1
080
Chemistry Letters Vol.37, No.10 (2008)
Dialkylzinc-accelerated ꢀ-Trifluoromethylation of Carbonyl Compounds Catalyzed
by Late-transition-metal Complexes
ꢀ
Yuichi Tomita, Yoshimitsu Itoh, and Koichi Mikami
Department of Applied Chemistry, Tokyo Institute of Technology, 2-12-1 Ookayama, Tokyo 152-8552
(Received July 11, 2008; CL-080686; E-mail: mikami.k.ab@m.titech.ac.jp)
Trifluoromethylation of ketone silyl enol ethers is found
prepared from EtZnCl (X ¼ Et) gave the ꢀ-trifluoromethyl
ketone 2a in 25% yield. When employing Et2Zn without using
n-BuLi, 2a was obtained in 49% yield.
Several group VIII transition metal complexes including Ir,
Rh, Pd, Ni, and Ru were thus screened (Table 1). Without group
VIII transition metal, ꢀ-trifluromethyl ketone 2a was obtained
only in 12% yield (Entry 1). In the presence of late-transition-
metal complexes, 2a was obtained in up to 69% yield. Especial-
ly, the Wilkinson catalyst gave the highest yield (69% yield)
(Entry 5).
Zinc reagents were also examined (Table 2). Lewis acidic
zinc dihalide such as Br2Zn, F2Zn did not give any product
(Entries 6–9). Only dialkylzinc reagents gave 2a in moderate
to good yields (Entries 2–4). Especially, Et2Zn provides the best
yield (69% yield) (Entry 3).
to be significantly accelerated by late-transition-metal catalysts
and dialkylzincs to give ꢀ-trifluoromethyl ketones in good
yields. Addition of dialkylzinc is the key to the high yielding
ꢀ
-trifluoromethylation of carbonyl compounds.
Trifluoromethylated compounds have attracted much atten-
1
tion in medicinal and agrochemical science. However, ꢀ-tri-
fluoromethylation of non-fluorinated carbonyl compounds has
remained difficult because of the specific physical properties
of the CF3 groups. The problem in the direct introduction of
the CF3 group to the ꢀ-position of ketones associates with the
2
low reactivity of ketone silyl enol ethers and unstable ꢀ-CF3
ketonic products under basic conditions. Another problem is
that the electronegativity of trifluoromethyl group is higher than
3
Finally, ketonic substrates were screened to show eventually
the wide scope of this catalytic ꢀ-trifluoromethylation (Table
ꢁꢁ
ꢁþ
that of iodine (CF3 –I ), which is the reason we have focused
4
15
on radical methodology for ꢀ-trifluoromethylation. However,
the radical methodology still has problems such as difficulty in
3). Five- to seven-membered cyclic ketones gave the ꢀ-CF3
products 2a–2f in good yields (Entries 1–6). Starting from
4
d
16
application to catalytic asymmetric reaction.
Currently, the catalytic asymmetric ꢀ-arylation of carbonyl
compounds catalyzed by late-transition-metal complexes has
ꢀ-methylcyclohexanone, the ꢀ-trifluoromethyl product 2f
bearing all carbon quaternary center could be obtained (Entry
1
7
1
8
6). Acyclic ketones also gave the products 2g and 2h in good
yields (Entries 7 and 8).
5
attracted much attention. Furthermore, the ꢀ-arylation of
carbonyl compounds proceeds under neutral conditions using
In conclusion, we have uncovered ꢀ-trifluoromethylation
of silyl enol ethers catalyzed by late-transition-metal complexes
6
,7
Reformatsky reagents or silyl enol ethers in the presence of
zinc salts6 in good yield. Therefore, we decided to develop
an efficient synthetic method for transition-metal-catalyzed ꢀ-
trifluoromethylation of silyl enol ethers or metal enolates under
,8
Table 1. Group VIII transition metal catalyses
Et2Zn (1.5 equiv)
OTMS
CF3I (ca. 10 equiv)
Metal Complex (Metal 2 mol%)
O
9
neutral conditions. This catalytic process is intriguing in terms
CF3
of ‘‘Umpolung’’10 of CF3 –I to give the formal ‘‘CF3 ’’ for
ꢁꢁ
ꢁþ
þ
THF, -78 °C
0 °C, 30 min
1a
2a
enolate alkylation by late transition metal complexes. A quite
recent report11 prompted us to report herein our own results.9
We first focused on an iridium-catalyzed ꢀ-trifluoromethyl-
ation of ketone enolates with CF3I (Scheme 1). Because the
Entry
1
Metal Complex
Yield/%a
12
—
1
2,13
2
3
4
Ir
[IrCl(cod)]2
IrCl(CO)(PPh3)2
IrI(CO)(dppe)
49
24
10
oxidative addition of CF3I
temperature has already been reported.
carried out in the presence of CF3I and [IrCl(cod)]2 (1 mol %) us-
with an iridium catalyst at low
1
3a,13b
The reaction was
1
4
ing zinc enolate of weak metal–fluoride interaction to prevent
defluorination of the ꢀ-CF3 product. The zinc enolate prepared
with Cl2Zn (X ¼ Cl) did not give any product. The zinc enolate
5
6
7
8
9
Rh
RhCl(PPh3)3
[RhCl(cod)]2
69
54
55
58
57
[Rh(OH)(cod)]2
[Rh(acac)(C2H4)2
[Rh(cod)2](SbF6)
1
) n-BuLi (1.0 equiv)
°C, 30 min
) XZnCl (1.0 equiv)
0
2
OZnX CF3I (ca. 10 equiv)
[IrCl(cod)]2 (1 mol%)
O
b
0
°C, 30 min
10
Pd
Pd(PPh3)4
Pd(dba)2
Pd(OAc)2
56
54
58
CF3
b
1
1
1
2
THF
-78 °C
0 °C, 2 h
X = Cl 0%
OTMS
2a
b
Zinc Enolate
=
Et 25%
Et2Zn (1.5 equiv)
CF3I (ca. 10 equiv)
IrCl(cod)]2 (1 mol%)
1
3
Ni
Ni(PPh3)4
60
O
[
1
a
CF3
1
15
4
Ru
RuCl2(PPh3)3
60
55
THF, -78 °C
0 °C, 30 min
2a
[RuCl2(benzene)]2
4
9%
a
Determined by 19F NMR using BTF as an internal standard.
Reaction time was 2 h.
b
Scheme 1. ꢀ-Trifluoromethylation of ketone using Ir complex.
Copyright ꢀ 2008 The Chemical Society of Japan