Asymmetric epoxidation using aqueous hydrogen peroxide as oxidant: bio-inspired construction of pentacoordinated Mn-salen complexes and their catalysis

Article abstract of DOI:10.1016/j.tetlet.2006.03.049

Pentacoordinated Mn-salen complexes 1 and 5 possessing an internal pyridine or N-methylimidazole ligand, respectively, were found to be efficient catalysts for asymmetric epoxidation of conjugated Z-olefins using aqueous hydrogen peroxide. In particular,

Full text of DOI:10.1016/j.tetlet.2006.03.049

Tetrahedron Letters 47 (2006) 3203–3207
Asymmetric epoxidation using aqueous hydrogen peroxide
as oxidant: bio-inspired construction of pentacoordinated
Mn–salen complexes and their catalysis
Hiroaki Shitama and Tsutomu Katsuki*
Department of Chemistry, Faculty of Science, Graduate School, Kyushu University 33, CREST,
Japan Science and Technology Agency (JST), 6-10-1, Hakozaki, Higashi-ku, Fukuoka 812-8581, Japan
Received 4 February 2006; revised 4 March 2006; accepted 9 March 2006
Available online 29 March 2006
Abstract—Pentacoordinated Mn–salen complexes 1 and 5 possessing an internal pyridine or N-methylimidazole ligand, respectively,
were found to be efficient catalysts for asymmetric epoxidation of conjugated Z-olefins using aqueous hydrogen peroxide. In par-
ticular, the epoxidation of chromene derivatives proceeded with high enantioselectivity greater than or equal to 97% ee.
ꢀ 2006 Elsevier Ltd. All rights reserved.
Asymmetric epoxidation using aqueous hydrogen per-
oxide is a topic of current interest and much effort has
oxo metal species through a synergetic push–pull mech-
anism:7 the cleavage of the O–O bond of intermediary
iron-hydroperoxo or -acylperoxo species is promoted
by coordination of an imidazole group and protonation
of the distal oxygen (in the case of acylperoxo species,
protonation is not necessarily needed due to high disso-
ciating ability of the acyloxy group)7c,d,8 and generates
the active iron-oxo species which undergoes oxidation.
In 1993, Berkessel et al. reported biomimetic asymmetric
epoxidation using a manganese(salalen) complex bear-
ing an imidazole substituent at the C7 carbon in the pres-
ence of aqueous 1% hydrogen peroxide, albeit with
moderate enantioselectivity,9 wherein the imidazole
group has been considered to coordinate at the apical po-
sition of the complex. Subsequently to this, Pietika¨inen
reported an intermolecular biomimetic version using a
Mn–salen complex, aqueous hydrogen peroxide, N-
methylimidazole, and ammonium acetate system.10,11
We independently disclosed Mn–salen catalyzed asym-
metric epoxidation using hydrogen peroxide under
almost identical conditions.12 During this study, we also
found that the epoxidation using hydrogen peroxide
proceeded even in the absence of a protonic substance,
if an excess amount of N-methylimidazole was used.
