Electronic effects on enantioselectivity in epoxidation catalyzed by D 4-symmetric chiral porphyrins

Article abstract of DOI:10.1016/j.tetasy.2004.10.029

Asymmetric epoxidation of various aromatic olefins was examined with our D₄-symmetric chiral porphyrin. The enantioselectivity was greatly improved upon when the substrates contained electron-deficient groups. Moreover, examination of electroni

Full text of DOI:10.1016/j.tetasy.2004.10.029

Tetrahedron:
Asymmetry
Tetrahedron: Asymmetry 15 (2004) 3861–3867
Electronic effects on enantioselectivity in epoxidation
catalyzed by D -symmetric chiral porphyrins
4
a
Hiroshi Nakagawa, Yoshihisa Sei, Kentaro Yamaguchi,
b
b
a
Tetsuo Nagano and Tsunehiko Higuchi
c,
*
a
Graduate School of Pharmaceutical Sciences, The University of Tokyo, Bunkyo-ku, Tokyo 113-0033, Japan
b
Chemical Analysis Center, Chiba University, Yayoi-cho, Inage-ku, Chiba 263-8522, Japan
Graduate School of Pharmaceutical Sciences, Nagoya City University, Tanabe-dori, Mizuho-ku, Nagoya 467-8603, Japan
c
Received 31 August 2004; accepted 14 October 2004
4
Abstract—Asymmetric epoxidation of various aromatic olefins was examined with our D -symmetric chiral porphyrin. The enan-
tioselectivity was greatly improved upon when the substrates contained electron-deficient groups. Moreover, examination of elec-
tronic effects in the porphyrin catalyst revealed that electron-deficient groups lowered and electron-donating groups raised the
enantioselectivity. Hammett plot analysis suggested that these electronic effects could be interpreted in terms of the Hammond
postulate.
ꢀ
2004 Elsevier Ltd. All rights reserved.
1
. Introduction
R
R
R
R
O
R
R
R
O
R
X
O
O
O
X
Metalloporphyrins are of great interest as oxidation cata-
lysts, for example, for epoxidation or hydroxylation.
Since application of chiral metalloporphyrins to asym-
metric epoxidation was first reported in 1983, many
chiral porphyrins designed to catalyze asymmetric reac-
O O
O
X
N
H
N
N
H
N
X
1
2
3
4
5
X=H
O
X=t-BuCO
X=Br
O
O
R
O
R
O
R
1
R R
X=Me
R
O
X=n-BuO
O
2
tions have been synthesized, with some of them show-
R R
R= -CH2OCH3
ing high enantioselectivity. In these asymmetric
reactions, electronic features have an important influ-
ence on the enantioselectivity. The addition of amines
as axial ligands is a convenient method for modifying
the electronic features, while the enantioselectivity was
enhanced by the addition of axial ligands in some por-
Figure 1. D
4
-symmetric porphyrins with electron-withdrawing or
donating groups.
of the porphyrin rings affect the enantioselectivity. Our
findings should lead to the development of superior cata-
lysts offering improved levels of asymmetric induction.
3
–6
phyrins.
In addition, the enantioselectivity was
greatly improved with NarutaÕs catalyst by utilizing
7
the p-stacking effect between porphyrin and substrate.
We were interested in the electronic effect in the epoxida-
8
tion catalyzed by our D -symmetric chiral porphyrin 1
4
2. Results and discussion
(
Fig. 1), because 1 showed much higher enantioselecti-
vity in the case of substrates with electron-deficient
groups. The electronic effects were examined in detail
by utilizing various styrene derivatives and porphyrins
Table 1 shows the results of the epoxidation of substi-
9
tuted styrenes with the 1–Fe(Br)/PhIO system. The
presence of electron-deficient substituents greatly in-
creased the enantioselectivity, with the highest enantio-
meric excess achieved in the case of 3-nitrostyrene
1
–5. Herein, we focus on how the electronic features
(78% ee, run 11). To evaluate the effects of the substitu-
*
Corresponding author. Tel.: +81 52 836 3435; fax: +81 52 836
3435; e-mail: higuchi@phar.nagoya-cu.ac.jp
ents, the results in Table 1 were analyzed by means of a
Hammett plot (Fig. 2). A linear correlation was
0
957-4166/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2004.10.029

3862
H. Nakagawa et al. / Tetrahedron: Asymmetry 15 (2004) 3861–3867
Table 1. Epoxidation of substituted styrenes with 1–Fe(Br)
of cis-oxide and trans-oxide formed. In the epoxidation
of 3-trifluoromethyl-trans-b-methylstyrene, the ratio of
cis-oxide/trans-oxide was slightly lower than in that of
trans-b-methylstyrene (0.05–0.01), while the enantio-
*
O
1
-Fe(Br) / PhIO
X
X
1
2
facial selectivity was markedly increased (2.3–3.7).
Therefore the main effect of these electron-deficient
groups is not due to shortening of the lifetime of the
benzyl cation intermediate. p-Stacking effects between
porphyrins and substrates may explain the improvement
of the ees, but such interaction would be sterically diffi-
cult and lead to deviation from the linear correlation of
Run
X
Ee (%)
Yield (%)
1
2
3
4
5
6
7
8
9
H
47
66
52
51
48
31
42
28
74
52
78
68
66
59
67
67
58
53
51
53
66
62
a
3-F
4-F
4-C1
4-Br
a
2-Me
3-Me
4-Me
7
Figure 2. The reason for the high enantioselectivity is
discussed later.
