Communication
compared with the phenyl group (3k–3l, Table 1). However,
excellent and comparable enantioselectivities were detected
under 10% and 20% catalyst loadings. It was obvious that
when phenol esters were replaced by phthalimide esters, the
catalyst loading and base loading could be decreased to 10%
and 2.0 equivalents, respectively, and that no further additive
was needed. These results revealed that esters 2 exhibited
a higher reactivity than 4-nitrophenol esters overall from the
point of catalyst loading, amount of base, and additive. As
esters as substrates are more stable than enals and more con-
venient to use, this formal [4+2] annulation provided a series
of products containing a single aryl substituent at the lactam
a-carbon, while NHC-bound enolates via enals[8a] could not be
obtained. On the other hand, this activation mode provided
trans products, while NHC-bound enolates via enals only pro-
vided cis products.
lated that an excess amount of DIPEA and conjugate acid
[DIPEA-H]+ may disturb the reaction to some extent. When the
mixture of the NHC-catalyzed reaction was washed either with
water or 1n HCl, a satisfactory yield of 4a was obtained
(Scheme 3). However, the yields were decreased sharply either
Scheme 3. Sequential transformation of the two steps.[a] ([a] Condition
a (achiral catalyst B’ instead of chiral catalyst B): 1 (0.10 mmol), 2
(0.15 mmol, 1.5 equiv), NHC catalyst B’ (10 mol%), DIPEA (2.0 equiv), CH2Cl2
(0.5 mL), r.t. for 24 h; [b] Isolated yield over two steps; the minor diastereo-
meric isomer was not detected.)
Next, we investigated whether NHPI could be used as a new
catalyst for the following transformation of the first main prod-
uct 3a. Thus, 3a (12:1 d.r.) was prepared as the substrate for
the designed reaction. When 3a was treated with NHPI and O2
in CHCl3, the sulfonyl migration product 4a was obtained in
only 16% yield (entry 1, Table 2). A higher yield was obtained
in the absence of Co(II) or by replacement of O2 (bubbles) with
an air atmosphere. This result revealed that the by-product
NHPI actually acted as a new catalyst for the second transfor-
mation, and this verified our hypothesis (Scheme 1).
Table 2. Exploring the reaction conditions for the N- to C-sulfonyl migra-
tion.[a]
Subsequent studies explored the possibility for the NHC/
NHPI catalytic cascade process (Table 3). We found that many
substrates could give the desired products successfully, and
electronic effects enormously affected the yields of the sulfonyl
migration products. When substituents on the aromatic ring R2
were electron-donating groups (4d–4e), the migration reac-
tion gave a higher yield; indeed, it is reasonable that an elec-
tron-rich system is beneficial for the oxidation process. No ob-
vious interference of the reaction was observed when substitu-
ents on the aromatic ring R1 were different groups (4b–4c).
The inverse trend was observed when different esters were
used under the same conditions (4 f–4h). In this process, the
enantiomeric excess values of products were all maintained
from the upstream products.
Entry
NHPI
20%
20%
–
50%
20%
20%
20%
20%
Co(OAc)2·4H2O
–
Solvent
CHCl3
CHCl3
CHCl3
CHCl3
CH2Cl2
(CH2Cl)2
CH3CN
CH3OH
Yield [%][b]
16
50
12
1
2
3
4
5c
6
7
8
5%
5%
5%
5%
5%
5%
5%
58
trace
trace
n.d.
This sulfonyl migration reaction may proceed via a radical
pathway. In order to support this idea, TEMPO was added to
the reaction, and the reaction was completely suppressed. This
result provides indirect evidence that the sulfonyl migration re-
action proceeds via a radical pathway[17] (Scheme 4).
n.d.
[a] Reaction conditions: 3a (0.1 mmol), solvent (2.5 mL). [b] Isolated yield;
the minor diastereomeric isomer was not detected in all cases. [c] The re-
action was performed at 608C.
In conclusion, we have developed sequential N-heterocyclic
carbene (NHC) catalysis and N-hydroxyphthalimide (NHPI) cat-
alysis in which the upstream by-product NHPI acts as the cata-
lyst for a downstream nitrogen-to-carbon sulfonyl migration
reaction. In the upstream catalytic system, only a low catalyst
loading was needed and no additive was required for the cycli-
zation of enolate azolium with a,b-unsaturated imines, reveal-
ing that this kind of esters exhibited a high reactivity. This reac-
tion proceeded to provide a high yield as well as satisfactory
enantioselectivity and diastereoselectivity. The downstream
catalytic process allowed for sulfonyl migration of the first
product, while retaining the enantiomeric excess. Further in-
vestigations on the mechanistic studies and other transforma-
when Co(OAc)2·4H2O was added as co-catalyst as the combina-
tion of O2 and CoII could accelerate the formation of a PINO
radical (entry 2, Table 2). Control experiments revealed that
NHPI is necessary for the transformation (entry 3, Table 2).
Pleasingly, increasing the amount of NHPI to 50 mol% gave
a higher yield (entry 4, Table 2), and the structure of 4a was
confirmed by X-ray analysis (CCDC 1059049).[16] However, other
solvents could not provide improved results (entries 5–8,
Table 2).
After the NHC catalytic reaction, Co(OAc)2·4H2O was added
and the reaction was performed under oxygen atmosphere.
However, only trace amounts of 4a were detected. We specu-
Chem. Asian J. 2016, 11, 193 – 197
195
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim