Angewandte
Communications
Chemie
cant limitations in the reaction scope and further ligand-
modifications as well as inconvenient multi-step substrate
syntheses are necessary.
Inspired by these reports we designed the first regiose-
lective intermolecular Cp*CoIII-catalyzed carboamination of
alkenes. The unique properties of the catalyst system obviate
the need for either a bidentate directing group or a bidentate
coupling partner (Scheme 1b). In this conceptually and
mechanistically distinct reaction, the Cp*CoIII-catalyst led to
a different product than its Cp*RhIII-counterpart. For the
rhodium catalyst, if the saturation of the nascent C(sp3)–metal
center is not achieved, b-H-elimination predominates and the
oxidative Heck product is formed. In the case of cobalt, no
saturation of the C(sp3)–metal center is required and the
carboamination is inherently favored over the b-H-elimina-
tion.
We began our studies using phenoxyacetamide (1) as it
features a redox-active directing group and n-butylacrylate
(2) as coupling partner. After an extensive screening, (see the
Supporting Information) we could achieve a completely
regioselective addition to the double bond and isolate the
desired product 3a in 82% yield using stoichiometric amounts
of 1 and 2 in the presence of catalytic amounts of [Cp*Co-
(CO)I2], AgSbF6, CsOAc, and K3PO4 (Scheme 2 and see the
Supporting information). When the same conditions were
applied with the [(Cp*RhCl2)2] catalyst instead of the
[Cp*Co(CO)I2] catalyst, the oxidative Heck product was
isolated in 53% yield and only trace amounts of the
carboamination product could be detected in the crude
reaction mixture by ESI-MS (see the Supporting Informa-
tion).
With the optimized reaction conditions in hand, we
investigated the scope of the carboamination reaction
(Scheme 2). Para-substituted phenoxyacetamides with elec-
tron-withdrawing functional groups, such as halogens (3b–
3d), esters (3e), or even the CF3 substituent (3 f), delivered
the desired amino acid derivatives in good to excellent yields.
Similar results could be obtained for para-substituted phe-
noxyacetamides bearing electron-donating functional groups
(3g, 3h). Using the disubstituted 3,4-dimethylphenoxyaceta-
mide, the corresponding product could be isolated in good
yield (60%) and excellent regioselectivity (3l). Furthermore,
2-N-(naphthalen-2-yloxy)acetamide could deliver the desired
amino acid derivative in a moderate yield of 47% (3i). Meta-
substituted electron-withdrawing (m-Br) as well as electron-
donating (m-Me) phenoxyacetamides delivered the corre-
sponding products in good yields and a single regioisomer was
obtained (3j, 3k). The absolute molecule structures for 3j and
3c were confirmed by X-ray crystallography (see Scheme 2
and the Supporting Information).[12] For the meta-methoxy-
and the meta-fluorophenoxyacetamides excellent yields could
be obtained, although a mixture of regioisomers was observed
(3n, 3o). The ortho-methylphenoxyacetamide delivered the
corresponding product in 30% yield (3m). The poor reac-
tivity is attributed to the great steric demand of the Cp*CoIII
catalyst, which results in sensitivity to steric bulk on the
substrates. No product formation was observed in the absence
of the Cp*CoIII catalyst. We further investigated the scope of
the N-protecting group. However, benzyloxycarbonyl- (Cbz),
Scheme 2. Variation of the phenoxyacetamide and acrylate. Isolated
yields are given. 1 (0.2 mmol), 2 (1.0 equiv), [Cp*Co(CO)I2]
(10 mol%), AgSbF6 (20 mol%), CsOAc (25 mol%), K3PO4 (25 mol%),
3 ꢂ MS (40 mg) in TFE (1 mL) at 408C for 22 h; [a] 0.4 mmol of 2 was
used. For crystallographic data see: Ref. [12].
benzoyl- (Bz), tosyl- (Ts), and pivaloyl- (Piv) protected
substrates were unsuccessful under the optimized reaction
conditions (see Supporting Information).
Several acrylates also led to the formation of the
corresponding amino acid derivatives. tert-Butylacrylate and
ethylacrylate delivered the desired products in reasonable
yields (3p, 3q). Moreover, using benzylacrylate the carboa-
minated product (3r) was formed in a synthetically useful
yield of 51%.
Further experiments were carried out to elucidate a pos-
sible reaction mechanism (Scheme 3). First, the kinetic
isotope effect was studied in competition and parallel experi-
ments (Scheme 3a). A KIE of 2.1 (competition experiment)
and kH/kD = 1.5 (parallel experiment) were observed. Based
À
on these results, we propose that the C H activation step is
2
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
These are not the final page numbers!