On the other hand, in the course of our study on asym-
metric catalysis of Mn–salen complexes, we had found
that asymmetry-inducing ability of a Mn–salen complex
is strongly related to its conformation13 and that intro-
duction of a coordinating substituent such as a carboxy-
late group at the ethylenediamine unit causes inversion
been directed to this field of chemistry.1 Since Julia
´
and co-workers reported chiral phase transfer-mediated
asymmetric epoxidation using aqueous hydrogen perox-
ide,2 many methodologies have been reported to date3
but only a few methods have attained high enantioselec-
tivity. Shi and co-workers reported that a fructose-
derived ketone catalyzed epoxidation using aqueous
hydrogen peroxide in acetonitrile in a highly enantio-
selective manner, albeit with moderate turnover number
(TON).4 Jørgensen and co-workers reported highly
enantioselective epoxidation of a,b-unsaturated alde-
hydes using an organocatalyst in the presence of aque-
ous hydrogen peroxide.5 We have recently disclosed
that a di-l-oxo Ti(salalen = half-reduced salen) complex
serves as the catalyst for highly enantioselective epoxi-
dation using 1 equiv of aqueous hydrogen peroxide with
high TON (up to 4500).6 On the other hand, the cataly-
sis of a typical oxidizing enzyme, cytochrome P-450,
that carries an ironporphyrin complex as its active site
has been well studied. As a part of the studies of
P-450’s catalysis with modeling porphyrin complexes, it
has been found that hydrogen peroxide or peracid serves
as the terminal oxidant and that the ironporphyrin com-
plex is oxidized with the oxidant to the corresponding
*
Corresponding author. Fax: +81 92 642 2607; e-mail: katsuscc@
mbox.nc.kyushu-u.ac.jp
0040-4039/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2006.03.049

3204
H. Shitama, T. Katsuki / Tetrahedron Letters 47 (2006) 3203–3207
type of Mn–salen complex is that the coordination of
R
(S)
N
the internal nucleophile always keeps one of the two api-
cal sites available for coordination of hydrogen peroxide.
Thus, we synthesized complexes 1–515 bearing a (2-pyr-
idyl)methyl, (4-phenyl-2-pyridyl)methyl, (1H-imidazol-
4-yl)methyl or (N-methylimidazol-4-yl)methyl group as
the nucleophilic substituent (Fig. 1) and first examined
the epoxidation of 2,2-dimethylchromene in the presence
of 3 equiv of aqueous hydrogen peroxide as the terminal
oxidant, because some of the epoxy compounds derived
from 2,2-disubstituted epoxychromane derivatives show
unique physiological activity such as potent hypertensive
activity of Cromakalim (Table 1).16,17
NH
N
O
Mn+
N
(S)
N
O
Ph
N
O
Ph
Mn+
(aS)
O
Ph
Ph
-
(aR)
PF6
Ph
-
PF6
2: R=
1: R=
4
N
N
NH
N
3: R=
5: R=
N
N
As expected, the reaction with 1 in dichloromethane
proceeded at room temperature with high enantioselec-
tivity of 95%, irrespective of the concentration of hydro-
gen peroxide, albeit with modest chemical yields (entry
1). The use of urea Æ hydrogen peroxide adduct instead
of using aqueous hydrogen peroxide slightly reduced
enantioselectivity (entry 2). On the other hand, lowering
the reaction temperature to 0 ꢁC somewhat improved
both enantioselectivity (up to 97% ee) and the chemical
yield (entry 3). Use of acetonitrile as solvent not only re-
duced chemical yield but also diminished enantioselec-
tivity (entry 4).18 It is noteworthy that the observed
sense of asymmetric induction by complex 1 is opposite
to that observed in the epoxidation with the (aR,S)-Mn–
salen complexes bearing a cyclohexanediamine unit.
This supports that the pyridyl substituent is coordinated
Figure 1.
of the conformation of the parent Mn–salen complex
and reverses the sense of asymmetric induction, together
with improvement of the TON of the catalyst.14
Although biomimetic approaches to asymmetric epoxi-
dation using hydrogen peroxide have hitherto achieved
moderate success, we considered that it should be a
promising one: based on the above-mentioned results,
a Mn–salen complex bearing a nucleophilic substituent
at its ethylenediamine unit would promote epoxidation
using hydrogen peroxide in a highly asymmetric atmo-
sphere without adding any external nucleophile, showing
high enantioselectivity. Another advantage of this
Table 1. Asymmetric epoxidation of 2,2-dimethylchromene catalyzed by 1–5a
O
O
O
catalyst (2.5 mol%)
oxidant
Entry
Catalyst
Oxidant
T (ꢁC)
Yield (%)b
ee (%)c
1
2
3
4e
5
6
7
8
9
10
11
12
13
1
30% H2O2 aq
UHPd
rt
rt
0
0
0
0
0
0
0
0
0
0
0
27
22
49–60
28
67
26
12
63
85
48
56
67
80
95
92
97
90
96
95
83
97
98
98
98
98
98
1
1
30% H2O2 aq
30% H2O2 aq
30% H2O2 aq
30% H2O2 aq
30% H2O2 aq
30% H2O2 aq
30% H2O2 aq
30% H2O2 aqg
30% H2O2 aqg,h
30% H2O2 aq.i
30% H2O2 aqj
1
2
3
4
5
5f
5f
5f
5f
5f
a All the reactions were carried out for 24 h with 2.5 mol % of catalyst in the presence of 3 equiv of oxidant in dichloromethane, unless otherwise
mentioned.
