(Salen)ruthenium-catalyzed desymmetrization of meso-Diols (2). Apical ligand effect on enantioselectivity

Article abstract of DOI:10.1246/cl.2003.480

Enantiotopos-selectivity of aerobic oxidation of meso-diols using chiral (nitrosyl)Ru(salen) complexes as catalyst was found to be dependent on the nature of their apical ligand. The oxidation with complex 5 possessing an apical hydroxo ligand as the cata

Full text of DOI:10.1246/cl.2003.480

480
Chemistry Letters Vol.32, No.6 (2003)
(Salen)ruthenium-catalyzed Desymmetrization of meso-Diols (2).
Apical Ligand Effect on Enantioselectivity
Hideki Shimizu and Tsutomu Katsukiꢀ
Department of Chemistry, Faculty of Science, Graduate School, Kyushu University 33 and
CREST, Japan Science and Technology (JST), Hakozaki, Higashi-ku, Fukuoka 812-8581
(Received February 14, 2003;CL-030126)
Enantiotopos-selectivity of aerobic oxidation of meso-diols
Table 1. Desymmetrization of various meso-diols using 3 as cat-
alysta
Entry Substrate Time/d Yield/%b % eec (% ee)d Configne
using chiral (nitrosyl)Ru(salen) complexes as catalyst was
found to be dependent on the nature of their apical ligand.
The oxidation with complex 5 possessing an apical hydroxo li-
gand as the catalyst showed moderate to good enantioselectivity
(up to 80% ee).
1
2
3
4
5
1
2.5
3
3
3
3
28
32
34
30
25
70
66
64
62
71
(67)
(59)
(65)
(63)
(66)
1R, 6Sf
1R, 4Sg
1R, 5Sg
6a
6b
6c
6d
h
-
1R, 6Si
aTwo mol% of the catalyst was used. bIsolated yield of lactol.
cDetermined by GLC analysis using optically active column (SU-
PELCO BETA-DEX-225) after its conversion to the correspond-
ing lactone. dThe enantioselectivity observed with complex 2 (ab-
Optically active lactols occur in many natural products as
subunits. Among various methods for the synthesis of optically
active lactols, desymmetrization (enantiotopos-selective oxida-
tion) of meso-diols is an attractive one from the synthetic point
of view, because various meso-diols are readily available. Horse
liver alcohol dehydrogenase is well known to undergo highly
selective desymmetrization of meso-diols;1 however, the pro-
ducts are not lactols but lactones. Amongst chemical meth-
ods,2;3 electrooxidation of prochiral or meso-diols is highly en-
antioselective, but the products are also lactones.3 Thus, for the
transformation of meso-diols to chiral lactols, a catalyst show-
ing not only enantiotopos selectivity but also chemoselectivity
is required.
e
stracted from Ref. 7). Determined by comparison of the specific
rotation after its conversion to the corresponding lactone.
fRef. 10a. gRef. 10b. hThe absolute configuration has not been de-
i
termined. Ref. 10c.
OH
OH
OH
OH
OH
(CH2)n
(CH2)n
(CH2)n
O
6a : n = 2
6b : n = 3
6c : n = 5
6d
7: n= 3
8: n= 5
O
We have recently demonstrated that a chiral (nitrosyl)(sa-
len)-ruthenium complex [hereafter denoted as (ON)Ru(salen)]
is an efficient catalyst for enantiomer-differentiating aerobic
oxidation of racemic secondary alcohols under photo-
irradiation,4;5 and (ON)Ru(salen) bearing tetramethylethylene-
diamine as its diamine unit catalyzes chemoselective aerobic
oxidation of 1,n-diols to give the corresponding lactols also un-
der photo-irradiated conditions.6 Based on these results, we ex-
pected that desymmetrization of meso-1,4-diols giving the cor-
responding optically active lactols would be realized by using a
chiral (ON)Ru(salen) bearing chiral quaternary carbons at its
ethylenediamine unit as the catalyst. Indeed, the oxidation of
1,2-bis(hydroxymethyl)cyclohexane 1 using complex 2 as the
catalyst showed moderate enantioselectivity (Scheme 1).7
We recently found that complex 2 was converted into com-
plex 3 bearing an apical hydroxo ligand upon its exposure to sil-
ica gel (FL100B, FUJI SILYSIA CHEMICAL LTD).8 Thus, we
examined desymmetrization of a series of 1,4-meso-diols with 3
as catalyst. The reactions were slow, but it is noteworthy that
complex 3 showed a slightly better enantioselectivity in the oxi-
dation of meso-diols (1, 6a, and 6d) than complex 2, while both
2 and 3 showed a similar level of enantioselectivity in the oxi-
dation of 6b and 6c (Table 1). Upon the oxidation of 6b and 6c,
the formation of a trace amount of mixed acetals (7 and 8) was
observed.9
In these aerobic oxidations, the alkyl moiety of the diol co-
ordinated to the ruthenium ion was considered to be directed to-
ward the space around C3 or C30.6a Therefore, we expected that
(ON)Ru(salen) bearing a bulkier substituent at C200 would show
higher asymmetric induction. Thus, we synthesized complex 4.
During chromatography on silica gel, complex 4 was found to
be partly transformed to hydroxo complex 5.8 Oxidation of diol
1 was examined by using 4 as the catalyst (Table 2). Contrary to
our expectation, the reaction with complex 4 showed slightly di-
minished enantioselectivity together with low chemical yield
(Entry 1). In contrast, the reaction with 5 showed better enan-
tioselectivity of 81% ee (Entry 2), albeit with low chemical
yield. Use of 4 mol% of 5 improved the chemical yield ca. twice
as many (Entry 3), and prolonged reaction time gave an accep-
OH
*
O
O
1
2, (2.0 mol%)
1
6
PDC, MS4A
CH2Cl2
OH
OH
*
*
*
*
O
air, hν, rt
CHCl3
6
1
67% ee, 65%
R
S
Me
3
Me
Me
Me
NO
Ru
X
NO
Ru
X
N
O
N
N
O
N
O
O
Ph Ph
3
Ar Ar
3'
3'
R
S
2"
2"
2: X= Cl
3: X= OH
4: Ar=
5: Ar=
p
-(C H )C H , X= Cl
6 5 6 4
p-(C H )C H , X= OH
6
5
6 4
Scheme 1.
Copyright ꢀ 2003 The Chemical Society of Japan

