Table 2. Substrate scope of the asymmetric conjugate addition of nitro-
olefins and acyclic enones.[a]
Next, a series of DPEN derivatives were synthesized and ex-
amined. Encouragingly, when the monoalkyl-protected pri-
mary-secondary diamines were used with PhCO2H as a co-
catalyst, the enantioselectivity increased slightly and the
yield was maintained (Table 1, entries 3–5). The enantiose-
lectivity also increased with an increase in steric hindrance
of the substituent on DPEN. The DPEN derivative L5, sub-
stituted with a monocyclohexyl group, which may offer suit-
able alkaline and steric hindrance, was found to be the best
candidate giving the desired allylic nitro compound in 33%
yield with 69% ee (Table 1, entry 5). However, the use of
disubstituted secondary diamine L2 resulted in a lower yield
than monoalkyl substituted diamine L3 (Table 1, entry 2 vs.
entry 3). Further investigation of the solvent showed that
toluene improves the enantioselectivity up to 76% ee, yet
the yield was low (Table 1, entry 6). Diethyl ether proved
more suitable and increased both the ee value and the yield
(42% yield, 81% ee of the syn product, Table 1, entry 7 vs.
entries 5 and 6).
Different Brønsted acids that might facilitate the activa-
tion of the enone through the iminium intermediate were
evaluated to improve the results. A series of achiral and
chiral acids were tested, and d-mandelic acid was found to
increase the enantioselectivity (43% yield, 94% ee)
(Table 1, entry 9 vs. entries 7 and 8). Additionally, the prod-
uct was obtained in 12% yield in the absence of the acidic
co-catalyst, which also indicated that the acid was crucial in
the activation process (Table 1, entry 10). To improve the
yield, a mixture of diamine L5 and d-mandelic acid was
treated with an additional achiral organic base. The base ad-
ditive was found to have a remarkable effect on the catalytic
activity by releasing the amine catalyst from the nitroalkene
adduct, thereby accelerating the catalytic cycle (Table 1, en-
tries 11–13). When 20 mol% of 4-dimethylaminopyridine
(DMAP) was used as the additive, the yield was improved
without any decrease in enantioselectivity (87% yield,
94% ee; Table 1, entry 13 vs. entries 11 and 12). Further-
more, when the catalyst loading was decreased to 10 mol%,
excellent results could also be obtained by increasing the
substrate concentration and prolonging the reaction time
(87% yield, 94% ee, Table 1, entry 14). For convenience, the
same result could be obtained by treating L5 with one
equivalent of d-mandelic acid to give the salt, which was
used directly as the catalyst (Table 1, entry 15 vs. entry 14).[7]
After evaluating a variety of conditions, a combination of
L5-d-mandelic acid (10 mol%), DMAP (10 mol%) as the
additive, and diethyl ether as the solvent was found to be
optimal (Table 1, entry 15).[8]
Entry
R1
R2
Yield
[%][b]
d.r.[c]
syn/anti
ee [%][d]
syn(anti)
ACHTUNGTRENNUNG
1
2
3
4
5
6
7
8
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
3-ClC6H4
2-furnyl
4-ClC6H4
4-BrC6H4
Ph
88 (3aa)
91 (3ab)
87 (3ac)
92 (3ad)
75 (3ae)
78 (3af)
93 (3ag)
90 (3ah)
70 (3ai)
75 (3aj)
84 (3ca)
90 (3ha)
93 (3 fa)
85 (3ga)
1.7:1
1.3:1
3.0:1
2.7:1
1.6:1
1.5:1
2.6:1
1.3:1
3.0:1
1.3:1
1.5:1
2.6:1
24:1
93 (94)
96 (90)
90 (89)
91 (91)
92 (90)
92 (93)
93 (89)
93 (89)
87 (89)
89 (91)
85 (89)
94 (94)
97
3-MeOC6H4
4-MeOC6H4
3-MeC6H4
3-CF3C6H4
3-ClC6H4
4-FC6H4
2-furnyl
2-thiophenyl
2-naphthyl
Ph
9
10
11[e]
12[e]
13[e]
14[e]
Ph
Ph
Ph
24:1
96[g]
15[e]
4-ClC6H4
93 (3 fi)
33:1
95
16[e]
17[e]
18[f]
19[f]
4-ClC6H4
4-ClC6H4
Ph
2-naphthyl
4-PhC6H4
Ph
90 (3 fj)
91 (3 fh)
61 (3ak)
60 (3al)
25:1
5:1
1.1:1
1.5:1
90
96
80 (81)
80 (83)
Ph
3-MeC6H4
[a] All reactions were carried out by using the salt (10 mol%), which was
prepared beforehand,[7] DMAP (10 mol%),
(0.1 mmol), and
1
2
(0.12 mmol) in Et2O (0.3 mL) at 258C for 48 h. [b] Total yield of isolated
product of the diastereomeric mixture of 3. [c] Determined by 1H NMR
spectroscopy. [d] Determined by HPLC analysis; the data in parenthesis
is the ee value of the minor isomer. [e] The reaction time was 36 h.
[f] Catalyst loading was 20 mol% and the reaction time was 72 h. [g] The
absolute configuration was determined by X-ray analysis as (4S,5S)-6-(4-
bromophenyl)-5-nitro-4-phenylhept-6-en-2-one.
the catalyst was also efficient (Table 2, entries 8 and 9).
Moreover, the conjugate addition of a fused-ring aromatic
enone with 1a proceeded well to give the desired product in
good yield with an excellent ee value (Table 2, entry 10). In
these cases, low diastereoselectivity was obtained due to a
poor spatial arrangement and the acidic deprotonation on
the nitro carbon atom (1.3:1–3:1 syn/anti, Table 2, entries 1–
10). This was a minor problem because the two diastereo-
mers could be separated with high optical purity. Subse-
quently, substituted nitroalkenes were also tested with (E)-
4-aryl-but-3-en-2-ones (Table 2, entries 11–17). Notably, the
diastereoselectivity greatly depended on the substituted po-
sition of the nitroalkenes; p-Br- and p-Cl-substituted nitroal-
kenes gave the corresponding products with high diastereo-
selectivities (up to 33:1) in good yields and with excellent
enantioselectivities (Table 2, entries 13–17 vs. entries 11 and
12). In addition, chalcones also performed well in the reac-
tion with good enantioselectivities in moderate yields
(Table 2, entries 18 and 19).
Under the optimized conditions (Table 1, entry 15), the
substrate scope of the crossed-conjugate addition of differ-
ent a,b-enones with b-methyl nitrostyrene (1a) was exam-
ined. In general, regardless of the electronic property or
steric hindrance of the substituent on the aromatic ring of
the enones, chiral allylic nitro compound 3 could be ob-
tained in good to excellent yields with excellent enantiose-
lectivities for both isomers (75–93% yield, 89–96% ee,
Table 2, entries 1–7). In the case of heteroaromatic enones,
Next, we expanded our attention to investigate the sub-
strate scope of the crossed-conjugate addition of cyclo-
2366
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 2365 – 2368