b Isolated yield.
c Determined by HPLC analysis using a chiral stationary phase column (Daicel Chiralpack OB-H; hexane/i-PrOH 9:1). Absolute configuration was
determined to be 3R,4R by chiroptical comparison (Ref. 19).
d 1 equiv of UHP was used as oxidant.
e Acetonitrile was used as a solvent.
f 5 mol % of 5 was used.
g 1 equiv of H2O2 was used.
h The reaction was carried out for 48 h.
i 1 equiv of H2O2 was added over the period of 8 h and the mixture was stirred for another 62 h.
j After the reaction was carried out with 1 equiv of H2O2 for 24 h, 2 equiv of H2O2 were added and the reaction mixture was stirred for another 24 h.

H. Shitama, T. Katsuki / Tetrahedron Letters 47 (2006) 3203–3207
3205
to the manganese ion and the conformation of the salen
ligand is reversed.14 Under the optimized conditions, we
next examined the epoxidation with complexes 2–5 as
catalyst (entries 5–8). It is noteworthy that all the
aS,S-complexes showed high enantioselectivity (entries
5, 6, and 8), while the aR,S-complex 4 showed consider-
ably inferior enantioselectivity (entry 7): in general, the
(aR,S)-Mn–salen complex bearing a cyclohexanedi-
amine unit is a superior catalyst for asymmetric epoxi-
dation to the corresponding (aS,S)-Mn–salen complex
and this fact also supports our hypothesis on the effect
of the nucleophilic substituent.14 Of the aS,S-complexes
(1–3, and 5), complex 5 was found to be the catalyst of
choice in terms of yield and enantioselectivity. The reac-
tion using 5 mol % of complex 5 and 3 equiv of aqueous
hydrogen peroxide at 0 ꢁC gave the desired epoxide of
Table 2. Asymmetric epoxidation using aqueous hydrogen peroxidea
Entry
Substrate
Product
Yield (%)
ee (%)
O
O
3R
R
R
4R
O
1
2
3
4
5
6; R = CN
7; R = Br
8; R = NO2
9; R = Me
10; R = OMe
R = CN
R = Br
R = NO2
R = Me
95b
98b
85b
80b
78b
99c,d
98e
99c,d
97e
R = OMe
98f
O
O
6
84b
84b
95h
92h
98e
11
O
O
O
7
97g
88i,j
90i,j
12
13
O
8
6S
5R
O
9k
Ph
Ph
4S
4S
3R
Ph
10k
94 (4:1)h,l
88m,n
O
14
15
O
3S
2R
Ph
11
58h
31o,j
Ph
1R
O
a All the reactions were carried out for 24 h with 5 mol % of 5 and 3 equiv of aq 30% H2O2 in dichloromethane at 0 ꢁC, unless otherwise mentioned.
b Isolated yield.
c Determined by HPLC analysis using a chiral stationary phase column (Daicel Chiralpack OJ-H; hexane/i-PrOH 7:3).
d Determined by chiroptical comparison (Ref. 19).
e Determined by HPLC analysis using a chiral stationary phase column (Daicel Chiralpack OJ-H; hexane/i-PrOH 9:1).
f Determined by HPLC analysis using a chiral stationary phase column (Daicel Chiralpack OB-H; hexane/i-PrOH 7:3).
g Determined by HPLC analysis using a chiral stationary phase column (Daicel Chiralpack OB-H; hexane/i-PrOH 99:1).
h Determined by 1H NMR (400 MHz) spectroscopic analysis.
