cles with average crystallite size of 7–8 nm (Figure S2 in the
Supporting Information). Owing to the nanocrystalline
nature of the support, the BET (Brunner–Emmett–Teller)
surface area of the as synthesized b-Ga O is very high
Table 1. Oxidative esterification of benzyl alcohol in the presence of
methanol over various gold catalysts.
[
a]
Entry
Catalysts
Conv.
%]
Selectivity [%]
Aldehyde
[
Ester
Acetal
2
3
2
À1
1
2
Au/b-Ga
Au/b-Ga
2
O
O
O
3
3
3
90
71
77.4
97
70
36.8
5.5
88
93.3
22.6
81.6
85
64.5
45
4.1
32.1
14.9
9.2
20.7
27.1
31.2
3.7
1.2
43.4
2.5
(
84 m g ). Surface acidity measurements by temperature-
2
-c
programmed desorption of NH (NH -TPD) and pyridine
3
3
3Au/
g-Ga
2
adsorption coupled with FTIR measurements (for details in
the Supporting Information, Table S1 and Figure S3) reveal
4
5
6
7
8
Au/Ga
3
Al
3
O
9
4.4
Au/TiO2
13.1
26.3
54.2
2.1
an abundance of surface Lewis acidity of medium strength
Au/Fe
Au/C
2
O
3
À1
12
93.9
with a site density of ca. 0.60 mmol NH (gcat) in the as
3
[
b]
Au/b-Ga
2
O
3
synthesized b-Ga O material, which is much higher than
2
3
[
(
a] Reaction conditions: benzyl alcohol (3.5 mmol) and Au catalysts
0.29 mol%) in methanol (10 mL) at 908C, 5 atm O , time=2 h; conver-
that previously reported for conventional calcination-de-
À1
2
rived b-Ga O (~0.24 mmol NH (gcat )).
2
3
3
sion (conv.) and selectivity were determined by GC by using n-nonane as
the external standard. [b] Results for the fifth run.
When gold nanoparticles were deposited onto the nano-
crystalline b-Ga O (see Supporting Information for the de-
2
3
[18]
tailed preparation procedure),
the XPS spectrum of the
0
Au 4 f7/2 core level shows a contribution from Au at a bind-
ing energy of 83.1 eV (Figure S4 in the Supporting Informa-
tion). This value is significantly lower than the reported
value (84.0 eV) for a bulk gold metal, suggesting the pres-
ence of a strong metal–support interaction in the Au/b-
Ga O system. The sole presence of metallic gold was also
activity for the oxidative esterification of benzyl alcohol.
Note that Au/C showed only very little activity under pres-
ent additive-free conditions.
After the catalytic oxidative esterification of benzyl alco-
hol was completed under the conditions in Table 1, the reac-
tion mixture was filtered to remove the Au/b-Ga O . It was
2
3
2
3
confirmed by the IR band of adsorbed CO (Figure S5 in
the Supporting Information). An almost identical XRD pat-
tern was obtained for Au/b-Ga O relative to b-Ga O , indi-
confirmed that no gold was present in the filtrate by induc-
tively coupled plasma atomic emission spectroscopy (ICP-
AES; detection limit of 2.5 ppb). In addition, the oxidation
process was terminated by the removal of Au/b-Ga O from
2
3
2
3
cating that the crystal structure of the nanocrystalline sup-
2
3
port is maintained. TEM analysis of the Au/b-Ga O catalyst
the reaction solution (Figure S6 in the Supporting Informa-
tion). These results indicate that any contribution to the ob-
served catalysis from gold species that leached into the reac-
tion solution can be ruled out and that the observed cataly-
sis is truly heterogeneous. The Au/b-Ga O catalyst could be
2
3
reveals randomly dispersed particles, and energy-dispersive
X-ray (EDX) analyses confirmed that the particles corre-
sponded to gold with an average diameter of about 2–4 nm.
It should be noted that, upon the introduction of gold nano-
2
3
particles, similar NH -TPD results were obtained for the
reused after aerobic oxidation, and both the catalytic activi-
ty and selectivity toward the oxidative esterification reac-
tions were retained (entry 8 in Table 1).
3
Au/b-Ga O catalyst (Table S1), thus revealing that the sur-
2
3
face acidity of b-Ga O was largely retained after deposition
2
3
of gold nanoparticles.
The gold nanoparticles supported on nanocrystalline b-
Oxidative esterifications of other primary alcohols were
examined by using the Au/b-Ga O -catalyzed protocol, and
2
3
Ga O were initially tested as a catalyst for the selective oxi-
the results are depicted in Table 2. A family of aromatic and
aliphatic primary alcohols, including benzyl alcohol, cinnam-
yl alcohol, and 1-octanol, can be converted into their corre-
sponding methyl, ethyl, or propyl esters with essentially
complete conversion and high selectivities (Table 2, en-
tries 1, 4, 5, 9, and 10), although longer reaction times are
required for the aliphatic alcohols. The results in Table 2
also indicate that the benzyl alcohols substituted with an
2
3
dation of 1-phenylethanol with O as the oxidant in the ab-
2
sence of solvent (see Supporting Information for experimen-
tal details). The Au/b-Ga O catalyst is found to be highly
2
3
effective for alcohol oxidation, with very high turnover fre-
À1
quencies (~15,000 h ) achievable for the exclusive synthesis
of acetophenone. Subsequent experiments exploring the oxi-
dative esterification of benzyl alcohol in the presence of
methanol have shown that the Au/b-Ga O3 catalyst was
electron-donating group (ÀOCH ) tend to be oxidized more
2
3
highly active and selective for the facile oxidative transfor-
easily than those with electron-withdrawing substituents, the
reaction rate for which was also slower than unsubstituted
benzyl alcohol (entries 2 and 3). Furthermore, it was estab-
lished that oxidative esterification of heteroaromatic pri-
mary alcohols such as furfuryl alcohol under the Au/b-
Ga O -catalyzed conditions also proceeded efficiently, and
[19]
mation of benzyl alcohol into methyl benzoate. The bene-
fit of using nanocrystalline b-Ga O as a bifunctional sup-
2
3
port for gold catalysts is evident when comparing the cata-
lytic activity of Au supported on conventional b-Ga O (Au/
2
3
b-Ga O -c), g-Ga O (Au/g-Ga O ) and gold reference cata-
2
3
2
3
2
3
2
3
lysts (Au/TiO , Au/Fe O and Au/C) supplied by the World
Gold Council (Table 1). A clear advantage of the Au/b-
the ester products were produced in good yields (>75%,
Table 2 entry 6). In contrast, only moderate selectivities to
the corresponding methyl esters was obtained for allylic pri-
mary alcohols such as crotonyl alcohol and allyl alcohol
under the same condition as described above (Table 2, en-
tries 7 and 8). The lower selectivity obtained for the oxida-
2
2
3
Ga O catalyst over Au/Ga Al O was also noticed when the
2
3
3
3
9
reaction was conducted by using Au/Ga Al O under other-
3
3
9
wise identical conditions. Among various Au catalysts
tested, the Au/b-Ga O catalyst showed the highest catalytic
2
3
7132
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 7131 – 7135