Gold–Titanium(IV) Oxide Plasmonic Photocatalysts Prepared by a

Aug 19, 2012 - (5) To 750 cm3 of an aqueous tetrachloroauric acid (HAuCl4) solution (0.49 mmol dm–3), 100 cm3 of an aqueous solution containing sodi...
0 downloads 0 Views 1MB Size
Article pubs.acs.org/Langmuir

Gold−Titanium(IV) Oxide Plasmonic Photocatalysts Prepared by a Colloid-Photodeposition Method: Correlation Between Physical Properties and Photocatalytic Activities Atsuhiro Tanaka,† Asako Ogino,† Moe Iwaki,† Keiji Hashimoto,† Akira Ohnuma,‡ Fumiaki Amano,‡ Bunsho Ohtani,‡ and Hiroshi Kominami*,† †

Department of Applied Chemistry, Faculty of Science and Engineering, Kinki University, Kowakae, Higashiosaka, Osaka 577-8502, Japan ‡ Catalysis Research Center, Hokkaido University, Sapporo 001-0021, Japan S Supporting Information *

ABSTRACT: Colloidal gold (Au) nanoparticles were prepared and successfully loaded on titanium(IV) oxide (TiO2) without change in the original particle size using a method of colloid photodeposition operated in the presence of a hole scavenger (CPH). The prepared Au nanoparticles supported on TiO2 showed strong photoabsorption at around 550 nm due to surface plasmon resonance (SPR) of Au and exhibited a photocatalytic activity in mineralization of formic acid in aqueous suspensions under irradiation of visible light (>ca. 520 nm). A linear correlation between photocatalytic activity and the amount of Au loaded, that is, the number of Au nanoparticles, was observed, indicating that the activity of Au/TiO2 plasmonic photocatalysts can be controlled simply by the amount of Au loading using the CPH method and that the external surface area of Au nanoparticles is a decisive factor in mineralization of formic acid under visible light irradiation. Very high reaction rates were obtained in samples with 5 wt % Au or more, although the rate tended to be saturated. The CPH method can be widely applied for loading of Au nanoparticles on various TiO2 supports without change in the original size independent of the TiO2 phase. The rate of CO2 formation also increased linearly with increase in the external surface area of Au. Interestingly, the TiO2 supports showed different slopes of the plots. The slope is important for selection of TiO2 as a material supporting colloidal Au nanoparticles.

1. INTRODUCTION Noble metal nanoparticles have been studied extensively because of their unique properties. These unique properties are associated with their strong photoabsorption in the visible light region, which is due to surface plasmon resonance (SPR).1 The unique plasmon absorbance features of these noble metal nanoparticles have been exploited for a wide variety of applications. However, there are only a few reports on application of SPR-induced absorption to photochemical reactions.2−4 Recently, it has been reported that electron transfer from gold (Au) nanoparticles to a semiconductor (TiO2) occurred under irradiation of visible light (λ = ca. 550 nm) due to SPR.2a,b Subsequently, some research groups reported that chemical reactions such as oxidation of aromatic alcohols2c and oxidation of 2-propanol to acetone2d,e and hydrogen production from alcohols2f,g over Au/TiO2 occurred under irradiation of visible light and concluded that these reactions were induced by SPR of Au nanoparticles. We have found that Au nanoparticles supported on cerium(IV) oxide (Au/CeO2) exhibited absorption due to SPR more strongly than did various Au/TiO2 samples with the same Au content and that Au/CeO2 exhibited a much higher level of activity than did Au/TiO2 samples in mineralization3a,b and partial oxidation3c of organic compounds. We also succeeded in © 2012 American Chemical Society

shifting SPR to a longer length in copper (Cu)-Au/CeO2 and mineralization of organic acids in an aqueous suspension of Cu−Au/CeO2 even under irradiation of visible light of λ = 700 nm.4 In the selective oxidation of benzyl alcohol to benzaldehyde in aqueous suspensions of Au/CeO2 under irradiation of green light from an LED, we investigated effects of the amount of Au on particle size of Au photodeposited on a CeO2 support, SPR-induced photoabsorption of Au/CeO2 and the rate of photocatalytic benzaldehyde formation. From these results, we found that the rate of benzaldehyde formation exhibited linear dependency on the external surface area of Au loaded on CeO2, that is, the external surface area of Au loaded on CeO2 rather than the amount of loaded Au is an important factor controlling the photocatalytic activity of Au/CeO2. Still, TiO2 attracted our interest for use as a support for Au nanoparticles because TiO2 is inexpensive and its physical properties can be easily controlled. Although the effectiveness of Au/TiO2 plasmonic photocatalysts for various reactions have been studied by some researchers, the key factor(s) of Au/TiO2 plasmonic photocatalysts in mineralization of organic comReceived: May 12, 2012 Revised: August 8, 2012 Published: August 19, 2012 13105