i Determined by HPLC analysis using a chiral stationary phase column (Daicel Chiralpack OB-H; hexane/i-PrOH 99.9:0.1).
j Determined by chiroptical comparison (Ref. 20).
k Complex 1 was used as catalyst.
l Product is a mixture of cis- and trans-epoxides. Numbers in parentheses are a ratio of cis- and trans-epoxides.
m Face selectivity. The face selectivity was calculated by the equation, face selectivity = [% ee(cis) · %(cis) + % ee(trans) · %(trans)]/100. % ee of cis-
isomer (3R,4S) = 83 and that of trans-isomer (3S,4S) = 96. Ees were determined by HPLC analysis using a chiral stationary phase column (Daicel
Chiralpack OJ-H; hexane/i-PrOH 99.9:0.1).
n Determined by chiroptical comparison (Refs. 6 and 11b).
o Determined by HPLC analysis using a chiral stationary phase column (Daicel Chiralpack IA; hexane/i-PrOH 99.9:0.1).

3206
H. Shitama, T. Katsuki / Tetrahedron Letters 47 (2006) 3203–3207
´
931; (b) Colonna, S.; Molinari, H.; Banfi, S.; Julia, S.;
98% ee in 85% yield (entry 9). The reaction with 1 equiv
of aqueous hydrogen peroxide was much slow and the
yield of the epoxide was reduced (entry 10). The elon-
gated reaction slightly increased the chemical yield (en-
try 11). Slow addition of 1 equiv of aqueous hydrogen
peroxide also slightly but significantly improved the
yield to 67% with the same enantioselectivity of 98%
ee (entry 12). However, when 2 equiv of aqueous hydro-
gen peroxide were further added after 24 h to the reac-
tion mixture including 1 equiv of aqueous hydrogen
peroxide, the yield amounted to 80% (entry 13). These
results suggest that hydrogen peroxide is slowly decom-
posed during the reaction, but the catalyst is proof
against the conditions.
Masana, J.; Alvarez, A. Tetrahedron 1983, 39, 1635–1641;
(c) Banfi, S.; Colonna, S.; Molinari, H.; Julia, S.; Guixer,
´
J. Tetrahedron 1984, 40, 5207–5211.
3. For recent reports on asymmetric epoxidation using
aqueous hydrogen peroxide as stoichiometric oxidant,
see: (a) Francis, M. B.; Jacobsen, E. N. Angew. Chem., Int.
Ed. 1999, 38, 937–941; (b) Stoop, R. M.; Bachmann, S.;
Valentini, M.; Mezzetti, A. Organometallics 2000, 19,
4117–4126; (c) Tse, M. K.; Do¨bler, C.; Bhor, S.; Klawonn,
M.; Ma¨gerlein, W.; Hugl, H.; Beller, M. Angew. Chem.,
Int. Ed. 2004, 43, 5255–5260; (d) Tse, M. K.; Klawonn,
M.; Bhor, S.; Do¨bler, C.; Anilkumar, G.; Hugl, H.;
Ma¨gerlein, W.; Beller, M. Org. Lett. 2005, 7, 987–990; (e)
Bhor, S.; Anilkumar, G.; Tse, M. K.; Klawonn, M.;
Do¨bler, C.; Bitterlich, B.; Grotevendt, A.; Beller, M. Org.
Lett. 2005, 7, 3393–3396.
Under the optimized conditions, the asymmetric epoxi-
dation of various 2,2-disubstituted chromene derivatives
was examined with 3 equiv of aqueous hydrogen perox-
ide, because the reaction with stoichiometric one was
slow, and the results are shown in Table 2. All the reac-
tions of 2,2-disubstituted chromene derivatives showed
high enantioselectivity greater than or equal to 97% ee,
irrespective of the electronic nature of the C6-substitu-
ent (entries 1–6). The epoxidation of trisubstituted olefin
12 also proceeded with good chemical yield and high
enantioselectivity (entry 7). The scope of the present
reaction is not limited to the epoxidation of chromene
derivatives. The epoxidation of 1,2-benzo-1,3-cyclohepta-
diene and (Z)-1-phenyl-3-penten-1-yne also proceeded
4. (a) Shu, L.; Shi, Y. Tetrahedron Lett. 1999, 40, 8721–8724;
(b) Shu, L.; Shi, Y. Tetrahedron 2001, 57, 5213–5218.
