Table 2. Ruthenium-catalyzed allylic amination of tertiary
allylic carbonates with aminesa
Table 3. Ruthenium-catalyzed allylic amination of α,α-diaryl
allylic acetates with aminesa
5 mol% [Cp*Ru(CH3CN)3][PF6]
5 mol% 5,5'-diMe-2,2'-bpy
5 mol% [Cp*Ru(CH3CN)3][PF6]
5 mol% 5,5'-diMe-2,2'-bpy
R
NR1R2
Ar
Ph
X
Ar NR1R2
Ph
R
OCO2Me
HNR1R2
HNR1R2
+
+
CH3CN, 80 ˚C, 19 h
toluene, 60 ˚C, 72 h
Ph
Ph
3m–v
X = OCO2Me: 1i
OAc: 1j–p
2a,e,g,h
2a–g
3a–l
1b–h
yieldb
(%)
entry
R
HNR1R2
product
yieldb
(%)
entry
Ar
HNR1R2
product
1
2
3
4
5
6
7
8
9
Et (1b)
Et (1b)
Et (1b)
Et (1b)
Et (1b)
Et (1b)
Et (1b)
nBu (1d)
iPr (1e)
morpholine (2a)
piperidine (2b)
pyrrolidine (2c)
HNBnMe (2d)
H2NBn (2e)
H2NiPr (2f)
H2NPh (2g)
morpholine (2a)
morpholine (2a)
3a
3b
3c
3d
3e
3f
3g
3h
3i
66
41
35
43
58
43
55
54
<2c
32
1c
2c
3
4
5
6
7
8
9
Ph (1i)
Ph (1j)
Ph (1j)
Ph (1j)
Ph (1j)
morpholine (2a)
morpholine (2a)
morpholine (2a)
HNEt2 (2h)
H2NBn (2e)
H2NPh (2g)
H2NBn (2e)
H2NBn (2e)
H2NBn (2e)
H2NBn (2e)
H2NBn (2e)
H2NBn (2e)
3m
3m
3m
3n
3o
3p
3q
3r
3s
3t
3u
3v
62
69
79
69
84
51
76
59
75
83
46
61
Ph (1j)
2-naphthyl (1k)
2-MeC6H4 (1l)
4-MeC6H4 (1m)
4-ClC6H4 (1n)
4-CF3C6H4 (1o)
4-PhC6H4 (1p)
10
11
12
MeOCH2 (1f) morpholine (2a)
Bn(1g)
CF3 (1h)
3j
3k
3l
10
11
12
morpholine (2a)
morpholine (2a)
38
<2c
aReaction conditions: 1b-h (0.3 mmol), 2a-g (0.6 mmol),
5 mol % of [Cp*Ru(CH3CN)3][PF6], and 5 mol % of 5,5¤-
diMe-2,2¤-bpy in toluene (1.0 mL) at 60 °C for 72 h. Isolated
aReaction conditions: 1i-p (0.3 mmol), 2a, 2e, 2g, and, 2h
(0.6 mmol), 5 mol % of [Cp*Ru(CH3CN)3][PF6], and 5 mol %
of 5,5¤-diMe-2,2¤-bpy in toluene (1.0 mL) at 80 °C for 19 h.
b
c
1
yield. Determined by H NMR of crude materials.
bIsolated yield. Toluene was used instead of CH3CN.
c
such as DMF, THF, DCE, toluene, and xylene (Entries 5-9); 3a
was obtained in 41% yield (88% conversion of 1a) when toluene
was used (Entry 8). We also observed 100% conversion of 1a
when the reaction time was increased to 72 h, but the yield of
3a increased only slightly (48% yield) (Entries 10 and 11). We
further examined the reaction by changing the leaving group of
the tertiary allylic compounds from acetate to methyl or tert-butyl
carbonate (1b and 1c, respectively) and obtained 3a in 66% yield
as a single regioisomer when using 1b (Entries 12 and 13). Then,
we established the optimal conditions as follows: reaction of 1b
with 2a using [Cp*Ru(CH3CN)3][PF6]/5,5¤-diMe-2,2¤-bpy in
toluene at 60 °C for 72 h. Furthermore, we confirmed that
1,3-diene was formed as a byproduct from 1b in 32% yield.
With the standard conditions (Table 1, Entry 12) in hand,
we next investigated the reaction of allylic carbonates 1b-h
with several amines 2a-g (Table 2). The reactions with cyclic
aliphatic amines 2b and 2c provided the desired amination
products 3b and 3c in moderate yields, while the reactions with
acyclic aliphatic secondary amines 2d gave the desired α,α-
disubstituted allylic amine in 43% yield (Entries 2-4). The
reactions of 1b with aliphatic and aromatic primary amines, such
as benzylamine (2e), isopropylamine (2f), and aniline (2g),
afforded the desired allylic amines 3e, 3f, and 3g in 58%, 43%,
and 55% yields, respectively (Entries 5-7). We next carried out
the reaction of other alkyl group-substituted allylic carbonates
1d-h with 2a (Entries 8-12). The reaction of n-butyl group
substituted allylic carbonate 1d also proceeded to furnish allylic
amine 3h (Entry 8). Unfortunately, 1e, which had an iso-propyl
group, did not give the desired aminated product 3i but produced
only a 1,3-diene (Entry 9).11 This was probably because 1e
underwent elimination more rapidly than amination due to the
increased steric hindrance as compared to that observed with the
ethyl group.12 Accordingly, the reactions of substrates including
a methylene carbon at the α-position with morpholine were also
attempted: 1f and 1g afforded the α,α-disubstituted allylic amines
in 32% and 38% yields, respectively (Entries 10 and 11). On the
other hand, allylic carbonate 1h with strong electron-withdraw-
ing groups not gave 3l (Entry 12).13
We next carried out the reaction of allylic carbonate 1i
bearing two aryl groups; since the reaction proceeded smoothly
in the presence of the [Cp*Ru(CH3CN)3][PF6]/5,5¤-diMe-2,2¤-
bpy catalyst, we changed the reaction time from 72 h to 19 h
(Table 3, Entry 1). Screening of the reaction conditions revealed
that the use of allylic acetate 1j, with acetonitrile instead of
toluene, led to an increase in the yield (Entries 2 and 3). Having
established the optimal reaction conditions, we next investigated
the scope of the allylic amination of 1j with other amines
(Entries 4-6). The reaction with aliphatic secondary or primary
amines such as 2h and 2e produced the desired products 3n and
3o in 69% and 84% yields, respectively (Entries 4 and 5). On the
other hand, aniline (2g) reacted with 1j to give the desired
product 3p in only moderate yield (Entry 6). Furthermore, the
reactions of other aromatic group-substituted carbonates with
benzylamine (2e) proceeded to afford the corresponding prod-
ucts in good yields (Entries 7-12). The reactions of 1k bearing a
2-naphthyl group and 1l bearing methyl groups at the ortho-
position of the aromatic rings proceeded to give the desired
aminated products in 76% and 59% yields, respectively (Entries
7 and 8). Both electron-donating (1m) and electron-withdrawing
(1n-p) substituents at the para-position of the phenyl rings were
well tolerated, and the corresponding amination products were
obtained in 46%-83% yields (Entries 9-12).
The details of the mechanism underlying the ruthenium-
catalyzed amination are still not clear. However, the active
catalyst is proposed to be a Cp*Ru(2,2¤-bipyridine)(CH3CN)14,15
complex on the basis of the ruthenium catalyst system
© 2020 The Chemical Society of Japan