3-CF
3
Next, we focused on the electronic features of the por-
phyrin ring, and examined the electronic effects on the
enantioselectivity of the catalysts. Substituents, which
could influence the electronic features of the catalysts,
were placed at the p-positions of the meso-phenyl groups
of 1 (Fig. 1) and the effects on the enantioselectivity
were examined. n-Butoxy, methyl, bromo, and pivaloyl
groups were introduced as substituents.
1
1
04-CF
1
3
a
3-NO
2
Ees and yields (based on PhIO) were determined by HPLC analysis.
a
1
3
Isolated yields. Ees were determined by H NMR with (+)-Eu(hfc) .
1
1
0
0
.2
.0
.8
.6
3
-NO2
major enantiomer
minor enantiomer
Styrene, 3-nitrostyrene, and trans-b-methylstyrene were
epoxidized in the presence of these iron porphyrins with
log
(
)
)
3-CF3
9
+
=
0.516σ +0.445
PhIO as the oxidant (Table 2). While electron-deficient
3-F
r=0.847
groups on the porphyrin lowered the enantiomeric ex-
cesses, electron-donating groups increased the enantio-
selectivity (for example, the n-butoxy-substituted
catalyst 5 gave 52% ee for styrene, 79% ee for 3-nitrosty-
rene and 45% ee for trans-b-methylstyrene) with the dif-
ference between 2–Fe(Br) and 5–Fe(Br) being 6–10% ee.
The cis-oxide was obtained in the epoxidation of
trans-b-methylstyrene with low yield (2–3%), and no
correlation was observed among the substituents, the
enantiomeric excesses [(1S,2R), 8–10% ee] and chemical
4-Cl
4-F
4-CF3
0
0
.4
.2
H
4-Br
(
3
-CH3
4
-CH3
0.0
-
0.50
-0.25
0.00
0.25
σ+
0.50
0.75
1
3
yields.
+
Figure 2. Correlation of optical yields of epoxides versus r of
substituents of styrenes.
The results in Table 2 were plotted against the r values.
As shown in Figure 4, linear relationships were observed
p
between the r values of the substituents of the porphy-
p
rins and log(major enantiomer/minor enantiomer) for
all three substrates. This can be rationalized in terms
of the Hammond postulate, as demonstrated for Jacob-
observed between log(major enantiomer/minor enantio-
+
mer) and r values. This suggests the existence of a benz-
yl cation intermediate, which is subject to racemization
1
2
1
0,11
sen catalysts. That is, lower reactivity leads to a com-
paratively late transition state and concomitantly higher
selectivity. The electron-donating groups of 4–Fe(Br)
(Fig. 3).
Thus, the improvement of the enantioselecti-
vity by electron-deficient groups might be rationalized
in terms of a shortening of the lifetime of this benzyl cat-
ion intermediate. To investigate the role of the putative
benzyl cation intermediate, 3-trifluoromethyl-trans-b-
methylstyrene was used, because the influence of Ca–
1
4
and 5–Fe(Br) attenuate the reactivity of the catalysts,
and the transition states would be more product-like
according to the Hammond postulate. Thus, catalysts
and substrates would have a more effective interaction
and the enantioselectivity with 4–Fe(Br) and 5–Fe(Br)
becomes higher. The jqj values of these substrates, which
are useful for investigating the Hammond postulate,
were slightly different. The jqj value of 3-nitrostyrene
was the largest, suggesting that the interaction between
catalyst and substrate was the greatest, because the low-
er reactivity of 3-nitrostyrene leads to a transition state
with the most product-like conformation. Higher enan-
tioselectivity of styrenes with electron-deficient groups
in Table 1 can also be explained by the Hammond pos-
C bond rotation could be quantified from the amounts
b
O
(
S)
+
O
O
Fe
(
R)
Racemize
6
,7
tulate. The jqj values in Figure 4 are smaller than the
+
Figure 3. Effect of carbocationic intermediate.
jq j value in Figure 2, because the substituents of the

H. Nakagawa et al. / Tetrahedron: Asymmetry 15 (2004) 3861–3867
3863
Table 2. Epoxidation of olefins catalyzed by bromo-[tetrakis(p-substituted phenyl)porphyrinato]iron
R1
* O R1
catalyst / PhIO
*
R2
R2
Substrate
Ee (%), yield (%)
2–Fe(Br)
3–Fe(Br)
1–Fe(Br)
4–Fe(Br)
5–Fe(Br)
a
3
9 (48)
40(43)
42 (45)
47 (68)
44 (46)
45 (49)
52 (66)
b
c
4
7
2 (72)
1 (68)
44 (65)
74 (70)
49 (65)
78 (70)
78 (61)
79 (63)
NO2
Ees and yields (based on PhIO) were determined by HPLC analysis except for 3-nitrostyrene.
a
(
1S,2S)-Epoxide was obtained as the major enantiomer. The cis-epoxide was also obtained in 2–3% yield (1S,2R), 8–10% ee.
S)-Epoxide was obtained.
b
c
(
1
Isolated yields. Ees were determined by H NMR with (+)-Eu(hfc)
3
. Absolute configuration of epoxide was not determined.
