Highly enantioselective cyclopropanation of styrenes and diazoacetates catalyzed by 3-oxobutylideneaminatocobalt(II) complexes, part 1. Designs of cobalt complex catalysts and the effects of donating ligands

Article abstract of DOI:10.1246/bcsj.74.2139

Highly enantioselective cyclopropanation of styrene derivatives and diazoacetates was effectively catalyzed by reasonably designed 3-oxobutylideneaminatocobalt(II) complexes, whose ligands were prepared from 1,2-dimesitylethyl-enediamine and alkyl 3-oxobutanoates. The steric demand of the diamine unit of the complexes seriously influenced the enantioselectivity, and the ester groups on their side chains somewhat improved the trans-selectivity. Addition of a catalytic amount of N-methylimidazole significantly accelerated the reaction and enhanced the enantioselectivity due to its coordination to the center cobalt atom of the complex as an axial ligand. Alcoholic or aqueous alcoholic solvents were also effective particularly for the cyclopropanation of 1-substituted 1-phenylethylenes to achieve high enantioselectivity in aqueous methanol.

Full text of DOI:10.1246/bcsj.74.2139

c. Jpn.
e Chemical © 2001 The Chemical Society of Japan
Bull. Chem. Soc. Jpn., 74, 2139–2150 (2001) 2139
ciety of Ja-
n
e Chemical
ciety of Ja-
n
Highly Enantioselective Cyclopropanation of Styrenes and Diazoacetates
Catalyzed by 3-Oxobutylideneaminatocobalt(Ⅱ) Complexes, Part 1.
Designs of Cobalt Complex Catalysts and the Effects of Donating
Ligands
01
Cobalt Catalyzed Asymmetric Cyclopropanation
3
5
9
0
T. Ikeno et al.
Taketo Ikeno, Mitsuo Sato, Hiroyuki Sekino, Asae Nishizuka, and Tohru Yamada*
SJA8
0
9-2673
Department of Chemistry, Faculty of Science and Technology, Keio University,
Hiyoshi, Kohoku-ku, Yokohama 223-8522
085
(
Received March 13, 2001)
01
Highly enantioselective cyclopropanation of styrene derivatives and diazoacetates was effectively catalyzed by rea-
0
1
1
sonably designed 3-oxobutylideneaminatocobalt(Ⅱ) complexes, whose ligands were prepared from 1,2-dimesitylethyl-
enediamine and alkyl 3-oxobutanoates. The steric demand of the diamine unit of the complexes seriously influenced the
enantioselectivity, and the ester groups on their side chains somewhat improved the trans-selectivity. Addition of a cata-
lytic amount of N-methylimidazole significantly accelerated the reaction and enhanced the enantioselectivity due to its
coordination to the center cobalt atom of the complex as an axial ligand. Alcoholic or aqueous alcoholic solvents were
also effective particularly for the cyclopropanation of 1-substituted 1-phenylethylenes to achieve high enantioselectivity
in aqueous methanol.
.
0
.3
Since the catalytic enantioselective cyclopropanation of ole-
1
fins with diazoacetates was first reported in 1966, various
transition metals and many varieties of optically active ligands
2
have been subjected to this benchmark test reaction, and some
of them have been applied to the industrial synthesis of pyre-
3
4
throids. For example, chiral (dioximato)cobalt(Ⅱ) and chiral
5
Schiff-base copper(Ⅱ) complex catalysts were originally pro-
posed as the catalyst for enantioselective cyclopropanation.
Recently, copper(Ⅰ) complexes with semicorrin, bis(oxazo-
6
Chart 1.
7
8
line) or bipyridine ligands, ruthenium(Ⅱ) complexes with
9
18
bis(oxazolinyl)pyridine ligands, and rhodium(Ⅲ) complexes
reaction as a Lewis Acid.
1
0
with proline derivative ligands were developed as effective
catalysts to provide an extremely high level of enantioselectiv-
ity. It was very recently reported that optically active
In the previous communications, we reported that the highly
enantioselective cyclopropanation of styrenes with diazoace-
tates proceeded when catalyzed by rationally designed co-
balt(Ⅱ) complexes having 1,2-dimesitylethylenediamine in the
diamine unit and ester moieties on the side chain and that addi-
1
1
(
salen)cobalt(Ⅲ) complexes were efficient catalysts for the
asymmetric cyclopropanation of styrene derivatives, and much
effort has been devoted to developing more effective cata-
1
9
tion of a catalytic amount of N-methylimidazole or alcoholic
solvent caused a remarkable effect on the reaction. In this ar-
12
20
lysts.
The optically active 3-oxobutylideneamines derived from
the corresponding 3-oxoalkanals and optically active 1,2-diar-
yl-1,2-ethylenediamines have been proposed to provide a new
class of effective ligands for catalytic enantioselective reac-
tions; manganese(Ⅲ) complexes with optically active 3-
oxoalkylideneaminato ligands have been employed as effective
catalysts for the aerobic and enantioselective epoxidation of
ticle, we would like to fully describe the details of the experi-
ments and to discuss the effects of additives and/or solvents in
the cobalt-complex-catalyzed cyclopropanation reaction.
Results and Discussion
Preliminary Experiments. Several experiments for enan-
tioselective cyclopropanation were performed on the reaction
of neat styrene with ethyl diazoacetate in the presence of 0.05
molar amount 3-oxobutylideneaminato cobalt complex 1a at
room temperature. The corresponding cyclopropanecarboxy-
late was obtained in 85% yield with moderate trans-selectivity
(trans:cis = 77:23) and with moderate enantioselectivity
(63% ee for trans-isomer, Eq. 1). This result encouraged us to
improve the reaction system by the design of the cobalt-com-
1
3
unfunctionalized olefins and the asymmetric aerobic oxida-
tion of sulfides to the corresponding optically active sulfox-
14
ides. The corresponding cobalt(Ⅱ) complexes, for example,
MPAC (1a) (Chart 1), effectively catalyzed the enantioselec-
1
5
tive borohydride reduction of prochiral aryl ketones, N-phos-
1
6
17
phinylimines, and α,β-unsaturated carboxamides.
They
also catalyzed an enantioselective hetero Diels–Alder