dx.doi.org/10.1021/la301944b | Langmuir 2012, 28, 13105−13111

Langmuir

Article

MT-150A (rutile, 100 m2 −1) were also used to examine the effects of type of TiO2 on photocatalytic activities. (a) CI method: TiO2 powder was suspended in 10 cm3 of an aqueous solution of colloidal Au nanoparticles in a glass dish and was evaporated to dryness at 333 K. (b) CS method: TiO2 was suspended in 20 cm3 of an aqueous solution of colloidal Au nanoparticles at room temperature for 30 min with stirring in a centrifuging tube. Ammonium chloride (ca. 1 g) was added to 20 cm3 of the mixture for 30 min with stirring. The resulting powder was washed repeatedly with distilled water and air-dried. (c) CP method: TiO2 powder was suspended in 20 cm3 of an aqueous solution of colloidal Au nanoparticles (1.39 mg-Au/20 cm3) in a test tube and the test tube was sealed with a rubber septum under argon (Ar). The mixture was photoirradiated at λ > 300 nm by a 400-W high-pressure mercury arc (Eiko-sha, Osaka, Japan) under Ar with magnetic stirring in a water bath continuously kept at 298 K. The resultant powder was washed repeatedly with distilled water and then dried at 310 K overnight under air. (d) CPH method: Before the photoirradiation in the CP method, an aqueous solution (20 cm3) of oxalic acid (50 μmol) was injected into the sealed test tube. Other operations and conditions were the same as those for the CP method. (e) PH method: TiO2 powder (198 mg) was suspended in water (10 cm3) in test tubes and the test tubes were sealed with rubber septums under Ar. Aqueous solutions of oxalic acid (50 μmol) and tetrachloroauric acid were injected into the sealed test tubes and then photoirradiated at λ > 300 nm by a 400-W high-pressure mercury arc under Ar with magnetic stirring in a water bath continuously kept at 298 K. The Au source was reduced by photogenerated electrons in the conduction band of TiO2, and Au metal was deposited on TiO2 particles, resulting in the formation of Au/TiO2. The resultant powder was washed repeatedly with distilled water and then dried at 310 K overnight under air. Hereafter, Au/TiO2 having X wt% of Au is designated as Au(X)/ TiO2 and the kind of TiO2 is shown in the parentheses after TiO2; for example, P25 TiO2 having 1.0 wt % Au is shown as Au(1.0)/ TiO2(P25). The amounts of Au loaded on TiO2 were determined by atomic absorption spectrometry (AAS) after dissolving Au fixed on TiO2 with aqua regia. Colloidal Au nanoparticles were almost quantitatively loaded on TiO2 when the CI, CS and CPH methods were used, whereas only 50% of Au was fixed on TiO2 when the CP method was used. Almost 100% of Au was introduced on TiO2 when the PH method was used. 2.3. Characterization. Diffuse reflectance spectra of Au/TiO2 samples were obtained with a UV−visible spectrometer (UV-2400, Shimadzu, Kyoto) equipped with a diffuse reflectance measurement unit (ISR-2000, Shimadzu). The morphology of Au/TiO2 samples was observed under a JEOL JEM-3010 transmission electron microscope (TEM) operated at 300 kV in the Joint Research Center of Kinki University. 2.4. Mineralization of Organic Acid in Aqueous Suspensions of Au/TiO2 under Irradiation of Visible Light. Dried Au/TiO2 powder (50 mg) was suspended in distilled water (5 cm3) in a test tube. The aqueous mixture was bubbled with oxygen (O2) and the test tube was sealed with a rubber septum. Formic acid (100 μmol) was injected into the suspension and then irradiated with visible light from a Xe lamp (Ushio, Tokyo) with an O-54 cutoff filter (AGC Techno Glass) and IR cutoff filter (Sigma Koki, CDLF-50S) with magnetic stirring in a water bath continuously kept at 298 K. Spectra and light intensity of the Xe lamp (filtered) were determined using a spectroradiometer USR-45D (Ushio, Tokyo). The intensity of light irradiated to the reaction mixture was 150 mW cm−2 in the range of 500 to 750 nm. The amount of carbon dioxide (CO2) in the gas phase of the reaction mixture was measured using a gas chromatograph (GC8A, Shimadzu) equipped with Porapak QS columns. No CO2 was observed in the blank test, i.e., photoirradiation of visible light to