´
5. Marigo, M.; Franzen, J.; Poulsen, T. B.; Zhuang, W.;
Jørgensen, K. A. J. Am. Chem. Soc. 2005, 127, 6964–6965.
6. Matsumoto, K.; Sawada, Y.; Saito, B.; Sakai, K.; Katsuki,
T. Angew. Chem., Int. Ed. 2005, 44, 4935–4939, Sala-
len = a hybrid salan/salen tetradentate ligand.
7. (a) Yuan, L. C.; Bruice, T. C. J. Am. Chem. Soc. 1986, 108,
1643–1650; (b) Battioni, P.; Renaud, J. P.; Bartoli, J. F.;
Artiles, M. R.; Fort, M.; Mansuy, D. J. Am. Chem. Soc.
1988, 110, 8462–8470; (c) Yamaguchi, K.; Watanabe, Y.;
Morishima, I. J. Am. Chem. Soc. 1993, 115, 4058–4065; (d)
Machii, K.; Watanabe, Y.; Morishima, I. J. Am. Chem.
Soc. 1995, 117, 6691–6697; (e) Ozaki, S.; Inaba, Y.;
Watanabe, Y. J. Am. Chem. Soc. 1998, 120, 8020–8025; (f)
Nam, W.; Lee, H. J.; Oh, S.-Y.; Kim, C.; Jang, H. G. J.
Inorg. Biochem. 2000, 80, 219–225; (g) Watanabe, Y.;
Ueno, T. Bull. Chem. Soc. Jpn. 2003, 76, 1309–1322, and
references cited therein.
8. Although P-450 possesses a thiol group as the apical
ligand, Watanabe et al. have reported that a myoglobin
mutant that has a imidazole group as the apical ligand
shows P-450’s activity and they have proposed that an
intermediary hydroperoxo species is converted to the
corresponding oxo species due to synergetic push–pull
effect by the apical imidazole and distal histidine groups
(Ref. 7g).
9. (a) Schwenkreis, T.; Berkessel, A. Tetrahedron Lett. 1993,
34, 4785–4788; (b) Berkessel, A.; Frauenkron, M.; Sch-
wenkreis, T.; Steinmetz, A.; Baum, G.; Fenske, D. J. Mol.
Cat. A: Chem. 1996, 113, 321–342; (c) Berkessel, A.;
Schwenkreis, T.; Frauenkron, M.; Steinmetz, A.; Scha¨tz,
N.; Prox, J. Peroxide Chem. 2000, 511–525.
10. (a) Pietika¨inen, P. Tetrahedron Lett. 1994, 35, 941–944; (b)
Pietika¨inen, P. Tetrahedron 1998, 54, 4319–4326.
11. For the review on asymmetric epoxidation using Mn–salen
complexes as catalyst in the presence of oxidant other than
hydrogen peroxide, see: (a) Katsuki, T. Coord. Chem. Rev.
1995, 140, 189–214; (b) Katsuki, T. J. Mol. Catal. A 1996,
113, 87–107; (c) Jacobsen, E. N.; Wu, M. H. Compr.
Asymmetric Catal. I–III 1999, 2, 649–677.
with high enantioselectivity. However, complex
1
showed better enantioselectivity in these reactions than
complex 5 (cf. entries 8 and 9). The epoxidation of
1,2-benzo-1,3-cycloheptadiene and (Z)-1-phenyl-3-pen-
ten-1-yne with 1 showed 90 and 86% ees, respectively
(entries 9 and 10). As in the epoxidation using usual
Mn–salen complex as catalyst, the epoxidation of
acyclic cis-enyne with 1 or 5 gave a mixture of cis-
and trans-epoxides (entry 10).11 In agreement with the
epoxidation with usual Mn–salen complex, epoxidation
of trans-b-methylstyrene was low enantioselective
(entry 11).
