J. Phys. Chem. B 2000, 104, 4911-4915
4911
Photocatalytic Decomposition of N2O into N2 and O2 at 298 K on Cu(I) Ion Catalysts Anchored onto Various Oxides. The Effect of the Coordination State of the Cu(I) Ions on the Photocatalytic Reactivity Masaya Matsuoka, Woo-Sung Ju, Kenzo Takahashi, Hiromi Yamashita, and Masakazu Anpo* Department of Applied Chemistry, Graduate School of Engineering, Osaka Prefecture UniVersity, Gakuen-cho 1-1, Sakai, Osaka 599-8531, Japan ReceiVed: NoVember 10, 1999; In Final Form: February 14, 2000
Cu(I) (Cu+) ion catalysts anchored onto various oxides (SiO2‚Al2O3, Al2O3, and SiO2) were prepared by the combination of an ion-exchange method and a thermovacuum treatment. XAFS (X-ray absorption fine structure) investigations revealed that Cu+ ions exist as linear two-coordinate Cu+ on SiO2‚Al2O3, while they exist as planar three-coordinate Cu+ on Al2O3 or SiO2. It was also found that the typical photoluminescence observed at 430 nm for the Cu+/(SiO2‚Al2O3) catalyst and that observed at 510 nm for the Cu+/Al2O3 and Cu+/SiO2 catalysts could be attributed to the radiative decay from the excited electronic state of the linear two-coordinate Cu+ ions and planar three-coordinate Cu+ ions, respectively. The addition of N2O onto the Cu+ ion catalyst leads to the quenching of the photoluminescence of Cu+, indicating that N2O interacts with the photoexcited Cu+ ion. UV irradiation of the catalysts in the presence of N2O led to the formation of N2 and O2 at 298 K, indicating that the photocatalytic decomposition of N2O proceeds on the Cu+ ion catalysts. The reaction was found to proceed more efficiently on the Cu+/(SiO2‚Al2O3) catalyst than on the Cu+/Al2O3 or Cu+/SiO2 catalysts, suggesting that the two-coordinate Cu+ species show higher activity for this reaction than the three-coordinate Cu+ species.
1. Introduction Copper ions exchanged within the zeolite cavities show various catalytic activities such as the direct decomposition of NO,1 the selective reduction of NO by hydrocarbons,2 the oxidation reaction of hydrocarbons3 or aromatics,4 and the cyclodimerization of butadiene.5 It has been also reported that Cu(I) (Cu+) ion catalysts supported on zeolites or silica show high photocatalytic activity for the decomposition of NO6-8 and N2O into N2 and O2 at 278 K.9,10 On these reactions, the catalytic activity of the catalyst is greatly affected by the local structure of the copper ions which is easily modified by changing the types of supports and loadings of the metal ions. Elucidating the local structure of the active species in the catalytic reactions is of great importance in the design of new types of catalysts which are highly efficient and selective. In situ photoluminescence studies of the catalysts are also very useful in investigating the surface structure as well as the nature of the excited states of highly dispersed metal oxides or metal ions due to the high sensitivity and nondestructive nature of this photoluminescence technique.11-13 Fundamentals and applications of this photoluminescence techniques to adsorption, catalysis and photocatalysis have been recently summarized by Anpo and Che.14,15 Utilizing these advantages, the local structures and redox behavior of Cu+ introduced into zeolites or anchored onto metal oxides have been investigated.16-23 This study deals with the ESR, XAFS (XANES and EXAFS), and photoluminescence investigations of the Cu+/(SiO2‚Al2O3), Cu+/Al2O3, and Cu+/ SiO2 catalysts prepared by a combination of an ion-exchange method and the thermovacuum treatment of Cu(II) (Cu2+) ion samples in order to clarify the effect of the kind of supports on the local environment of the Cu+ ions. The effects of the coordination geometry of the Cu+ ions on the wavelength
regions of the corresponding observed photoluminescence bands and on their photocatalytic reactivity for the decomposition of N2O at 298 K have also been investigated. 2. Experimental Section Cu2+/(SiO2‚Al2O3), Cu2+/Al2O3, and Cu2+/SiO2 samples were prepared by ion-exchange with an aqueous (Cu(NH3)4)2+ solution. After being washed with water and drying in air at 373 K, the copper loadings of the samples were determined by an inductively coupled plasma emission spectrometer: the loading of copper cations as metal was 0.76 wt % for SiO2‚ Al2O3, 0.55 wt % for Al2O3, and 0.50 wt % for SiO2, respectively. The SiO2‚Al2O3 binary oxide (SiO2/Al2O3 mol ratio ) 24.6) was prepared by coprecipitation of the desired amounts of a mixed solution of Si(OC2H5)4, AlCl3‚6H2O, ethanol and water by the addition of an aqueous solution of ammonia. Al2O3 and SiO2 (Aerosil 130) for the supports were supplied by the Japan Aerosil Corporation. A quartz cell with window and furnace sections connected to a vacuum system was used for pretreatments and the in situ measurements of the spectra. Prior to spectroscopic measurements, the Cu2+ ion samples were evacuated at the desired temperatures for 1 h. The photoluminescence spectra of the catalyst were recorded at 298 K with a Shimadzu RF-501 spectrofluorophotometer. The ESR spectra were recorded at 298 K using a JES-RE2X spectrometer operating in the X-band mode and equipped with a dual cavity. The XAFS spectra (XANES and EXAFS) were obtained at the BL-10B facility of the Photon Factory at the High Energy Accelerator Research Organization, Tsukuba. Si(311) channelcut crystal was used to monochromatize the X-rays from the 2.5 GeV electron storage ring. The Cu K-edge absorption spectra were recorded in the transmission mode at 298 K under vacuum. Photon energy was calibrated by characteristic preedge peaks
10.1021/jp9940001 CCC: $19.00 © 2000 American Chemical Society Published on Web 05/03/2000
4912 J. Phys. Chem. B, Vol. 104, No. 20, 2000
Matsuoka et al.
Figure 1. XANES (a) and Fourier transformed EXAFS (FT-EXAFS) (a′) spectra of the Cu2+/(SiO2‚Al2O3) sample evacuated at 298 K.
TABLE 1: Results of the Curve-Fitting of Cu K-Edge EXAFS Data for Cu2+ or Cu+ Ion Catalysts Anchored onto Various Oxides catalyst
tempa(K)
shell
R (Å)b
CNc
σ2 (Å2)d
Cu /(SiO2‚Al2O3) Cu+/(SiO2‚Al2O3) Cu+/Al2O3 Cu+/SiO2 CuO (reference) Cu2O (reference)
298 973 973 973
Cu-N Cu-O Cu-O Cu-O Cu-O Cu-O
2.00 1.92 1.90 1.86 1.96 1.85
4.1 (Cu-N) 1.9 (Cu-O) 2.9 (Cu-O) 3.1 (Cu-O) 4 (Cu-O) 2 (Cu-O)
0.0020 0.0024 0.0020 0.0035
2+
a
temp: evacuation temperature. bR: Bond distances. c CN: Coordination number. d σ2: Debye-Waller factor.
in the absorption spectrum of a Cu foil (8980 eV). The normalized spectra were obtained by procedures described in previous papers,24 and Fourier transformations were performed by k3-weighted EXAFS oscillations in the range of 3-12 Å-1. The curve fittings of the EXAFS data were carried out by employing the iterative nonlinear least-squares method of Levernberg24 and the empirical backscattering parameter sets extracted from the shell features of the copper compounds. Photocatalytic reactions were performed using a high-pressure Hg lamp and water filter at 298 K. The reaction products were analyzed by gas chromatography. 3. Results and Discussion ESR and XAFS Investigations of the Cu+ Ion Catalysts. The Cu2+/(SiO2‚Al2O3) sample degassed at 298 K gives a broad and axial ESR signal (g| ) 2.25, A| ) 169 G) due to the typical square planar tetraammine copper(II) complex: (Cu(NH3)4)2+.25-27 The signal is anisotropic at 298 K, indicating that the tetraammine copper(II) complex is fixed onto the surface of the SiO2‚ Al2O3 binary oxide. Figure 1 a shows the Cu K-edge XANES spectrum of the Cu2+/(SiO2‚Al2O3) sample degassed at 298 K. The XANES spectrum exhibits a well-separated weak preedge band due to the 1s-3d transition A, as well as an intense band due to the 1s-4p transition B,C.28-32 The presence of the band A attributed to the 1s-3d transition indicates that the copper species exist in a divalent state.33 The band B attributed to the 1s-4pz transition can be observed as a shoulder of the intense band C attributed to the 1s-4px,y transition accompanied by their weak shake-down bands B′,C′ induced by the ligand effect.28-32 These peaks are typical for the Cu2+ ion having an unoccupied d orbital (3d9) exhibiting planar or distorted octahedral symmetry.30,31 Figure 1a′ shows the Cu K-edge Fourier transformed EXAFS (FT-EXAFS) spectrum of the Cu2+/(SiO2‚Al2O3) sample. The FT-EXAFS spectrum of the sample exhibits an intense peak at around 1.5 Å (without any corrections for phase shift) due to the neighboring N atoms, as suggested by the ESR measurements. As shown in Table 1, the curve fitting of the peak yields a Cu-N bond length of (R) ) 2.00 Å and a coordination number
Figure 2. (left) XANES and (right) FT-EXAFS spectra of the (a,a′) Cu+/(SiO2‚Al2O3), (b, b′) Cu+/Al2O3, and (c, c′) Cu+/SiO2 catalysts.
