Angewandte
Chemie
À
or in the presence of complex with multiple CO ligands
(Table 1, entries 1–3) when using tert-butyl peroxide as an
oxidizing reagent. On the other hand, the use of a ruthenium
catalyst bearing acac, cod, phosphine (with mono CO), or
arene ligands all led to the desired mono- and bisalkylation
products 4a and 5a, respectively (Table 1, entries 4–8).
The use of other peroxides are less effective (Table 1,
entries 9–13). Decreasing the reaction temperature (Table 1,
entry 14) or the amount of the catalyst (Table 1, entry 15)
decreased the product yield. With 4 equivalents of peroxide,
80% of the desired product was obtained in a 1:1 ratio of the
mono- and biscycloalkylation products (Table 1, entry 16). As
a control, the reaction was run in the absence of cyclooctane
and no product was obtained (Table 1, entry 18). Interest-
ingly, a methylated product was obtained in good yield and
regioselectivity (Table 1, entry 18) when benzene was used as
the solvent.
chelation-directed C H activation to generate intermediate
B.[10] The reaction of B with cycloalkane 2 and peroxide 3[11]
generates intermediate C and an alcohol. Finally, reductive
elimination of intermediate C generates arene–cycloalkane
coupling product 4 and regenerates active ruthenium catalyst
A. Subsequent reaction of monocycloalkylation product 4
leads to biscycloalkylation product 5. As experimental
evidence to support the proposed mechanism, no [Ru]–alkyl
intermediate was detected in a stoichiometric reaction
between the ruthenium complex, cyclooctane, and the
peroxide. In addition, deuterium isotope experiments with
mono- and bisdeuterated 2-phenylpyridines 6 and 7 showed a
With the optimized conditions in hand, other arenes and
cycloalkanes were investigated (Table 2). The reaction occur-
red exclusively on the non-nitrogen atom-containing aromatic
ring. Various electron-withdrawing and electron-donating
substituents did not affect the reaction significantly (Table 2,
entries 2–8). meta-Substituted phenylpyridines almost exclu-
sively afforded the monocycloalkylation products (Table 2,
entries 9–11). Other N-heteroaromatic compounds also cou-
pled with cycloalkanes regioselectively (Table 2, entries 12
and 13). Cyclohexane and cycloheptane also reacted
smoothly with 2-phenylpyridine to give the desired
large negative kinetic isotope effect (in both cases only
nondeuterated products 4a and 5a, respectively, were
obtained, and no deuterated product was observed), which
suggests that there is a fast equilibrium between starting
material 1 and intermediate B, with the formation of
intermediate C being the rate-limiting step.[12]
À
In conclusion, a novel C C bond formation based on the
À
À
direct oxidative C H/C H coupling involving arenes and
cycloalkanes has been developed. The scope and detailed
mechanism of this reaction are under additional investigation.
À
À
C H/C H coupled products (Table 2, entries 14 and 15).
Surprisingly, when the seemingly more reactive cyclooctene
was used instead of cyclooctane under the standard con-
ditions, the desired alkylation product was obtained in less
than 30% yield in addition to unidentified products; and most
of the 2-phenylpyridine remained unreacted.
A tentative mechanism to rationalize the product forma-
tion is illustrated in Scheme 2. The active ruthenium catalyst
(A) reacts with 2-phenylpyridine (and other arenes) by a
Experimental Section
A representative experimental procedure (4a and 5a): An oven-dried
reaction vessel was charged with [{Ru(p-cymene)Cl2}2] (6.1 mg,
0.01 mmol), 2-phenylpyridine (1a, 31 mg, 0.2 mmol), tert-butyl per-
oxide (3a, 117 mg, 0.8 mmol), and cyclooctane (2a, 0.6 mL,
4.5 mmol). The reaction vessel was then sealed and the resulting
solution was stirred at 1358C for 16 h. After cooling to room
temperature, the resulting mixture was filtered through a short silica
gel plug in a pipette by using methylene chloride as the eluent. The
volatiles were removed in vacuo and the residue was purified by
column chromatography (SiO2, hexane/ethyl acetate 9:1) to give 4a
(21 mg, 39%) and 5a (27 mg, 36%) as pale yellow oils.
4a IR (liquid film): n˜ = 3050, 2918, 2852, 1580, 1461, 1120,
1
775 cmÀ1; H NMR (400 MHz) d = 8.70–8.68 (m, 1H), 7.86 (s, 1H),
7.75–7.69 (m, 3H), 7.37 (t, J = 7.6 Hz, 1H), 7.26–7.23 (m, 1H), 7.20–
7.16 (m, 1H), 2.90–2.85 (m, 1H), 1.93–1.77 (m, 6H), 1.69–1.59 ppm
(m, 8H); 13C (100 MHz) d = 158.0, 151.3, 149.8, 139.6, 136.9, 128.9,
127.8, 126.0, 124.4, 122.2, 120.9, 45.0, 35.1, 27.1, 26.6, 26.3 ppm;
HRMS calcd for C19H23N + H 266.1903, found 266.1904.
5a IR (liquid film): n˜ = 2912, 2839, 1580, 1435, 782 cmÀ1; 1H NMR
(400 MHz) d = 8.68–8.66 (m, 1H), 7.71–7.69 (m, 2H), 7.57 (d, J =
1.6 Hz, 2H), 7.20–7.16 (m, 1H), 7.05 (s, 1H), 2.84–2.79 (m, 2H), 1.91–
1.57 ppm (m, 28H); 13C (100 MHz) d = 158.5, 151.3, 149.8, 139.4,
136.8, 126.6, 123.2, 122.0, 121.0, 45.1, 35.2, 27.1, 26.6, 26.4 ppm;
HRMS calcd for C27H37N + H 376.2999, found 376.3003.
The experiments in Table 2 were carried out analogously. All
products were purified by column chromatography and characterized
by NMR spectroscopy and standard/high-resolution mass spectrom-
etry.
Scheme 2. Proposed mechanism for the ruthenium-catalyzed cycloalky-
Received: April 2, 2008
lation of arenes mediated withperoxides.
Published online: July 9, 2008
Angew. Chem. Int. Ed. 2008, 47, 6278 –6282
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim