarkable, unique enhancement of the photocatalytic performance as compared to corresponding bulk metal oxides such as TiO2 , Cu2 O, Ag2 O or Pr6 O11 , since photo-formed electron–hole pair states contribute directly to the reactions, without charge separation. It should also be emphasized that by utilizing the physicochemical properties of various zeolites such as the pore size diameter or the channel structure, it becomes possible to control the photocatalytic activity as well as the selectivity of the reaction product, which is well demonstrated in the results of photocatalytic reduction of CO2 with H2 O on Ti-containing mesoporous zeolites. It can thus be seen that the use of zeolites is one of the most promising approaches in designing efficient local structures for photocatalysts at the molecular level for the development of effective photocatalytic systems to reduce and eliminate global air and water pollution.

References [1] M. Anpo, in: P. Tundo, P. Anastas (Eds.), Green Chemistry, Oxford University Press, 2000, p. 1. [2] M. Anpo, in: A. Corma, F.V. Melo, S. Mendioroz, J.L.G. Fierro (Eds.), Proceedings of the 12th International Congress on Catalysts, Granada, Studies in Surface Science and Catalysis, vol. 130, 2000, p. 157. [3] D.F. Ollis, H. AlEkabi (Eds.), Photocatalytic Purification and Treatment of Water and Air, Elsevier, Amsterdam, 1993. [4] E. Pelizzetti, M. Schiavello (Eds.), Photochemical Conversion and Storage of Solar Energy, Kluwer Academic Publishers, Dordrecht, 1991. [5] Y. Yoshida, M. Matsuoka, S.C. Moon, H. Mametsuka, E. Suzuki, M. Anpo, Res. Chem. Intermed. 26 (2000) 567.