1.2
performed on a chiral column (CHIRALPAK AS,
)
4
6mm · 250mm, Daicel Chemical Industries) to quan-
1
0
.0
.8
tify enantiomeric excesses. Scheme 1 shows the synthetic
route to the p-substituted benzaldehydes.
3
-nitrostyrene
ρ=-0.232, r=0.95
styrene
ρ=-0.166, r=0.99
0
.6
3.1. (R,R,R,R)-9-Formyl-2,3,6,7-tetramethoxymethyl-
1,4,5,8-tetraoxa-1,2,3,4,5,6,7,8-octahydroanthracene 6
trans-β-methylstyrene
ρ=-0.104, r=0.99
0
0
.4
.2
(
Prepared by the same method as Ref. 8.
-
0.40 -0.20 0.00 0.20 0.40
σp
3
2
7
.2. (R,R,R,R)-9-(2,2-Dimethyl-1-(R,S)-hydroxypropyl)-
,3,6,7-tetramethoxymethyl-1,4,5,8-tetraoxa-1,2,3,4,5,6,
,8-octahydroanthracene 7
Figure 4. Correlation of optical yields of epoxides versus r
substituents of porphyrins. Log(enantiofacial selectivity) was used as
p
of
the vertical axis for trans-b-methylstyrene. Enantiofacial selecti-
vity = [(1S,2S) + (1R,2S)]/[(1R,2R) + (1S,2R)].
10
To a solution of aldehyde 6 (266mg, 0.67mmol) in THF
6mL) was added tert-butylmagnesium bromide 1.0M
(
THF solution (1.4mL, 1.4mmol) at 0ꢁC under an argon
atmosphere, and the mixture stirred at room tempera-
ture for 15h. Saturated NH Cl (aq) was added, followed
4
porphyrins 1–5 are far from the central metal and the
electronic effect would be smaller. In general, porphyrins
with electron-deficient groups would be preferable as
oxidation catalysts, since their reactivity is higher and
the porphyrins become more resistant to self-oxida-
tion. However these groups also lower the selectivity.
Modifications that lead to milder reactivity, such
as introduction of electron-donating groups into the
catalyst, would be expected to yield higher
enantioselectivity.
by extraction with EtOAc. The organic layer was dried
over Na SO and evaporated. Purification by silica gel
column chromatography (EtOAc/n-hexane) afforded 7
2
4
1
15
as a colorless solid (200mg, 0.44mmol, 66% yield); H
NMR (CDCl ) d 0.96 (9H, s), 3.38, 3.40 (12H, m),
3
3.6–3.7 (8H, m), 4.1–4.3 (4H, m), 4.83 (1H), 6.45 (1H).
3.3. (R,R,R,R)-9-Pivaloyl-2,3,6,7-tetramethoxymethyl-
1,4,5,8-tetraoxa-1,2,3,4,5,6,7,8-octahydroanthracene 8
3
. Experimental
To a mixture of substrate 7 (181mg, 0.40mmol), N-
methylmorpholine N-oxide (139mg, 1.2mmol) and
˚
molecular sieves 4A (300mg) in CH CN (3mL) was
All solvents were purified by standard methods. Unless
otherwise noted, reagents were obtained from commer-
cial suppliers and used without further purification.
Melting points were determined on a cover glass with
3
added tetrapropylammonium perruthenate (14mg,
0.10mmol), and the mixture was stirred for 5h at
40ꢁC. The mixture was passed through a short silica
gel column (EtOAc) and purification of the product by
silica gel column chromatography (EtOAc/n-hexane)
an electrothermal melting point apparatus and are
uncorrected. H NMR chemical shifts in CDCl were re-
1
3
1
ported relative to internal TMS. H NMR chemical
afforded 8 as colorless solid (154mg, 0.34mmol, 86%
1
shifts in CD Cl and C NMR chemical shifts were
3
1
yield); H NMR (CDCl ) d 1.20(9H, s), 3.36 (6H, s),
2
2
3
reported relative to the solvent peak. HPLC was
3.40(6H, s), 3.67 (8H, m), 4.14 (4H, m), 6.50(1H, s);

3864
H. Nakagawa et al. / Tetrahedron: Asymmetry 15 (2004) 3861–3867
HO
O
O
CHO
R
O
O
O
R
R
R
R
O
O
R
R
R
O
O
O
R
R
R
O
O
O
O
R
R
t-BuMgBr
TPAP, NMO
Cl2CHOCH3
SnCl4
O
O
O
R
R
O
6
R
CHO
7 y. 66%
8 y. 86%
R = CH2OCH3
9 y. 75%
CHO
Br2
R
R
O
O
O
R
O
R
Br
1
0 y. 67%
Me
Me
R
R
O
O
O
O
R
R
R
R
O
O
O
R
R
BF3-OEt2
NaBH3CN
Cl2CHOCH3
SnCl4
O
CHO
1
1 y. 90%
1
2 y. 79%
OH
O
O
1
2
) m-CPBA
) NH3
R
O
O
O
R
R
R
R
R
R
O
O
O
O
O
O
O
R
n-BuBr
K2CO3
Cl2CHOCH3
SnCl4
R
O
R
R
O
R
CHO
1
3 y. 48%
14 y. 95%
1
5 y. 91%
Scheme 1. Synthetic scheme for p-substituted benzaldehydes. Reagents and conditions: TPAP tetra-n-propylammonium perruthenate; NMO
N-methylmorpholine N-oxide.
