Communication
double 1,3-dipolar cycloaddition of azomethine ylides (1) with
cyclopentadienone (2) (Scheme 1).[9–11] Azomethine ylides (1)
are readily available from commercial amino acid esters and al-
dehydes. However, cyclopentadienone (2) is only accessible by
means of multistep procedures[12] and the low stability of dien-
one 2 severely complicates its use.[13] Intrigued by a recent
finding of Kumar et al.[14] for bioinspired oxygenation of acti-
vated CÀH bonds, we considered that an efficient in situ syn-
thesis of 2 could be achieved by means of CÀH bond oxidation
of the cyclopentadiene methylene group. The CÀH bonds in
the methylene group of 3 have relatively low bond dissocia-
tion energy such that the desired keto group could be selec-
tively installed by oxidation of 3. It would be desirable to
employ oxygen as an environmentally benign, highly atom-
economical oxidant, and for an aerobic oxidation, the high ac-
tivation energy of oxygen a catalyst is required.[15] Copper-
based catalysts were successfully used in the aerobic oxidation
of CÀH bonds with molecular oxygen, and notably, copper
salts may also serve as catalysts of the enantioselective 1,3-di-
polar cycloaddition of azomethine ylides (1) and various dipo-
larophiles. In light of these preconditions we envisaged the de-
velopment of a cascade reaction in which a chiral copper cata-
lyst is used for the aerobic oxidation of cyclopenta-
Scheme 1. Cascade enantioselective cycloaddition.
while application of P,N- or P,P-bidentate ligands was not suc-
cessful. With the exception of Cu(OAc)2 all tested copper salts
(including copper(I) and copper(II) sources) resulted in compa-
rable results with respect to enantioselectivity (Table 1, entries
13–15). Copper(II) triflate led to product 4a formation with
lower yield. Interestingly, also with AgOAc the desired product
4a was obtained with 92% enantioselectivity, (Table 1, entry
16), however, in only 36% yield. Decreasing the catalyst load-
diene (3) to cyclopentadienone (2) followed by
double enantioselective 1,3-dipolar cycloaddition of
azomethine ylides (1) to form the 5,5,5-tricyclic prod-
uct (4) (Scheme 1).
Table 1. Screening of reaction conditions.[a]
To establish the proposed cascade, we initially in-
vestigated the reaction of glycine ester imine 1a and
cyclopentadiene 3 in the presence of triethylamine
under air using a complex of copper(I) with (R)-Fesul-
phos[16] 5a as catalyst. To our delight we obtained
the desired double cycloaddition product 4a with
moderate yield (50%) but excellent diastereo- and
enantioselectivity (ee 97%, Table 1, entry 1). It is nota-
ble, that a S,P-ligand, such as (R)-Fesulphos, can be
successfully employed in the cascade reaction, be-
cause the aerobic oxidation of CÀH bonds usually re-
quires dinitrogen-based ligands. Furthermore, over-
oxidation of the desired product 4a was not detect-
ed. Under an argon atmosphere (Table 1, entry 2) the
yield of the double cycloaddition product 4a sub-
stantially decreased while under an oxygen atmos-
phere (Table 1, entry 3) the yield increased to 74%
without altering the enantioselectivity.
Entry
Catalyst
Ligand
Solvent
atm
d.r.[b]
Yield
[%][c]
ee
[%][d]
1
2
3
4
5
6
7
8
CuBF4
CuBF4
CuBF4
CuBF4
CuBF4
CuBF4
CuBF4
CuBF4
CuBF4
CuBF4
CuBF4
CuBF4
CuPF6
CuOAc2
CuOTf2
AgOAc
CuBF4
5a
5a
5a
5a
5a
5a
5a
5b
CH2Cl2
CH2Cl2
CH2Cl2
THF
air
Ar
O2
O2
O2
O2
O2
O2
O2
O2
O2
O2
O2
O2
O2
O2
O2
>95:5
>95:5
>95:5
n.d.
n.d.
>95:5
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
>95:5
n.d.
>95:5
>95:5
n.d.
50
20
74
traces
traces
58[e]
n.d.
n.d.
n.d.
traces
traces
n.d.
69
traces
50
97
97
97
n.d.
n.d.
95
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
94
PhMe
MeOH
MeCN
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
9
5 c
5d
10
11
12
13
14
15
16
17[f]
(R)-BINAP
(R)-SEGPHOS
Encouraged by these results, we investigated vari-
ous solvents for the cascade reaction (Table 1, entries
3–7). While in toluene and THF traces of the desired
product 4a were observed after 16 hours, in acetoni-
trile no conversion occurred (Table 1, entries 4–5, 7).
In methanol a high reaction rate was observed and
full conversion with respect to a-iminoester 1a
within one hour was detected. However, the yield of
the desired product 4a was lower than in CH2Cl2
(Table 1, entry 6). Investigation of different chiral li-
gands were (Table 1, entries 8–12) revealed that, no-
tably, again (R)-Fesulphos 5a yielded the best results
5a
5a
5a
5a
5a
n.d.
96
92
36
traces
n.d.
[a] Reaction conditions: chiral ligand 5 (5.5 mol%), catalyst (5 mol%), Et3N (50 mol%),
glycine ester imine 2a (1 equiv, 0.20 mmol) and cyclopentadiene 3 (0.70 mmol) in sol-
vent (0.05m) at ambient temperature. [b] Determined by 1H NMR spectroscopy of
crude reaction mixture. [c] Yield of isolated pure product 4a after chromatography on
silica gel. [d] Determined by HPLC analysis on a chiral stationary phase. [e] Isolated
after 1 h. [f] Using 2.5 mol% of catalyst and 2.75 mol% of ligand 5a. n.d. = not deter-
mined; CuBF4 = Cu(CH3CN)4BF4; CuPF6 = Cu(CH3CN)4PF6.
&
&
Chem. Eur. J. 2015, 21, 1 – 6
2
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!