250

M. Matsuoka, M. Anpo / Journal of Photochemistry and Photobiology C: Photochemistry Reviews 3 (2003) 225–252

[6] S.C. Moon, H. Mametsuka, E. Suzuki, M. Anpo, Chem. Lett. (1998) 117. [7] M. Anpo, Catal. Surv. Jpn. 1 (1997) 169, and references therein. [8] M. Anpo, Y. Kubokawa, Res. Chem. Intermed. 8 (1987) 105. [9] A. Corma, Chem. Rev. 97 (1997) 2373, and references therein. [10] G. Centi, B. Wichterlova, A.T. Bell (Eds.), Catalysis by Unique Metal Ion Structures in Solid Matrices, NATO Science Series, vol. 13, 2001, and references therein. [11] B. Notari, Adv. Catal. 41 (1996) 253, and references therein. [12] N.J. Turro, X.G. Lei, S. Jockusch, W. Li, Z.Q. Liu, L. Abrams, M.F. Ottaviani, J. Org. Chem. 67 (2002) 2606. [13] J. Shailaja, J. Sivaguru, S. Uppili, A. Joy, V. Ramamurthy, Microporous Mesoporous Mater. 48 (2001) 319. [14] M. Anpo (Ed.), Photofunctional Zeolites, NOVA Publishers Inc., 2000, and references therein. [15] H. Nishiguchi, M. Anpo, J. Photochem. Photobiol. A 77 (1994) 183. [16] M. Anpo (Ed.), Surface Photochemistry, Wiley, New York, 1996, and references therein. [17] M. Anpo, S. Dohshi, M. Takeuchi, K. Ikeue, Curr. Opin. Solid State Mater. Sci. 6 (2002), in press. [18] H. Yamashita, Y. Ichihashi, M. Anpo, M. Hashimoto, C. Louis, M. Che, J. Phys. Chem. 100 (1996) 16041. [19] H. Yamashita, Y. Ichihashi, S.G. Zhang, Y. Matsumura, Y. Souma, T. Tatsumi, M. Anpo, Appl. Surf. Sci. 121 (1997) 305. [20] J.L. Zhang, Y. Hu, M. Matsuoka, H. Yamashita, M. Minagawa, H. Hidaka, M. Anpo, J. Phys. Chem. B 105 (2001) 8395. [21] S. Higashimoto, M. Matsuoka, S.G. Zhang, H. Yamashita, O. Kitao, H. Hidaka, M. Anpo, Microporous Mesoporous Mater. 48 (2001) 329. [22] M. Anpo, S.G. Zhang, H. Mishima, M. Matsuoka, H. Yamashita, Catal. Today 39 (1997) 159. [23] S. Higashimoto, K. Nishimoto, T. Ono, M. Anpo, Chem. Lett. (2000) 1160. [24] W.S. Ju, M. Matsuoka, M. Anpo, Catal. Lett. 71 (2001) 91. [25] H. Yamashita, S. Kawasaki, Y. Ichihashi, M. Harada, M. Takeuchi, M. Anpo, G. Stewart, M.A. Fox, C. Louis, M. Che, J. Phys. Chem. B 102 (1998) 5870. [26] H. Yamashita, M. Honda, M. Harada, Y. Ichihashi, M. Anpo, T. Hirao, N. Itoh, N. Iwamoto, J. Phys. Chem. B 102 (1998) 10707. [27] H. Yamashita, Y. Ichihashi, M. Harada, G. Stewart, M.A. Fox, M. Anpo, J. Catal. 158 (1996) 97. [28] M. Anpo, T. Suzuki, E. Giamello, M. Che, New Developments in Selective Oxidation, Elsevier, Amsterdam, 1990, p. 683. [29] M. Anpo, Y. Shioya, M. Che, Res. Chem. Intermed. 17 (1992) 15. [30] M. Anpo, T. Suzuki, Y. Kubokawa, F. Tanaka, S. Yamashita, J. Phys. Chem. 88 (1984) 5778. [31] M. Anpo, M. Kondo, S. Coluccia, C. Louis, M. Che, J. Am. Chem. Soc. 111 (1989) 8791. [32] K. Ikeue, S. Nozaki, M. Ogawa, M. Anpo, Catal. Lett. 80 (2002) 111. [33] K. Ikeue, S. Nozaki, M. Ogawa, M. Anpo, Catal. Today 74 (2002) 241. [34] H. Yamashita, K. Ikeue, T. Takewaki, M. Anpo, Top. Catal. 18 (2002) 95. [35] K. Ikeue, H. Yamashita, M. Anpo, T. Takewaki, J. Phys. Chem. B 105 (2001) 8350. [36] M. Anpo, M. Matsuoka, Y. Shioya, H. Yamashita, E. Giamello, C. Morterra, M. Che, H.H. Patterson, S. Webber, S. Ouellette, M.A. Fox, J. Phys. Chem. 98 (1994) 5744. [37] M. Anpo, M. Matsuoka, H. Mishima, H. Yamashita, Res. Chem. Intermed. 23 (1997) 197. [38] M. Matsuoka, E. Matsuda, K. Tsuji, H. Yamashita, M. Anpo, Chem. Lett. (1995) 375. [39] M. Matsuoka, E. Matsuda, K. Tsuji, H. Yamashita, M. Anpo, J. Mol. Catal. 107 (1996) 399. [40] M. Matsuoka, W.S. Ju, K. Takahashi, M. Anpo, J. Phys. Chem. B 104 (2000) 4911.