of CN ) 4.1, suggesting that the square planar tetraammine copper(II) complex is formed on the SiO2‚Al2O3 surface. The bond length is in good agreement with the crystallographic Cu-N distance (2.03 Å) in the tetraammine copper(II) sulfate reference and the Cu-N distance (2.00 Å) of the tetraammine copper(II) species in aqueous solution,34 which is longer than the average Cu-O bond in the square planar CuO4 grouping of copper(II) oxides (1.96 Å). Figure 2 a,a′ shows the Cu K-edge XANES and FT-EXAFS spectra of the Cu+/(SiO2‚Al2O3) catalyst which was prepared by the evacuation of the original Cu2+/(SiO2‚Al2O3) sample at 973 K. As shown in Figure 2a, the Cu+/(SiO2‚Al2O3) catalyst exhibits a very strong and sharp band B due to the 1s-4pz transition. This band is clearly separated from the band C attributed to the 1s-4px,y transition. In a planar or a linear geometry, the 1s-4pz transition is not affected by the ligands, therefore, the copper compounds having these geometries exhibit a strong and sharp band B attributed to the 1s-4pz transition.28-32 Band B attributed to the 1s-4pz transition, without any shakedown bands, is intense enough to identify the copper species as the isolated Cu+ ions with linear two-coordination geometry.28,29 As shown in Figure 2a′, FT-EXAFS investigations of the Cu+/ (SiO2‚Al2O3) catalyst show a peak at around 1.5 Å due to the neighboring O atoms. A peak due to the neighboring Cu atoms cannot be observed, indicating that Cu+ are anchored onto the SiO2‚Al2O3 binary oxide in an isolated state. As shown in Table 1, the curve fitting of the Cu-O peak yields a Cu-O bond length of R ) 1.92 Å and a coordination number of CN ) 1.9, indicating that a two-coordinate Cu+ ion is formed on the SiO2‚ Al2O3 surface, as suggested by the XANES investigation. Figure 3 shows the effects of the evacuation temperature of the Cu2+/ (SiO2‚Al2O3) sample on the relative intensity of the ESR signal due to Cu2+ and the relative intensity of the XANES band due to the 1s-4pz transition of Cu+. As can be seen in Figure 3, the temperature dependence of the intensity of the ESR and XANES band clearly indicate that the Cu2+ species are
Photocatalytic Decomposition of N2O
Figure 3. Effect of the degassing temperature of the Cu2+/(SiO2‚Al2O3) sample on the intensity of the ESR signal due to Cu2+ and on the intensity of the preedge peak of the XANES spectrum due to Cu+.
Figure 4. The photoluminescence spectra of the (a) Cu+/(SiO2‚Al2O3), (b) Cu+/Al2O3, and (c) Cu+/SiO2 catalysts and the effect of the addition of N2O on the photoluminescence spectrum of the (d) Cu+/(SiO2‚Al2O3) catalyst. The addition of N2O was carried out at 298 K. N2O pressure (in Torr): (a) 0, (d) 20.
chemically reduced to isolated Cu+ ions by the evacuation of the Cu2+/(SiO2‚Al2O3) sample at temperatures higher than 573 K. Figure 2b,c show the Cu K-edge XANES spectra of the Cu+/ Al2O3 and Cu+/SiO2 catalysts which were prepared by the evacuation of the original Cu2+/Al2O3 and Cu2+/SiO2 samples at 973 K. The XANES spectra of Cu+/Al2O3 and Cu+/SiO2 exhibit a sharp band (B) due to the 1s-4pz transition without the corresponding shake-down peak, indicating the formation of the Cu+ species on the supports. However, the intensity of the band is smaller as compared with that of the Cu+/(SiO2‚ Al2O3) catalyst, suggesting that Cu+ ions with planar threecoordinate geometry are formed on these supports.28-32 As shown in Figure 2b′,c′, peaks due to the neighboring Cu atoms are not observed in both cases, indicating that Cu+ exist in an isolated state. As shown in Table 1, the curve-fitting analysis of the Cu-O peak observed in the FT-EXAFS spectra (b′,c′) of these catalysts yield a Cu-O bond length of R ) 1.90 Å and a coordination number of CN ) 2.9 for Cu+/Al2O3, and a Cu-O bond length of R ) 1.86 Å and a coordination number of CN ) 3.1 for Cu+/SiO2, showing that three-coordinate Cu+ ions are formed on these catalysts. The Photoluminescence Investigation of the Cu+ Ion Catalysts. As shown in Figure 4 a, after the evacuation of the Cu2+/(SiO2‚Al2O3) sample at temperatures above 573 K, the photoluminescence due to the Cu+ ion at around 430 nm can be observed upon irradiation at around 300 nm. The intensity of the photoluminescence spectrum increases when the evacuation temperature of the sample increases and reaches a
J. Phys. Chem. B, Vol. 104, No. 20, 2000 4913 maximum at 973 K. Considering that most of the copper species exist as linear two-coordinate Cu+, as indicated by the results of XAFS, it can be concluded that the absorption (excitation) band at around 300 nm and the photoluminescence band at around 430 nm are attributed to the electronic excitation of the linear two-coordinate Cu+ ion (3d10 f 3d94s1) and its reverse radiative deactivation (3d94s1 f 3d10), respectively.16-23 On the other hand, it was found that the shape and peak position of the photoluminescence are significantly affected when the type of oxide support is changed from the SiO2‚Al2O3 binary oxide to Al2O3 or SiO2. As shown in Figure 4, the Cu+/ Al2O3 and Cu+/SiO2 catalysts exhibit photoluminescence spectra at around 510 nm upon irradiation at around 300 nm. There are two assignments for the photoluminescence observed in the longer wavelength regions compared with the photoluminescence band of the Cu+ gas (456 nm).35 The photoluminescence band at 545 nm observed for the Cu+-doped Na+-β′′-alumina is attributed to the radiative decay process of the Cu+-Cu+ dimer, i.e., the σ(4s) f σ*(3d) electronic transition.18,19 On the other hand, the photoluminescence band observed at 540 nm in the case of the Cu+/ZSM-5 catalyst is attributed to the radiative decay process of the Cu+ monomer adjacent to one framework Al atom, i.e., the 3d94s1 f 3d10 electronic transition.20-23 Considering that most of the Cu+ ions supported on Al2O3 and SiO2 exist as isolated planar three-coordinate Cu+, as indicated by the results of XAFS, it can be concluded that the absorption (excitation) band at around 300 nm and the photoluminescence band at around 510 nm are attributed to the electronic excitation of the planer three-coordinate Cu+ monomer (3d10 f 3d94s1) and its reverse radiative deactivation (3d94s1 f 3d10), respectively, in the case of the Cu+/Al2O3 and Cu+/ SiO2 catalysts. These results are in good agreement with those reported by Texter et al. showing that the photoluminescence at 540 nm observed for the Cu+/Y zeolite is attributed to the radiative deactivation process (3d94s1f3d10) of the Cu+ monomer in the C3V coordination field.16,17 Changes in the Coordination States of Copper Cations during Thermovacuum Treatment. As discussed above, Cu2+/ (SiO2‚Al2O3) degassed at 298 K shows a typical ESR signal assigned to the tetraammine copper(II) complex, while Cu2+/ Al2O3 and Cu2+/SiO2 exhibit broad ESR signals (Cu2+/Al2O3: g| ) 2.31, A| ) 157 G, g⊥ ) 2.06; Cu2+/SiO2: g| ) 2.27, A| ) 168 G, g⊥ ) 2.06) due to the distorted octahedral Cu2+ having ammonia and water molecules as ligands.25,36 Evacuation at 723 K leads to changes in the ESR spectral profiles and the appearance of the well-resolved, structured signals, indicating that the ligand sphere of the Cu2+ site changed from NH3 and H2O to lattice O2- ions of the substrates. The ESR parameters of the Cu2+ samples evacuated at 723 K are listed in Table 2. Cu2+/(SiO2‚Al2O3) shows an ESR signal (g| ) 2.314 and A| ) 157 G) due to the Cu2+ in a square pyramidal coordination sphere37 after evacuation at 723 K. A further increase in the evacuation temperature higher than 723 K leads to a decrease in the intensity of the signal with very minor changes in its ESR parameters, and at the same time, an increase in the intensity of the photoluminescence at 430 nm can be observed. This indicates that the square pyramidal Cu2+ is reduced to linear two-coordinate Cu+. Quite similar results have been reported for the Cu2+/ZSM-5 system by Deˇdecˇek et al.21 showing that Cu2+ ion exchanged in the vicinity of the two Al atoms (g| ) 2.31, A| ) 150-160 G; square pyramidal) is reduced by H2 treatment to a Cu+ moiety identified by its photoluminescence at 450 nm, which may be essentially the same photoluminescence moiety at 430 nm observed for Cu+/(SiO2‚Al2O3), thus
4914 J. Phys. Chem. B, Vol. 104, No. 20, 2000
Matsuoka et al.
TABLE 2: Coordination Parameters of Cu Ions: Characteristic ESR of Cu2+ and Photoluminescence of Cu+
a
substrates
ESR parametersa
SiO2‚Al2O3
g| ) 2.314, A| ) 157 G, g⊥ ) 2.05
Al2O3
g| ) 2.310, A| ) 159 G, g⊥ ) 2.06
SiO2
g1 ) 2.519, g2 ) 2.150, g3 ) 2.050 (without hyperfine structure)
coordination of Cu2+
photoluminescence of Cu+ (nm)b
coordination of Cu+
square pyramidal square pyramidal tetrahedral
430
linear twocoordinate planar threecoordinate planar threecoordinate
510 510
Samples were evacuated at 723 K. b Catalysts were prepared by evacuation of Cu2+ samples at 973 K.
Figure 5. Time profiles of the photocatalytic decomposition reaction of N2O into N2 and O2 at 298 K on the (a) Cu+/(SiO2‚Al2O3), (b) Cu+/ Al2O3, and (c) Cu+/SiO2 catalysts.
suggesting that the two-coordinate Cu+ on SiO2‚Al2O3 locates near the two Al sites (-AlO(SiO)2Al-). On the other hand, Cu2+/Al2O3 degassed at 723 K shows an ESR signal due to the square pyramidal Cu2+ (g| ) 2.310 and A| ) 159 G),37 while Cu2+/SiO2 shows an ESR signal g1 ) 2.519, g2 ) 2.150, and g3 ) 2.050 without the hyperfine structure which is typical to Cu2+ in a tetrahedral crystal field.38 As shown in Table 2, it is evident that both the Cu2+ moiety on Al2O3 and SiO2 is reduced to the three-coordinate Cu+. These results show that the coordination state of the Cu moiety drastically changes during the reduction