2
140 Bull. Chem. Soc. Jpn., 74, No. 11 (2001)
Cobalt Catalyzed Asymmetric Cyclopropanation
a)
Table 2. Amount of N-Methylimidazole
Molar
amount
Entry
Yield/% trans:cis ee (trans)/%
b)
1
2
3
4
5
6
7
8
0
1
2
4
8
62
97
94
91
98
88
94
quant
76:24
76:24
76:24
76:24
79:21
76:24
76:24
79:21
75
84
84
82
84
82
85
85
16
32
solvent
(
1)
a) Reaction conditions: Five hundredth molar amount
of catalyst 1a and N-methylimidazole in THF at 25 °C
for 2 h. b) Molar amount of N-methylimidazole vs
Co(Ⅱ) complex.
plex catalysts and the optimization of the reaction conditions
for achieving high diastereo- and enantioselectivity.
Solvents were first examined in the reaction of styrene with
t-butyl diazoacetate in the presence of 0.05 molar amount of
cobalt(Ⅱ) complex 1a. As shown in Table 1, the yields in 10 h
depended on the solvents. In a nonpolar solvent such as tolu-
coordinated to the cobalt atom as an axial ligand to accelerate
the present cyclopropanation reaction. Several nitrogen-con-
taining compounds were then screened to identify an appropri-
ate additive to reveal that N-methylimidazole was the promi-
nent accelerator of the reaction. Two molar amounts of the N-
methylimidazole versus the cobalt complex greatly enhanced
the reaction, and the yield of the product was remarkably im-
proved. As shown in Fig. 1, in the absence of the N-methylim-
idazole, the reaction was not completed in 4 h; only 60% yield
of the product was obtained. On the contrary, in the presence
of N-methylimidazole, the reaction was completed in 4 h to af-
ford the product in about 90% yield. Along with rate enhance-
ment, the enantiomeric excess of the product was improved. In
the absence of N-methylimidazole, enantiomeric excesses of
the product ranged from 70% ee to 75% ee, whereas they were
over 80% ee in the presence of N-methylimidazole.
The loading amount of the N-methylimidazole versus the
cobalt complex was next examined. As shown in Table 2, no
effect on the diastereoselectivity was observed either in the
presence or the absence of N-methylimidazole. Compared
with the result of 2 molar amounts of N-methylimidazole (En-
try 3), the yield and enantiomeric excess of the product were
lower without N-methylimidazole (Entry 1), and it was re-
vealed that an equimolar amount of N-methylimidazole was
sufficient for acceleration (Entry 2). On the other hand, almost
no additional improvement was observed when more than 2
molar amounts of N-methylimidazole were added (Entries 4–
8). It was noteworthy that, when the reaction was performed in
N-methylimidazole solvent, the reaction was smoothly com-
pleted in 2 h with the remaining enantio- and diastereoselectiv-
ity.
ene or CH
1
5
2
Cl
–3), while the yield was increased up to 60% in THF (Entry
). Rate enhancement of the reaction was observed in THF but
O (Entry 4). It was noteworthy that the diastereo-
2
, yields of the product were below 40% (Entries
not in Et
2
and enantioselectivity of the product was also improved to
some extent when THF was used as a solvent.
Effect of Axial Ligand. In most of the X-ray structures of
3
-oxobutylideneaminato and salencobalt(Ⅱ) and (Ⅲ) complex-
es reported previously, donating ligands, such as alcohols,
2
1
THF, or pyridine, were found in their axial positions. It is
noted that these axial ligands often exert considerable influ-
ence on the catalytic activities. The effects of the axial donor
ligand were reported for manganese complex-catalyzed oxida-
2
2
tions. For example, pyridine N-oxide and 2-methylimidazole
accelerated the reaction rate and improved the enantioselectiv-
ity in salenmanganese-catalyzed epoxidation using iodosyl-
22c
benzene as a terminal oxidant, and reversal of the absolute
configuration of the product was observed by addition of N-
methylimidazole in salenmanganese-catalyzed aerobic oxida-
22e
tion.
Based on this knowledge, we deduced that the THF
Table 1. Solvent Effect on the Reaction of Styrene with
a)
t-Butyl Diazoacetate
b)
c)
d)
Entry Solvent Yield/%
trans:cis
ee (trans)/%
1
2
3
4
5
—
Toluene
26
18
40
21
60
71:29
70:30
75:25
68:32
76:24
62
58
57
55
75
A variety of additives was scrutinized for the present cyclo-
propanation (Table 3). N-Phenyl and N-benzylimidazole
showed an effect similar to that of N-methylimidazole both on
the yield and the enantioselectivity (Entries 2 and 3). Pyridine
and 5-methylimidazole increased the enantiomeric excess,
while yields of the product were lower (Entries 4 and 5). The
catalyst was deactivated, probably because these compounds
coordinated the cobalt atom strongly enough to occupy both
axial sides of the complex. Other nitrogen-containing aromat-
ic compounds produced slightly worse results compared to that
without additives (Entries 6–9). It seems that these heterocy-
CH
Et
THF
2
Cl
2
2
O
a) Reaction conditions: Five hundredth molar amount of
catalyst 1a at 25 °C for 10 h (see the general procedure).
b) Isolated yield based on t-butyl diazoacetate. c) Deter-
mined by GC. d) Determined by HPLC analysis after
reduction of the isolated products into the corresponding
alcohols (Trans: Daicel Chiralcel OD-H and/or OB-H,
hexane/2-propanol).

T. Ikeno et al.
Bull. Chem. Soc. Jpn., 74, No. 11 (2001) 2141
Fig. 1. Time-course of the cyclopropanation reaction in the presence and absence of N-methylimidazole.
Fig. 2. Relation between yields and enantiomeric excesses for nitrogen-containing heteroaromatics.
cles could not coordinate the cobalt complex for steric rea-
matics as additives. A tendency was observed in which the ad-
ditives affording the product in a higher yield realized a higher
enantiomeric excess. This figure clearly indicates that imida-
zole derivatives prominently enhanced both the yield and the
enantiomeric excess.
23
sons. Aliphatic nitrogen compounds were not employed as
effective additives (Entries 10–13). In the presence of a thiol
derivative, the reaction soon stopped, and a small amount of
di-t-butyl fumarate and di-t-butyl maleate were detected as by-
products (Entry 16). This was in great contrast to the result
where no dimeric products were obtained using nitrogen com-
pounds as additives. Pyridine N-oxide (Entry 17) remarkably
enhanced the trans-selectivity, but the enantiomeric excess was
slightly lower than that in the presence of N-methylimidazole.
Figure 2 shows the relation between the yield and the enan-
tiomeric excess using various nitrogen-containing heteroaro-
Design of the Cobalt(Ⅱ) Complex Catalyst. Based
on
the above-mentioned observation, standard reaction conditions
were determined as follows: 5.0 mmol of styrene and 1.0
mmol of t-butyl diazoacetate in THF (1 mL) in the presence of
0.05 molar amount of cobalt(Ⅱ) complex and 0.1 molar
amount of N-methylimidazole. A variety of the cobalt com-
plexes was examined under the standard conditions (Table 4).

2
142 Bull. Chem. Soc. Jpn., 74, No. 11 (2001)
Cobalt Catalyzed Asymmetric Cyclopropanation
a)
Table 3. Various Additives for Cyclopropanation
ee/%
b)
Entry
Additive
Yield/%
94
trans:cis
76:24
77:23
75:25
73:27
74:26
trans
cis
1
2
3
4
5
84
85
85
84
85
80
77
82
83
81
78
83
27
43
6
7
8
9
36
47
34
73:27
74:26
72:28
71
75
77
82
82
75
37
72
62
72
74
73:27
72:28
78:22
77:23
75:25
74
52
76
76
81
84
62
78
77
87
1
1
1
1
0
1
2
3
1
1
4
5
62
62
75:25
75:25
76
76
80
82
1
1
6
7
complex mixture
quant
82:18
80
80
1
1
8
9
57
62
76:24
76:24
76
75
79
84
a) Reaction conditions: Five hundredth molar amount of catalyst 1a in THF at 25 °C for 2 h.
b) Two molar amount of additive vs catalyst were employed.
The cobalt(Ⅱ) complexes 1a, 7a, and 8a were prepared from
the optically active 1,2-diphenylethylenediamine, 1,2-bis(3,5-
dimethylphenyl)ethylenediamine, and 1,2-dimesitylethylene-
diamine, respectively. The cobalt(Ⅱ) complexes 1a and 7a cat-
alyzed the reaction to afford the resulting cyclopropanecarbox-
ylates with 84–85% ee and 75–76% trans-selectivity, whereas
complex 8a improved the enantiomeric excess of the product
to 90% ee but decreased the trans-selectivity (68:32) (Entries
1–3). The cobalt(Ⅱ) complexes 9a, 10a, and 11a with a small-
er side chain (acetyl in 9a–11a vs 2,4,6-trimethylbenzoyl in

T. Ikeno et al.
Bull. Chem. Soc. Jpn., 74, No. 11 (2001) 2143
a)
Table 4. Various Cobalt(Ⅱ) Complex Catalysts for Enantioselective Cyclopropanation
Reaction
time/h
Entry
Catalyst
Temp/°C
Yield/%
94
trans:cis
ee (trans)/%
1
25
2
76:24
84
2
3
4
5
6
7
8
9
25
50
25
25
40
25
25
40
2
20
2
quant
89
75:25
68:32
76:24
78:22
83:17
81:19
83:17
90:10
85
90
70
61
96
66
63
96
quant
quant
80
2
10
2
quant
94
2
4.5
97
1
0
40
22.5
99
91:9
96
a) Reaction conditions: Five hundredth molar amount of catalyst and 0.1 molar amount (2.0 molar amount vs Co(Ⅱ) complex)
of N-methylimidazole in THF.
1
a, 7a, and 8a) were then subjected to the cyclopropanation
selectivity was also enhanced to 83%. These observations sug-
gested that the sterically demanding diamine would be effec-
tive for improving the enantioselection. Various cobalt(Ⅱ)
complexes with ester moieties on the side chains were then
synthesized and examined. When complexes 12a and 13a with
the respective cyclopentyl and t-butyl esters were used as cata-
(
Entries 4–6). When the complex catalyst 9a or 10a was em-
ployed, the enantiomeric excess of the product was lower than
that using complex catalyst 1a or 7a. On the contrary, in the
reaction catalyzed by complex 11a, the enantiomeric excess of
the product was remarkably improved to 96% ee and the trans-

