Angewandte
Chemie
Table 1: Investigation of background isomerization with Cu salts during asymmetric
conjugate addition to (Z)-4a or (E)-4a.
the starting material, 7% olefin isomerization was
observed (Table 1, entry 10). Similarly,
Entry Reaction mixture[a]
T [8C] (Z)-4a 2a
(E)-4a
[(MeCN)4Cu]BF4, CuTC, and CuOAc also induced
low levels of isomerization (Table 1, entries 11–13).
These four Cu salts were then further evaluated
in the actual ACA reactions (Table 2). The ACA
reaction was very sluggish with Cu(OCOCF3)2·xH2O
(Table 2, entry 3), CuTC (entry 5), or CuOAc
[%][b]
[%][b] [%][b,c]
1
2
3
4
5
6
7
8
9
10
11
12
13
(Z)-4a+(CuOTf)2·PhH
25
25
97
99
43
34
47
70
97
100
99
7
0
0
0
0
0
0
0
0
0
0
0
0
0
3
1
(Z)-4a+(CuOTf)2·PhH+L1
(Z)-4a+(CuOTf)2·PhH+Me2Zn
(E)-4a+(CuOTf)2·PhH+Me2Zn
À30
57
66
53
30
3
0
1
93
2
5
À30
(Z)-4a+Cu(OCOCF3)2·xH2O+Me2Zn À30
(entry 7).
Reactions
with
[(MeCN)4Cu]BF4
(Z)-4a+(MeCN)4CuOTf+Me2Zn
(Z)-4a+(MeCN)4CuPF6 +Me2Zn
(Z)-4a+(MeCN)4CuPF6
À30
À30
25
(Table 2, entry 4) or [(MeCN)4Cu]OTf (entry 6)
were not satisfactory in terms of their enantioselec-
tivity and/or reactivity. To our delight, the use of
[(MeCN)4Cu]PF6 led to superior enantioselectivity
as compared to the reaction with (CuOTf)2·PhH.
This observation is consistent with the background-
isomerization results in Table 1, thus suggesting that
(Z)-4a+(MeCN)4CuPF6 +L1
(E)-4a+(MeCN)4CuPF6 +Me2Zn
(Z)-4a+(MeCN)4CuBF4 +Me2Zn
(Z)-4a+CuTC+Me2Zn
25
À30
À30
À30
À30
98
95
96
(Z)-4a+CuOAc+Me2Zn
4
[a] Reaction conditions: 4 mol% Cu, L1 (4 mol%), Me2Zn (1.2 equiv), 65 h. [b] The [(MeCN)4Cu]PF6 indeed significantly suppressed the
yield is based on the consumption of the substrate and was determined by HPLC
analysis. [c] (E)-4a was isolated and fully characterized (see the Supporting
Information for details).
undesired olefin isomerization and that (Z)-4a was
more advantageous as a substrate than isomer (E)-
4a with regard to both reactivity and enantioselec-
course of the reaction. It has been reported previ-
Table 2: Screening of Cu salts in the ACA of (Z)-4a or (E)-4a.[a]
ously[7] that olefin isomerization could be responsible
Entry Substrate CuI (mol%)
L1
t
Conv.
ee
for lower enantioselectivities in conjugate addition
reactions owing to the different reactivity and/or
selectivity of the (E)- and (Z)-isomers, as illustrated
in Scheme 4. Indeed, in our case, opposite enantio-
selectivities were observed for the reactions of (E)-
and (Z)-4a (Scheme 3).
