to the corresponding anthracenes together with small amounts
of the ketones (entries 5-7). Not only benzylic C-H bonds
but also allylic C-H bonds were also efficiently dehydro-
genated to afford the corresponding aromatized products
(entries 9 and 10). Heterocyclic amines, 1,2,3,4-tetrahydro-
quinoline, and indoline gave the corresponding dehydroge-
nated products (entries 11 and 12). Remarkably, the N-pro-
tected indoline of N-benzylindoline was dehydrogenated to
N-benzylindole with protection of benzyl group (entry 14).
The rates of alcohol oxidations were much faster than those
of the oxidation of alkylarenes to the ketones. For example,
the oxidation rate of fluoren-9-ol to fluoren-9-one at 403 K
was 8.23 mM min-1 and was 20 times faster than that of
fluorene to fluoren-9-one. When the reaction of 9-hydroxy-
9,10-dihydroanthracene was carried out under reaction
conditions C in Table 2, 9-hydroxy-9,10-dihydroanthracene
was quantitatively converted within 10 min, and anthracene,
anthrone, and anthraquinone were obtained in 91, 7, and 2%
yields, respectively.13 Al2O3 could also efficiently catalyze
the dehydration of 9-hydroxy-9,10-dihydroanthracene to give
99% yield of anthracene as the sole product under the same
reaction conditions,13 suggesting that the dehydration may
mainly be promoted by the Al2O3 support. Further, it was
confirmed that Ru(OH)x/Al2O3 did not catalyze the oxidation
of anthracene to anthrone and anthraquinone. These facts
suggest that alcohols are formed as intermediate products
that are rapidly dehydrogenated or dehydrated to the corre-
sponding oxygenated or dehydrogenated products, respec-
tively, in the present alkylarene oxidation.
Table 2. Oxidation of Various Alkylarene Derivatives with
Molecular Oxygen Catalyzed by Ru(OH)x/Al2O3
a
The reaction rates and selectivities for the oxidation of
xanthene and 9,10-dihydroanthracene were not affected by
the addition of radical scavenger 2,6-di-tert-butyl-4-meth-
ylphenol (2 mol % with respect to the substrate). In addition,
the skeletal isomerization of cyclopropyl and cyclobutyl rings
did not take place under the present conditions.14 These facts
show that free radical intermediates are not involved in the
present oxidation. The competitive oxidation of alkylarenes
of xanthene, 9,10-dihydroanthracene, fluorene, triphenyl-
methane, and diphenylmethane was carried out. When the
relative rates were plotted against the heterolytic C-H bond
energy, the homolytic C-H bond energy, ionization poten-
tial, and pKa (Figure 1 and Figure S2, Supporting Informa-
tion), a fairly good correlation was observed between relative
rates versus the heterolytic C-H bond energy, suggesting
that the formation of carbocation-type transition state via the
hydride abstraction is involved in the present oxidation.
Here, we propose the possible mechanism for the Ru(OH)x/
Al2O3-catalyzed oxidation of alkylarenes in Scheme 1. This
catalytic oxidation can be divided into three steps: Initially,
the reaction of ruthenium hydroxide species I with an
alkylarene proceeds to form ruthenium hydride species II
and the intermediate alcohol (step 1). The hydride species
then reacts with molecular oxygen (step 2). The reoxidation
of the hydride species likely proceeds through the insertion
a Reaction conditions A: substrate (1 mmol), Ru(OH)x/Al2O3 (2 mol
%), PhCF3 (6 mL), 373 K, under 1 atm of molecular oxygen. Reaction
conditions B: substrate (1 mmol), Ru(OH)x/Al2O3 (5 mol %), o-dichlo-
robenzene (6 mL), 443 K, under 1 atm of molecular oxygen. Reaction
conditions C: Substrate (1 mmol), Ru(OH)x/Al2O3 (2 mol %), p-xylene (6
mL), 403 K, under 1 atm of molecular oxygen. Conversion and selectivity
were determined by GC with an internal standard. Carbon balance for each
reaction was greater than 95%. Values in parentheses are isoleted yields.
b These experiments used a recycled catalyst. The reaction conditions were
the same as those for the first runs with a fresh catalyst. The initial rates
for the recycled runs were the same as that for the first run with a fresh
catalyst. c 5H-Dibenzo[a,d]cyclohepten-5-one was formed (12% selectivity).
d Anthraquinone (5% selectivity) and anthrone (2% selectivity) were formed.
e 10-Methylanthrone was formed (8% selectivity). f 10-Phenylanthrone was
formed (18% selectivity).
oxygenation at the benzylic position proceeded selectively
to give the corresponding diaryl ketones in excellent yields
(entries 1 and 3). Larger scale oxidation of xanthene (20
mmol scale, Ru: 0.1 mol %) showed a turnover frequency
(TOF) of 278 h-1, a turnover number (TON) of 921, and a
yield of 92%. These values are the highest among those
reported for oxygenation of alkylarenes using 1 atm of
molecular oxygen so far: H5PV2Mo10O40 (TOF 3.2 h-1, TON
58 for xanthene),9a N-hydroxyphthalimide (TOF 0.5 h-1,
TON 10 for xanthene),8 Bu4NHSO4/NaOH (TOF 8.5 h-1,
TON 5 for xanthene),9b Ru-Co-Al-CO3 hydrotalcite (TOF
4.5 h-1, TON 9 for xanthene),9c Co Shiff base complex/2-
methylpropanal (TOF 0.5 h-1, TON 10 for fluorene).7
9,10-Dihydroanthracene derivatives were smoothly converted
(13) Under anaerobic conditions, the dehydration of 9-hydroxy-9,10-
dihydroanthracene to anthracene proceeded with the same rate as that in
the presence of molecular oxygen.
(14) (a) Yamaguchi, K.; Mizuno, N. Angew. Chem., Int. Ed. 2002, 41,
4538. (b) Yamaguchi, K.; Mizuno, N. Angew. Chem., Int. Ed. 2003, 42,
1480. (c) Yamaguchi, K.; Mizuno, N. Chem. Eur. J. 2003, 41, 4353.
Org. Lett., Vol. 6, No. 20, 2004
3579