1
3
C NMR (CDCl ) d 26.7, 44.8, 59.5, 71.1, 71.2, 73.1,
NMR (CDCl ) d 3.43 (6H, s), 3.45 (6H, s), 3.77 (8H, m),
3
13
4.28 (4H, m), 10.43 (1H, s); C NMR (CDCl ) d 59.7,
3
3
7
3.3, 105.0, 119.3, 132.8, 136.7, 211.2.
59.8, 70.8, 70.9, 73.8, 73.9, 112.6, 121.3, 134.9, 139.0,
187.3.
3
.4. (R,R,R,R)-9-Formyl-10-pivaloyl-2,3,6,7-tetramethoxy-
methyl-1,4,5,8-tetraoxa-1,2,3,4,5,6,7,8-octahydroanthra-
cene 9
3.6. (R,R,R,R)-9-Methyl-2,3,6,7-tetramethoxymethyl-
1
,4,5,8-tetraoxa-1,2,3,4,5,6,7,8-octahydroanthracene 11
A solution of 8 (147mg, 0.32mmol) in CH Cl (7mL)
2
2
was cooled to 0ꢁC under an argon atmosphere. SnCl
To a solution of aldehyde 6 (398mg, 1.0mmol) and bor-
on trifluoride diethyl etherate (0.39mL) in THF (5mL)
was added NaBH CN (165mg, 2.5mmol). The mixture
3
4
(
(
76lL, 0.64mmol) and dichloromethyl methyl ether
35lL, 0.39mmol) were sequentially added dropwise
via a syringe and the reaction mixture refluxed for 2h.
The reaction mixture was poured into ice water and
extracted with CH Cl . The organic layer was washed
was refluxed for 15h and then cooled to room tempera-
ture and poured into EtOAc. The organic layer was
washed with saturated NaHCO (aq) and brine. The or-
2
2
3
with saturated NaHCO aqueous solution and brine,
3
ganic layer was dried over Na SO and evaporated.
2
4
and dried over Na SO . The solvent was evaporated
2
Purification of the residue by silica gel column chroma-
tography (EtOAc/n-hexane) afforded 11 as a yellow solid
4
and purification of the residue by silica gel column
chromatography (EtOAc/n-hexane) afforded 9 as a
1
(345mg, 0.90mmol, 90% yield); H NMR (CDCl ) d
3
1
yellow solid (117g, 0.24mmol, 75% yield); H NMR
2.09 (3H, s), 3.41 (6H, s), 3.43 (6H, s), 3.69 (8H, m),
1
3
(
(
CDCl ) d 1.24 (9H, s), 3.36 (6H, s), 3.42 (6H, s), 3.64
4.16 (4H, m), 6.38 (1H, s); C NMR (CDCl ) d 8.2,
3
3
4H, dd, J = 3.8Hz, 11.1Hz), 3.73 (4H, m), 4.21 (4H,
59.5, 59.6, 71.4, 71.5, 73.0, 73.1, 102.1, 114.2, 135.1,
136.4.
1
3
m), 10.42 (1H, s); C NMR (CDCl ) d 26.6, 44.8,
3
5
2
9.7, 70.9, 71.0, 73.9, 108.8, 120.7, 132.4, 139.1, 187.3,
11.0.
3.7. (R,R,R,R)-9-Formyl-10-methyl-2,3,6,7-tetramethoxy-
methyl-1,4,5,8-tetraoxa-1,2,3,4,5,6,7,8-octahydroanthra-
cene 12
3
.5. (R,R,R,R)-10-Bromo-9-formyl-2,3,6,7-tetrameth-
oxymethyl-1,4,5,8-tetraoxa-1,2,3,4,5,6,7,8-octahydroan-
thracene 10
Prepared by the same method as 9 (79% yield, yellow so-
1
lid); H NMR (CDCl ) d 2.15 (3H, s), 3.43 (6H, s), 3.44
3
1
3
To a solution of aldehyde 6 (398mg, 1.0mmol) in CCl4
(
(6H, s), 3.74 (8H, m), 4.22 (4H, m), 10.41 (1H, s);
C
5mL) was added bromine (56lL, 1.1mmol) via a syr-
NMR (CDCl ) d 8.6, 59.6, 71.2, 71.3, 72.9, 73.4, 109.1,
3
inge under an argon atmosphere at 0ꢁC. The reaction
mixture was stirred for 2h and poured into water. The
organic layer was separated and the aqueous layer ex-
tracted with CH Cl . The combined organic layer was
117.7, 134.9, 136.6, 200.1.
3.8. (R,R,R,R)-9-Hydroxy-2,3,6,7-tetramethoxymethyl-
1,4,5,8-tetraoxa-1,2,3,4,5,6,7,8-octahydroanthracene 13
2
2
dried over Na SO and evaporated. Purification by silica
2
4
gel column chromatography (EtOAc/n-hexane) afforded
1
A solution of aldehyde 6 (800mg, 2.0mmol) and
m-CPBA (693mg, 4.0mmol) in CH Cl (20mL) was
1
0 as a yellow solid (321mg, 0.67mmol, 67% yield); H
2
2

H. Nakagawa et al. / Tetrahedron: Asymmetry 15 (2004) 3861–3867
3865
refluxed for 7h. The reaction mixture was evaporated
carefully and the residue dissolved in EtOAc. This solu-
tion was washed with saturated NaHCO3 (aq) and
brine, and dried over Na SO . The solvent was evapo-
(EtOAc). Purification was carried out by preparative
TLC (EtOAc/n-hexane) with 1 being obtained as a
1
brown solid (189mg, 0.11mmol, 37% yield); H NMR
(CDCl ) d ꢀ2.71 (2H, br s), 2.66 (24H, s), 3.04 (8H,
2
4
3
rated and the residue dissolved in EtOH (40mL) and
3% NH (aq 24mL). The reaction mixture was stirred
dd, J = 4.6Hz, 11.1Hz), 3.16 (8H, dd, J = 3.7Hz,
11.0Hz), 3.41 (24H, s), 3.71 (16H, d, J = 4.4Hz), 3.97
2
3
1
3
at room temperature under an argon atmosphere for
5h and evaporated. 2M HCl was added to the residue
(8H, m), 4.31 (8H, m), 7.02 (4H, s), 8.78 (8H, s);
C
1
NMR (CDCl ) d 59.0, 59.5, 70.5, 71.4, 73.1, 73.2,
3
and the aqueous layer extracted with EtOAc. The organic
layer was washed with saturated NaHCO3 (aq) and
brine, dried over Na SO and evaporated. Purification
105.6, 106.9, 108.7, 109.2, 120.2, 136.7, 137.7; MS
(FAB, 3-nitrobenzyl alcohol) m/z (relative intensity)
1783 (100), 1784 (32); HRMS (FAB, 3-nitrobenzyl alco-
hol) Calcd for C H N O : 1782.7103. Found:
2
4
of the residue by silica gel column chromatography
(
9
2
110
4
32
EtOAc/n-hexane) afforded 13 as a colorless solid
1782.6995; UV (CH CN) kmax 415nm (e =
3
1
ꢀ1 ꢀ1
(
371mg, 0.96mmol, 48% yield); H NMR (CDCl ) d
239,000cm
M
), 510 (17,000), 584 (6200).
3
3
.39 (12H, s), 3.6–3.7 (8H, m), 4.20(4H, m), 5.51 (1H,
1
3
br s), 6.12 (1H, s); C NMR (CDCl ) d 59.4, 59.5,
3
3
.12. 5,10,15,20-Tetrakis[(R,R,R,R)-10-pivaloyl-2,3,6,7-
7
1.1, 71.2, 73.2, 73.3, 95.6, 125.6, 133.7, 137.0.
tetramethoxymethyl-1,4,5,8-tetraoxa-1,2,3,4,5,6,7,8-octa-
hydroanthracen-9-yl]porphyrin 2
3
1
.9. (R,R,R,R)-9-n-Butoxy-2,3,6,7-tetramethoxymethyl-
,4,5,8-tetraoxa-1,2,3,4,5,6,7,8-octahydroanthracene 14
Prepared by the same method as 1 (39% yield, brown
1
solid); H NMR (CDCl ) d ꢀ2.74 (2H, br s), 1.49
3
A mixture of phenol 13 (231mg, 0.60mmol), n-butyl
bromide (161lL, 1.5mmol), and K CO (207mg,
1
an argon atmosphere. The reaction mixture was poured
into water and extracted with EtOAc. The organic layer
was dried over Na SO and evaporated. Purification of
(36H, s), 2.61 (24H, s), 3.01 (8H, m), 3.16 (8H, dd,
J = 3.7Hz, 11.0Hz), 3.37 (24H, s), 3.67 (16H, m), 3.97
2
3
1
3
.5mmol) in DMF (5mL) was stirred at 80ꢁC under
(8H, m), 4.29 (8H, m), 8.81 (8H, s); C NMR (CDCl )
3
d 27.0, 33.4, 59.0, 59.5, 70.3, 71.1, 73.2, 73.3, 95.0, 102.2,
108.6, 115.3, 131.3, 137.6, 148.5, 191.1; MS (FAB,
3-nitrobenzyl alcohol) m/z (relative intensity) 2118
(100); HRMS (FAB, 3-nitrobenzyl alcohol) Calcd for
2
4
the residue by silica gel column chromatography
EtOAc/n-hexane) afforded 14 as a colorless solid
(
C
(CH
112
H
142
N
4
O
36: 2118.9404. Found: 2118.9390; UV
M ), 510
1
ꢀ1 ꢀ1
(
251mg, 0.57mmol, 95% yield); H NMR (CDCl ) d
3
CN) kmax 416nm (e = 272,000cm
3
0
(
.95 (3H, 7.3Hz), 1.51 (2H, m), 1.69 (2H, m), 3.41
6H, s), 3.42 (6H, s), 3.68 (8H, m), 4.04 (2H, y,
(16,000), 584 (5000).
1
3
J = 6.6Hz), 4.18 (4H, m), 6.29 (1H, s); C NMR
CDCl ) d 13.8, 19.0, 32.1, 59.5, 71.2, 72.7, 73.1, 73.3,
3.13. 5,10,15,20-Tetrakis[(R,R,R,R)-10-bromo-2,3,6,7-tetra-
(
3
methoxymethyl-1,4,5,8-tetraoxa-1,2,3,4,5,6,7,8-octahydro-
anthracen-9-yl]porphyrin 3
9
9.4, 131.1, 136.4, 136.9.