2
144 Bull. Chem. Soc. Jpn., 74, No. 11 (2001)
Cobalt Catalyzed Asymmetric Cyclopropanation
lysts, the enantiomeric excess was decreased, but the trans-se-
lectivity was enhanced compared to that achieved by the com-
plex catalyst 1a (Entries 7 and 8). These results indicated that
the ester side chain would be effective for improving trans-se-
lectivity. Hence, we designed and synthesized cobalt(Ⅱ) com-
plexes 14a and 15a having ester side chains and an optically
active 1,2-dimesitylethylenediamine unit. In the presence of
these complexes, the cyclopropanation was smoothly complet-
ed and the products were obtained in 96% ee with 90% trans-
selectivity (catalyst 14a, Entry 9) and in 96% ee with 91%
trans-selectivity (catalyst 15a, Entry 10), respectively.
Table 7. Loading Amount of the Cobalt(Ⅱ) Complex Cata-
lyst
a)
Entry
Amount
Yield/% trans:cis ee (trans)/%
b,c)
/
mol%
1
2
3
4
10
5.0
3.0
1.0
95
97
83
54
90:10
90:10
89:11
82:18
97
96
97
95
a) Reaction conditions: catalyst 14a and 0.1 molar
amount (2.0 molar amount vs Co(Ⅱ) complex) of N-
methylimidazole in THF at 40 °C for 3.5–38 h. b)
Mol% means10− molar amount against t-butyl diazo-
acetate.
Examination of Reaction Conditions. Using rationally
designed catalyst 14a, the temperature dependence of the dias-
tereo- and enantioselectivity was studied (Table 5). Whereas
the reaction at 30 °C proceeded slowly (Entry 1), at 40 and 50
2
served under the conditions of a 0.03 molar amount of catalyst
load, its yield was slightly decreased (Entry 3). When 0.01
molar amount of catalyst was employed, the yield and trans-
selectivity of the product were lowered but the ee of the trans-
product was retained (Entry 4). Based on these experiments, it
was noted that loading 0.05 molar amount of the catalyst was
required for achieving a high yield.
Cyclopropanation of Various Diazoacetates and Alkenes.
The cobalt(Ⅱ) complex catalysts 14a and 15a were successful-
ly applied to the enantioselective cyclopropanation of various
diazoacetates with styrene. In the presence of 0.05 molar
amount of cobalt(Ⅱ) complex 14a or 15a, methyl, ethyl, cyclo-
hexyl, and t-butyl diazoacetates (16, 2, 17, and 18, Entries 1–4
in Table 8) were reacted with styrene to afford the correspond-
ing cyclopropanecarboxylates in high yield with high diastere-
oselectivity (81–91%) and very high enantioselectivity (94–
°
C high diastereoselectivity and enantioselectivity were
achieved, and the reaction time was substantially shortened
Entries 2 and 3). It was noteworthy that the high temperature,
(
even the THF reflux temperature, induced high enantioselec-
tivity in the cobalt(Ⅱ) complex 14a-catalyzed cyclopropana-
tion (Entry 4).
The loading amount of styrene versus diazoacetate was then
examined (Table 6). Compared to the result using 5 molar
amounts of styrene, the reaction with 3 molar amounts of sty-
rene produced similar results in yield, trans-selectivity, and
enantiomeric excess (Entries 2 and 3). When an equimolar
amount of styrene was used, the yield was decreased but the
trans-selectivity and enantiomeric excess were maintained
(
Entry 1).
The catalytic amount of the cobalt complex was examined
Table 7). When 0.1 molar amount of the catalyst was used,
(
9
6% ee). It was reported that the high enantiomeric and dias-
the diastereo- and enantioselectivity were not improved (Entry
). Although the diastereo- and enantioselectivity were pre-
tereomeric excesses of the products were generally achieved
1
5,6
using bulky alkyl diazoacetates, such as menthyl, dicyclo-
7d
12,24
hexylmethyl, or 2,6-di-t-butyl-4-methylphenyl groups,
Table 5. Temperature Effect on Trans- and Enantioselectiv-
a)
and that only a few complex catalysts realized highly enanti-
oselective cyclopropanation with methyl or ethyl diazoace-
ity
2
Entry Temp. Time/h Yield/% trans:cis ee (trans)
tate. In the present cyclopropanation, the steric size of the
/
°C
/%
alkyl groups of the diazoacetates had little influence on enanti-
oselection. Even if methyl diazoacetate was used, 94% ee of
the product was achieved by employing the cobalt complex
14a as a catalyst.
Thus, the procedure mentioned above was successfully ap-
plied to various styrene derivatives, such as 4-chloro- and 4-
methoxystyrens, and 2-vinylnaphthalene, to afford the corre-
sponding cyclopropanecarboxylates in high yields with 96%
1
2
3
4
30
40
50
40
4.5
3.5
4.5
66
97
quant
90
90:10
90:10
89:11
85:15
97
96
96
89
reflux
a) Reaction conditions: Five hundredth molar amount of
catalyst 14a and 0.1 molar amount (2.0 molar amount vs
Co(Ⅱ) complex) of N-methylimidazole in THF.
a)
a)
Table 8. Cyclopropanation of Various Diazoacetates
Table 6. Amount of Styrene
ee (trans)
Molar
amount
Entry
Diazoacetate
Yield/% trans:cis
Entry
Yield/% trans:cis ee (trans)/%
b)
/
%
1
2
3
4
quant
96
83:17
81:19
86:14
91:9
94
95
94
96
1
2
3
1
3
5
49
85
97
89:11
89:11
90:10
96
97
96
94
a) Reaction conditions: Five hundredth molar amount
of catalyst 14a and 0.1 molar amount (2.0 molar
amount vs Co(Ⅱ) complex) of N-methylimidazole in
THF at 40 °C for 4.5–9 h. b) Against t-butyl diazo-
acetate.
quant
a) Reaction conditions: Five hundredth molar amount of
catalyst 14a and 0.1 molar amount (2.0 molar amount vs
Co(Ⅱ) complex) of additive in THF at 50 °C for 2–17 h.

