Angewandte
Chemie
support (see the Supporting Information) scarcely exhibits
reaction enhancement. These findings suggest that homoge-
neously mixed Au–Cu alloy particles are necessary for this
effect.
The enhanced activity of the Au–Cu alloy catalyst is due
to visible light reducing the surface Cu atoms that have been
oxidized by O2. This maintains the Au–Cu alloying effect and
promotes aerobic oxidation without catalyst deactivation.
This is confirmed by the time-dependent change in the
amount of acetone formed during aerobic oxidation of
*
2-propanol with Au0.7C0.3/P25. As shown in Figure 4 ( ), the
Figure 5. ESR spectra of Au0.7Cu0.3/P25 samples (a–d) and a Au1/P25
sample (e) measured at 77 K. a) The Au0.7Cu0.3/P25 catalyst was
treated with O2 (20 Torr) for 3 h in the dark. b) Sample (a) was
irradiated by visible light for 3 h. c) 2-Propanol (10 Torr) was added to
sample (b) in the dark. d) Sample (c) was irradiated by visible light for
3 h. e) The Au1/P25 catalyst was treated with O2 (20 Torr) for 3 h and
irradiated by visible light for 3 h. The g=2.005 signal is due to the
electrons trapped by oxygen vacancy site of TiO2, and the g=1.982
signal is due to the Ti3+ site on TiO2.[22]
Figure 4. Time-dependent change in the amount of acetone formed by
aerobic oxidation of 2-propanol with the Au0.7Cu0.3/P25 catalyst under
*
*
visible-light irradiation ( ), in the dark ( ), and in the dark for 8 h
~
followed by visible-light irradiation ( ). Reaction conditions are
identical to those in Figure 3.
Figure S4, visible-light irradiation of Au0.3Cu0.7/P25 with
a lower amount of Au shows a much lower decrease in the
Cu2+ signal than that of Au0.7Cu0.3/P25. Moreover, Cu1/P25
exhibits almost no signal decrease. These tendencies are
consistent with the SPR intensity of the catalysts (Figure 2),
suggesting that the reduction of Cu2+ is promoted by the light
absorption of surface Au species in the alloy. This is further
confirmed by the action spectrum for aerobic oxidation of 2-
propanol with Au0.7Cu0.3/P25 obtained by irradiation of
monochromatic light. As shown in Figure 6, a good correla-
tion is observed between the SPR band of the catalyst and the
apparent quantum yield for acetone formation (FAQY). This
suggests that plasmon-activated surface Au atoms indeed
promote the reduction of oxidized surface Cu atoms.
reaction in the dark is effective in the early stage (ca. 2 h), but
the activity decreases with time. This is because the surface
Cu atoms are oxidized by O2 during the reaction, thus
eliminating the alloying effect.[16] In contrast, under photo-
*
irradiation ( ), the rate of acetone formation in the early
stage is similar to that obtained in the dark, but the activity is
maintained even after 18 h. This suggests that visible-light
irradiation suppresses catalyst deactivation. To further clarify
this effect, the reaction was carried out in the dark for 8 h and
~
continued under visible-light irradiation ( ). The catalytic
activity is successfully regenerated by photoirradiation.
Activity regeneration by visible-light irradiation is due to
the reduction of oxidized surface Cu atoms with alcohol as an
electron donor. ESR analysis confirms this. Figure 5a shows
the spectrum of Au0.7Cu0.3/P25 measured at 77 K after
exposure to O2 in the dark. Strong signals, which are assigned
to Cu2+, are observed at g? = 2.03–2.06,[18] indicating that
surface Cu atoms are oxidized by O2. As shown in Figure 5b,
visible-light irradiation of sample (a) does not show any
spectral change. As shown in Figure 5c, exposure of sample
(b) to 2-propanol (10 Torr) in the dark still does not show
spectral change. However, as shown in Figure 5d, visible-light
irradiation of sample (c) leads to complete disappearance of
the Cu2+ signals. GC analysis of the sample detects the
formation of acetone and water. This suggests that visible-
light irradiation of the alloy catalyst reduces oxidized surface
Cu atoms with alcohol as an electron donor.
As shown in Figure 7, plasmon-activated surface Au
atoms promote the intraband transition of 6 sp eꢀ.[28] The
Figure 6. Action spectrum for aerobic oxidation of 2-propanol on the
Au0.7Cu0.3/P25 catalyst. The apparent quantum yield for acetone
formation (FAQY) was calculated with the following equation: FAQY
(%)=[{(YvisꢀYdark)ꢀ2}/(photon number entered into the reaction
vessel)]ꢀ100, where Yvis and Ydark are the amounts of acetone formed
(mmol) under light irradiation and dark conditions, respectively.
The effectiveness of the reduction of oxidized surface Cu
atoms depends on the amount of Au in the alloy. As shown in
Angew. Chem. Int. Ed. 2013, 52, 5295 –5299
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5297