3
.10. (R,R,R,R)-10-n-Butoxy-9-formyl-2,3,6,7-tetrameth-
Prepared by the same method as 1 (47% yield, brown
) d ꢀ2.74 (2H, br s), 2.63
1
solid); H NMR (CDCl
(24H, s), 3.13 (8H, dd, J = 4.5Hz, 11.1Hz), 3.20(8H,
dd, J = 3.5Hz, 11.1Hz), 3.44 (24H, s), 3.76 (16H,
oxymethyl-1,4,5,8-tetraoxa-1,2,3,4,5,6,7,8-octahydroanth-
racene 15
3
1
3
Prepared by the same method as 9 (91% yield, yellow
solid); H NMR (CDCl ) d 0.95 (3H, s), 1.47 (2H, m),
m), 3.99 (8H, m), 4.39 (8H, m), 8.77 (8H, s);
C
1
NMR (CDCl ) d 59.0, 59.8, 70.2, 71.0, 73.1, 73.8, 99.8,
3
3
1
.73 (2H, m), 3.41 (6H, s), 3.43 (6H, s), 3.72 (8H, m),
.17 (2H, t, J = 6.4Hz), 4.25 (4H, m), 10.36 (1H, s);
108.8, 105.0, 118.5, 130.1, 134.9, 137.9; MS (FAB,
3-nitrobenzyl alcohol) m/z (relative intensity) 2096
(100); HRMS (FAB, 3-nitrobenzyl alcohol) Calcd for
4
1
3
C NMR (CDCl ) d 13.8, 18.9, 32.1, 59.5, 59.7, 70.9,
3
7
1.0, 72.7, 73.6, 109.0, 130.5, 139.8, 142.3, 187.0.
C H107Br N O32: 2095.3602. Found: 2095.3796; UV
92 4 4
ꢀ1
ꢀ1
M ), 509
(
CH CN) k
416nm (e = 254,000cm
3
max
(16,400), 583 (5200).
3
.11. 5,10,15,20-Tetrakis[(R,R,R,R)-2,3,6,7-tetramethoxy-
methyl-1,4,5,8-tetraoxa-1,2,3,4,5,6,7,8-octahydroanthracen-
-yl]porphyrin 1
9
3.14. 5,10,15,20-Tetrakis[(R,R,R,R)-10-methyl-2,3,6,7-
tetramethoxymethyl-1,4,5,8-tetraoxa-1,2,3,4,5,6,7,8-octa-
hydroanthracen-9-yl]porphyrin 4
A solution of aldehyde 6 (456mg, 1.1mmol) in CHCl3
114mL) was purged with argon gas for 30min. Under
an argon atmosphere, boron trifluoride diethyl etherate
48lL, 0.38mmol) and then pyrrole (79lL, 1.1mmol)
(
Prepared by the same method as 1 (32% yield, brown
1
(
solid); H NMR (CDCl ) d ꢀ2.69 (2H, br s), 2.46
3
were added slowly via a syringe and the mixture stirred
for 1h. Chloranil (211mg, 0.86mmol) was added to the
mixture and the reaction mixture was refluxed for
another 1h. The reaction mixture was then cooled to
room temperature and triethylamine (53lL) and a
pad of silica gel added. The mixture was evaporated to
dryness and passed through a short silica gel column
(12H, s), 2.64 (24H, s), 3.07 (8H, dd, J = 4.8Hz,
11.0Hz), 3.17 (8H, dd, J = 3.9Hz, 11.0Hz), 3.42 (24H,
s), 3.73 (16H, d, J = 4.4Hz), 3.94 (8H, m), 4.31 (8H,
1
3
m), 8.76 (8H, s); C NMR (CDCl ) d 8.7, 59.1, 59.6,
3
70.5, 71.5, 72.8, 73.0, 109.4, 111.1, 114.6, 116.4, 134.8,
137.3, 142.2; MS (FAB, 3-nitrobenzyl alcohol) m/z (rela-
tive intensity) 1839 (100), 1840 (37); HRMS (FAB,

3866
H. Nakagawa et al. / Tetrahedron: Asymmetry 15 (2004) 3861–3867
ꢀ1
ꢀ1
M ), 511
3
-nitrobenzyl alcohol) Calcd for C H N O :
9
UV (CH CN) k
415nm (e = 71,500cm
6
118
4
32
3
max
1
838.7762. Found: 1838.7720; UV (CH CN) k
417
max
(11,900).
3
ꢀ1
ꢀ1
M ), 511 (18,500), 586 (5500).
nm (e = 275,000cm
3
6
.19. Bromo-[5,10,15,20-tetrakis[(R,R,R,R)-10-methyl-2,3,
,7-tetramethoxymethyl-1,4,5,8-tetraoxa-1,2,3,4,5,6,7,8-
3
.15. 5,10,15,20-Tetrakis[(R,R,R,R)-10-n-butoxy-2,3,6,7-
tetramethoxymethyl-1,4,5,8-tetraoxa-1,2,3,4,5,6,7,8-octa-
hydroanthracen-9-yl]porphyrin 5
octahydroanthracen-9-yl]porphyrinato]iron(III) 4–Fe(Br)
Prepared by the same method as 1–Fe(Br) (61% yield,
brown solid); MS (FAB, 3-nitrobenzyl alcohol) m/z (rel-
ative intensity) 1893 (100); HRMS (FAB, 3-nitrobenzyl
alcohol) Calcd for C H FeN O : 1892.6922. Found:
Prepared by the same method as 1 (32% yield, brown so-
1
lid); H NMR (CDCl ) d 2.71 (2H, br s), 1.10(12H, t,
3
J = 7.4Hz), 1.71 (8H, m), 1.94 (8H, m), 2.63 (24H, s),
3.08 (8H, dd, J = 4.9Hz, 11.1Hz), 3.17 (4H, dd,
J = 4.0Hz, 11.1Hz), 3.41 (24H, s), 3.74 (16H, m), 3.95
9
6
116
4
32
1
cm
892.6775; UV (CH CN) k
416 nm (e = 72,400
max
3
ꢀ
1
ꢀ1
M ), 511 (12,600).