T. Ikeno et al.
Bull. Chem. Soc. Jpn., 74, No. 11 (2001) 2145
Table 9. Enantioselective Cyclopropanation of Various Al-
kenes
the diastereo- and enantioselectivities were considerably im-
proved to 83% trans-selectivity and 93% ee. An aqueous alco-
holic solvent was also effective (Entry 6). The cyclopropana-
tion in aqueous methanol (5%(v/v) water) smoothly proceeded
to afford the product in good trans-selectivity with high enanti-
omeric excess. It was recently reported that the addition of
water remarkably decreased the enantiomeric excess of the
product during the cyclopropanation catalyzed by dirhodi-
um(Ⅱ) tetrakis[(S)-1-(4-t-butylphenylsulfonyl)pyrrolidine-2-
a)
b)
Entry
1
Alkene
Yield/% trans:cis ee (trans)/%
93
85
95
47
90:10
87:13
87:13
47:53
96
96
2
3
4
96
2
5
carboxylate] [Rh
of a coordinating solvent, i.e., nitroethane, reduced the diaste-
reo- and enantioselectivity in the Cu(OTf) -bisoxazoline-cata-
2 4
(TBSP) ]. It was also reported that the use
c)
99 (93)
2
2
6
lyzed cyclopropanation. The opposite effect of water and al-
cohol was observed for the present 3-oxobutylideneaminatoco-
a) Reaction conditions: Five handredth molar amount of cat-
alyst 14a and 0.1 molar amount (2.0 molar amount vs Co(Ⅱ)
complex) of N-methylimidazole in THF at 40–50 °C for
2
7
balt catalyzed cyclopropanation reaction.
4
.5–30 h. b) Five molar amount vs t-butyl diazoacetate
These observations were explained as follows: the tetraden-
tate ligand of the 3-oxobutylideneaminatocobalt complex pro-
duced a rigid square planar structure around the cobalt atom,
were employed. c) Ee of the cis-isomer is shown in paren-
thesis.
28
and the structure of the complex was almost independent of
solvent coordination. Therefore, the coordination of a donor
solvent on a vacant axial position would directly lead to acti-
ee (Entries 1–3 in Table 9). 2-Phenylstyrene was also convert-
ed into the corresponding product with 99% ee (for the trans-
product, Entry 4), while the yield and the diastereoselectivity
were poor.
Cobalt(Ⅱ) Complex-Catalyzed Cyclopropanation in Pro-
tic Solvents. Unexpectedly, an interesting solvent-effect was
discovered by reexamination of the solvent in the cobalt(Ⅱ)
complex-14a catalyzed reactions (Table 10). It was pointed
out that the reaction proceeded very slowly in a nonpolar sol-
29
vating the carbene–carbon located at the other axial position.
Among various alcoholic solvents and other polar solvents
studied (Table 11), methanol was the most effective alcohol for
achieving excellent yield and selectivity (Entry 1). In an alco-
hol with a longer alkyl chain, the reaction was slowed down
(
Entries 2–4). The enantioselectivities were affected by the
alkyl size of the alcohol; i.e., the ee decreased in order, metha-
nol (93% ee) > primary alcohol (88–90% ee) > secondary al-
cohol (85–86% ee) > tertiary alcohol (81% ee). As shown in
3
vent without N-methylimidazole, such as CHCl or toluene,
and that the reaction proceeded slightly faster in anhydrous
THF (Table 1 and Table 10). As shown in Entry 8 of Table 3, a
catalytic amount of water did not influence the reaction. Sur-
prisingly, it was found that the reaction was significantly accel-
erated in aqueous THF solvent (5%(v/v) water) (Entry 4). It
should be pointed out here that excess water did not inhibit the
present cyclopropanation but remarkably enhanced the reac-
tion rate to improve the diastereo- and enantioselectivity com-
pared with that in a nonpolar solvent. Methanol was examined
as the solvent instead of aqueous THF (Entry 5), and the reac-
tion proceeded smoothly and was completed in 7 h. Besides,
30
Fig. 3, the plot of the dielectric constant of alcohols vs enan-
tioemeric excesses clearly indicated that the enantioemeric ex-
cess was related to the dielectric constant of the alcohols. Us-
ing an alcohol with a larger dielectric constant produced a
higher enantiomeric excess of the product. These observations
should also be ascribed to the stronger coordination of the al-
cohol to the cobalt atom and to increased electron density on
2 4
the carbene carbon. During the [Rh (TBSP) ]-catalyzed cyclo-
propanation in supercritical and liquid solvent, the enantiose-
lectivity of the product was affected by the dielectric constant
2
5
of the nonpolar solvent. However, the present results clearly
showed the opposite tendency.
Table 10. Various Solvents for Enantioselective Cyclopro-
Examination of other solvents indicated that the polarity of
the solvents was not the only crucial factor for improving the
a)
panation
3
1
present reaction.
Dimethylformamide was also a suitable
solvent (Entry 11), while the product was obtained in low yield
with low ee in AcOEt, acetone, and nitromethane (Entries 8, 9,
and 12) or the reaction resulted in a complex mixture in aceto-
nitrile (Entry 10). Precise study of the reaction mechanism
will be required for clarifying the solvent effects.
The synthetic utilities of the alcoholic solvent for cobalt-cat-
alyzed cyclopropanation were demonstrated. Various mono-
substituted styrene derivatives were converted to the corre-
sponding carboxylates in good-to-high yield with 90–96% ee
Entry
Solvent
CHCl
Toluene
THF
Time Yield trans:cis
ee
/%
b)
/h
/%
1
2
3
4
5
6
3
34
34
10
8
7
7
40
31
39
58
88
89
70:30
72:28
77:23
78:22
83:17
83:17
76
76
87
88
93
92
c)
2
THF + H O
MeOH
MeOH + H
(
Entries 2–4 in Table 12). The superiority of methanol to the
c)
2
O
THF/N-methylimidazole system was observed for the cyclo-
propanation of 1,1-disubstituted ethylenes. Only a 47% yield
of the product for 2-phenylpropene was obtained when the cy-
clopropanation was performed in THF in the presence of a cat-
a) Reaction conditions: Five hundredth molar amount of
catalyst 14a at 50 °C. b) Enantiomeric excesses of trans-
products. c) Five %(v/v) water.

2
146 Bull. Chem. Soc. Jpn., 74, No. 11 (2001)
Cobalt Catalyzed Asymmetric Cyclopropanation
a)
Table 11. Various Alcoholic and Polar Solvents for Enantioselective Cyclopropanation
b)
c)
Entry
Solvent
ε
Time/h
Yield/%
trans:cis
ee/%
1
2
3
4
5
6
7
8
9
MeOH
EtOH
32.6
24.3
20.1
17.1
18.3
15.0
10.9
6.02
20.7
36.2
36.7
38.6
7
88
79
60
66
66
52
62
18
83:17
76:24
79:21
80:20
73:27
73:27
74:26
69:31
69:31
93
90
88
89
86
85
81
60
77
17
26
26
26
26
26
26
26
32
26
26
n-PrOH
n-BuOH
i-PrOH
c-HexOH
t-BuOH
CH
CH
CH
DMF
CH NO
3
COOEt
COCH
CN
3
3
29
complex mixture
quant
10
11
12
3
82:18
62:38
94
51
3
2
16
a) Reaction conditions: Five hundredth molar amount of catalyst 14a at 50 °C.
b) Dielectric constant at 25 °C (see Ref. 30). c) Ee of trans-products.
Fig. 3. The dependence of the optical yield of the product on the dielectric constant of the alcohols.
1
9
alytic amount of N-methylimidazole, although the reaction
smoothly proceeded in methanol and the enantiomeric excess
of the products was very high (Entry 5). Other 1,1-disubstitut-
ed ethylenes, such as 3,4-dihydro-1(2H)methylenenaphthalene
and the 1,1-diphenylethylenes, were also effectively converted
to the products with high enantiomeric excesses (Entries 6–8).
The reaction mechanism involving diastereo- and enantio-
selections of the present cyclopropanation will be described
using a computational method in a later paper.
diamine unit of the complex remarkably influenced the enanti-
oselectivity and its side chain influenced the diastereoselectivi-
ty. The cobalt(Ⅱ) complexes composed of sterically demand-
ing 1,2-dimesitylethylenediamine and ester side chains afford-
ed the cyclopropanecarboxylates in excellent enantio- and high
trans-selectivity (96% ee, trans:cis = 91:9) in the presence of
a catalytic amount of N-methylimidazole. Diazoacetates with
small alkyl groups such as methyl and ethyl esters were tolera-
ble for obtaining the product with high enantiomeric excess
(
94% ee for methyl and 95% ee for ethyl). Although catalytic
Conclusion
loading of water and alcohol had no effect on the reaction, the
cyclopropanation was remarkably accelerated in aqueous or al-
coholic solvents, even without N-methylimidazole, and the di-
astereo- and enantioselectivities were also improved. In aque-
ous methanol, 1,1-disubstituted ethylenes were smoothly cy-
clopropanated with high enantiomeric excesses.
Highly enantioselective cyclopropanation of styrenes was
achieved in the presence of the rationally designed 3-oxobutyl-
ideneaminatocobalt(Ⅱ) complex catalyst. Addition of N-meth-
ylimidazole had a significant effect on enhancing the reactivity
and improving the enantiomeric excess. The steric size of the