pounds has not been investigated in detail. Kowalska et al. reported2d that particle sizes of TiO2 and Au are the key factors for high level of activity. However, a possible reason for the high level of activity of their samples with larger Au particle size is the wider wavelength range, leading to a larger number of absorbed photons.2d In most studies on Au/TiO2 plasmonic photocatalysts, the photodeposition method was used for loading Au nanoparticles on the TiO2 support. This method utilizes photogenerated electrons for reduction of the Au source. Since Au nanoparticles are believed to be formed on “reduction sites” of the TiO2 surface, effective photocatalytic reduction is expected for prepared Au/TiO2. However, the photodeposition method often complicates understanding of the effect of Au loading amount on the activity of Au/TiO2 plasmonic photocatalysts because both the size and the number of Au nanoparticles simultaneously vary with the amount of Au loaded. Discussion of the effect of type of TiO2 on activities of Au/TiO2 plasmonic photocatalysts is generally difficult because properties of Au nanoparticles loaded are greatly affected by various factors of TiO2 such as particle (crystallite) size, crystallinity, crystalline phase and impurity of TiO2. If the size of each Au nanoparticle loaded on the TiO2 support is unchanged when the amount of Au is increased, the difference in the photocatalytic activities can be attributed to change in the number of Au nanoparticles. If Au nanoparticles having the same size are loaded on different TiO2 supports without change in the size, the difference in the photocatalytic activities can be attributed to the difference in properties of TiO2 supports. In this study, we examined preparation of Au/TiO2 having a sharp size distribution of Au nanoparticles without agglomeration by using various methods. Using colloid photodeposition methods in the absence and presence of a hole scavenger, Au nanoparticles were successfully loaded on representative commercial TiO2 samples having different properties without change in the original size of Au. Here we report the preparation of Au/TiO2 with various amounts of Au loading and the correlations between amount of Au, TiO2 properties and photocatalytic activity in mineralization of organic acids under irradiation of visible light. We examined the reaction by using strictly limited visible light (500−750 nm in wavelength) in order to rule out the contribution of the original photocatalytic activity of TiO2, which can be excited with UV light, as proposed by Yuzawa et al.2g

2. EXPERIMENTAL SECTION 2.1. Preparation of Colloidal Au Nanoparticles. Colloidal Au nanoparticles were prepared using the method reported by Frens.5 To 750 cm3 of an aqueous tetrachloroauric acid (HAuCl4) solution (0.49 mmol dm−3), 100 cm3 of an aqueous solution containing sodium citrate (39 mol dm−3) was added. The solution was heated and boiled for 1 h. After the color of the solution had changed from deep blue to deep red, the solution was boiled for a further 30 min. After cooling the solution to room temperature, Amberlite MB-1 (ORGANO, 60 cm3) was added to remove excess sodium citrate. After 1-h treatment, MB-1 was removed from the solution using a glass filter. 2.2. Preparation of Au/TiO2 Samples. Four methods, that is, (a) colloid impregnation (CI), (b) colloid salting-out (CS), (c) colloid photodeposition (CP), and (d) colloid photodeposition with a hole scaveger (CPH), were used to load colloidal Au nanoparticles on TiO2 supports. The conventional photodeposition with hole scavenger (PH) method was also used for preparation of Au/TiO2 to compare with Au/TiO2 preapred by the CPH method. In most of the experiments, Degussa P25 (mixture of anatase and rutile, 50 m2 −1) was used as the support for Au nanoparticles, and other commercial TiO2 powders, that is, Ishihara ST-01 (anatase, 300 m2 −1) and Tayca 13106

dx.doi.org/10.1021/la301944b | Langmuir 2012, 28, 13105−13111

Langmuir

Article

aqueous suspension of the Au/TiO2 samples prepared by CPH method in the absence of formic acid.

