Preparation of Au/TiO₂ exhibiting strong surface plasmon resonance effective for photoinduced hydrogen formation from organic and inorganic compounds under irradiation of visible light

Abstract

Au/TiO₂ samples exhibiting stronger photoabsorption at around 550 nm due to surface plasmon resonance were prepared by using a multi-step photodeposition method and the samples exhibited higher levels of activity for H₂ production from various compounds such as 2-propanol, ethanol and ammonia in aqueous suspensions under visible light irradiation.

---

The development of a clean and renewable energy carrier has become a subject of high priority. Hydrogen (H₂) production by semiconductor photocatalysts under irradiation of visible light has attracted much attention because it offers a promising way for clean, low-cost and environmentally friendly production of H₂ by solar light.¹ Recently, it has been reported that electron transfer from gold (Au) nanoparticles to titanium(IV) oxide (TiO₂) occurred under irradiation of visible light (λ = ca. 550 nm) due to surface plasmon resonance (SPR).^2a,b Some research groups have since found that chemical reactions, i.e., oxidation of organic compounds, occurred in an aqueous suspension of Au supported on TiO₂ (Au/TiO₂), a new type of catalytic material responding to visible light different from semiconductor photocatalysts, and it was concluded that these reactions were induced by SPR of supported Au nanoparticles.^2c–e 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 the activity levels of Au/TiO₂ samples in mineralization^3a,b and partial oxidation^3c of organic compounds. We also succeeded in shift of 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 Very recently, Silva et al.⁵ reported production of H₂ from EDTA or methanol in aqueous suspensions of Au/TiO₂ (in which Degussa P 25 TiO₂ was used) under irradiation of visible light and proposed that H₂ was formed by reduction of protons with electrons in the conduction bands of TiO₂ injected from Au nanoparticles. Since, unfortunately, their Au/TiO₂ samples were slightly deactivated within several hours of photoirradiation, understanding photocatalytic properties of Au/TiO₂ and preparation of Au/TiO₂ samples with high stability and high levels of activity are highly important. As is apparent from several studies, supported Au nanoparticles (Au/TiO₂ and Au/CeO₂) have various possibilities as photocatalytic materials utilizing visible light.

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 and we think that a combination of Au nanoparticles exhibiting strong SPR and a TiO₂ sample suitable for Au fixation, electron transfer and substrate adsorption is important for higher levels of activity in photoinduced reactions. However, effects of the method for Au deposition and size distribution of Au particles on the activity in photoinduced reactions have not been reported, although there are many papers reporting Au photodeposition on TiO₂ and the effects on (thermo)catalytic activities of Au/TiO₂. In this study, by using a multi-step (MS) photodeposition method⁴ in which addition of metal sources and photodeposition of the metals on semiconductor particles were repeated several times to obtain a desired metal loading, we succeeded in preparation of Au nanoparticles supported on TiO₂⁶ (MS-Au/TiO₂) exhibiting SPR absorption stronger than that of Au/TiO₂ prepared by using a single-step photodeposition method (SS-Au/TiO₂). Here we report a larger rate of photoinduced H₂ production from various substrates (2-propanol, ethanol, methanol, ammonia and benzyl alcohol) in aqueous suspensions of MS-Au/TiO₂ samples under irradiation of visible light.

Fig. 1(a) shows a TEM photograph of SS-(1.00)Au/TiO₂. The amount of Au loaded was generally 1.00 wt% and is shown in parentheses before Au if necessary. Small Au particles were observed in the TEM photograph of SS-(1.00)Au/TiO₂ and the average diameter of Au particles was determined to be 1.2 nm, suggesting that the SS photodeposition method can be used to highly disperse Au nanoparticles on the surface of TiO₂. Fig. 1(b)–(e) show TEM photographs of MS-Au/TiO₂ samples on which 0.25, 0.50, 0.75 and 1.00 wt% of Au was loaded by repeating 0.25 wt% loading of Au. Two types of Au particles having different sizes were observed in the 0.50, 0.75 and 1.00 wt% samples, although the number of larger particles was less than that of smaller particles. As shown in Fig. 1 and 2, the average sizes of larger Au particles gradually increased with increase in times of Au loading. These results indicate that the Au sources added after the first photodeposition in the MS photodeposition method were deposited on Au particles previously formed on the TiO₂ surface, resulting in growth of the Au particles. The original size of some of the Au particles formed at the first deposition was preserved and the average particle sizes of MS-(1.00)Au/TiO₂ samples were determined to be 1.4 nm and 13 nm, respectively (ESI†, Fig. S1).

Fig. 1 TEM image of (a) SS-(1.00)Au/TiO₂, (b) MS-(0.25)Au/TiO₂, (c) MS-(0.50)Au/TiO₂, (d) MS-(0.75)Au/TiO₂, (e) MS-(1.00)Au/TiO₂, and (f) sample obtained by post-calcination of MS-(1.00)Au/TiO₂ at 500 °C.