T. Ikeno et al.
Bull. Chem. Soc. Jpn., 74, No. 11 (2001) 2147
Table 12. Enantioselective Cyclopropanation of Various Al-
kenes in Aqueous Methanol
Preparation of Optically Active 3-Oxobutylideneaminato
a)
Ligands: The ligands 1b, 7b, and 8b were prepared by the re-
13
ported method from 3-oxo-2-(2,4,6-trimethylbenzoyl)butanal
Entry
1
Alkene
Yield/% trans:cis ee (trans)/%
and (1S,2S)-1,2-diarylethylenediamines.
(
1S,2S)-N,Nꢀ-Bis[3-oxo-2-(2,4,6-trimethylbenzoyl)butyli-
89
83
74
67
88
85
83:17
84:16
77:23
86:14
49:51
28:72
92
96
1
dene]-1,2-diphenylethylenediamine (1b): H NMR (400 MHz)
δ 1.92 (6H, s), 1.98 (6H, s), 2.26 (6H, s), 2.63 (6H, brs), 4.41 (2H,
d, J = 7.2 Hz), 6.72-6.74 (4H, m), 6.98–7.00 (6H, m), 7.24–7.26
2
3
4
5
6
13
(
6H, m), 11.98 (2H, brs); C NMR (100 MHz) δ 19.1, 21.1, 31.8,
91
6
1
1
9.4, 112.6, 126.5, 128.0, 128.2, 128.6, 128.9, 133.3, 133.4,
35.4, 137.4, 161.0, 196.2, 201.1; IR (KBr) 2971, 2917, 1615,
90
−1
454, 1405, 1353, 1299, 1252, 1200, 981, 849, 697 cm . Found:
C, 78.55; H, 6.91; N, 4.39%. Calcd for C42
H
44
28
N
2
O
4
: C, 78.72; H,
b)
b)
97 (51)
96 (84)
6
.92; N, 4.37%. Mp 210.4–211.2 °C. [α]
CHCl ).
1S,2S)-1,2-Bis(3,5-dimethylphenyl)-N,Nꢀ-bis[3-oxo-2-(2,4,
D
+72.3° (c 0.507,
3
(
1
6
-trimethylbenzoyl)butylidene]ethylenediamine (7b):
H
NMR (400 MHz) δ 1.82 (6H, s), 1.99 (6H, s), 2.23–2.25 (18H, m),
7
8
99
95
—
—
90
89
2
.66 (6H, brs), 4.34 (2H, d, J = 6.8 Hz), 6.63 (4H, s), 6.70 (4H, s),
1
3
6
1
1
2
.83–6.95 (4H, m), 11.91 (2H, br); C NMR (100 MHz) δ 19.0,
9.2, 21.1, 21.3, 31.8, 69.5, 112.5, 124.1, 128.0, 130.2, 133.3,
33.4, 135.4, 137.3, 138.6, 161.3, 196.2, 200.9; IR (KBr) 2921,
860, 1620, 1407, 1352, 1307, 1257, 1200, 1169, 980, 849, 802
−
1
+
a) Reaction conditions: Five handredth molar amount of
catalyst 14a in MeOH −5% H O at 50 °C for 7–26 h. b) Ee
of cis-isomer is shown in parenthesis.
cm
.
52 2 4
Calcd for C46H N O : (M ), 696.3927. Found: m/z
2
2
2
6
96.3914. Mp 211.3–213.3 °C. [α]
D 3
+106.3° (c 1.006, CHCl ).
(
1S,2S)-N,Nꢀ-Bis[3-oxo-2-(2,4,6-trimethylbenzoyl)butyli-
1
dene]-1,2-dimesitylethylenediamine (8b): H NMR(400 MHz)
δ 1.42 (6H, s), 2.02 (6H, s), 2.10 (6H, s), 2.18 (6H, s), 2.23 (6H,
brs), 2.28 (6H, s), 2.64 (6H, brs), 4.84 (2H, brs), 6.54 (2H, s), 6.73
Experimental
13
General: The melting points were measured on an Electro-
thermal IA9100 apparatus or a Rigaku DSC-8230 apparatus (dif-
ferential scanning calorimetry, DSC) and are uncorrected.
(2H, s), 6.79 (4H, s), 7.07 (2H, brs), 11.52 (2H, brs); C NMR
(100 MHz) δ 19.01, 19.05, 19.9, 20.4, 20.8, 21.2, 31.6, 61.6,
112.7, 127.7, 128.1, 129.38, 129.44, 131.4, 133.0, 133.6, 134.8,
(
a) Spectrometers: Infrared (IR) spectra were recorded on a
136.7, 137.5, 138.3, 140.7, 161.2, 196.3, 200.2; IR (KBr) 2921,
1624, 1404, 1350, 1305, 1257, 1200, 1171, 983, 849, 820 cm−
1
.
JASCO Model FT/IR-410 infrared spectrometer on KBr pellets or
liquid film on NaCl. H NMR spectra and C NMR spectra were
measured on a JEOL Model GX-400 spectrometer using CDCl as
a solvent and with tetramethylsilane as an internal standard.
High-resolution mass spectra were obtained with a HITACHI M-
013. FAB mass spectra were recorded with a JEOL MStation
1
13
Found: C, 79.27; H, 7.70; N, 3.83. Calcd for C48
H
56
N
2
O
4
: C,
2
4
79.52; H, 7.79; N, 3.86%. Mp 171.4–173.2 °C. [α]
D
+27.5° (c
3
3
1.019, CHCl ).
(1S,2S)-N,Nꢀ-Bis(2-acetyl-3-oxobutylidene)-1,2-diphenyleth-
1
8
ylenediamine (9b): H NMR (400 MHz) δ 2.09 (6H, s), 2.50
JMS-700 spectrometer using 3-nitrobenzyl alcohol (NBA) as a
matrix agent.
(6H, s), 4.57–4.62 (2H, m), 7.11–7.14 (4H, m), 7.30–7.36 (6H,
13
m), 7.55 (2H, d, J = 12.7 Hz), 11.97 (2H, br); C NMR (100
MHz) δ 27.0, 31.9, 70.2, 112.1, 126.8, 128.8, 129.0, 135.4, 158.8,
194.0, 200.9; IR (KBr) 3445, 3029, 2925, 1613, 1398, 1250, 1194,
984, 694, 592 cm− . Found: C, 72.28; H, 6.52; N, 6.48%. Calcd
(
b) Chromatography: For thin-layer chromatography (TLC)
analysis throughout this work, Merck precoated TLC plates (silica
gel 60 F254, 0.25 mm) were used. The products were purified by
preparative column chromatography on silica gel (silica gel 60N).
High-performance liquid chromatography (HPLC) analyses were
performed with a Shimadzu LC-6A chromatograph using an opti-
cally active column (Chiralcel OB-H, OD-H, and Chiralpak AD
columns, Daicel Ltd., Co.); the peak areas were obtained with a
Shimadzu Chromatopack CR-4A or a Varian Dynamax MacInte-
grator.
1
for C26
°C. [α]
H
28
N
2
O
4
: C, 72.20; H, 6.53; N, 6.48%. Mp 209.3–210.3
2
7
D
+199.9° (c 1.036, CHCl ).
3
(1S,2S)-N,Nꢀ-Bis(2-acetyl-3-oxobutylidene)-1,2-bis(3,5-di-
methylphenyl)ethylenediamine (10b): H NMR (400 MHz) δ
2.06 (6H, s), 2.27 (12H, s), 2.50 (6H, s), 4.49–4.54 (2H, m), 6.73
1
(4H, s), 6.95 (2H, s), 7.46 (2H, d, J = 12.2 Hz), 11.80 (2H, br);
1
3
C NMR (100 MHz) δ 21.3, 26.9, 31.9, 70.2, 111.9, 124.6, 130.3,
(
c) Optical Rotations: Optical rotations were measured with
135.3, 138.7, 159.0, 194.0, 200.7; IR (KBr) 3448, 2921, 1632,
1397, 1313, 1245, 984, 851, 601 cm− . Found: C, 73.81; H, 7.38;
1
a JASCO DIP-370 digital polarimeter.
Styrenes: Styrene (3), p-chlorostyrene (19), p-methoxysty-
rene (20), 2-phenylpropene (22), and 1,1-diphenylethylene (24),
were purchased from Tokyo Kasei Kogyo (TCI) Co., Ltd.. 2-Vi-
nylnaphthalene (21) was purchased from Aldrich Co.. 3,4-Dihy-
N, 5.74%. Calcd for C30
H
36
N
2
O
4
: C, 73.74; H, 7.43; N, 5.73%.
27
Mp 217.5–219.3 °C. [α]
D
+196.2° (c 1.041, CHCl ).
3
(1S,2S)-N,Nꢀ-Bis(2-acetyl-3-oxobutylidene)-1,2-dimesityl-
1
ethylenediamine (11b): H NMR (400 MHz) δ 1.74 (6H, brs),
2.22 (12H, s), 2.46 (6H, s), 2.73 (6H, brs), 5.17–5.23 (2H, m),
3
2
dro-1(2H)-methylenenaphthalene (23) and 1,1-bis(4-methylphe-
3
3
nyl)ethylene (25) were each prepared by the reported method.
6.66 (2H, brs), 6.91 (2H, brs), 7.66 (2H, d, J = 12.2 Hz), 11.89
3
4
13
Diazoacetates: Methyl diazoacetate (16), ethyl diazoace-
(2H, br); C NMR (100 MHz) δ 20.3, 20.8, 21.5, 27.1, 31.8, 62.9,
3
4
24
tate (2), cyclohexyl diazoacetate (17) and t-butyl diazoacetate
(
112.5, 129.4, 129.8, 131.4, 138.3, 158.5, 194.1, 200.5; IR (KBr)
2
4
−1
18) were prepared according to the literature.
3239, 2970, 1627, 1576, 1398, 1312, 1248, 1193, 799, 590 cm .