Chemistry Letters Vol.32, No.6 (2003)
481
Table 2. Desymmetrization of meso-diol 1 with 4 or 5 as catalyst
what factor determines the coordinating behavior of each meso-
diol.
Entry
Catalysta
Time/d
2.5
2.5
2.5
7
Yield/%b
% ee
62
81
82
Confign
1S, 6R
1S, 6R
1S, 6R
1S, 6R
1
2
3
4
4
5
5c
5c
9
In conclusion, we were able to achieve good to high enan-
tioselective aerobic oxidation of meso-diols by using an
(ON)(hydroxo)(salen)ruthenium complex as the catalyst under
photo-irradiation. The nature of the apical ligand was found
to influence the reaction pathway of the present aerobic oxida-
tion. Further study on the mechanism of the present aerobic oxi-
dation is in progress.
14
33
80
80
aThe catalyst (2 mol%) was used, unless otherwise mentioned.
bIsolated yield of lactol. Four mol% of the catalyst was used.
c
Table 3. Desymmetrization of various meso-diols using 5 as cat-
alysta
References and Notes
Entry
Substrate
Time/d
Yield/%b
% ee
74
66
63
Confign
1S, 4R
1S, 5R
-
1S, 6R
1
a) G. Fantin and P. Pedrini, in ‘‘Asymmetric Oxidation Reactions,’’ ed.
by T. Katsuki, Oxford University Press, Oxford (2001), p 200. b) J. B.
Jones, in ‘‘Asymmetric Synthesis,’’ ed. by J. D. Morrison, Academic
Press, New York (1984), Vol. 5, p 309.
1
2
3
4
6a
6b
6c
6d
7
7
7
7
64
78
80
68
c
75
2
a) Y. Ishii, K. Suzuki, T. Ikariya, M. Saburi, and S. Yoshikawa, J. Org.
Chem., 51, 2822 (1986). b) Y. Ishii, K. Osakada, T. Ikariya, and M.
Saburi, Chem. Lett., 1982, 1179. c) Asymmetric Oppenauer oxidation-
like desymmetrization of meso-diols (up to 50% ee) has recently been re-
ported: M. Ito, A. Osaku, and T. Ikariya, The 81st Annual Meeting of the
Chemical Society of Japan, Tokyo, March 2002, Abstr., No. 3 G4-14. d)
During the reviewing process of this manuscript, iridium-catalyzed
asymmetric Oppenauer oxidation of meso-diol was reported: T. Suzuki,
K. Morita, Y. Matsuo, and K. Hiroi, Tetrahedron Lett., 44, 2003 (2003).
a) Y. Yanagisawa, Y. Kashiwagi, F. Kurashima, J.-i. Anzai, T. Osa, and
J. M. Bobbitt, Chem. Lett., 1996, 1043. b) H. Tanaka, Y. Kawanami, K.
Goto, and M. Kuroboshi, Tetrahedron Lett., 42, 445 (2001).
K. Masutani, T. Uchida, R. Irie, and T. Katsuki, Tetrahedron Lett., 41,
5119 (2000).
aFour mol% of the catalyst was used. bIsolated yield of lactol.
cThe absolute configuration has not been determined.
table chemical yield without reducing enantioselectivity (Entry
4).11
Thus, we examined the oxidation of other meso-diols with
complex 5 as the catalyst (Table 3). The same trend in enantios-
electivity as observed in the reaction with 3 was again observed:
the oxidations of diols 6a and 6d proceeded with better enan-
tioselectivity than that with complex 2, while the oxidations
of diols 6b and 6c showed a similar level of selectivity to that
with 2 or 3.
3
4
5
Enantiomer-differentiating aerobic oxidation of racemic alcohols has re-
cently been reported: a) D. R. Jensen, J. S. Pugsley, and M. S. Sigman, J.
Am. Chem. Soc., 123, 7475 (2001). b) E. M. Ferreira and B. M. Stoltz, J.
Am. Chem. Soc., 123, 7725 (2001).
Although the mechanism of asymmetric induction of the
oxidations is unclear at present, we were intrigued by the
above-described trend in enantioselectivity of the reactions with
(hydroxo)(ON)Ru(salen) complexes (3 and 5) as the catalyst. It
is known that hydroxo and alkoxo ligands are readily exchange-
able with alcohol, and metallosalen complexes bearing such an
exchangeable ligand readily adopt a cis-b structure when a bi-
dentate ligand is coordinated.12;13 We have proposed that the re-
actions using complex 2 as the catalyst proceed via intermediate