Fig. 2 The average size of larger and smaller Au nanoparticles loaded on MS-Au/TiO₂ having different Au loadings. Values in the figure are the amounts and the times of Au loading.

Fig. 3(a) shows intensity of photoabsorption at 550 nm of MS-Au/TiO₂ having different Au loadings and Fig. 3(b) shows absorption spectra of SS- and MS-(1.00)Au/TiO₂ samples. Photoabsorption at 550 nm was attributed to SPR of the supported Au particles as reported previously.^2–5 It should be noted that MS-(1.00)Au/TiO₂ exhibited much stronger photoabsorption than that of SS-(1.00)Au/TiO₂, although the amounts of Au loaded on TiO₂ were the same. This result suggests that the intensity of photoabsorption due to SPR of Au was affected by the size of Au nanoparticles. Kowalska et al. reported that particle size and photoabsorption peak of Au supported on TiO₂ were affected by properties of TiO₂.^2d,e

Fig. 3 (a) Intensity of photoabsorption of MS-Au/TiO₂ having different Au loadings (circles) and SS-(1.00)Au/TiO₂ (square), (b) absorption spectra of SS-(1.00)Au/TiO₂ (broken line) and MS-(1.00)Au/TiO₂ (solid line), and light intensity of visible light irradiated to reaction systems from a Xe lamp with a Y-48 filter (right axis). Values in the figure are the amounts and the times of Au loading.

Fig. 4(a) shows time courses of evolution of H₂ from 2-propanol in aqueous suspensions of SS- and MS-(1.00)Au/TiO₂ under irradiation of visible light in the absence of oxygen. For comparison, results for Au-free TiO₂ are also shown in the figure. Just after irradiation of visible light, H₂ was evolved from suspensions of SS- and MS-Au/TiO₂, while no H₂ was formed from the suspension of Au-free TiO₂. Since H₂ increased linearly with photoirradiation time, the rates of H₂ evolution were determined to be 0.15 and 2.7 μmol h⁻¹, respectively, i.e., MS-Au/TiO₂ exhibited a rate of H₂ formation ca. 20-times larger than that of SS-Au/TiO₂. MS-(0.50)Au/TiO₂ and MS-(0.75)Au/TiO₂ with smaller Au contents exhibited rates (0.44 and 0.98 μmol h⁻¹, respectively) larger than that of SS-(1.00)Au/TiO₂. Intense SPR absorption of MS-Au/TiO₂ due to the larger Au particles is one of the main reasons of the extraordinarily large activity for H₂ production. However, the difference in activities of two Au/TiO₂ samples was much larger than the difference in SPR absorption (Fig. 3). It has been reported that Au particles having a diameter of around 2 nm loaded on TiO₂ worked as reduction centers for H₂ evolution in addition to SPR photoabsorption.⁵ Post-calcination has been widely applied to control the particle size of Au loaded on various supports. When MS-Au/TiO₂ was calcined at 500 °C in air, smaller Au particles having an average diameter of 1.4 nm disappeared (Fig. 1f), suggesting that these particles were unified with larger Au particles. As shown in Fig. 4(b), the rate of H₂ formation decreased with elevating calcination temperature. These results suggest that smaller Au particles of MS-Au/TiO₂ uncalcined might have important roles such as charge separation and H₂ evolution other than photoabsorption, although we do not have any other evidence. The yields of H₂ reached 3.0 and 54.2 μmol after 20 h in the case of SS- and MS-Au/TiO₂, respectively, while the yields of acetone in the liquid phase were 3.98 and 54.9 μmol, respectively. Good agreement with yields of H₂ and acetone in both samples shows that stoichiometric dehydrogenation of 2-propanol occurred as expressed in eqn (1).

(CH₃)₂CHOH → (CH₃)₂CO + H₂ (1)

No gas was evolved in the dark between 20 and 26 h, indicating that no thermocatalytic H₂ formation occurred in both cases of SS- and MS-Au/TiO₂ under the present conditions. To evaluate the stability of Au/TiO₂ in H₂ production from 2-propanol, Au/TiO₂ was used again. Irradiation of visible light to the reaction mixtures again induced evolution of H₂ in both cases and the formations continued from 26 h to 46 h without deactivation. The total amount of H₂ (ca. 105 μmol) was ca. 42-times larger than the amount of Au (2.5 μmol, corresponding to 1.0 wt%) loaded on TiO₂, indicating that the H₂ production observed under the present irradiation conditions was not a quantitative reagent reaction between Au and 2-propanol but a (photo)catalytic reaction.

Fig. 4 (a) Time courses of evolution of H₂ from 2-propanol in aqueous suspensions of TiO₂ (triangles), SS-(1.00)Au/TiO₂ (open circles) and MS-(1.00)Au/TiO₂ (closed circles) under irradiation of visible light from a Xe lamp with a Y-48 filter and (b) effect of post-calcination temperature on the rate of H₂ evolution.

