Chloroform is not solvent but activator for cobalt complex catalyst of enantioselective borohydride reduction

Article abstract of DOI:10.1246/cl.2007.26

For the enantioselective borohydride reduction of carbonyl compounds catalyzed by the optically active ketoiminatocobalt complexes, chloroform has been employed as a unique solvent for achieving a high enantioselectivity, whereas it was found that a catal

Full text of DOI:10.1246/cl.2007.26

26
Chemistry Letters Vol.36, No.1 (2007)
Chloroform Is Not Solvent but Activator for Cobalt Complex Catalyst
of Enantioselective Borohydride Reduction
Ai Kokura,1 Saiko Tanaka,1 Haruna Teraoka,2 Atsushi Shibahara,2 Taketo Ikeno,1
Takushi Nagata,2 and Tohru YamadaÃ1
1Department of Chemistry, Faculty of Science and Technology, Keio University, Hiyoshi, Kohoku-ku, Yokohama 223-8522
2Catalysis Science Laboratory, Mitsui Chemicals, Inc., Nagaura, Sodegaura, Chiba 299-0265
(Received September 7, 2006; CL-061036; E-mail: yamada@chem.keio.ac.jp)
For the enantioselective borohydride reduction of carbonyl
vert various ketones into the corresponding reduced product with
a high enantioselectivity.
compounds catalyzed by the optically active ketoiminatocobalt
complexes, chloroform has been employed as a unique solvent
for achieving a high enantioselectivity, whereas it was found that
a catalytic amount of chloroform effectively activated the pres-
ent catalytic system to convert various ketones into the corre-
sponding reduced product with a high ee in the THF solvent.
The loading amount of chloroform was examined in detail
(Table 1). Although in pure THF solvent, the reduced product
was obtained with 41% ee, in the presence of 200 equiv. of
chloroform vs the cobalt complex, the ee of the reduced product
was increased to 82% ee. The remarkable effect of the additive
was observed under the conditions of 100, 25, and 5 equiv. of
chloroform vs the cobalt complex in THF as the mother solvent,
to afford the products with 84, 85, and 85% ee, respectively.
Finally, it was found that only 1.5 equiv. of chloroform vs the
cobalt complex effectively maintained the enantioselectivity at
81% ee (Entry 4). These results revealed that chloroform was
Previously, it was reported by our laboratory that the enan-
tioselective borohydride reduction was efficiently catalyzed by
the optically active ketoiminatocobalt(II) complexes to convert
a wide variety of ketones into the corresponding optically active
alcohols in high to excellent yields and enantioselectivities.1 The
drawback in the present reduction system has been pointed out
that chloroform is a unique solvent that achieves the high effi-
ciency and high enantioselectivity. For example, the catalytic
enantioselective reduction of valerophenone in a pure THF
solvent afforded the reduced product with 41% ee, while with
91% ee in the pure chloroform solvent. Due to the troublesome
regulation regarding on their waste, halocarbon solvents includ-
ing chloroform are apt to be prevented from being employed as a
solvent in practical applications and non-halogenated solvents
have been desired to be employed for the catalytic enantioselec-
tive borohydride reduction. During the detailed examination of
the reaction system, a small amount of dichloromethane was de-
tected by GC-MS analysis after the reaction was completed. This
observation suggested that chloroform could react with the hy-
dride in this reduction system2 as well as be employed as the
suitable solvent. The key reactive intermediate of the borohy-
dride reduction catalyzed by the cobalt complexes was recently
proposed to be the dichloromethylcobalthydride with a sodium
cation (Figure 1) based on both experimental and theoretical
studies.3 It was revealed that chloroform was not the solvent,
but the reactant that activates the cobalt complex catalyst. In this
letter, we report that a catalytic amount of chloroform effectively
activated the catalytic reduction system in THF solvent to con-
Table 1. The loading amount of chloroform
OH
O
1 mol % cobalt(II) complex 1a
NaBH4, EtOH,
HO
O
THF, 0 °C
Entrya
Amount of CHCl3/equiv.b
Yield/%c
ee/% eed
1
2
3
4
5
solvent
25
80
97
83
80
74
91
85
85
81
41
5
1.5
none
aReaction conditions: 0.50 mmol substrate, 0.005 mmol Co catalyst,
and 0.75 mmol modified NaBH4 (0.75 mmol NaBH4, 0.75 mmol EtOH,
and 10.3 mmol THFA) in THF 8 mL. Reaction time was 15 h.
bEquivalent vs cobalt complex. cIsolated yield. dDetermined by HPLC
analysis (Chiralpak OD-H).
Table 2. Various halocarbons as activators of Co complex
1 mol %cobalt(II) complex 1a
OH
O
5 mol % activator
NaBH4, EtOH,HO
THF, -20 °C
O
Entrya
Activator
Yield/%
ee/% ee
1
CCl4
CBr4
90
87
98
93
95
91
87
89
93
53
47
51
64
43
81
80
88
91
2
H
H
3
CHBr3
CHI3
O
H
4
1.993
5
ICH2Cl
CBrCl3
CH2Cl2
CHCl3
N
1.586
O
2.267
2.180
Na+
6
Co
N
N
N
O
7
O
2.180
Co
8b
9c
3.026
O
O
O
Cl
C
(R,R)-Co complex 1a
aReaction conditions: 0.50 mmol substrate, 0.005 mmol Co cata-
lyst, and 0.75 mmol modified NaBH4 (0.75 mmol NaBH4, 0.75
mmol EtOH, and 10.3 mmol THFA) in THF 8 mL. Reaction time
H
diatance Å
Cl
Figure 1. The cobalt complex catalyst and the proposed TS of cobalt-
catalyzed borohydride reduction.
c
was 15 h. bTHF (13 mL) was used. 2 mol % cat. was used.
Copyright Ó 2007 The Chemical Society of Japan

Chemistry Letters Vol.36, No.1 (2007)
27
CH3
not only the appropriate reaction solvent, but also an effective
activator of the ketoiminatocobalt complex to realize a highly
enantioselective catalytic reduction with sodium borohydride.
For the following investigations, the conditions of adding 5
equiv. of chloroform were adopted.
Ar
Ar
complex 1a : Ar =
complex 1b : Ar =
R =
R =
CH3
H3C
H3C
CH3
CH3
CH3
N
N
O
R
O
R
O
Co
O
H3C
CH3
Various halocarbons other than chloroform were examined
as the activator for the catalytic enantioselective reduction
(Table 2). In the presence of tetrahalocarbons, CCl4 and CBr4,
the reduction smoothly proceeded, but their enantioselectivities
were 53 and 47% ee, respectively (Entries 1 and 2). The halo-
forms, CHBr3 and CHI3, were not effective activators to im-
prove the enantioselectivities (Entries 3 and 4). Although chlo-
roiodomethane was not effective (Entry 5), bromotrichloro-
methane improved the ee of the reduced product (Entry 6).
Dichloromethane (Entry 7) was also effective for improving
the product ee as did chloroform (Entry 8). After optimization
of the reaction conditions, the catalytic enantioselective reduc-
tion in THF afforded a product with a 91% ee (Entry 9), which
was the same as that obtained in pure chloroform solvent. The
examination suggested that halogenated methanes containing
more than two chlorine atoms were effective activators, and
these observations were quite compatible with the proposed
structure in Figure 1.
Thus, the optimized procedure was successfully applied to
the enantioselective borohydride reduction catalyzed by the op-
tically active cobalt(II) complexes (Figure 2) in THF solvent. In
the presence of 5 mol % of chloroform and 1 mol % of the cobalt
complex, various ketones were subjected to the enantioselective
borohydride reduction (Table 3). For the isopropyl phenyl ke-
tone (Entry 1), cyclohexyl phenyl ketone (Entry 2), and cyclo-
propyl phenyl ketone (Entry 3) were converted to the corre-
sponding alcohols in high yields with higher than 90% ee. Tetra-
lone (Entry 4) and 2,2-dimethylchromanone (Entry 5) were also
reduced into the corresponding alcohols with 85 and 82% ee,
respectively. This reduction system can be applied for the
enantioselective reduction of ortho-fluorinated benzophenones4
(Entry 6) and diphenylphosphinylimine derivatives (Entries 7
and 8). The diaryloylmethanes were converted to the corre-