[41] M. Anpo, M. Matsuoka, H. Yamashita, Catal. Today 35 (1997) 177. [42] A. Miyamoto, D. Medhanayn, T. Inui, Appl. Catal. 28 (1986) 89. [43] T. Inui, D. Medhanavyn, P. Praserthdam, K. Fukuda, T. Ukawa, A. Sakamoto, A. Miyamoto, Appl. Catal. 18 (1985) 311. [44] S. Bordiga, S. Coluccia, C. Lamberti, L. Marchese, A. Zecchina, F. Boscherini, F. Buffa, F. Genoni, G. Leofaniti, G. Petrini, G. Vlaic, J. Phys. Chem. 98 (1994) 4125. [45] F. Cavani, F. Trifiro, P. Jiru, K. Habersberger, Z. Tvaruzkova, Zeolites 8 (1998) 12. [46] M. Anpo, M. Sunamoto, M. Che, J. Phys. Chem. 93 (1989) 1187. [47] H.H. Patterson, J. Cheng, S. Despres, M. Sunamoto, M. Anpo, J. Phys. Chem. 95 (1991) 8813. [48] M. Anpo, I. Tanahashi, Y. Kubokawa, J. Phys. Chem. 84 (1980) 3440. [49] M.S. Rigutto, H. Van Bekkum, Appl. Catal. 68 (L1) (1991). [50] G. Centi, S. Perathoner, F. Trifiró, A. Aboukais, C.F. Aiss, M. Gueltin, J. Phys. Chem. 96 (1992) 2617. [51] S.G. Zhang, M. Ariyuki, S. Higashimoto, H. Yamashita, M. Anpo, Microporous Mater. 21 (1998) 621. [52] M. Matsuoka, S. Higashimoto, H. Yamashita, M. Anpo, Res. Chem. Intermed. 26 (2000) 85. [53] T. Tanaka, H. Yamashita, R. Tsuchitani, T. Funabiki, S. Yoshida, J. Chem. Soc., Faraday Trans. 184 (1988) 2987. [54] M. Iwamoto, H. Furukawa, K. Matsukami, T. Takenaka, S. Kagawa, J. Am. Chem. Soc. 105 (1983) 3719. [55] S. Dzwigaj, M. Matsuoka, R. Franck, M. Anpo, M. Che, J. Phys. Chem. B 102 (1998) 6309. [56] S. Dzwigaj, M. Matsuoka, M. Anpo, M. Che, J. Phys. Chem. B 104 (2000) 6012. [57] S. Higashimoto, S.G. Zhang, H. Yamashita, Y. Matsumura, Y. Souma, M. Anpo, Chem. Lett. (1997) 1127. [58] S. Higashimoto, M. Matsuoka, H. Yamashita, M. Anpo, O. Kitao, H. Hidaka, M. Che, E. Giamello, J. Phys. Chem. B 104 (2000) 10288. [59] M. Anpo, S.G. Zhang, S. Higashimoto, M. Matsuoka, H. Yamashita, Y. Ichihashi, Y. Matsumura, Y. Souma, J Phys. Chem. B 103 (1999) 9295. [60] C. Louis, M. Che, M. Anpo, J. Catal. 141 (1993) 453. [61] W. Zhang, J. Wang, P.T. Tanev, T.J. Pinnavaia, J. Chem. Soc., Chem. Commun. (1996) 979. [62] S. Higashimoto, R. Tsumura, M. Matsuoka, H. Yamashita, M. Che, M. Anpo, Stud. Surf. Sci. Catal. 140 (2001) 315. [63] S. Higashimoto, R. Tsumura, S.G. Zhang, M. Matsuoka, H. Yamashita, C. Louis, M. Che, M. Anpo, Chem. Lett. (2000) 408. [64] H. Yamashita, K. Yoshizawa, M. Ariyuki, S. Higashimoto, M. Che, M. Anpo, J. Chem. Soc. Chem. Commun. (2001) 435. [65] H. Yamashita, M. Ariyuki, S. Higashimoto, S.G. Zhang, J.S. Chang, S.E. Park, J.M. Lee, Y. Matsumura, M. Anpo, J. Synchrotron Radiat. 6 (1999) 453. [66] W. Zhang, P.T. Tanev, T.J. Pinnavaia, Chem. Commun. (1996) 979. [67] D. Wei, N. Yao, G.L. Haller, Stud. Surf. Sci. Catal. 121 (1999) 239. [68] B.M. Weckhuysen, A.A. Verberckmoes, A.L. Buttiens, R.A. Schoonheydt, J. Phys. Chem. B 98 (1994) 579. [69] M. Anpo, I. Takahashi, Y. Kubokawa, J. Phys. Chem. 86 (1982) 1. [70] M.F. Hazenkamp, G. Blasse, J. Phys. Chem. 96 (1992) 3442. [71] T. Inoue, A. Fujishima, S. Konishi, K. Honda, Nature 277 (1979) 673. [72] J.C. Hemminger, R. Carr, G.A. Somorjai, Chem. Phys. Lett. 57 (1978) 100. [73] M. Halmann, in: M. Grätzel (Ed.), Energy Resources through Photochemistry and Catalysis, Academic Press, New York, 1983, p. 507. [74] M. Anpo, H. Yamashita, in: M. Schiavello (Ed.), Heterogeneous Photocatalysis, Wiley, Chichester, 1997, p. 133. [75] M. Anpo, K. Chiba, J. Mol. Catal. 74 (1992) 207. [76] H. Yamashita, H. Nishiguchi, N. Kamada, M. Anpo, Y. Teraoka, H. Hatano, S. Ehara, K. Kikui, L. Palmisano, A. Sclafani, M. Schiavello, M.A. Fox, Res. Chem. Intermed. 20 (1994) 815.