process, the extent being dependent on the kind of substrates. It can be considered that linear two-coordinate Cu+ is easily stabilized in substrates composed of silica and alumina such as zeolites32 which have a strong negative charge on their framework, while on inert substrates, such as Al2O3 or SiO2, the Cu+ moiety needs to be coordinated by more oxygen to compensate their positive charge by the framework oxygen. The Excited States of Cu+ Ion Catalysts and Their Reaction with N2O. The interaction of N2O with the photoexcited Cu+ ion was investigated by means of the photoluminescence measurements. As shown in Figure 4d, the addition of N2O onto the Cu+/(SiO2‚Al2O3) catalyst led to a quenching of the photoluminescence and, at the same time, the shortening of the lifetime from 72.4 to 69.3 µs, indicating that N2O interacts with the photoexcited Cu+ ion. Figure 5 shows the time profiles of the photocatalytic decomposition of N2O on the Cu+/(SiO2‚ Al2O3), Cu+/Al2O3 and Cu+/SiO2 catalysts. UV irradiation of the Cu+ ion catalyst in the presence of N2O led to the formation of N2 and O2 with an increase in the UV irradiation time, while under dark conditions, the formation of N2 and O2 could not be observed. Moreover, over the Cu2+/(SiO2‚Al2O3)(Ox) catalyst (prepared by the calcination of the Cu2+/(SiO2‚Al2O3) sample at 773 K in air followed by evacuation at 773 K), the reaction proceeds at a rate of less than 10% of that on the Cu+/(SiO2‚ Al2O3) catalyst. These results clearly indicate that the Cu+ ion catalyst acts as an efficient photocatalyst for the direct decom-
position of N2O. UV irradiation of the catalysts in the presence of N2O through an UV-27 cut filter (λ > 270 nm) led to the formation of N2 and O2, while under UV irradiation through the UV-32 cut filter (λ>320 nm), no reaction products were formed. These results suggest that UV wavelengths shorter than 320 nm, where the absorption band of the Cu+ ions locate (300 nm), is the most effective, i.e., the photoexcited state of the Cu+ ion plays a significant role in the reaction. Although it is difficult to determine the actual quantum yields in such very small powdered systems, the results in Figure 5 clearly show that Cu+/(SiO2‚Al2O3) exhibits higher activity for the photocatalytic decomposition of N2O than the Cu+/Al2O3 or Cu+/SiO2 catalysts, suggesting that the linear two-coordinate Cu+ shows higher activity than the planar three-coordinate Cu+. This can be attributed to the high intensity and long lifetime (72.4 µs) of the photoluminescence of the linear two-coordinate Cu+ observed for Cu+/(SiO2‚Al2O3). Figure 5 also shows the O2/N2 ratio in the photocatalytic decomposition products on the Cu+ ion catalysts. It is evident that the O2/N2 ratio is much higher on the Cu+/(SiO2‚Al2O3) catalyst than on the other two catalysts, indicating that linear two-coordinate Cu+ can release O2 much easier than planar three-coordinate Cu+. This may be one of the reasons for the higher photocatalytic activity of twocoordinate Cu+ than that of planar three-coordinate Cu+. The photocatalytic activity of Cu+/ZSM-5 (prepared by the evacuation of the Cu2+/ZSM-5 at 923 K) for N2O decomposition was also investigated. It has been reported that Cu+/ZSM-5 has the unique feature that it can easily release oxygen at low temperatures.1 The photocatalytic activity of the Cu+/ZSM-5 catalyst where the majority of Cu moieties are linear twocoordinate Cu+ (ref 32) is found to be 1.5 times higher than that of Cu+/(SiO2‚Al2O3), i.e., 5.7 mmol (g of Cu)-1 h-1 (yield of N2), while the O2/N2 ratio in the reaction products is 0.32, which is also greater than that of Cu+/(SiO2‚Al2O3) (O2/N2 ) 0.25), indicating that the O2/N2 ratio is closely related to the stability of the Cu+ ion under oxygen atmosphere, one of the key factors affecting the photocatalytic activity for N2O decomposition. It should also be noted that Cu+ ion catalysts show photocatalytic activity for the decomposition of NO into N2 and O2, while it has been found that the reaction rate increases in the following order: Cu+/SiO2 (yield of N2; 20 µmol (g of Cu)-1 h-1) < Cu+/Al2O3 (24 µmol (g of Cu)-1 h-1) < Cu+/(SiO2‚ Al2O3) (30 µmol (g of Cu)-1 h-1) < Cu+/ZSM-5 (41 µmol (g of Cu)-1 h-1). It has also been elucidated by dynamic ESR investigations that the electron transfer of the 4s electron of the photoexcited Cu+ into the anti-π-orbital of NO plays a significant role in this reaction.6,7 Since the rate of the photocatalytic decomposition of N2O on these catalysts increases exactly in the same order observed in the reaction with NO, it can be concluded that the photocatalytic decomposition of N2O proceeds under a similar mechanism proposed in the case of NO, i.e., the 4s electron transfer from the photoexcited Cu+
Photocatalytic Decomposition of N2O into the antibonding molecular orbital (LUMO) of N2O initiates the decomposition of N2O to produce N2 and O2. From these results, it was found that the linear two-coordinate Cu+ is selectively produced on the SiO2‚Al2O3 binary oxide, which has a higher extent of coordinative unsaturation than the planar three-coordinate Cu+ species on Al2O3 or SiO2. The linear two-coordinate Cu+ exhibits a photoluminescence with a high yield and a long lifetime, as well as the highest and most unique photocatalytic reactivity for the decomposition of N2O into N2 and O2 with a high O2/N2 ratio. The efficient quenching of the photoluminescence of two-coordinate Cu+ in the presence of N2O suggests that in the photocatalytic decomposition reaction of N2O, the electron transfer from the photoexcited Cu+ ion into the antibonding molecular orbital of N2O initiates the decomposition of N2O into N2 and an adsorbed oxygen atom, Oads, which easily recombines to form O2. 4. Conclusions The local structure of the Cu+ catalysts anchored onto various oxides (SiO2‚Al2O3, Al2O3, and SiO2) prepared by a combination of an ion-exchange method and a thermovacuum treatment were clarified by means of in situ ESR, XAFS, and photoluminescence measurements. It was found that the thermovacuum treatment of the as-prepared Cu2+ samples at 973 K leads to the reduction of Cu2+ to the Cu+. XAFS and photoluminescence results suggest that, by using a SiO2‚Al2O3 binary oxide as a support, linear two-coordinate Cu+ can be selectively produced, exhibiting a typical photoluminescence spectrum at around 430 nm upon irradiation at 300 nm, while by using Al2O3 or SiO2 as a support, planar three-coordinate Cu+ ions are produced, exhibiting a photoluminescence spectrum at around 510 nm. The addition of N2O on Cu+ catalysts led to the efficient quenching of the photoluminescence of Cu+, indicating that N2O interacts with the photoexcited Cu+ ion. The Cu+ ion catalysts decompose N2O into N2 and O2 at 298 K under UV irradiation of the absorption band of the Cu+ ion at around 300 nm, indicating that the photoexcited state of the Cu+ ions plays a significant role in the photocatalytic decomposition of N2O. Comprehensive investigations into the relationship between the local structures and the photocatalytic reactivities of the Cu+/ oxide catalysts showed that two-coordinate Cu+ species show higher activity for the photocatalytic decomposition of N2O than three-coordinate Cu+ species, clearly indicating the importance of the coordinative unsaturation of the active sites. References and Notes (1) Iwamoto, M.; Yahiro, H.; Tanda, K.; Mizuno, N.; Mine, Y.; Kagawa, S. J. Phys. Chem. 1991, 95, 3727. (2) Sato, S.; Yu-u, Y.; Yahiro, H.; Mizuno, N.; Iwamoto, M. Appl. Catal. 1991, 70, L1.