To fully understand how severe the isomerization
was, we investigated the background isomerization
with (Z)-4a and (E)-4a (Table 1).[6] When (Z)-4a
(1 equiv) and (CuOTf)2·PhH (2 mol%) with or
without chiral ligand L1 (4 mol%) were stirred in
toluene in the absence of Me2Zn, olefin isomer-
ization was minimal, even at room temperature
(Table 1, entries 1 and 2). However, in the reaction
mixture of pure (Z)-4a (1 equiv) with (CuOTf)2·PhH
(2 mol%) and Me2Zn (1.2 equiv), isomer (E)-4a was
formed in 57% yield and 43% of (Z)-4a remained
after 65 h at À308C (Table 1, entry 3). When pure
[mol%] [h] [%][b]
[%][c]
1
2
3
4
5
6
7
8
9
(Z)-4a
(E)-4a
(Z)-4a
(Z)-4a
(Z)-4a
(Z)-4a
(Z)-4a
(Z)-4a
(E)-4a
(CuOTf)2·PhH (2)
(CuOTf)2·PhH (2)
Cu(OCOCF3)2·xH2O (4)
[(MeCN)4Cu]BF4 (4)
CuTC (4)
[(MeCN)4Cu]OTf (4)
CuOAc (4)
[(MeCN)4Cu]PF6 (4)
[(MeCN)4Cu]PF6 (4)
4
4
4
4
4
4
4
4
4
24 100 (96) 65 (S)
24
65
65
65
24
65
65
96
32 (26) 76 (R)
5 (3)
42 (42) 50 (S)
5 (1)
10 (98) 63 (S)
8 (5)
–
–
–
98 (95) 95 (R)
16 (15) 85 (R)
[a] Reactions were carried out in toluene at À308C. [b] The reported conversion is
based on the consumption of the substrate and was determined by HPLC analysis.
The number in parenthesis indicates the HPLC assay yield. [c] The ee value was
determined by HPLC on a chiral stationary phase. The absolute configuration was
determined by our internal program and by analogy.
(E)-4a was used in place of (Z)-4a in this control experiment,
isomer (Z)-4a was formed in 34% yield (Table 1, entry 4).
These observations, coupled with the fact that under the
standard reaction conditions with (CuOTf)2·PhH, 2a was
formed with a moderate ee value of 65%, suggested that
nitroolefin isomerization during the course of ACA is
probably the cause of the poor enantioselectivity. Accord-
ingly, if we could suppress the undesired nitroolefin isomer-
ization during the ACA reaction, we might be able to improve
the enantioselectivity significantly. We suspected that the
nature of the Cu salt could play a critical role in the observed
olefin isomerization and examined the effect of a variety of
Cu salts on olefin isomerization. As with (CuOTf)2·PhH, the
use of Cu(OCOCF3)2·xH2O or [(MeCN)4Cu]OTf resulted in
significant olefin isomerization (Table 1, entries 5 and 6). In
contrast, the use of a catalytic amount of [(MeCN)4Cu]PF6
and Me2Zn (1.2 equiv) induced only 3% olefin isomerization
(Table 1, entry 7). Under the same conditions with (E)-4a as
tivity (Table 2, entries 8 and 9). Thus, through the use of
[(MeCN)4Cu]PF6 as the catalyst and the (Z)-nitroalkene 4a as
the substrate, dramatically improved enantioselectivity was
observed in the ACA reaction with Me2Zn. Interestingly,
when (CuOTf)2·PhH was used as the catalyst, the ACA of
(Z)-4a gave the S enantiomer of 2a as the major product,
whereas the ACA of (E)-4a gave the R enantiomer as the
major product. However, when [(MeCN)4Cu]PF6 was used as
the catalyst, ACA of either (Z)-4a or (E)-4a gave (R)-2a as
the major product, thus suggesting that nucleophilic attack by
Me2Zn from the Re face was favored, irrespective of the
nitroolefin geometry.
To explore the scope of the [(MeCN)4Cu]PF6–L1-cata-
lyzed ACA reaction of (Z)-nitroalkenes with Me2Zn, we first
needed an efficient and stereoselective method for the
synthesis of Z nitroalkenes. These compounds have been
synthesized previously by the b-elimination of mesylates,[8a]
by the oxidative elimination of b-phenylselenides[8b] or b-
Angew. Chem. Int. Ed. 2014, 53, 12153 –12157
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim