N.Ch. Maity et al. / Journal of Catalysis 277 (2011) 123–127
125
about 5-h reaction time in THF, formation of the product 1b0 is
epoxidation of STR as representative substrate was carried with
catalysts 1a and 1b under identical reaction conditions (Table S1,
entries 1, 8–10, 17, 24–26). Among these, CH2Cl2 (Table S1, entries
1 and 17) was found to be the best solvent. It is reported in the lit-
erature [20,22] that pyridine N-oxide derivatives greatly influence
catalyst turnover and enantioselectivity in Mn(III) salen catalyzed
epoxidation of alkenes with NaOCl. According to these studies, N-
oxides prevent the formation of catalytically inactive Mn–O–Mn
species [26] as well as improve the transportation of HOCl from
aqueous phase to the organic phase. Hydrophobic N-oxides e.g.,
4-phenylpyridine N-oxide (4-PPyNO) and 4-(3-phenylpropyl) pyr-
idine N-oxide (4-PPPyNO) have been reported to stabilize the reac-
tive Mn@O intermediate [27]. In our optimized condition for the
epoxidation of styrene, the use of 4-PPyNO and 4-PPPyNO as axial
bases with catalysts 1a and 1b (Table S1, entries 11, 12, 27, 28) had
no obvious advantage over the simple and inexpensive PyNO. This
finding is in contrast to the results obtained with the use of Mn(III)
salen complex [27] but is in line with the use of built-in phase
transfer Mn(III) salen complexes reported by us [28,29]. A control
experiment for the epoxidation of STR conducted without PyNO
gave only 40–48% styrene oxide in 3 h with low enantioselectivity
(Table S1, ee 18–22%; entries 13, 29), demonstrated the positive
role of PyNO as an axial base.
We also explored the oxidant-UHP in the epoxidation of STR
with catalysts 1a and 1b in the presence of different N-oxides
viz., PyNO, 4-PPyNO, and 4-PPPyNO at 3 °C in CH2Cl2:CH3OH 1:1
(Table S1, entries 14–16, 30–32). Excellent conversion (>99%) to
styrene oxide was achieved in 5–6 h with both the catalysts, how-
ever, catalyst 1b imparted better ee (ꢀ59%) when compared to the
catalyst 1a (ee, ꢀ41%), which fared better with UHP than NaOCl.
The optimized epoxidation condition (as per entries 1 and 2; Ta-
ble 1) was applied for other non-functionalized alkenes, to see the
general applicability of catalysts 1a and 1b using PyNO as an axial
base with NaOCl as an oxidant. Excellent conversion to epoxides
(>99%) and ee (Table 1) was observed for all the substrates (entries
1–14) in 3–8 h at 0 °C. As far as enantioselectivity is in concern, the
catalyst 1b worked much better for STR (ee, 59%) and IND (ee, 91%)
than the catalyst 1a, which gave ee 33% and 75%, respectively.
However, in the case of chromenes both the catalysts imparted
1
optimum (ꢀ75%). The H NMR and LCMS data for ligand 1b0 show
the presence of monomer (n = 0), dimer (n = 1) and trimeric species
in 80%, 16% and 4%, abundance, respectively. These mono-, di- and
trimeric salen species were separated by neutral alumina column
chromatography with benzene and acetone (7:3) as eluent. Ab-
sence of aldehyde peak in NMR indicates that all the three species
were cyclic. This mixture of species was used as such for making
the chiral macrocyclic Mn(III) salen complex, as complexes of
above separated ligands gave nearly similar performance. The
monomeric 1a0 gave single crystals suitable for X-ray structure
determination from acetone in 2–3 days, which further established
its cyclic structure (Fig. S1) [24]. The ligand 1a0/1b0 on reaction
with Mn(OAc)2ꢁ4H2O and subsequent air oxidation yielded the de-
sired complex 1a/1b in quantitative yield. Numerous attempts to
get single crystals of complexes 1a and 1b failed; hence, we ex-
plored the probable structure of 1a using energy minimization op-
tion (Fig. S2) in Material Studio (version 4.2). The crystallographic
data of ligand 1a0 were used for simulating the structure of com-
plex 1a, which lacked C2-symmetry (or distorted C2-symmetry)
as against stepped conformation [25] of the Jacobsen’s Mn(III)
salen complex. The cause of this distortion is due to the shorter
length of the linker group. Both the complexes 1a and 1b
(5 mol%) were used to catalyze the enantioselective epoxidation
of STR as a representative substrate using NaOCl as an oxidant in
the presence of PyNO as an axial base, and the results are given
in Table S1. Excellent yield (>99%) of styrene oxide was achieved
in 3 h with both the catalyst. However, the ee of styrene oxide
was superior with catalyst 1b (Table S1, entry 17). Epoxidation of
STR was conducted with 2.5–10 mol% catalyst loading and the re-
sults indicated that 5 mol% (Table S1, entries 1–3, 17–19) is opti-
mum at 0 °C. Epoxidation of STR over a temperature range of ꢂ5
to 20 °C at 5 mol% catalyst loading concluded that 0 °C is optimum
reaction temperature (Table S1, entries 5–7, 21–23). Having opti-
mized catalyst loading and reaction temperature next, we opti-
mized reaction medium, as solvent is known to play a crucial
role in the enantioselective epoxidation of non-functionalized al-
kenes [17,22]. In view of this, the effect of solvents viz. CH2Cl2,
dichloroethane, CHCl3 and mixture of (CH2Cl2 + CH3CN) on the
Table 1
Asymmetric epoxidation of alkenes catalyzed by 1a and 1b with NaOCl as an oxidant.
R
R'
catalyst (5 mol%), NaOCl (1.5 mmol)
PyNO (0.12 mmol), 0 oC
R
R'
*
*
H
H
O
Entrya
Catalyst
Substrate
STR
% Yieldb
Time (h)
ee (%)
Configuration
TOF ꢃ 10ꢂ3 sꢂ1f
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
1a
1b
1a
1b
1a
1b
1a
1b
1a
1b
1a
1b
1a
1b
>99
>99
>99
3
3
4.5
4
6
5
6
5
8
7
6
5
6
5
9
2
33c
59c
75d
91d
86e
84e
93e
92e
73e
77e
90e
91e
93e
95e
92
S
R
1.83
1.83
1.22
1.38
0.92
1.10
0.92
1.10
0.69
0.79
0.92
1.10
0.92
1.10
0.37
4.17
IND
1S,2R
1R,2S
3S,4S
3R,4R
3S,4S
3R,4R
3S,4S
3R,4R
3S,4S
3R,4R
3S,4S
3R,4R
3S,4S
3S,4S
Cy-CHR
CHR
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
60
MeO-CHR
NO2-CHR
CN-CHR
Jacobson’s catalyst
Katsuki’s catalyst
CN-CHR
CHR
75
99
a
b
c
d
e
f
Reaction condition: catalyst (5 mol% in 1 ml CH2Cl2), substrate (0.625 mmol), pyridine N-oxide (0.12 mmol), NaOCl (1.5 mmol).
Determined on GC.
Chiral capillary column GTA-type.
Chiral HPLC OB column.
Chiral HPLC OD column.
Turn over frequency is calculated by the expression [product]/[catalyst] ꢃ time (sꢂ1).