2
148 Bull. Chem. Soc. Jpn., 74, No. 11 (2001)
Cobalt Catalyzed Asymmetric Cyclopropanation
+
+
Calcd for C32
5
H
41
N
2
O
4
: (M+H) , 517.3066.
Found: m/z
+180.7° (c 1.003, CHCl ).
1S,2S)-N,Nꢀ-Bis(2-cyclopentyloxycarbonyl-3-oxobutyli-
(M+H) . Mp 286.7 °C (DSC).
2
5
17.3090. Mp 165.8–167.3 °C. [α]
D
3
(1S,2S)-N,Nꢀ-Bis(2-cyclopentyloxycarbonyl-3-oxobutyli-
dene)-1,2-diphenylethylenediaminatocobalt(Ⅱ) (12a): MS
(FAB ) m/z 630 (M+H) . Mp 262.2 °C (DSC).
(1S,2S)-N,Nꢀ-Bis(2-t-butoxycarbonyl-3-oxobutylidene)-1,2-
diphenylethylenediaminatocobalt(Ⅱ) (13a): MS (FAB ) m/z
606 (M+H) . Mp 263.6 °C (DSC).
(1S,2S)-N,Nꢀ-Bis(2-cyclopentyloxycarbonyl-3-oxobutyli-
dene)-1,2-dimesitylethylenediaminatocobalt(Ⅱ) (14a): MS
(FAB ) m/z 714 (M+H) . Mp 249.3 °C (DSC).
(1S,2S)-N,Nꢀ-Bis(2-t-butoxycarbonyl-3-oxobutylidene)-1,2-
(
1
+
+
dene)-1,2-diphenylethylenediamine (12b): H NMR(400 MHz)
δ 1.55–1.72 (12H, m), 1.81–1.92 (4H, m), 2.48 (6H, s), 4.59–4.63
2H, m), 5.17–5.21 (2H, m), 7.02–7.09 (4H, m), 7.25–7.33 (6H,
m), 7.78 (2H, d, J = 12.7 Hz), 11.90 (2H, br); C NMR (100
MHz) δ 23.8, 30.9, 32.7, 69.5, 76.2, 101.6, 126.9, 128.5, 128.7,
35.4, 158.8, 166.2, 199.5; IR (KBr) 3446, 2964, 1697, 1632,
+
(
1
3
+
1
1
−
1
+
+
573, 1415, 1299, 1240, 1056, 699, 578 cm
.
Calcd for
+
C
34
H
41
2
N O
6
: (M+H) , 573.2965. Found: m/z 573.2981. Mp
+77.64° (c 1.138, CHCl
1S,2S)-N,Nꢀ-Bis(2-t-butyloxycarbonyl-3-oxobutylidene)-
,2-diphenylethylenediamine (13b): H NMR (400 MHz) δ
.46 (18H, s), 2.47 (6H, s), 4.55–4.60 (2H, m), 7.04–7.06 (4H, m),
.27–7.29 (6H, m), 7.77 (2H, d, J = 12.7 Hz), 11.85 (2H, br);
NMR (100 MHz) δ 28.5, 31.2, 69.7, 79.8, 102.7, 127.0, 128.6,
28.8, 135.6, 158.9, 166.1, 199.7; IR (KBr) 3442, 3008, 2978,
688, 1625, 1391, 1245, 1164, 1064, 802, 692 cm . Calcd for
: (M ), 548.2886. Found: m/z 548.2902. Mp 143.4–
+100.5° (c 1.009, CHCl ).
D 3
1S,2S)-N,Nꢀ-Bis(2-cyclopentyloxycarbonyl-3-oxobutyli-
dene)-1,2-dimesitylethylenediamine (14b):
2
5
+
7
4.6–75.7 °C. [α]
(
D
3
).
dimesitylethylenediaminatocobalt(Ⅱ) (15a): MS (FAB ) m/z
690 (M+H) . Mp 227.6 °C (DSC).
+
1
1
1
7
General Procedure for the Enantioselective Cyclopropana-
tion Reaction to t-butyl (1R,2S)-cis- and (1S,2S)-trans-2-phen-
ylcyclopropane-1-carboxylate (26 and 27): To a solution of
the cobalt complex 1a (35.0 mg, 0.05 mmol) in THF (1.0 mL) was
added styrene (0.57 mL, 5.0 mmol) and t-butyl diazoacetate (148
13
C
1
1
−
1
2
µL, 1.0 mmol) at 25 °C under a N atmosphere, and then N-meth-
+
C
32
H
40
N
2
O
6
ylimidazole (8 µL, 0.1 mmol) was added to the solution. GC anal-
ysis after 2 h indicated that the diazoacetate was completely con-
sumed and that the corresponding t-butyl 2-phenylcyclopropane-
1-carboxylate was produced in a 76:24 ratio mixture of trans and
cis isomers. The products, t-butyl cis- and trans-2-phenylcyclo-
2
4
1
44.4 °C. [α]
(
1
H
NMR (400
MHz) δ 1.54–1.78 (16H, m), 1.81–1.90 (6H, m), 2.21 (6H, s),
.43 (6H, s), 2.65 (6H, brs), 5.17–5.26 (4H, m), 6.63 (2H, brs),
.88 (2H, brs), 7.89 (2H, d, J = 12.7 Hz), 11.79 (2H, br);
36
11b
2
6
propane-1-carboxylate (26 and 27 ) and cis- and trans-2-
phenylcyclopropylmethanol (28 and 29) were identified with the
13
37
C
NMR (100 MHz) δ 20.1, 20.7, 21.1, 23.7, 30.8, 32.8, 62.1, 76.3,
corresponding materials reported previously. The reaction mix-
ture was purified with silica-gel column chromatography to obtain
the product in 98% yield. The mixture of cis and trans 2-phenyl-
1
3
01.9, 129.5, 130.1, 131.3, 137.9, 158.7, 166.5, 198.8; IR (KBr)
480, 2965, 2871, 1700, 1632, 1574, 1415, 1240, 1055, 794 cm−
: C,
+48.51°
1
.
Found: C, 73.05; H, 8.06; N, 4.12%. Calcd for C40
H
52
N
2
O
6
4
cyclopropane-1-carboxylate was treated with LiAlH to afford the
2
4
7
(
3.14; H, 7.98; N, 4.26%. Mp 73.1–76.1 °C. [α]
c 1.022, CH Cl ).
1S,2S)-N,Nꢀ-Bis(2-t-butoxycarbonyl-3-oxobutylidene)-1,2-
D
corresponding mixture of cis- and trans-2-phenylcyclopropyl-
methanol (28 and 29) and then HPLC analysis of the reduced mix-
ture revealed that the trans isomer was 84% ee (Daicel Chiralcel
OD-H, 5% 2-propanol in hexane, flow 1.0 mL/min, 12.9 min (ma-
jor), 17.9 min (minor)) and that the cis isomer was 85% ee (Daicel
Chiralcel OB-H, 5% 2-propanol in hexane, flow 1.0 mL/min, 7.2
min (minor), 8.0 min (major)).
2
2
(
1
dimesitylethylenediamine (15b): H NMR (400 MHz) δ 1.50
18H, s), 1.67 (6H, br), 2.20 (6H, s), 2.42 (6H, s), 2.67 (6H, brs),
(
5
1
2
1
1
.17–5.19 (2H, m), 6.61 (2H, brs), 6.87 (2H, brs), 7.89 (2H, d, J =
1
3
2.7 Hz), 11.75 (2H, br); C NMR (100 MHz) δ 20.1, 20.8, 21.2,
8.5, 31.2, 62.2, 79.8, 120.8, 129.5, 130.3, 131.3, 137.9, 159.0,
66.3, 198.9; IR (KBr) 3423, 2975, 2929, 1701, 1632, 1573, 1415,
2-Phenylcyclopropane-1-carboxylate derivatives: t-Butyl
36
cis-2-(4-chlorophenyl)cyclopropane-1-carboxylate (30), t-butyl
−
1
11b
366, 1245, 1167, 1058, 792 cm
.
Calcd for C38
H
53
N
2
O
6
:
trans-2-(4-chlorophenyl)cyclopropane-1-carboxylate (31), t-bu-
+
36
(
M+H) , 633.3904. Found: m/z 633.3945. Mp 74.2–75.4 °C.
tyl cis-2-(4-methoxyphenyl)cyclopropane-1-carboxylate (32),
2
7
[
α]
Preparation of Optically Active 3-Oxobutylideneaminatoco-
balt(Ⅱ) Complexes: Cobalt(Ⅱ) complexes 1a, 7a, and 8a were
D
+52.82° (c 1.059, CHCl
3
).
t-butyl
trans-2-(4-methoxyphenyl)cyclopropane-1-carboxylate
t-butyl cis-2-(2-naphthyl)cyclopropane-1-carboxylate
t-butyl trans-2-(2-naphthyl)cyclopropane-1-carboxylate
(35), t-butyl (1R*,2S*)-cis-2-methyl-2-phenylcyclopropane-1-
8
(33),
38
(34),
2
8
38
prepared by the reported method.
1S,2S)-N,Nꢀ-Bis[3-oxo-2-(2,4,6-trimethylbenzoyl)butyli-
dene]-1,2-diphenylethylenediaminatocobalt(Ⅱ) (1a): MS
36
(
carboxylate (36), t-butyl (1R*,2R*)-trans-2-methyl-2-phenylcy-
36
clopropane-1-carboxylate (37), t-butyl 2,2-diphenylcyclopro-
+
+
39
(
FAB ) m/z 698 (M+H) . Mp 299.2 °C (DSC).
1S,2S)-1,2-Bis(3,5-dimethylphenyl)-N,Nꢀ-bis[3-oxo-2-
2,4,6-trimethylbenzoyl)]ethylenediaminato]cobalt(Ⅱ) (7a):
pane-1-carboxylate (38), ethyl cis-2-phenylcyclopropane-1-car-
6b
(
boxylate (39), ethyl trans-2-phenylcyclopropane-1-carboxylate
6
b
40
(
(40), methyl cis-2-phenylcyclopropane-1-carboxylate (41), and
+
+
40
MS (FAB ) m/z 754 (M+H) . Mp 334.1 °C (DSC).
1S,2S)-N,Nꢀ-Bis[3-oxo-2-(2,4,6-trimethylbenzoyl)butyli-
dene]-1,2-dimesitylethylenediaminatocobalt(Ⅱ) (8a): MS
methyl trans-2-phenylcyclopropane-1-carboxylate (42) were
(
identified with the corresponding material reported previously.
t-Butyl (1R*,2R*)-spiro[cyclopropane-1,1ꢀ(2ꢀH)-3ꢀ,4ꢀ-dihy-
+
+
1
(
FAB ) m/z 782 (M+H) . Mp 248.8 °C (DSC).
1S,2S)-N,Nꢀ-Bis(2-acetyl-3-oxobutylidene)-1,2-diphenyl-
dronaphthalene]-2-carboxylate (43): H NMR (400 MHz) δ
(
1.45 (9H, s), 1.45–1.48 (1H, m), 1.52 (1H, dd, J = 5.1, 8.1 Hz),
+
ethylenediaminatocobalt(Ⅱ) (9a): MS (FAB ) m/z 490 (M+
H) . Mp 293.5 °C (DSC).
1.70–1.80 (1H, m), 1.83–2.00 (4H, m), 2.87–2.90 (2H, m), 6.70–
+
13
6.73 (1H, m), 7.08–7.13 (3H, m); C NMR (100 MHz) δ 21.8,
(
1S,2S)-N,Nꢀ-Bis(2-acetyl-3-oxobutylidene)-1,2-bis(3,5-di-
22.3, 27.6, 28.3, 28.6, 30.6, 33.5, 80.3, 121.7, 125.5, 126.1, 128.9,
137.7, 139.8, 170.3; IR (neat) 2977, 2929, 1715, 1455, 1392,
+
methylphenyl)ethylenediaminatocobalt(Ⅱ) (10a): MS (FAB )
m/z 546 (M+H) . Mp 268.6 °C (DSC).
+
−1
1366, 1147, 841, 760, 742 cm . Found: C, 79.05; H, 8.62%.
2
8
(
1S,2S)-N,Nꢀ-Bis(2-acetyl-3-oxobutylidene)-1,2-dimesityl-
Calcd for C17
H
22
O
2
: C, 79.03; H, 8.58%. [α]
D
+226° (c 0.251,
3
5
+
ethylenediaminatocobalt(Ⅱ) (11a):
MS (FAB ) m/z 574
CHCl ). Enantiomeric excess was determined by Chiralcel OB-H
3