(
(
8H, m), 4.32 (8H, m), 4.39 (8H, t, J = 6.6Hz), 8.76
1
3
8H, s); C NMR (CDCl ) d 14.0, 19.3, 32.4, 59.0,
3
3.20. Bromo-[5,10,15,20-tetrakis[(R,R,R,R)-10-n-butoxy-
2,3,6,7-tetramethoxymethyl-1,4,5,8-tetraoxa-1,2,3,4,5,6,
7,8-octahydroanthracen-9-yl]porphyrinato]iron(III) 5–Fe(Br)
5
9.5, 70.4, 71.3, 72.7, 73.0, 73.8, 109.0, 113.8, 130.9,
32.1, 133.4, 136.9, 137.7; MS (FAB, 3-nitrobenzyl alco-
1
hol) m/z (relative intensity) 2074 (100); HRMS (FAB,
3
2
4
-nitrobenzyl alcohol) Calcd for C108H142N O :
4
36
Prepared by the same method as 1–Fe(Br) (54% yield,
brown solid); MS (FAB, 3-nitrobenzyl alcohol) m/z
070.9404. Found: 2070.9316; UV (CH CN) k
3
max
ꢀ1
ꢀ1
17nm (e = 237,000cm
M
), 511 (16,700), 586
(relative intensity) 2124 (100); HRMS (FAB, 3-nitro-
benzyl alcohol) Calcd for C108H140N O : 2124.8597.
(
5100).
4
36
Found: 2124.8650; UV (CH CN) kmax 416nm
3
3
.16. Bromo-[5,10,15,20-tetrakis[(R,R,R,R)-2,3,6,7-tetra-
ꢀ1 ꢀ1
(
e = 74,700cm
M
), 512 (12,900).
methoxymethyl-1,4,5,8-tetraoxa-1,2,3,4,5,6,7,8-octahydro-
anthracen-9-yl]porphyrinato]iron(III) 1–Fe(Br)
3.21. Typical procedure for asymmetric epoxidation of
aromatic olefins
To a mixture of porphyrin 1 (150mg, 84 lmol) and
iodine (213mg, 0.84mmol) in toluene (15mL) was
added Fe(CO)5 (1.9mL, 15mmol) via a syringe
under an argon atmosphere. The mixture was ref-
luxed for 5h and then passed through an alumina
short column (CH Cl ). The porphyrin fraction was
A mixture of styrene (28lL, 0.25mmol) and a catalyst
0.25lmol) in dry toluene (500lL) was cooled to
20ꢁC under an argon atmosphere. Iodosylbenzene
5.5mg, 25lmol) was added and the reaction stirred
(
ꢀ
(
2
2
for 3h. Triphenylphosphine (33mg, 0.13mmol) in tolu-
ene (100lL) was then added to stop the reaction. The
reaction mixture was analyzed by HPLC or the epoxide
washed with 5% aqueous HBr solution and dried with
KBr. The solvent was evaporated and recrystallization
(
(
CH Cl /n-hexane) afforded 1–Fe(Br) as a brown solid
2 2
88mg, 46lmol, 55% yield); MS (FAB, 3-nitrobenzyl
was purified by silica gel column chromatography and
analyzed by H NMR with (+)-Eu(hfc) . The absolute
1
alcohol) m/z (relative intensity) 1838 (100), 1839 (31);
HRMS (FAB, 3-nitrobenzyl alcohol) Calcd for
C H FeN O : 1836.6296. Found: 1836.6250; UV
3
configuration was determined by a comparison with an
authentic sample.
9
2
108
4
32
ꢀ1
ꢀ1
M ), 511
(
CH CN) kmax 414nm (e = 59,700cm
3
(
10,300).
Acknowledgements
3
2
7
.17. Bromo-5,10,15,20-tetrakis((R,R,R,R)-10-pivaloyl-
,3,6,7-tetramethoxymethyl-1,4,5,8-tetraoxa-1,2,3,4,5,6,
,8-octahydroanthracen-9-yl)porphyrinato]iron(III) 2–Fe(Br)
This work was supported in part by Grants-in-aid for
Scientific Research (No. 12793009 and No. 11470494)
from the Ministry of Education, Culture, Sports, Sci-
ence and Technology, Japan.
Prepared by the same method as 1–Fe(Br) (58% yield,
brown solid); MS (FAB, 3-nitrobenzyl alcohol) m/z
(
benzyl alcohol) Calcd for C112H140FeN O : 2172.8597.
relative intensity) 2174 (100); HRMS (FAB, 3-nitro-
4
36
References
Found: 2172.8603; UV (CH CN) kmax 415 nm
3
ꢀ
1
ꢀ1
M ), 511 (13,100).