sponding diols with a ca. 3:1 dl selectivity and high enantioselec-
tivity (Entry 9). The 2-alkyl substituted 1,3-diketones were
reduced to selectively afford the anti-alcohol derivatives and
their ee’s were excellent (Entries 10–13). For the enantioselec-
tive reduction along with a dynamic kinetic resolution, this re-
duction system was effective (Entry 14).
Figure 2. The structure of the cobalt complexes.
Table 3. Catalytic enantioselective borohydride reduction in THF
1 mol %cobalt(II) complex
O
OH
5 mol % CHCl3
R
R
NaBH4, EtOH,
THF, -20 °C
HO
O
Entrya,b
1
Reduced product Time/h
Yield/%i ee/%eej
OH
15
80
91
OH
2
15
89
95
92
90
OH
3c
18
OH
4c
5
15
87
85k
82k
OH
quant
9
O
OH
F
77k
85
6
25
15
10
97
88
96
NHP(O)Ph2
NHP(O)Ph2
7c,d
8
95
H3CO
OH OH
9e,f
40
16
77
92
dl:meso = 76:24
OH O
OH O
10e,g
95k
84
MeO
OMe
87k
11e,g
12e,g
39
16
81
87
Ph
OH O
It is noted that a catalytic amount of chloroform effectively
activated the cobalt-catalyzed borohydride reduction system that
converts various ketones into the corresponding reduced product
with a high ee in THF solvent.
96
OH O
13e,g
14h
15
60
90
91
94k
82k
Dedicated to Prof. Teruaki Mukaiyama on the occasion of
his 80th birthday.
OH O
OEt
dynamic kinetic resolution
References
aReaction conditions: 0.50 mmol substrate, 0.005 mmol Co catalyst, and
0.75 mmol modified borohydride (0.75 mmol NaBH4, 0.75 mmol EtOH,
and 10.3 mmol THFA) in THF 8 mL. bCobalt complex 1a was employed
1
T. Nagata, K. Yorozu, T. Yamada, T. Mukaiyama, Angew. Chem., Int.
Ed. Engl. 1995, 34, 2145; T. Yamada, T. Nagata, K. D. Sugi, K. Yorozu,
T. Ikeno, Y. Ohtsuka, D. Miyazaki, T. Mukaiyama, Chem. Eur. J. 2003,
9, 4485.
Q. Chen, L. G. Marzilli, N. B. Pahor, L. Randaccio, E. Zangrando, Inorg.
Chim. Acta 1988, 144, 241.
I. Iwakura, M. Hatanaka, A. Kokura, H. Teraoka, T. Ikeno, T. Nagata, T.
Yamada, Chem. Asian J. 2006, 1, 656.
A. Kokura, S. Tanaka, T. Ikeno, T. Yamada, Org. Lett. 2006, 8, 3025.
c
for Entries 1, 2, 4, 5, 8–12, and 1b for Entries 3, 6, and 7. À40 ꢀC,
e
2.0 mol % Co catalyst, 10 mol % CHCl3. dIn toluene. 5.0 mol % Co cata-
2
3
4
lyst, 25 mol % CHCl3. f5.0 equiv. NaBH4. g0.8 equiv. NaBH4. h4.0 mol %
Co catalyst, 20 mol % CHCl3, 1.2 equiv. modified borohydride, 1.0 equiv.
NaOMe. iIsolated yield. jDetermined by HPLC analysis (Chiralpak
AD-H). kDetermined by HPLC analysis (Chiralcel OD-H).
Published on the web (Advance View) December 2, 2006; doi:10.1246/cl.2007.26