T. Ikeno et al.
Bull. Chem. Soc. Jpn., 74, No. 11 (2001) 2149
(
1
2% 2-propanol in hexane, flow 1.0 mL/min), 12.3 min (major),
3.5 min (minor), after the ester 43 was converted to the corre-
sponding alcohol by LiAlH
and M. Scott, Angew. Chem., Int. Ed. Engl., 31, 430 (1992).
8
9
K. Ito and T. Katsuki, Tetrahedron Lett., 34, 2661 (1993).
a) H. Nishiyama, Y. Itoh, H. Matsumoto, S.-B. Park, and K.
4
.
t-Butyl (1R*,2S*)-Spiro[cyclopropane-1,1ꢀ(2ꢀH)-3ꢀ,4ꢀ-dihy-
dronaphthalene]-2-carboxylate (44): H NMR (400 MHz) δ
.08 (9H, s), 1.11–1.14 (1H, m), 1.20–1.22 (1H, m), 1.78 (1H, dd,
Itoh, J. Am. Chem. Soc., 116, 2223 (1994). b) H. Nishiyama, Y.
Itoh, Y. Sugawara, H. Matsumoto, K. Aoki, and K. Itoh, Bull.
Chem. Soc. Jpn., 68, 1247 (1995).
1
1
J = 5.9, 7.6 Hz), 1.83–1.91 (1H, m), 2.00–2.03 (1H, m), 2.06–
10 a) M. P. Doyle, B. D. Brandes, A. P. Kazala, R. J. Poeters,
M. B. Jarstfer, L. M. Watkins, and C. T. Eagle, Tetrahedron Lett.,
31, 6613 (1990). b) M. P. Doyle, W. R. Winchester, J. A. A.
Hoorn, V. Lynch, S. H. Simonsen, and R. Ghosh, J. Am. Chem.
Soc., 115, 9968 (1993). c) H. M. L. Davies, P. R. Bruzinski, D. H.
Lake, N. Kong, and M. J. Fall, J. Am. Chem. Soc., 118, 6897
(1996). d) H. M. L. Davies, D. G. Stafford, B. D. Doan, and J. H.
Houser, J. Am. Chem. Soc., 120, 3326 (1998).
11 a) T. Fukuda and T. Katsuki, Synlett, 1995, 825. b) T.
Fukuda and T. Katsuki, Tetrahedron, 53, 7201 (1997).
12 M. M.-C. Lo and G. C. Fu, J. Am. Chem. Soc., 120, 10270
(1998).
13 T. Nagata, K. Imagawa, T. Yamada, and T. Mukaiyama,
Bull. Chem. Soc. Jpn., 68, 1455 (1995).
14 T. Nagata, K. Imagawa, T. Yamada, and T. Mukaiyama,
Bull. Chem. Soc. Jpn., 68, 3241 (1995).
15 T. Nagata, K. Yorozu, T. Yamada, and T. Mukaiyama, An-
gew. Chem., Int. Ed. Engl., 34, 2145 (1995).
16 K. D. Sugi, T. Nagata, T. Yamada, and T. Mukaiyama,
Chem. Lett., 1997, 493.
1
3
2
.16 (2H, m), 2.80–2.96 (2H, m), 7.08 (4H, m); C NMR (100
MHz) δ 16.1, 21.9, 27.8, 29.3, 31.0, 33.5, 35.1, 79.8, 124.6, 125.7,
1
1
C
26.1, 127.7, 135.2, 139.4, 169.2; IR (neat) 2976, 2931, 1725,
−
1
456, 1391, 1366, 1232, 1148, 847, 789, 741 cm . Calcd for
+
28
H O
17 22 2
: (M ), 258.1620. Found: m/z 258.1620. [α]
D
+72.5°
3
(c 0.306, CHCl ). Enantiomeric excess was determined by Chiral-
cel OD-H (2% 2-propanol in hexane, flow 1.0 mL/min), 11.6 min
minor), 13.1 min (major), after the ester 44 was converted to the
corresponding alcohol by LiAlH
(
4
.
t-Butyl 2,2-bis(4-methylphenyl)cyclopropane-1-carboxylate
1
(
8
2
45): H NMR (400 MHz) δ 1.24 (9H, s), 1.47 (1H, dd, J = 4.9,
.1 Hz), 2.04 (1H, dd, J = 4.9, 5.9 Hz), 2.27 (3H, s), 2.29 (3H, s),
.40 (1H, dd, J = 5.9, 8.1 Hz), 7.03–7.07 (4H, m), 7.10–7.12 (2H,
m), 7.21–7.23 (2H, m); CNMR (100 MHz) δ 20.2, 21.0, 21.2,
7.9, 30.1, 38.9, 80.2, 127.2, 128.7, 128.9, 129.7, 135.7, 136.1,
37.3, 142.4, 169.7; IR (neat) 2977, 2925, 1729, 1515, 1366,
150, 1090, 849, 811, 550 cm− . Found: C, 81.84; H, 8.02%.
1
3
2
1
1
1
2
5
Calcd for C22
H
26
O
2
: C, 81.95; H, 8.13%. [α]
D
+161° (c 0.622,
CHCl
3
). Enantiomeric excess was determined by Chiralcel OD-H
(
1
5% 2-propanol in hexane, flow 1.0 mL/min), 7.9 min (minor),
0.2 min (major), after the ester 45 was converted to the corre-
17 T. Yamada, Y. Ohtsuka, and T. Ikeno, Chem. Lett., 1998,
1129.
sponding alcohol by LiAlH
4
.
18 a) T. Yamada, S. Kezuka, T. Mita, and T. Ikeno, Heterocy-
cles, 52, 1041 (2000). b) S. Kezuka, T. Mita, N. Ohtsuki, T. Ikeno,
and T. Yamada, Chem. Lett., 2000, 824. c) S. Kezuka, T. Mita, N.
Ohtsuki, T. Ikeno, and T. Yamada, Bull. Chem. Soc. Jpn., 74, 1333
(2001).
References
1
a) H. Nozaki, S. Moriuti, H. Takaya, and R. Noyori, Tetra-
hedron Lett., 1966, 5239. b) H. Nozaki, H. Takaya, S. Moriuti,
and R. Noyori, Tetrahedron, 24, 3655 (1968).
19 a) T. Yamada, T. Ikeno, H. Sekino, and M. Sato, Chem.
Lett., 1999, 719. b) T. Ikeno, M. Sato, and T. Yamada, Chem.
Lett., 1999, 1345.
2
a) M. P. Doyle, M. A. McKervey, and T. Ye, “Modern Cata-
lytic Methods for Organic Synthesis with Diazo Compounds,”
John Wiley & Sons, New York (1998). b) M. P. Doyle and M. N.
Protopopova, Tetrahedron, 54, 7919 (1998). c) M. P. Doyle and
D. C. Forbes, Chem. Rev., 98, 911 (1998).
20 T. Ikeno, A. Nishizuka, M. Sato, and T. Yamada, Synlett,
2001, 406.
21 For examples: a) M. Calligaris, G. Nardin, L. Randaccio,
and G. Tauzher, Inorg. Nucl. Chem. Lett., 9, 419 (1973). b) M.
Calligaris, G. Nardin, and L. Randaccio, Inorg. Nucl. Chem. Lett.,
8, 477 (1972). c) M. Calligaris, D. Minichelli, G. Nardin, and L.
Randaccio, J. Chem. Soc., A, 1970, 2411. d) R. S. Gall, J. F.
Rogers, W. P. Schaefer, and G. G. Christoph, J. Am. Chem. Soc.,
98, 5135 (1976). e) W. P. Schaefer, R. Waltzman, and B. T. Huie,
J. Am. Chem. Soc., 100, 5062 (1978).
22 For examples: a) P. Battioni, J. P. Renaud, J. F. Bartoli, M.
Reina-Artiles, and D. Mansuy, J. Am. Chem. Soc., 110, 8462
(1988). b) P. L. Anelli, S. Banfi, F. Montanari, and S. Quici, J.
Chem. Soc., Chem. Commun., 1989, 779. c) R. Irie, Y. Ito, and T.
Katsuki, Synlett, 1991, 265. d) T. Schwenkreis and A. Berkessel,
Tetrahedron Lett., 34, 4785 (1993). e) T. Yamada, K. Imagawa, T.
Nagata, and T. Mukaiyama, Bull. Chem. Soc. Jpn., 67, 2248
(1994).
23 When N-methylimidazole was added to a solution of the
cobalt(Ⅱ) complex catalyst and diazoacetate, the color of the solu-
tion immediately turned brown. Whereas, when 2-methylimida-
zole was added to the solution, the color changed little.
24 M. P. Doyle, V. Bagheri, T. J. Wandless, N. K. Harn, D. A.
Brinker, C. T. Eagle, and K.-L. Loh, J. Am. Chem. Soc., 112, 1906
(1990).
3
4
T. Aratani, Pure Appl. Chem., 57, 1839 (1985).
a) A. Nakamura, A. Konishi, Y. Tatsuno, and S. Otsuka, J.
Am. Chem. Soc., 100, 3443, 6544 (1978). b) A. Nakamura, A.
Konishi, R. Tsujitani, M. Kudo, and S. Otsuka, J. Am. Chem. Soc.,
1
00, 3449 (1978). c) A. Nakamura, Pure Appl. Chem., 50, 37
(
1978).
5
a) T. Aratani, Y. Yoneyoshi, and T. Nagase, Tetrahedron
Lett., 1975, 1707. b) T. Aratani, Y. Yoneyoshi, and T. Nagase, Tet-
rahedron Lett., 1977, 2599. c) T. Aratani, Y. Yoneyoshi, and T.
Nagase, Tetrahedron Lett., 1982, 685.
6
a) H. Fritschi, U. Leutenegger, and A. Pfaltz, Angew.
Chem., Int. Ed. Engl., 25, 1005 (1986). b) H. Fritschi, U.
Leutenegger, and A. Pfaltz, Helv. Chim. Acta, 71, 1553 (1988). c)
U. Leutenegger, G. Umbricht, C. Fahrni, and P. von Matt, and A.
Pfaltz, Tetrahedron, 48, 2143 (1992). d) A. Pfaltz, Acc. Chem.
Res., 26, 339 (1993).
7
a) R. E. Lowenthal, A. Abiko, and S. Masamune, Tetrahe-
dron Lett., 31, 6005 (1990). b) D. Müller, G. Umbricht, B. Weber,
and A. Pfaltz, Helv. Chim. Acta, 74, 232 (1991). c) D. A. Evans,
K. A. Woerpel, M. M. Hinman, and M. M. Faul, J. Am. Chem.
Soc., 113, 726 (1991). d) R. E. Lowenthal and S. Masamune, Tet-
rahedron Lett., 32, 7373 (1991). e) D. A. Evans, K. A. Woerpel,
25 D. C. Wynne, M. M. Olmstead, and P. G. Jessop, J. Am.