J. Phys. Chem. B, Vol. 104, No. 20, 2000 4915 (3) Kevan, L.; Yu, J. S. Res. Chem. Intermed. 1991, 15, 67. (4) Nagao, M.; Nishiyama, S.; Tsuruya, S.; Masai, M. Inorg. Chem. 1992, 31, 4662. (5) Maxwell, I. E.; Downing, R. S.; van Langen, S. A. J. J. Catal. 1980, 61, 485. (6) Anpo, M.; Matsuoka, M.; Shioya, Y.; Yamashita, H.; Giamello, E.; Morterra, C.; Che, M.; Patterson, H. H.; Webber, S.; Ouellette, S.; Fox, M. A. J. Phys. Chem. 1994, 98, 5744. (7) Anpo, M.; Nomura, T.; Kitao, T.; Giamello, E.; Murphy, D.; Che, M.; Fox, M. A. Res. Chem. Intermed. 1991, 15, 225. (8) Negishi, N.; Matsuoka, M.; Yamashita, H.; Anpo, M. J. Phys. Chem. 1993, 97, 5211. (9) Ebitani, K.; Morokuma, M.; Kim, J. H.; Morikawa, A. J. Catal. 1993, 141, 725. (10) Ebitani, K.; Morokuma, M.; Kim, J. H.; Morikawa, A. J. Chem. Soc., Faraday Trans. 1994, 90, 377. (11) Anpo, M.; Kondo, M.; Coluccia, S.; Louis, C.; Che, M. J. Am. Chem. Soc. 1989, 111, 8791. (12) Anpo, M.; Aikawa, N.; Kubokawa, Y.; Che, M.; Louis, C.; Giamello, E. J. Phys. Chem. 1985, 89, 5017. (13) Dzwigaj, S.; Matsuoka, M.; Franck, R.; Anpo, M.; Che, M. J. Phys. Chem. B 1998, 102, 6309. (14) Anpo, M.; In Heterogeneous Catalysis; Ertl, G., Kno¨zinger, H., Weitkamp, J., Eds.; VCH, A Wiley Company: New York, 1997; p 664. (15) Anpo, M.; Che, M. AdV. Catal. 1999, 44, 119. (16) Texter, J.; Storme, D. H.; Herman, R. G.; Klier, K. J. Phys. Chem. 1977, 81, 333. (17) Storme, D. H.; Klier, K. J. Phys. Chem. 1980, 84, 981. (18) Barrie, J. D.; Dunn, B.; Stafsudd, O. M.; Nelson, P. J. Luminesc. 1987, 37, 303. (19) Barrie, J. D.; Dunn, B.; Hollingsworth, G.; Zink, J. I. J. Phys. Chem. 1989, 93, 3958. (20) Deˇdecˇek, J.; Wichterlova´, B. J. Phys. Chem. 1994, 98, 5721. (21) Deˇdecˇek, J.; Sobalı´k, Z.; Tvaru˚zˇkova´, Z.; Kaucky´, D.; Wichterlova´, B. J. Phys. Chem. 1995, 99, 16327. (22) Wichterlova´, B.; Deˇdecˇek, J.; Sobalı´k, Z.; Vondrova´, A.; Klier, K. J. Catal. 1997, 169, 194. (23) Deˇdecˇek, J.; Wichterlova´, B. J. Phys. Chem. B 1997, 101, 10233. (24) Tanaka, T.; Yamashita, H.; Tsuchitani, R.; Funabiki, T.; Yoshida, S. J. Chem. Soc., Faraday Trans. 1 1988, 84, 2987. (25) Turkevich, J.; Ono, Y.; Soria, J. J. Catal. 1972, 25, 44. (26) Vansant, E. F.; Lunsford, J. H. J. Phys. Chem. 1972, 76, 2860. (27) Flentge, D. R.; Lunsford, J. H.; Jacobs, P. A.; Uytterhoven, J. B. J. Phys. Chem. 1975, 79, 354. (28) Kau, L. S.; Spira-Solomon, D. J.; Penner-Halm, J. E.; Hodgson, K. O.; Solomon, E. I. J. Am. Chem. Soc. 1987, 109, 6433. (29) Kau, L. S.; Hodgson, K. O.; Solomon, E. I. J. Am. Chem. Soc. 1989, 111, 7103. (30) Kosugi, N.; Yokoyama, K.; Asakura, K.; Kuroda, H. Chem. Phys. 1984, 91, 249. (31) Kosugi, N.; Kondo, H.; Tajima, H.; Kuroda, H. Chem. Phys. 1989, 135, 149. (32) Yamashita, H.; Matsuoka, M.; Tsuji, K.; Shioya, Y.; Anpo, M.; Che, M. J. Phys. Chem. 1996, 100, 397. (33) Nomura, M.; Kazusaka, A.; Kakuta, N.; Ukisu, Y.; Miyahara, K. Chem. Phys. Lett. 1985, 122, 538. (34) Lamble, G.; Moen, A.; Nicholson, D. G. J. Chem. Soc., Faraday Trans. 1994, 90, 2211. (35) McClure, D. S. Solid State Phys. 1959, 9, 339. (36) Narayana, M.; Kevan, L. J. Chem. Soc., Faraday Trans. 1 1986, 82, 213. (37) Kucherov, A. V.; Slinkin, A. A.; Kondratev, D. A.; Bondarenko, T. N.; Rubinstein, A. M.; Minachev, K. M. Zeolites 1985, 5, 320. (38) Tominaga, H.; Ono, Y.; Keii, T. J. Catal 1975, 40, 197.