To extend the possibility of Au/TiO₂ for H₂ formation under visible light, various compounds soluble in water were used as substrates. Ethanol, glycerin and ammonia were chosen as a typical biomass, a by-product in biomass utilization (saponification and transesterification of fats and oils) and biomass waste that is mainly included in excretion, respectively. Table 1 summarizes rates of H₂ production from these compounds in aqueous suspensions of SS- and MS-Au/TiO₂ under visible light irradiation. Evolution of H₂ was observed in all compounds, indicating that biomass-related compounds are strong candidates as an H₂ source when Au/TiO₂ is used under visible light irradiation. Larger rates of H₂ formation were obtained in all of the Au/TiO₂ samples prepared using the MS photodeposition method for Au loading. In the reaction of ammonia, 10 μmol of H₂ was evolved (entry 5) along with 3.2 μmol of N₂. Since the amount of N₂ was in good agreement with that of H₂, ammonia was decomposed stoichiometrically to H₂ and N₂ as shown in eqn (2).

2NH₃ → N₂ + 3H₂ (2)

In the reaction of benzyl alcohol in an aqueous suspension of MS-Au/TiO₂, 4.6 μmol of H₂ was formed over a period of 25 h (entry 6). Since almost the same amount of benzaldehyde (4.2 μmol) was formed in the liquid phase, it can be concluded that dehydrogenation of benzyl alcohol to benzaldehyde occurred as shown in eqn (3) under the present conditions.

C₆H₅CH₂OH → C₆H₅CHO + H₂ (3)

Since benzaldehyde is used widely in various chemical industries, this reaction has two meanings, i.e., evolution of H₂ and selective production of benzaldehyde under irradiation of visible light.

Table 1 Photocatalytic H₂ generation from various substrates in aqueous suspensions of (1.00)Au/TiO₂ under irradiation of visible light

Entry	Substrate^a	Time/h	H2 production/μmol
			SS	MS
a 50 vol% alcohol solutions and 14 wt% ammonia solution. b External quantum efficiency was roughly estimated to be 0.05% based on incident photons from 470 to 720 nm.
1	2-Propanol	5	0.73	13^b
2	Methanol	5	0.84	17
3	Ethanol	5	0.59	12
4	Glycerin	10	0.63	14
5	Ammonia	10	0.73	10
6	Benzyl alcohol	25	0.34	4.6

---

In conclusion, we succeeded in the formation of H₂ from various compounds including biomass-related compounds (2-propanol, methanol, ethanol, ammonia and benzyl alcohol) in aqueous suspensions of Au/TiO₂ under visible light irradiation. By using the MS photodeposition method, Au/TiO₂ samples exhibiting stronger photoabsorption at around 550 nm due to SPR and having higher levels of activity for H₂ production were prepared.

Notes and references

1. (a) I. Tsuji, H. Kato and A. Kudo, Angew. Chem., Int. Ed., 2005, 44, 3565; (b) K. Maeda, K. Teramura, D. Lu, T. Takata, N. Saito, Y. Inoue and K. Domen, Nature, 2006, 440, 295; (c) K. Maeda, A. Xiong, T. Yoshinaga, T. Ikeda, N. Sakamoto, T. Hisatomi, M. Takashima, D. Lu, M. Kanehara, T. Setoyama, T. Teranishi and K. Domen, Angew. Chem., Int. Ed., 2010, 49, 4096; (d) X. Chen, S. Shen, L. Guo and S. S. Mao, Chem. Rev., 2010, 110, 6503.
2. (a) Y. Tian and T. Tatsuma, J. Am. Chem. Soc., 2005, 127, 7632; (b) A. Furube, L. Du, K. Hara, R. Katoh and M. Tachiya, J. Am. Chem. Soc., 2007, 129, 14852; (c) S. Naya, M. Teranishi, T. Isobe and H. Tada, Chem. Commun., 2010, 46, 815; (d) E. Kowalska, R. Abe and B. Ohtani, Chem. Commun., 2009, 241; (e) E. Kowalska, O. O. P. Mahaney, R. Abe and B. Ohtani, Phys. Chem. Chem. Phys., 2010, 12, 2344.
3. (a) H. Kominami, A. Tanaka and K. Hashimoto, Chem. Commun., 2010, 46, 1287; (b) H. Kominami, A. Tanaka and K. Hashimoto, Appl. Catal., A, 2011, 397, 121; (c) A. Tanaka, K. Hashimoto and H. Kominami, Chem. Commun., 2011, 47, 10446.
4. A. Tanaka, K. Hashimoto and H. Kominami, ChemCatChem, 2011, 3, 1619.
5. C. G. Silva, R. Juarez, T. Marino, R. Molinari and H. Garcia, J. Am. Chem. Soc., 2011, 133, 595.
6. (a) H. Kominami, M. Kohno, Y. Takeda, M. Inoue, T. Inui and Y. Kera, Ind. Eng. Chem. Res., 1999, 38, 3925; (b) H. Kominami, S.-y. Murakami, J.-i. Kato, Y. Kera and B. Ohtani, J. Phys. Chem. B, 2002, 106, 10501.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental section and Fig. S1. See DOI: 10.1039/c2cy20108a

---