2
150 Bull. Chem. Soc. Jpn., 74, No. 11 (2001)
Cobalt Catalyzed Asymmetric Cyclopropanation
Chem. Soc., 122, 7638 (2000).
J. M. Fraile, J. I García, J. A.Mayora, and T. Tarnai, J. Mol.
Cat. A: Chem., 144, 85 (1999).
A similar effect of protic solvents was independently re-
ported in Ru-pybox catalyzed cyclopropanation. S. Iwasa, F.
Takezawa, Y. Tuchiya, and H. Nishiyama, Chem. Commun., 2001,
correlation between the dielectric constant and stereoselectivities
could not be observed among different types of solvents. On the
contrary, these factors should be similar or have a relation with the
dielectric constant in a series of alcohols. Therefore, the correla-
tion between the dielectric constant and enantioselectivity could
be observed.
2
6
2
7
5
9.
2
32 M. M. Campbell, N. Abbas, and M. Sainsbury, Tetrahe-
dron, 41, 5637 (1985).
8
T. Yamada, T. Nagata, T. Ikeno, Y. Ohtsuka, A. Sagara, and
T. Mukaiyama, Inorg. Chim. Acta, 296, 86 (1999).
As an alternative possibility of the acceleration effect of the
33 C. F. H. Allen and S. Converse, Org. Synth., Coll. Vol. 1,
226 (1941).
2
9
axial ligand, it should be considered that the transformation of the
diazoacetate into the cobalt carbene complex could be accelerated
by the coordination of N-methylimidazole to the cobalt(Ⅱ) com-
plex.
34 E. B. Womack and A. B. Nelson, Org. Synth., Coll. Vol. 3,
392 (1955).
35 S. Ohba, T. Nagata, and T. Yamada, Acta Crystallogr., Sect.
E, E57, m124 (2001).
3
0
A. J. Gordon, and A. F. Ford, “The Chemist’s Companion
36 T. Uchida, R. Irie, and T. Katsuki, Tetrahedron, 56, 3501
(2000).
37 B. Scholl and H.-J. Hansen, Helv. Chim. Acta, 69, 1936
(1986).
A Handbook of Practical Data, Techniques, and References,” John
Wiley & Sons, New York (1972).
3
1
Not only the polarity of the solvents, but the other factors,
such as the donating and accepting nature of the solvent depended
on its heteroatom and/or functional group of coordination, the en-
vironment around the complex influenced by the solvent, and the
side reactions between the solvent and the diazoester or between
the solvent and the cobalt carbene complex, would play important
roles for the enhancement of the reaction rate and the improve-
ment of the stereoselectivities. Therefore, it is reasonable that the
38 P. Müller, C. Baud, D. Ené, S. Motallebi, M. P. Doyle, B.
D. Brandes, A. B. Dyatkin, and M. M. See, Helv. Chim. Acta, 78,
459 (1995).
39 A. V. Bedekar, E. B. Koroleva, and P. G. Andersson, J. Org.
Chem., 62, 2518 (1997).
40 P. E. Krieger and J. A. Landgrebe, J. Org. Chem., 43, 4447
(1978).