M. Matsuoka, M. Anpo / Journal of Photochemistry and Photobiology C: Photochemistry Reviews 3 (2003) 225–252 [77] H. Yamashita, N. Kamada, H. He, K. Tanaka, S. Ehara, M. Anpo, Chem. Lett. (1994) 855. [78] M. Anpo, H. Yamashita, Y. Ichihashi, S. Ehara, J. Electroanal. Chem. 21 (1995) 396. [79] F. Saladin, L. Forss, I. Kamber, J. Chem. Soc., Chem. Commun. (1995) 533. [80] M. Anpo, N. Aikawa, Y. Kubokawa, J. Chem. Soc., Chem. Commun. (1984) 644. [81] M. Anpo, N. Aikawa, Y. Kubokawa, M. Che, C. Louis, E. Giamello, J. Phys. Chem. 89 (1985) 5017. [82] M. Anpo, N. Aikawa, Y. Kubokawa, M. Che, C. Louis, E. Giamello, J. Phys. Chem. 89 (1985) 5689. [83] M. Anpo, T. Shima, T. Fujii, M. Che, Chem. Lett. (1987) 65. [84] M. Anpo, H. Yamashita, K. Ikeue, Y. Fujii, S.G. Zhang, Y. Ichihashi, D.R. Park, Y. Suzuki, K. Koyano, T. Tatsumi, Catal. Today 44 (1998) 327. [85] H. Yamashita, Y. Fujii, Y. Ichihashi, S.G. Zhang, K. Ikeue, D.R. Park, K. Koyano, T. Tatsumi, M. Anpo, Catal. Today 45 (1998) 221. [86] S.G. Zhang, Y. Fujii, H. Yamashita, K. Koyano, T. Tatsumi, M. Anpo, Chem. Lett. (1997) 659. [87] M. Anpo, H. Yamashita, Y. Ichihashi, Y. Fujii, M. Honda, J. Phys. Chem. B 101 (1997) 2632. [88] H. Courbon, P. Pichat, J. Chem. Soc., Faraday Trans. 180 (1984) 3175. [89] H. Bosch, F. Janssen, Catal. Today 2 (1988) 369. [90] M. Iwamoto, H. Yahiro, N. Mizuno, W.X. Zhang, Y. Mine, H. Furukawa, S. Kagawa, J. Phys. Chem. 96 (1992) 9360. [91] L.S. Kau, K.O. Hodgson, E.I. Solomon, J. Am. Chem. Soc. 111 (1989) 103. [92] N. Kosugi, H. Kondo, H. Tajima, H. Kuroda, Chem. Phys. 135 (1989) 149. [93] H. Yamashita, M. Matsuoka, K. Tsuji, Y. Shioya, M. Anpo, M. Che, J. Phys. Chem. 100 (1996) 397. [94] M. Anpo, M. Matsuoka, K. Hanou, H. Mishima, H. Yamashita, H.H. Patterson, Coord. Chem. Rev. 171 (1998) 175. [95] E. Giamello, D. Murphy, G. Magnacca, C. Morterra, Y. Shioya, T. Nomura, M. Anpo, J. Catal. 136 (1992) 510. [96] J. Valyon, W.K. Hall, J. Phys. Chem. 97 (1993) 7054. [97] D.S. McClure, Solid State Phys. 9 (1959) 399. [98] J.D. Barrie, B. Dunn, G. Hollingsworth, J.I. Zink, J. Phys. Chem. 93 (1989) 3958. [99] J. Dedecek, Z. Sobalik, Z. Tvaruzkova, D. Kaucky, B. Wichterlova, J. Phys. Chem. 99 (1995) 16327. [100] J. Dedecek, B. Wichterlova, J. Phys. Chem. 98 (1994) 5721. [101] J. Texter, D.H. Strome, R.G. Herman, K. Klier, J. Phys. Chem. 81 (1977) 333. [102] Z. Sojka, M. Che, E. Giamello, J. Phys. Chem. 101 (1997) 4831. [103] C.C. Chao, J.H. Lunsford, J. Phys. Chem. 76 (1972) 1546. [104] C. Naccache, M. Che, T.Y. Ben, Chem. Phys. Lett. 13 (1972) 109. [105] M. Anpo, M. Che, Adv. Catal. 44 (2000) 119. [106] T. Baba, Y. Ono, Zeolites 7 (1987) 292. [107] Y. Ono, T. Baba, K. Kanae, S.G. Seo, J. Chem. Soc. Jpn. 7 (1988) 985. [108] P.A. Jacobs, J.B. Uytterhoeven, H.K. Beyer, J. Chem. Soc., Chem. Commun. (1977) 128. [109] G. Calzaferri, S. Hug, T. Hugentobler, B. Sulzberger, J. Photochem. 26 (1984) 109. [110] G.A. Ozin, F. Hugues, J. Phys. Chem. 86 (1982) 5174. [111] S. Sato, Y. Yu-u, H. Yahiro, N. Mizuno, M. Iwamoto, Appl. Catal. 70 (1991) L1. [112] T. Miyadera, A. Abe, G. Muramatsu, K. Yoshida, in: Proceedings of the Symposium on Environmental Conscious Materials and Proceedings of the 3rd IUMRS International Conference on Advanced Materials, Tokyo, 1993, p. 405. [113] C.E. Moore, Atomic Energy Levels, vol. 3, National Bureau of Standards, Washington, DC, 1971, p. 48.