In conclusion, we were able to disclose that manga-
nese(III)–salen complexes bearing a nucleophilic substi-
tuent at the diamine unit, especially complexes 1 and 5,
serve as efficient catalysts for epoxidation of conjugated
cis-olefins using aqueous 30% hydrogen peroxide as the
terminal oxidant. The role of the nucleophilic substitu-
ent is considered to be twofold: regulation of the
conformation of the manganese(III) complexes and
acceleration of the conversion of the hydroperoxo inter-
mediate to the oxo species. These results demonstrate
that bio-inspired approach is a potential entry to asym-
metric epoxidation using aqueous hydrogen peroxide.
12. Irie, R.; Hosoya, N.; Katsuki, T. Synlett 1994, 255–256.
13. (a) Hashihayata, T.; Ito, Y. N.; Katsuki, T. Synlett 1996,
1079–1081; (b) Hashihayata, T.; Ito, Y. N.; Katsuki, T.
Tetrahedron 1997, 53, 9541–9552; (c) Katsuki, T. Adv.
Synth. Catal. 2002, 344, 131–147.
References and notes
14. Ito, Y. N.; Katsuki, T. Tetrahedron Lett. 1998, 39, 4325–
4328.
15. Diamine units bearing a nucleophilic substituent were
synthesized from L-serine or L-histidine in conventional
manner. The pathway for the synthesis of 5 is shown
below.21
1. (a) Lane, B. S.; Burgess, K. Chem. Rev. 2003, 103, 2457–
2474; (b) Noyori, R.; Aoki, M.; Saito, K. Chem. Commun.
2003, 1977–1986.
´
2. (a) Julia, S.; Masana, J.; Vega, J. C. Angew. Chem. 1980,
92, 968–969; Angew. Chem., Int. Ed. Engl. 1980, 19, 929–

H. Shitama, T. Katsuki / Tetrahedron Letters 47 (2006) 3203–3207
3207
CO2Me
NH
O
PhCH2OH, (i-Pr)2NEt
NH
N
N
CH3CN, reflux, 2 days
N
N+
I-
NH2
HO
Ref. 21
O
60% (3 steps)
N
N
O
1) NaBH4, MeOH, 0 oC, 3 h
1) NaN3, DMF, 60 oC, 24 h
2) H2, Pd/C, MeOH, r.t., 24 h
N
2) MsCl, Et3N, CH2Cl2, r.t., 3 h
MeO
NH
MsO
NH
Cbz
Cbz
84% (2 steps)
79%
N
1) Mn(OAc)2 4H2O, EtOH, r.t.
5
N
CHO
2)
H2N
NH2
, 60 o
C
OH
Ph
30% (2 steps)
85%
3) NaPF6
16. Bergmann, R.; Gericke, R. J. Med. Chem. 1990, 33, 492–
504.
of the epoxide was determined by HPLC analysis under
the condition described in the footnote to Table 1.
17. General procedure for asymmetric epoxidation: Mn–salen
complex 5 (2.7 mg, 2.5 lmol) and 2,2-dimethylchromene
(0.1 mmol) were dissolved in dichloromethane (1 ml).
After the addition of aqueous 30% hydrogen peroxide
(0.3 mmol) at 0 ꢁC, the resultant mixture was stirred for
24 h. The solvent was removed in vacuo and the residue
was purified by chromatography on silica gel (pentane/
Et2O 20:1) to give the corresponding epoxide. The ee value
18. The reaction in AcOEt was also slow and gave the
corresponding epoxide of 95% ee in 9% yield after 24 h
and the reaction in THF was sluggish.
19. Lee, N. H.; Muci, A. R.; Jacobsen, E. N. Tetrahedron Lett.
1991, 32, 5055–5058.
20. Tian, H.; She, X.; Yu, H.; Shu, L.; Shi, Y. J. Org. Chem.
2002, 67, 2435–2446.
21. Jain, R.; Cohen, L. A. Tetrahedron 1996, 52, 5363–5370.