A (X = Cl).7 We speculated that some meso-diols (6b and 6c)
would coordinate to the metal center through one hydroxy
group to give an intermediate A (X = OH), while other meso-
diols (1, 6a, and 6d) would make chelates (B) (Figure 1). It was
considered that the reactions through intermediate A would
show a similar level of enantioselectivity irrespective of the
salen and apical ligands of the catalysts: because of high confor-
mational freedom of the coordinated diol, a slight modification
of the salen ligand had minimal effect on the enantioselectivity.
On the other hand, the reaction via B was considered to show
better enantioselectivity when the coordination sphere of the
ruthenium ion is asymmetrically well defined, because the che-
lated diol is conformationally restricted. These considerations
agree with the experimental results, but it is unclear at present
6
a) A. Miyata, M. Furukawa, R. Irie, and T. Katsuki, Tetrahedron Lett.,
43, 3481 (2002). b) A. Miyata, M. Murakami, R. Irie, and T. Katsuki,
Tetrahedron Lett., 42, 7067 (2001).
7
8
H. Shimizu, K. Nakata, and T. Katsuki, Chem. Lett., 2002, 1080.
Complexes 2 and 4 were completely converted into complexes 3 and 5,
respectively, when they were exposed to silica gel for 12 h. On the other
hand, treatment of 3 and 5 with trimethylsilyl chloride gave 2 and 4, re-
spectively: Complexes 2-5 gave satisfactory 1H NMR and IR spectra, re-
spectively. IR spectra of 3 and 5 showed the absorption at ca. 3540 cmꢁ1
characteristic to OH group, while those of 2 and 4 did not show such ab-
sorption.
9
Formation of these unusual acetals was considered to be attributed to the
presence of hydrochloric acid derived from decomposition of chloro-
form, and use of non-fresh chloroform increased the formation of the
acetals.
10 a) I. J. Jakovac, H. B. Goodbrand, K. P. Lok, and J. B. Jones, J. Am.
Chem. Soc., 104, 4659 (1982). b) I. J. Jakovac, G. Ng, K. P. Lok, and
J. B. Jones, J. Chem. Soc., Chem. Commun., 1980, 515. c) H.-J. Gais,
K. L. Lukas, W. A. Ball, S. Braun, and H. J. Lindner, Liebigs Ann.
Chem., 1986, 687.
11 Typical experimental procedure is exemplified by the desymmetrization
of meso-diol 1: 1 (14.4 mg, 0.1 mmol) and 5 (4.6 mg, 4.0 mol%) were dis-
solved in anhydrous CHCl3 (0.5 mL). The solution was stirred under ir-
radiation with a halogen lamp in air for 7 days at ambient temperature.
The mixture was directly chromatographed on silica gel (hexane/ethyl
acetate = 1/1) to give the corresponding lactol (11.4 mg, 80%). To a sus-
ꢀ
pension of the lactol and molecular sieves 4 A (120 mg) in anhydrous
CH2Cl2 (0.5 mL) was added pyridinium dichromate (60 mg, 0.16 mmol).
The mixture was stirred overnight at room temperature, diluted with hex-
ane/EtOAc (4/1) and filtered through a pad of silica gel. The filtrate was
concentrated under reduced pressure. The ee of the resulting lactone was
determined by GLC analysis.
S
OH
(CH )
2 n
N
O
N
O
(CH )
Me
Me
2 n
HO
Ru
N
N
N
O
N
O
OH
-
-
O
Ar
O
O
12 a) B. Saito and T. Katsuki, Tetrahedron Lett., 42, 3873 (2001). b) B. Sai-
to and T. Katsuki, Tetrahedron Lett., 42, 8333 (2001). c) A. Watanabe, T.
Uchida, K. Ito, and T. Katsuki, Tetrahedron Lett., 43, 4481 (2002).
13 It has been reported that [(ON)Ru(salen)(H2O)](SbF6) complex under-
goes replacement of its apical ligand readily to give
[(ON)Ru(salen)(R2S)](SbF6) of a cis-b structure, when it is treated with
a sulfide: A. A. Sauve and J. T. Groves, J. Am. Chem. Soc., 124, 4770
(2002).
N
N
Ar
Ru
X
O
O
B
S
A (X= Cl or OH)
Ar= Ph or p-(C H )C H
6 5 6 4
Figure 1.
Published on the web (Advance View) May 6, 2003;DOI 10.1246/cl.2003.480