251

[114] A.N. Truklin, S.S. Etsin, A.V. Shendrik, Izv. Akad. Nauk. SSSR, Ser. Fiz. 40 (1976) 2329. [115] J. Texter, R. Kellerman, T. Gonsiorwski, J. Phys. Chem. 90 (1986) 2118. [116] G.A. Ozin, H. Huber, Inorg. Chem. 17 (1978) 155. [117] G.A. Ozin, F. Hugues, S.M. Matter, D.F. McIntosh, J. Phys. Chem. 87 (1983) 3445. [118] M. Matsuoka, H. Yamashita, M. Anpo, Catal. Catal. (Shokubai) 36 (1994) 76. [119] C. Chao, J.H. Lunsford, J. Phys. Chem. 78 (1974) 1174. [120] M. Matsuoka, W.-S. Ju, M. Anpo, Chem. Lett. 6 (2000) 626. [121] K. Ebitani, M. Morokuma, J.H. Kim, A. Morikawa, J. Catal. 141 (1993) 725. [122] K. Ebitani, M. Morokuma, J.H. Kim, A. Morikawa, J. Chem. Soc., Faraday Trans. 90 (1994) 377. [123] K. Ebitani, Y. Hirano, A. Morikawa, J. Catal. 157 (1995) 262. [124] G.H. Dieke, Spectra and Energy Levels of Rare Earth Ions in Crystals, Interscience, New York, 1968. [125] W.W. Piper, J.A. DeLuca, F.S. Ham, J. Lumin. 8 (1974) 344. [126] K. Ebitani, Y. Hirano, A. Morikawa, Catal. Catal. (Shokubai) 37 (1995) 416. [127] K. Ebitani, Y. Hirano, A. Morikawa, Catal. Catal. (Shokubai) 38 (1996) 177.