nanoparticles occurred during the loading processes and that fixation of Au nanoparticles on the TiO2 support as those before loading, i.e., without change in the size, is difficult. These results suggest that Au nanoparticles were subjected to a large stress during the processes of these methods, resulting in change in the particle size. Figure 2c and d shows TEM photographs of Au/TiO2 prepared by the CP and CPH methods, respectively. In contrast to Au/TiO2 prepared by the CI and CS methods, Au nanoparticles were successfully fixed on TiO2 as those before loading, that is, keeping the original particle size. Very recently, loading of Au nanoparticles on TiO2 using the photodeposition method (similar to the CP method used in this study) was reported by Yuzawa et al., although an increase in the size of Au nanoparticles was observed after Au loading on TiO2.2g As mentioned in the experimental procedure, Au loading was insufficient when the CP method was used. The only difference between the CPH and CP methods is whether oxalic acid was added or not before photoirradiation. Oxalic acid is generally used to remove holes and to increase the rate of reduction in photodeposition of metals and photocatalytic reactions. Other hole scavengers such as formic acid and methanol were also used for loading of Au nanoparticles on TiO2. The results obtained by using the CP method suggest that a strong reductive condition was required to fix a large amount of Au nanoparticles on the surface of TiO2. Figure 2e shows a TEM photograph of Au/TiO2 prepared by the PH method. Fine Au nanoparticles having an average diameter of ca. 10−30 nm were observed as many researchers have already reported.2d,e,g Since colloidal Au nanoparticles were quantitatively loaded on P25 TiO2 without change in particle size only by using the CPH method as mentioned above, Au(X)/TiO2(P25) samples having various Au loading amounts (X) were prepared by the CPH method. Figure 3 shows TEM photographs of Au(X)/ TiO2(P25) samples (X = 0.1, 0.5, 1.0 and 2.0). The average diameter of Au particles of Au(1.0)/TiO2(P25) was determined

3. RESULTS AND DISCUSSION 3.1. Effects of Methods on Loading of Colloidal Au Nanoparticles on TiO2 Supports. Figure 1 shows a TEM

Figure 1. TEM image of particles (left) and size distribution (right) of Au particles.

photograph and distribution of colloidal Au nanoparticles, revealing that Au nanoparticles have an average particle size of 12.8 nm within a relatively sharp distribution with a standard deviation of 1.4 nm. Figure 2 shows TEM photographs of Au(0.5)/TiO2(P25) samples prepared by the five methods. The size of Au nanoparticles was changed after loading of Au on P25 TiO2 by the CI and CS methods as shown in Figure 2a and b (and Figures S1a and b, Supporting Information), respectively, indicating that division and/or aggregation of Au

Figure 2. TEM images of Au(0.5)/TiO2(P25) samples prepared by (a) CI, (b) CS, (c) CP, (d) CPH and (e) PH methods.

Figure 3. TEM images of Au(X)/TiO2(P25) samples (X = (a) 0.1, (b) 0.5, (c) 1.0 and (d) 2.0) prepared by the CPH method. 13107

dx.doi.org/10.1021/la301944b | Langmuir 2012, 28, 13105−13111

Langmuir

Article

to be 12.7 nm, which is in good agreement with the average diameter (12.8 nm, see Figure 1) of original colloidal Au nanoparticles before Au loading. Similar results were obtained for Au(X)/TiO2(P25) samples with larger X. It should be noted that the average particle size was not changed even though the amount of Au loading was increased up to 2.0 wt % as shown in Figure 4a. On the other hand, the average diameter

Figure 5. (a) Absorption spectra of Au(X)/TiO2(P25) samples (X = 0, 0.1, 0.5, 1.0, 2.0, 3.0, 3.5, 5.0, 6.0 and 7.0) prepared by the CPH method and visible light irradiated to reaction systems (formic acid in aqueous suspensions of Au/TiO2 samples) from a Xe lamp with an O54 filter and (b) effect of amount of Au on the Kubelka−Munk function at 550 nm of Au(X)/TiO2(P25) samples.

No clear changes in other physical properties such as crystalline phase and specific surface area were observed for a series of the Au(X)/TiO2(P25) samples. Therefore, the increase in Au content (and thereat, the change in interaction between Au nanoparticles) can be attributed to the changes in SPRphotoabsorption of a series of the Au(X)/TiO2(P25) samples. 3.3. Mineralization of Organic Acid in Aqueous Suspensions of Au/TiO2 Prepared by the CPH Method. Figure 6 shows time courses of evolution of CO2 from formic

Figure 4. Effect of Au loading amounts (X) on distribution of Au nanoparticles of Au(X)/TiO2(P25) samples prepared by (a) CPH and (b) PH methods.

of Au nanoparticles of Au(X)/TiO2(P25) samples prepared by the PH method increased with increase in X as shown in Figure 4b. These results indicate that only the number of Au nanoparticles increased with increase in the amount of Au loading when the CPH method was used and that effect(s) of the size of Au nanoparticle can be ruled out in discussion about SPR-photoabsorption and photocatalytic activities of Au(X)/ TiO2(P25) samples prepared by the CPH method with different amounts of Au loading. 3.2. Characterization of Au/TiO2 Prepared by the CPH Method. Figure 5a shows absorption spectra of Au(X)/ TiO2(P25) samples (X = 0, 0.1, 0.5, 1.0, 2.0, 3.0, 3.5, 5.0, 6.0, and 7.0 wt %) prepared using the CPH method. Although TiO2 exhibited only an absorption at λ < 400 nm due to the band gap excitation, another photoabsorption was observed in the spectrum of Au/TiO2 at around 550 nm, which was attributed to SPR of the supported Au nanoparticles.2,3 SPR-photoabsorption was observed in the spectra of a series of the Au(X)/TiO2(P25) samples and the photoabsorption increased with increase in X. This tendency is more clearly shown in Figure 5b, in which the Kubelka−Munk function at 550 nm of Au(X)/TiO2(P25) samples was plotted against X. The Kubelka−Munk function at 550 nm increased almost linearly until X = 3.0 and tended to be slightly saturated after X = 3.0.

Figure 6. Time courses of evolution of CO2 from formic acid in aqueous suspensions of Au(X)/TiO2(P25) samples (X = 0.1, 0.5, 1.0, 2.0, 3.0, 3.5, 5.0, 6.0 and 7.0) prepared by the CPH method under irradiation of visible light.

acid in aqueous suspensions of Au(X)/TiO2(P25) (X = 0.1− 7.0) under irradiation of visible light from a Xe lamp with an O54 filter at 298 K. Visible light irradiated to the reaction system is shown in Figure 5a. Just after irradiation of visible light, CO2 was evolved and formation of CO2 continued almost linearly with irradiation time, indicating that formic acid was decomposed to CO2 (eq 1) under visible light irradiation. HCOOH + 1/2O2 = CO2 + H 2O

(1)

To elucidate the applicability of Au/TiO2 prepared by the CPH method, mineralization of oxalic acid and acetic acid in aqueous suspensions of Au(5)/TiO2(P25) was also examined under the same condition and the results are shown in Figure S2 (Supporting Information). In both cases, CO2 was linearly evolved just after irradiation of visible light as well as formic acid. These results show that oxalic acid and acetic acid were also decomposed to CO2 as well as formic acid under visible light irradiation in the presence of Au(5.0)/TiO2(P25). On the other hand, no CO2 was formed in control experiments, that is, 13108

dx.doi.org/10.1021/la301944b | Langmuir 2012, 28, 13105−13111

Langmuir

Article

neither photochemical reaction of organic acid in the absence of Au/TiO2, photocatalytic reaction induced by bare TiO2 nor dark reaction (thermocatalytic reaction) by Au/TiO2 at 298 K occurred. From the linear part of plot shown in Figure 6, the reaction rate of each Au(X)/TiO2(P25) prepared by the CPH method was calculated and the effect of the amount of Au (X) on the rate of CO2 evolution is shown in Figure 7. The rate of CO2

Table 1. Various Properties of Au(X)/TiO2(P25) Samples Prepared by the CPH Method Xa/ wt%

NAub/ 1012

SAuc/ 10−3 m2

0.1 0.5 1.0 2.0 3.0 3.5 5.0 6.0 7.0

0.28 1.4 2.8 5.6 8.4 9.9 14 17 20

0.60 3.0 6.0 12 18 21 30 36 42

1-Rd

(1-R)/ NAu/ (1013)−1

rCO2e/ μmol h−1

rCO2/SAu/μmol h−1(10−3)−1

0.53 0.74 0.80 0.89 0.89 0.90 0.94 0.92 0.93

19 5.3 2.8 1.6 1.1 0.91 0.67 0.54 0.47

0.88 4.8 11 25 31 34 46 50 55

1.5 1.6 1.9 2.1 1.7 1.6 1.5 1.3 1.1

a

Au loading. bNAu: number density of Au nanoparticles in 50 mg of Au/TiO2. cExternal surface area of Au nanoparticles in 50 mg of Au/ TiO2. These values were calculated from the average size of the Au nanoparticles (12.8 nm) and density of Au metal (19.32 g cm−3). d Photoabsorption at 550 nm due to SPR. eRate of CO2 evolution from formic acid. Au/TiO2: 50 mg, Light intensity: 150 mW cm−2 in the range of 500−750 nm.

Figure 7. Effect of Au loading amounts (X) on the rate of evolution of CO2 from formic acid in aqueous suspensions of Au(X)/TiO2(P25) samples prepared by the CPH method (■) and PH method (○) under irradiation of visible light.

spherical. Photoabsorption at 550 nm due to SPR (1-R) and ratio of 1-R to NAu ((1-R)/NAu) are shown in Table 1. The latter can be used as an indication of the efficiency of Au nanoparticles for SPR-photoabsorption. The value of (1-R)/ NAu decreased with increase in X. In contrast to the decrease in efficiency in SPR-photoabsorption, almost constant values of rate of CO2 formation per surface area of Au nanoparticles (rCO2/SAu) were observed independent of the increase in X up to 3.5, as expected from Figure 7. Similar results have been obtained in the selective oxidation of benzyl alcohol to benzaldehyde in aqueous suspensions of Au/CeO2 plasmonic photocatalysts,3c although reaction systems and method for Au loading were different in this study. We proposed that oxidation of organic compounds such as formic acid and benzyl alcohol occurred on the surface of Au nanoparticles fixed on CeO2.3b,c Therefore, we conclude that the external surface area of Au nanoparticles had a decisive effect on the reaction rate in the present system. 3.5. Effect of the Kind of TiO2 Support. To investigate effects of the types of TiO2 support, Au nanoparticles were loaded on representative anatase and rutile TiO2 powders (Ishihara ST01, 300 m2 g−1 and Tayca MT-150A, 100 m2 g−1, respectively) using the CPH method in the same manner as that for preparation of Au/TiO2(P25). Figure 8 shows TEM photographs of Au(0.5)/TiO 2 (ST01) and Au(0.5)/ TiO2(MT150A). These photographs can be compared with that of Au(0.5)/TiO2(P25) shown in Figure 3b. The sizes of Au nanoparticles in Au(0.5)/TiO2(ST01) and Au(0.5)/ TiO2(MT150A) observed in the photographs were in agreement with the average diameter of the original Au nanoparticles before loading. These results indicate that the CPH method can be widely applied for loading of Au nanoparticles on various TiO2 supports without change in the original size independent of the TiO2 phase. Since Au(X)/TiO2(ST01) and Au(X)/ TiO2(MT150A) samples having various values of X were successfully prepared, their photoabsorption and photocatalytic activity in mineralization of formic acid under irradiation of visible light were investigated as well as Au(X)/TiO2(P25). External surface areas of Au nanoparticles per 50 mg of Au(X)/ TiO2(ST01) and Au(X)/TiO2(MT150A) (SAu) were calculated from X and the average size of Au nanoparticles, and

evolution increased almost linearly with increase in the amount of Au loading up to X = 3.5. Very large reaction rates were obtained in the samples with X > 5, although the rate tended to be saturated against the increase in X. These results show that activity of Au/TiO2 plasmonic photocatalysts can be controlled simply by the amount of Au loading using the CPH method. For comparison, Au/TiO2 samples with various Au contents were prepared by the PH method. The activities of Au(X)/ TiO2(P25) samples prepared by the PH method with X = 0.5 and 1.0 were slightly larger than those of Au/TiO2 with the same X prepared by the CPH method as shown in Figure 7; however, the reaction rate of Au(X)/TiO2(P25) prepared by the PH method was saturated even at X = 1.0 and further increase in X decreased activities. This behavior was in contrast to the reaction rate of Au(X)/TiO2(P25) prepared by the CPH method, which increased until at least X = 7.0. These results indicate that the PH method might be useful for preparation of Au/TiO2 with small Au contents; however, this method cannot be applied for preparation of Au/TiO2 with large amounts of Au loading. 3.4. Correlation between Photocatalytic Activity and Amount of Au Loading. As shown in Figure 7, the rate of CO2 evolution increased almost linearly with increase in X up to 3.5. Since Au nanoparticles were fixed on TiO2 by the CPH method as those before loading, the number of Au particles simply increased with increase in X. This means that the external surface area of Au nanoparticles also simply increased with increase in X. Therefore, the linear correlation between photocatalytic activity of Au(X)/TiO2(P25) samples and X suggests that the external surface area of Au nanoparticles is one of important factors controlling the activity of an Au/TiO2 plasmonic photocatalyst under the present reaction conditions. Table 1 summarizes the number and surface area of Au nanoparticles (NAu and SAu, respectively) per 50 mg of Au/ TiO2 calculated from the average size of Au nanoparticles (13 nm) and the density of Au metal (19.32 g cm−2) on the basis of the assumption of that all of the Au nanoparticles were 13109

