.
Angewandte
Communications
unknown. Inspired by the previous report, we envisaged that
the utilization of 2-hydroxypyridine might allow entry to the
intermolecular asymmetric allylic dearomatization because
there exists a tautomeric equilibrium between 2-hydroxypyr-
a significant increase in yield and enantioselectivity (entry 2)
[
14]
when the Alexakis ligand (S,S,S )-1b was used.
The
a
reaction with 1c also led to a satisfactory result (entry 3).
However, the use of (R,R )-1d or (R )-1e afforded the desired
a
a
[
10]
idine and 2-pyridone.
This methodology will provide
product (4aa) in moderate enantioselectivity but with dis-
appointing yield and regioselectivity (entries 4 and 5). Next,
we turned our attention to examining the effect of the base.
Notably, the reaction did not occur at all in the absence of an
additional base (entry 6, Table 1). Meanwhile, the use of
a direct approach to the asymmetric synthesis of N-substi-
tuted 2-pyridones bearing a stereocenter a to N (Scheme 1c).
Herein, we report our results from this study.
The initial experiment was performed with 2-hydroxypyr-
idine (2a) and cinnamyl methyl carbonate (3a) in the
a weak base such as Et N could only give trace amounts of
3
[
11,12]
presence of a well-developed iridium catalytic system
product (entry 7). In contrast, a stronger organic base such as
including [{Ir(cod)Cl} ] (2 mol%) and the Feringa phosphor-
DBU and various inorganic bases such as tBuOLi, K PO ,
2
3
4
amidite (S,S,S )-1a (4 mol%) in tetrahydrofuran at 508C. By
K CO , and NaH could all be tolerated, thus affording the
2 3
a
using Cs CO as the base, the desired N-alkylation process
desired products in good yields and excellent enantioselec-
tivity (entries 8–12). Among them, Cs CO was found to be
the optimal base (entry 2). Then, varying the amount of
2
3
proceeded in 96% conversion to give the branched product
aa in 84% yield and 94% ee without observation of the O-
2
3
4
alkylation product (entry 1, Table 1). It is likely that the
corresponding h -allyl/Ir intermediate is a soft electrophile,
Cs CO3 indicated that increasing the loading of base to
1 equivalent led to a slightly lower yield. However, 2 equiv-
2
3
which is more easily attacked by the soft 2-pyridone nitrogen
nucleophile rather than the hard oxygen anion of 2-hydrox-
ypyridine. The feasibility of N attack of 2a encouraged us
to investigate other chiral phosphoramidite ligands. To our
delight, the reaction proceeded much more quickly with
alents of Cs CO3 had a detrimental effect on both the
2
conversion and regioselectivity (entries 13 and 14). Above all,
the best conditions were obtained as the following: reaction of
3a with 2 equivalents of 2a in THF with 2 mol% of
[{Ir(cod)Cl} ], 4 mol% of (S,S,S )-1b, and 40 mol% of
[13]
2
a
15]
[
Cs CO at 508C (entry 2, Table 1).
2
3
Under the optimized reaction conditions, the substrate
[
a]
Table 1: Investigation of the reaction conditions.
scope of this intermolecular allylic amination with 2-hydrox-
ypyridines was examined. Firstly, various allylic carbonates
were tested using 2a as the nucleophile. As summarized in
Scheme 2, for the aryl allylic carbonates, substituents on
different positions of the phenyl moiety (4-MeOC H , 4-
6
4
MeC H , 4-FC H , 4-ClC H , 4-BrC H , 3-MeOC H , 3-
6
4
6
4
6
4
6
4
6
4
MeC H , 3-ClC H ) were well tolerated, and their corre-
6
4
6
4
sponding products 4ab–ai were obtained in 80–98% yields
with 98–99% ee. Notably, the generally unfavorable ortho-
substituted substrate was also compatible with this trans-
formation, and 4aj was obtained in 77% yield and 94% ee. To
our delight, heteroaryl allylic carbonates were also suitable
substrates, thus delivering the amination products 4ak and
4al with excellent regioselectivity and ee values, and good
yields upon isolation. In addition to aromatic substituents,
aliphatic allylic carbonates (Me, nPr) were also tested. The
reaction of methyl crotyl carbonate occurred smoothly, thus
affording satisfactory results (4am/5am), but the reaction of
n-propyl-substituted allylic carbonate gave rise to moderate
chemoselectivity and regioselectivity, with excellent enantio-
selectivity (N/O = 86:14; 4an/5an). No notable improvement
on the chemo-, regio-, and enantioselectivity was observed
[
b]
Entry Ligand Base
t [h] Conv. 4aa/5aa
4aa
[
b]
[c]
[d]
[%]
Yield [%] ee [%]
1
2
3
4
5
6
7
8
9
0
1
2
1a
1b
1c
1d
1e
1b
1b
1b
1b
1b
1b
1b
1b
1b
Cs CO3 19
96
95:5
95:5
96:4
48:52
18:82
–
84
94
96
26
6
94
98
93
77
60
–
96
98
99
99
98
99
99
98
2
Cs CO3 1.5 >95
2
Cs CO3 6.5 >95
2
Cs CO3 19
82
26
<5
16
2
Cs CO3 19
2
–
22
22
1.5
–
Et N
–
14
83
79
79
85
87
84
61
3
DBU
tBuOLi 22
K3PO4 1.5
92 >97:3
85
85
94:6
94:6
95:5
94:6
94:6
89:11
[
12g,16]
using [{Ir(dbcot)Cl}2]
(N/O = 82:18; 4an/5an) instead of
1
1
1
1
1
K2CO3 1.5 >95
NaH 1.5 >95
Cs CO3 1.5 >95
[{Ir(cod)Cl} ]. Notably, the bulky t-butyl-substituted allylic
2
[
17]
carbonate was found to be unreactive. Next, the generality
of the reaction with respect to the 2-hydroxypyridine
derivatives using the cinnamyl carbonate 3a as the electro-
philic counterpart was explored. We were pleased to discover
that the reactions of substrates with a range of substituents on
the pyridine moiety, including those bearing either an
electron-donating (5-Me, 4-Me, 3-OH) or electron-withdraw-
ing group (5-Cl, 5-Br, 3-Cl, 3-Br), were highly efficient with
excellent regioselectivity (4/5: 87:13–95:5) and enantioselec-
[
e]
f]
3
2
[
4
Cs CO3 22
2
89
[
a] Reaction conditions: 2 mol% of [{Ir(cod)Cl} ], 4 mol% of 1, 40 mol%
of base, 200 mol% of 2a and 100 mol% of 3a in 2.0 mL THF at 508C.
Unless noted, N/O> 95:5. [b] Determined by H NMR analysis of the
crude reaction mixture. [c] Yield of isolated 4aa. [d] The ee value of 4aa
was determined by HPLC analysis. [e] 100 mol% of Cs CO was used as
the base. [f] 200 mol% of Cs CO was used as the base. cod=1,5-
cyclooctadiene, DBU=1,8-diazabicyclo[5.4.0]undec-7-ene, THF=tetra-
2
1
2
3
2
3
hydrofuran.
tivity (4ba–4ha). The reaction of 5-NO pyridone (2i) and 3a
2
2
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 5
These are not the final page numbers!