Masaya Matsuoka was born in 1969 in Osaka, Japan. He studied at the Osaka Prefecture University, where he obtained his PhD in applied chemistry studying on the photocatalysis on the transition metal ion exchanged zeolites in 1997. During 1997 he worked as a postdoctral fellow at the Université Pierre et Marie Curie, Paris. In 1998 he again joined the Osaka Prefecture University as a research associate. His current research interests include the development of the visible light responsive photocatalysts and their applications for the environmental purifications as well as the H2 production from pure water.

Masakazu Anpo, born on 15 August 1946, completed his PhD from Graduate School of Engineering, Osaka Prefecture University in 1975 and is presently a professor at the same university. His main areas of research are photochemistry of ketones adsorbed on solid surfaces such as zeolites and within zeolite cavities (heterogeneous photochemistry); photocatalysis of metal oxides such as TiO2 and ZnO; preparation of visible light-responsive TiO2 thin films; the local structures and photoreactivities of various transition metal oxides and cations highly dispersed within zeolite framework structures. He was offered the post of research associate at the National Research Council of Canada, University, Pierre et Marie Curie in Paris, Torino University, East China University of Science and Technology (Shanghai), Tokyo Institute of Technology, Nagoya University, The Tokyo University, Kyusyu University, etc. He was given the Japan Photochemical Society Award in 1994; the 28th Mitsubishi Foundation Award in 1998; and the Award of the Ueda Science Promotion in 2001. He was the plenary and invited lecturer at the Gordon Conference on Solid State Chemistry, Plymouth, USA; the 8th International Conference on Solar Energy Conversion and Storage, Palermo, Italy; the 12th International Conference on Catalysis, The Taniguchi Foundation, Kobe, Japan; the 10th International Conference on Solar Energy Conversion and Storage, Interlaken, Swiss; International Conference on Green Chemistry, Venice, Italy; IUPAC Congress

252

M. Matsuoka, M. Anpo / Journal of Photochemistry and Photobiology C: Photochemistry Reviews 3 (2003) 225–252

on Organic Chemistry, Warsaw, Poland; the 12th International Congress on Catalysis, Granada, Spain; the 6th International Conference on Solar Energy and Applied Photochemistry, Cairo, Egypt; the 13th International Zeolites Conference, Montpellier, France; the 3rd International Mesostructured Materials Symposium, Jeju, Korea; and many more. His works have been published in Hikarishokubai (Photocatalysis), Asakurashoten, Tokyo, Japan; Photochemistry on Solid Surfaces, Elsevier, Amsterdam, The Netherlands; Surface Photochemistry, Wiley, London, UK; Green Chemistry, Introduction, Ai-Pi-Shi, Tokyo, Japan; Photofunctional Zeolites, Nova Sciences, New York, USA; Green Chemistry, Oxford University Press, UK; Photocatalysis—Science and Technology, Kodansha Scientific Publications, Japan; and many more. His research works have

been published in total of 292 original papers, 51 books, 66 reviews and 55 proceedings. Anpo is the editor and member of the editorial board for various publications, such as Research on Chemical Intermediates, VSP, Zeist, The Netherlands; Catalysis Surveys from Japan, Baltzer Scientific Publications, The Netherlands; Applied Catalysis, A: General, Elsevier, The Netherlands; Catalysis Letters, Kluwer Academic Publishers, The Netherlands; Topics in Catalysis, Kluwer Academic Publishers, The Netherlands; The International Journal of Photoenergy (IJP), Photoerngy Center, Cairo, Egypt; Advances in Technology of Materials and Materials Processing Journal (ATM), NSW, Australia; Journal of Photochemistry and Photobiology, Elsevier, The Netherlands; Current Opinion in Solid State and Materials Science, Elsevier, The Netherlands.