(
e = 79,700cm
1. Groves, J. T.; Myers, R. S. J. Am. Chem. Soc. 1983, 105,
5
791.
. For reviews, see: (a) Collman, J. P.; Zhang, X.; Lee, V. J.;
2
3
.18. Bromo-[5,10,15,20-tetrakis((R,R,R,R)-10-bromo-2,3,
Uffelman, E. S.; Brauman, J. I. Science 1993, 261, 1404;
6,7-tetramethoxymethyl-1,4,5,8-tetraoxa-1,2,3,4,5,6,7,8-
octahydroanthracen-9-yl)porphyrinato]iron(III) 3–Fe(Br)
(
(
b) Besse, P.; Veschambre, H. Tetrahedron 1994, 50, 8885;
c) Campbell, L. A.; Kodadek, T. J. Mol. Catal. A–Chem.
1
996, 113, 293; (d) Marchon, J.-C.; Ramasseul, R.
Prepared by the same method as 1–Fe(Br) (55%
yield, brown solid); MS (FAB, 3-nitrobenzyl alco-
hol) m/z (relative intensity) 2153 (100), 2155 (52);
HRMS (FAB, 3-nitrobenzyl alcohol) Calcd for
C H Br FeN O : 2148.2717. Found: 2148.2878;
In Handbook of Porphyrins and Phthalocyanines; Kadish,
K. M., Smith, K. M., Guilard, R., Eds.; Academic: New
York, 2003; Vol. 11, p 75; (e) Naruta, Y. In Metallopor-
phyrins in Catalytic Oxidations; Sheldon, R. A., Ed.;
Marcel Dekker: New York, 1994; p 241.
9
2
104
4
4
36

H. Nakagawa et al. / Tetrahedron: Asymmetry 15 (2004) 3861–3867
3867
3
. Collman, J. P.; Lee, V. J.; Kellen-Yuen, C. J.; Zhang, X.;
Ibers, J. A.; Brauman, J. I. J. Am. Chem. Soc. 1995, 117,
12. Palucki, M.; Finney, N. S.; Pospisil, P. J.; G u¨ ler, M. L.;
Ishida, T.; Jacobsen, E. N. J. Am. Chem. Soc. 1998, 120,
948.
6
92.
. Gross, Z.; Ini, S. J. Org. Chem. 1997, 62,
514.
4
5
6
7
13. The major enantiomer of the cis-oxide had a (1S,2R)-
configuration, which was derived from minor benzyl
cation intermediate. The cis/trans ratio at the ring closure
step should be different between the major pathway and
the minor one, because the major and minor intermediates
are diastereomers and therefore the lifetime and the steric
circumstances between these intermediates would differ.
There would be a tendency that the favored intermediate
selectively affords the major diastereomer of the product,
while the minor intermediate preferentially gives the minor
diastereomer: Zhang, W.; Lee, N. H.; Jacobsen, E. N. J.
Am. Chem. Soc. 1994, 116, 425.
14. In iron porphyrin-catalyzed oxidation with iodosylarenes,
a Ph-IO-ligated complex of the catalyst is proposed to be a
major active oxidant in the presence of olefins. For the
mechanism of the oxidation by metalloporphyrin with
PhIO, see: (a) Collman, J. P.; Chien, A. S.; Eberspacher, T.
A.; Brauman, J. I. J. Am. Chem. Soc. 2000, 122, 11098; (b)
Nam, W.; Lim, M. H.; Lee, H. J.; Kim, C. J. Am. Chem.
Soc. 2000, 122, 6641; (c) Machii, K.; Watanabe, Y.;
Morishima, I. J. Am. Chem. Soc. 1995, 117, 6691; (d) Nam,
W.; Valentine, J. S. J. Am. Chem. Soc. 1990, 112, 4977.
15. Berkessel, A.; Kaiser, P.; Lex, J. Chem. Eur. J. 2003, 9,
4746.
5
. Lai, T.-S.; Ng, K.-H.; Liu, H.-Y.; Chang, C.-K.; Yeung,
L.-L. Synlett 2002, 1475.
. Liu, S.-Q.; P e´ caut, J.; Marchon, J. C. Eur. J. Inorg. Chem.
2
002, 7, 1823.
. (a) Naruta, Y.; Tani, F.; Ishihara, N.; Maruyama, K. J.
Am. Chem. Soc. 1991, 113, 6865; (b) Naruta, Y.; Ishihara,
N.; Tani, F.; Maruyama, K. Bull. Chem. Soc. Jpn. 1993,
66, 158.
. Nakagawa, H.; Sei, Y.; Yamaguchi, K.; Nagano, T.;
Higuchi, T. J. Mol. Catal. A–Chem. 2004, 219,
21.
. Only a small amount of by-product via H -migration
phenylacetaldehyde derivatives) was detected in each
reaction (<2%).
0. Traylor, T. G.; Xu, F. J. Am. Chem. Soc. 1988, 110,
953.
8
9
2
ꢀ
(
1
1
1
1. In Fig. 2, the plot of 4-substituted styrenes deviates from
the linear correlation rather than that of 3-substituted
styrenes. The deviation would be caused by the steric or
electronic interaction between the substituents at the
4-position of styrenes and methoxymethyl moieties of
the porphyrin.