Angewandte
Chemie
[
a]
Table 1: Selected optimization studies.
Entry 2a
F source
Promoter
(equiv)
PTC
(mol%)
Yield ee
[
b]
[c]
[d]
[
equiv] (equiv)
[%]
[%]
1
2
3
4
5
6
7
8
9
1
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
1.5
1.1
HF (2.0)
HF (2.0)
HF (2.0)
KF (2.0)
KF (2.0)
–
A (10)
B (50)
41
46
49
32
52
65
86
88
91
83
84
96
95
96
97
98
98
98
99
99
MsOH (0.5)
CCl CO H (0.5) B (50)
CCl CO H (3.5) B (50)
CF CO H (3.5) B (50)
3
2
3
2
3
2
KHF (1.0) CF CO H (2.0) B (50)
2
3
2
KHF (1.5) CF CO H (2.5) B (50)
2
3
2
KHF (1.5) CF CO H (2.5) B (10)
2
3
2
KHF (1.5) CF CO H (2.5) B (10)
2
3
2
0
KHF (1.5) CF CO H (2.5) B (10)
2 3 2
[
(
[
a] Reaction conditions: 1a (83.0 mmol, 1.0 equiv), [{Ir(cod)Cl}2]
4.0 mol%), (S)-L (16.0 mol%), dioxane (0.17 mL), 258C for 24 h.
b] Phase-transfer catalyst (PTC): A: nBu NHSO ; B: nBu NBr. [c] Deter-
4
4
4
1
mined by H NMR integration relative to the internal standard (1,3,5-
trimethoxybenzene). Ratio of branched-to-linear products determined by
1
H NMR integration and was always >50:1. [d] Determined by SFC on
Scheme 2. Scope of the direct, enantioselective, allylic alkynylation of
a chiral stationary phase. Absolute configuration determined by
correlation.
aromatic allylic alcohols. All reactions were carried out on a 0.25 mmol
scale under the standard conditions (see Table 1, entry 9). Yield is of
the isolated product after purification by chromatography. Regioselec-
1
(
entries 8–10). Thus, under optimal conditions including the
tivity (reported in brackets) was determined by the H NMR spectros-
copy of the reaction mixture. ee values were determined by super-
critical fluid chromatography (SFC) on a chiral stationary phase. [a]
use of 10 mol% of nBu NBr and 1.5 equivalents of 2a,
4
alkynylated product was furnished in 91% yield and 99% ee
in 6 h (entry 9).
1
.1 equiv of 2a, 1.1 equiv of KHF and 1.6 equiv of CF CO H were
2
3
2
used.
We next conducted a series of experiments to investigate
the scope of allylic alcohols with potassium phenylethynyltri-
fluoroborate (2a), as summarized in Scheme 2. A wide range
of alkoxy-substituted (products 3b–3d) and halogenated
Finally, we tested the catalytic system with a series of aliphatic
alkynes. A number of cycloalkyl- (products 3t and 3u) and
alkyl-substituted alkynyltrifluoroborates (products 3v–3z)
proved to be good substrates for the substitution reaction.
Importantly, various functional groups, such as a terminal
alkyne (product 3x) or chloride (product 3y), are chemically
inert in this Ir-catalyzed process, thus providing opportunities
for further elaboration of the products.
(
products 3e and 3 f) aromatic substrates afforded the
corresponding alkynes in good yields with high regio- and
stereoselectivity. Although substrates bearing electron-with-
drawing substituents (products 3g–3i) were less reactive than
the former, the observed selectivity was nonetheless excel-
lent. In the case of highly electron-rich aromatics (products
3
b, 3d, 3k, and 3l), high efficiency and enantiocontrol was
To showcase the synthetic utility of the process, we
applied the alkynylation chemistry to the synthesis of
AMG 837 (6), a potent GPR40 receptor agonist developed
achieved even when using reduced amounts of 2a, KHF , and
2
trifluoroacetic acid. It is noteworthy that allylic alcohol 1g,
containing a carbaldehyde, reacted exclusively at the allylic
position to afford the corresponding aldehyde 3g. In addition,
other aromatics and heteroaromatics, such as naphthyl,
thiophene, and indole (products 3j–3l) proved to be good
substrates for the reaction.
Subsequently, a range of potassium alkynyltrifluorobo-
rates were evaluated to examine the generality of the method
with respect to this substrate (Scheme 3). In this regard, allylic
phenylalkynylation proceeded smoothly with methoxy-,
alkyl-, and halogen-substituted benezenetrifluoroborates
[14]
by Amgen for the treatment of type 2 diabetes (Scheme 4).
Along these lines, allylic alcohol 4 was subjected to the
substitution reaction under the optimized conditions using
potassium 1-propynyltrifluoroborate (2w) and provided
enyne 5 in 72% yield and excellent selectivity (40:1
branched/linear, > 99% ee). Next, we tested the practicability
and robustness of the described process by conducting the
reaction on a larger scale (1.54 g, 4.0 mmol of 4), applying
lower Ir/(S)-L catalyst loadings (3/12 mol%), in technical
grade solvent, and open to air. Gratifyingly, only a slight
decrease in yield and selectivity was observed (67%, 98% ee).
Subsequent chemoselective hydroboration of enyne 5, fol-
lowed by Jones oxidation of the corresponding alcohol,
(
products 3m–3p). Furthermore, other aromatic systems
can be successfully employed, as exemplified by 3-thienyl-
substituted alkyne (product 3q). In addition to aromatic
alkynes, conjugated enynes (products 3r and 3s) were also
tolerated, and the corresponding alkynes were obtained in
good yields with excellent regio- and enantioselectivity.
[15]
furnished 6 in 74% yield (> 99% ee).
In summary, we have developed an Ir-catalyzed alkynyl-
ation that allows the direct and highly enantioselective
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3
These are not the final page numbers!