dx.doi.org/10.1021/la301944b | Langmuir 2012, 28, 13105−13111

Langmuir

Article

Figure 8. TEM images of (a) Au(0.5)/TiO2(ST01) and (b) Au(0.5)/TiO2(MT150A) samples prepared by the CPH method.

supporting colloidal Au nanoparticles, although the slope reflects many factors of physical and chemical properties of Au/ TiO2 and TiO2 and it is not clear which factor is predominant.

photoabsorption at 550 nm and the rate of CO2 formation were plotted against SAu (Figure 9). For comparison, results for



CONCLUSIONS Colloidal gold (Au) nanoparticles were loaded on titanium(IV) oxide (TiO2) supports by using various methods. When colloid impregnation and colloid salting-out methods were used, division and/or aggregation of Au nanoparticles occurred. When colloid photodeposition in the presence of a hole scavenger (CPH) was used, Au nanoparticles were successfully fixed on TiO2 as those before loading, that is, keeping the original particle size. The rate of CO 2 evolution in mineralization of formic acid in aqueous suspensions of Au/ TiO2 under irradiation of visible light (>ca. 520 nm) increased almost linearly with increase in the amount of Au loading, indicating that the activity of Au/TiO2 plasmonic photocatalysts can be controlled simply by the amount of Au loading using the CPH method. This means that the external surface area of Au nanoparticles also simply increased with increase in the amount of Au and suggests that the external surface area of Au nanoparticles is one of important factors controlling the activity of an Au/TiO2 plasmonic photocatalyst under the present reaction conditions. Very large reaction rates were obtained in the samples with 5 wt % Au or more, although the rate tended to be saturated. On the other hand, the reaction rate of Au/TiO2(P25) prepared by the conventional photodeposition in the presence of hole scavenger method was saturated even at 1.0 wt % and further increase in the amount of Au decreased activities. The CPH method can be widely applied for loading of Au nanoparticles on various TiO2 supports without change in the original size independent of the TiO2 phase. The rates of CO2 formation for a series of Au/ TiO2(ST01) and Au/TiO2(MT150A) samples also increased linearly with increase in SAu; however, three TiO2 supports showed different slopes of the plots. The slope is important for selection of TiO2 as a material supporting colloidal Au nanoparticles, although the slope reflects many factors of physical and chemical properties of Au/TiO2 and TiO2.

Figure 9. Effects of external surface area of Au nanoparticles loaded on representative TiO2 supports on (a) [1-reflection] at 550 nm and (b) rate of CO2 evolution from formic acid in aqueous suspensions of various Au/TiO2 samples under irradiation of visible light.

Au(X)/TiO2(P25) are also shown in the same figure. The same tendency of SPR-photoabsorption was observed for the three TiO2 supports, that is, the SPR-photoabsorption drastically increased at a small SAu (0.60 × 10−3 m2, corresponding to X = 0.1) and tended to be saturated at around SAu = 3.0 × 10−3 m2, corresponding to X = 0.5. However, slight differences were observed in the three TiO2 supports, suggesting that SPRphotoabsorption was slightly affected by the kind of TiO2 support probably due to change in interactions between Au nanoparticles and the TiO2 surface, although we cannot find a clear correlation between physical and chemical properties of TiO2 and SPR-photoabsorption. The rates of CO2 formation for a series of Au/TiO2(P25), Au/TiO2(ST01) and Au/TiO2(MT150A) samples increased linearly with increase in SAu, as shown in Figure 9b. It is interesting that Au/TiO2(P25), Au/TiO2(ST01) and Au/ TiO2(MT150A) samples exhibited different slopes of the plots. The slope is important for selection of TiO2 as a material



ASSOCIATED CONTENT

S Supporting Information *

Supplemental figures. This material is available free of charge via the Internet at http://pubs.acs.org. 13110

dx.doi.org/10.1021/la301944b | Langmuir 2012, 28, 13105−13111

Langmuir



Article

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This work was partly supported by a Grant-in-Aid for Scientific Research (No. 23560935) from the Ministry of Education, Culture, Sports, Science, and Technology (MEXT) of Japan and by the Cooperative Research Program of Catalysis Research Center, Hokkaido University (Grant # 10A0006 and 12B1006).



REFERENCES

(1) Ghosh, S. K.; Pal, T. Interparticle Coupling Effect on the Surface Plasmon Resonance of Gold Nanoparticles: From Theory to Applications. Chem. Rev. 2007, 107, 4797−4862. (2) (a) Tian, Y.; Tatsuma, T. Mechanisms and Applications of Plasmon-Induced Charge Separation at TiO2 Films Loaded with Gold Nanoparticles. J. Am. Chem. Soc. 2005, 127, 7632−7637. (b) Furube, A.; Du, L.; Hara, K.; Katoh, R.; Tachiya, M. Ultrafast Plasmon-Induced Electron Transfer from Gold Nanodots into TiO2 Nanoparticles. J. Am. Chem. Soc. 2007, 129, 14852−14853. (c) Naya, S.; Teranishi, M.; Isobe, T.; Tada, H. Light Wavelength-Switchable Photocatalytic Reaction by Gold Nanoparticle-Loaded Titanium(IV) Dioxide. Chem. Commun. 2010, 46, 815−817. (d) Kowalska, E.; Abe, R.; Ohtani, B. Visible Light-Induced Photocatalytic Reaction of GoldModified Titanium(IV) Oxide Particles: Action Spectrum Analysis. Chem. Commun. 2009, 241−243. (e) Kowalska, E.; Mahaney, O. O. P.; Abe, R.; Ohtani, B. Visible-Light-Induced Photocatalysis Through Surface Plasmon Excitation of Gold on Titania Surfaces. Phys. Chem. Chem. Phys. 2010, 12, 2344−2355. (f) Silva, C. G.; Juarez, R.; Marino, T.; Molinari, R.; Garcia, H. Influence of Excitation Wavelength (UV or Visible Light) on the Photocatalytic Activity of Titania Containing Gold Nanoparticles for the Generation of Hydrogen or Oxygen from Water. J. Am. Chem. Soc. 2011, 133, 595−602. (g) Yuzawa, H.; Yoshida, T.; Yoshida, H. Gold Nanoparticles on Titanium Oxide Effective for Photocatalytic Hydrogen Formation under Visible Light. Appl. Catal., B 2012, 115, 294−302. (3) (a) Kominami, H.; Tanaka, A.; Hashimoto, K. Mineralization of Organic Acids in Aqueous Suspensions of Gold Nanoparticles Supported on Cerium(IV) Oxide Powder under Visible Light Irradiation. Chem. Commun. 2010, 46, 1287−1289. (b) Kominami, H.; Tanaka, A.; Hashimoto, K. Gold Nanoparticles Supported on Cerium(IV) Oxide Powder for Mineralization of Organic Acids in Aqueous Suspensions under Irradiation of Visible Light of λ = 530nm. Appl. Catal., A 2011, 397, 121−126. (c) Tanaka, A.; Hashimoto, K.; Kominami, H. Selective Photocatalytic Oxidation of Aromatic Alcohols to Aldehydes in an Aqueous Suspension of Gold Nanoparticles Supported on Cerium(IV) Oxide under Irradiation of Green light. Chem. Commun. 2011, 47, 10446−10448. (4) Tanaka, A.; Hashimoto, K.; Kominami, H. Gold and Copper Nanoparticles Supported on Cerium(IV) Oxide-A Photocatalyst Mineralizing Organic Acids under Red Light Irradiation. ChemCatChem 2011, 3, 1619−1623. (5) Frens, G. Controlled Nucleation for the Regulation of the Particle Size in Monodisperse Gold Suspensions. Nat. Phys. Sci. 1973, 241, 20−22.

13111

dx.doi.org/10.1021/la301944b | Langmuir 2012, 28, 13105−13111