Solvent effects on catalytic epoxidation of alkenes by tetra-n- butylammonium periodate and (Tetraarylporphyrinato)manganese(III)

Article abstract of DOI:10.1002/ejic.200800253

The epoxidation of cyclooctene and cis-stilbene was performed by tetra-n-butylammonium periodate (nBu₄NIO₄) in the presence of (tetraarylporphyrinato)manganese(III) [Mn(por)] and imidazole (im) in various CH₂Cl₂

Full text of DOI:10.1002/ejic.200800253

FULL PAPER
DOI: 10.1002/ejic.200800253
Solvent Effects on Catalytic Epoxidation of Alkenes by Tetra-n-
butylammonium Periodate and (Tetraarylporphyrinato)manganese(III)
Daryoush Mohajer*[a] and Massoumeh Abbasi[a]
Keywords: Manganese / Porphyrinoids / Homogeneous catalysis / Epoxidation / Solvent effects / Oxidation
The epoxidation of cyclooctene and cis-stilbene was per-
formed by tetra-n-butylammonium periodate (nBu4NIO4)
in the presence of (tetraarylporphyrinato)manganese(III)
[Mn(por)] and imidazole (im) in various CH2Cl2/alcohol sol-
vents (alcohol = CH3OH, C2H5OH, n-C3H7OH, i-C3H7OH, t-
C4H9OH). In accord with the coordinating abilities of the
alcohols to [Mn(por)], the epoxidation yields increased from
CH2Cl2/CH3OH to CH2Cl2/t-C4H9OH. In the epoxidation of
cis-stilbene in the presence of (acetato)(tetraphenylpor-
phyrinato)manganese(III) [Mn(tpp)(OAc)], the cis- to trans-
stilbene oxide ratio increased consistently with the bulk of
the alcohol in CH2Cl2/alcohol solvents. Also, it was found
that the interaction of [Mn(tpp)(OAc)] and (acetato)(tetra-
mesitylporphyrinato)manganese(III) [Mn(tmp)(OAc)] with
nBu4NIO4 in the presence of imidazole but in the absence of
alkenes yields [Mn(tpp)(O)(im)]+ and [Mn(tmp)(O)(im)]+
complexes. The formation of the Mn=O species for the small
linear alcohols was faster than that for the larger ones,
whereas the stability of the Mn=O complex was greater in
the presence of the bulkier alcohols. Attempts were made to
explain these effects.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2008)
bond cleavage.[4] Furthermore, the product ratios for com-
petitive epoxidation of cis- and trans-stilbene have been
correlated with the acidity of the alcohol.[5] In contrast, the
generation, characterization, and reactivity studies of
(oxido)(porphyrinato)manganese intermediates, under vari-
ous reaction conditions, have received considerable atten-
tion due to their significant roles in the oxygenation of hy-
drocarbons and as models for P-450 enzymes.[6]
In this report, we demonstrated that the epoxidation
yields of cyclooctene and cis-stilbene by a mild single-oxy-
gen donor tetra-n-butylammonium periodate (nBu4NIO4)[7]
in the presence of (tetraarylporphyrinato)manganese(III)
[Mn(por)], Figure 1, and imidazole (im) are critically de-
pendent upon the nature of the alcohols in CH2Cl2/alcohol
solvents. Apparently, the coordinating abilities of the
alcohols and their steric and hydrogen-bonding properties
can make great differences in terms of the epoxidation
yields, cis- to trans-stilbene oxide ratios, and the formation
and stabilization of [Mn(por)(O)(im)]+ species. The advan-
tage of utilizing neutral nBu4NIO4 rather than H2O2 or m-
chloroperoxybenzoic acid as oxidant in this study is that
nBu4NIO4 is a nonhydrogen-bond donor species and would
not interfere with the hydrogen bonding of the alcohols. For
the case in which the oxidants contain strong hydrogen-
bond donor O–H group(s), the hydrogen bonding of the
alcohols are apparently disguised by those of the oxidants
and consequently no definitive conclusions can be drawn
concerning their effects.
Introduction
Since the pioneering work of Groves and coworkers on
the development of synthetic metalloporphyrin models of
cytochrome P-450, considerable efforts have been devoted
to the understanding of the factors that may control the
activity and selectivity of alkene epoxidation by using vari-
ous metalloporphyrin catalytic systems.[1] Particular atten-
tion has been given to the ratios of cis/trans-stilbene oxide
obtained from cis-stilbene epoxidation as a probe for the
stereoselectivity of P-450 model systems.[1b,2] The exact ra-
tio of the cis to trans epoxides in these cases, which depends
upon the nature of the catalyst and the reaction conditions,
has been taken as evidence for the presence of a transient
radical species in the general reaction scheme that allows
limited rotation about the former alkene bond prior to ep-
oxide ring closure.[1b]
Recent studies have also indicated that the composition
of alcohol/acetonitrile and alcohol/CH2Cl2 have important
effects on the epoxidation reactions by hydrogen peroxide
and iron porphyrins.[3] It has been proposed that alcohols
function as general acid catalysts and can facilitate O–O
[a] Department of Chemistry, Shiraz University,
Shiraz 71454, Iran
Fax: +98-711-2280926
E-mail: dmohajer@chem.susc.ac.ir
Supporting information for this article is available on the
WWW under http://www.eurjic.org or from the author.
3218
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2008, 3218–3224

Solvent Effects on Catalytic Epoxidation of Alkenes
n-C3H7OH Ͻ i-C3H7OH Ͻ t-C4H9OH. This seems to be
quite consistent with the reversed order of coordinating
abilities of the alcohols to [Mn(por)].[9]
In order to see how the presence of coordinating CH3OH
versus noncoordinating t-C4H9OH can affect the epoxid-
ation yields, we performed two different series of epoxid-
ation reactions. In the first series, the only variable was the
ratio of im/[Mn(tpp)(OAc)], whereas in the second series,
the amounts of the alcohols were changed at a fixed
im/[Mn(tpp)(OAc)] ratio.
It was found that the cyclooctene epoxidation yields were
quite dependent upon the im/[Mn(tpp)(OAc)] ratio in the
various solvents (Table 2). They initially increased to a
maximum value by increasing the im/[Mn(tpp)(OAc)] ratio,
and they then decreased at higher ratios (Figure 2). The in-
teraction of imidazole with [Mn(tpp)(OAc)] is expected to
lead to the formation of both the catalytically active 1:1
[Mn(tpp)(OAc)(im)] and the catalytically inactive 2:1
[Mn(tpp)(im)2]OAc complexes in equilibrium with
[Mn(tpp)(OAc)].[10] Before the maximum yield was reached,
the concentration of the 1:1 complex continuously increased
by adding imidazole, whereas beyond the maximum yield
the further addition of imidazole led to the formation of
the inactive 2:1 complex at the expense of the 1:1 complex,
which caused a reduction in the epoxidation yields.
Figure 1. (Tetraarylporphyrinato)manganese(III) complexes used
in this study.
Results and Discussion
Cyclooctene Epoxidation
The results of cyclooctene epoxidation by nBu4NIO4 in
the presence of four [Mn(por)] units in CH2Cl2 containing
various alcohols with different acidities (pKa values) and
steric properties are presented in Table 1. The epoxidation
reactions proceeded with absolute selectivity (100%) and
yielded only cyclooctene oxide. A control experiment for
the “best” catalytic system indicated that the % of effec-
tively transferred oxygen was 57 % in 5 h and 89% after
24 h (Table 1, Entry 5). In the absence of imidazole the
epoxidation yields were very low for both CH2Cl2/alcohol
solvents (Ͻ7%) and neat CH2Cl2 (Ͻ3%; Table 1, Entries 7,
8). Consideration of the reaction times and the epoxidation
yields clearly suggested the following order of catalytic ac-
tivities for [Mn(por)]: [Mn(tpfpp)(OAc)] Ͻ [Mn(tdcpp)-
(OAc)] Ͻ [Mn(tmp)(OAc)] Ͻ [Mn(tpp)(OAc)]. Accordingly,
the electron-poor and/or bulky [Mn(por)] are less-effective
catalysts than the electron-rich ones with less steric hin-
drance.[8] In contrast, the epoxidation yields for [Mn(por)]
were found to be directly related to the steric bulk of the
Table 2. Cyclooctene epoxidation yields (%) for different im/
[Mn(tpp)(OAc)] ratios in various solvents.[a]
Solvent
im/[Mn(tpp)(OAc)] molar ratio
0
1
5
10
25
50 100 200
CH2Cl2
CH2Cl2/CH3OH
CH2Cl2/t-C4H9OH
Ͻ3 40
68
24
78
45
70
65
56
69
83
37
61
72
24
50
67
7
5
10
49
81 100 98
[a] The molar ratio of [Mn(tpp)(OAc)]/im/cyclooctene/periodate
was 1/x/83/167; (x = 0, 1, 5, 10, 25, 50, 100, 200) with a concentra-
tion of [Mn(tpp)(OAc)] of 3ϫ10–3 . The reactions were carried
alcohols and increased in the order CH3OH Ͻ C2H5OH Ͻ out for 5 h at 22Ϯ2 °C.
Table 1. Epoxidation of cyclooctene by nBu4NIO4 in the presence of [Mn(por)] catalysts in various solvents.[a,b]
[c]
Entry
Solvent
pKa
[Mn(tpp)(OAc)]
Yield / %, 5 h
[Mn(tmp)(OAc)]
Yield / %, 5 h
[Mn(tdcpp)(OAc)]
Yield / %, 24 h
[Mn(tpfpp)(OAc)]
Yield / %, 24 h
1
2
3
4
5
6
7
8
CH2Cl2/CH3OH[d]
CH2Cl2/C2H5OH[d]
CH2Cl2/n-C3H7OH[d]
CH2Cl2/i-C3H7OH[d]
CH2Cl2/t-C4H9OH[d]
15.3
15.9
16.1
17.1
19
45
58
69
17
25
34
42
41
45
49
57
76
11
14
17
28
38
86
100 (57)[e]
78
60
[f]
CH2Cl2
58
5 (Ͻ2)[h]
0
56
16
CH2Cl2/CH3OH[g]
7 (Ͻ3)[h]
Ͻ3
5 (0)[h]
0
6 (0)[h]
0
[i]
CH2Cl2
[a] The molar ratio for [Mn(por)]/im/cyclooctene/periodate was 1:10:83:167 with a concentration of [Mn(por)] of 3ϫ10–3  at 22 Ϯ2 °C.
[b] The epoxidation yields (%) were measured relative to starting cyclooctene. All the reactions were run at least in triplicate, and the
data represent an average of these reactions with Ϯ8–10%. [c] The pKa values were obtained from ref.[12] [d] The reactions were carried
out in CH2Cl2 (1 mL)/alcohol (0.5 mL). [e] The yield for a control experiment at the molar ratio of [Mn(tpp)(OAc)]/im/cyclooctene/
periodate, 1:10:83:83. The % yield increased with time to 74% (8 h) and 89% (24 h), and then the reaction was discontinued. [f] The
reactions were performed in CH2Cl2 (1 mL) with no alcohol present. [g] The reactions were carried out in CH2Cl2 (1 mL)/CH3OH
(0.5 mL) with no imidazole. [h] The numbers in the parentheses are for CH2Cl2/t-C4H9OH, under similar conditions. [i] The reactions
were carried out in neat CH2Cl2 (1 mL) with no imidazole.
Eur. J. Inorg. Chem. 2008, 3218–3224
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
3219

D. Mohajer, M. Abbasi
FULL PAPER
Figure 2. Cyclooctene epoxidation yield (%) vs. im/[Mn(tpp)(OAc)]
molar ratio (or vs. number of mmols of imidazole).
Figure 3. Cyclooctene epoxidation yields (%) vs. number of µL of
CH3OH or t-C4H9OH in CH2Cl2 (1 mL).
amount of CH3OH. However, the solvation process would
reach its saturation level at some concentration of CH3OH
(≈50 µL) and t-C4H9OH (≈100 µL). Beyond these limits,
further addition of the coordinating CH3OH should cause
a reduction in the catalytic activity of [Mn(tpp)(OAc)],
whereas for the noncoordinating t-C4H9OH no particular
effect is expected and observed.
It is notable that the epoxidation curves for both neat
CH2Cl2 and CH2Cl2/t-C4H9OH solvents generally dis-
played sharper increases and decreases than that of CH2Cl2/
CH3OH (Figure 2). This is presumably related to the coor-
dination of CH3OH to [Mn(tpp)(OAc)] to form
[Mn(tpp)(OAc)(CH3OH)] and [Mn(tpp)(CH3OH)2]OAc;[9]
hence, the formation of both catalytically active
[Mn(tpp)(OAc)(im)] and inactive [Mn(tpp)(im)2]OAc spe-
cies was hindered. However, the greater epoxidation yields
observed in CH2Cl2/t-C4H9OH as solvent than that ob-
tained in neat CH2Cl2 at all im/[Mn(tpp)(OAc)] ratios must
reflect the important solvation effects of t-C4H9OH.[9]
In contrast, the cyclooctene epoxidation yields varied
very differently by analogous changes in the amounts of
CH3OH and t-C4H9OH in CH2Cl2 (Table 3). The addition
of 10 to 100 µL of t-C4H9OH into the reaction solution
gradually increased the epoxidation yields to a fixed maxi-
mum value, and further addition of the alcohol had no ef-
fect (Figure 3). In the case of CH3OH, the maximum epox-
idation yield was achieved at a much smaller amount of the
alcohol (≈25 µL) and more addition of CH3OH (ca.
Ͼ50 µL) decreased the epoxidation yields.
cis-Stilbene Epoxidation
The results of epoxidation of cis-stilbene in the presence
of varied [Mn(por)] in various CH2Cl2/alcohol solvents are
given in Table 4. Although the oxidation of cis-stilbene by
the [Mn(tpp)(OAc)] catalyst led to various mixtures of cis-
and trans-stilbene oxides, its reactions with other [Mn(por)]
catalysts proceeded with absolute stereoselectivity and pure
cis-stilbene oxide was the sole product. Again, the epoxid-
ation yields were very sensitive to the nature of the alcohols
and increased in the order CH3OH Ͻ C2H5OH Ͻ n-
C3H7OH Ͻ i-C3H7OH Ͻ t-C4H9OH for every [Mn(por)].
A novel finding was that the cis/trans-stilbene oxide ratio
in the epoxidation of cis-stilbene by [Mn(tpp)(OAc)] was
dependent upon the molecular size of the alcohol and in-
creased as the steric bulk of the alcohol increased. For the
linear alcohols, the ratios were rather close (≈14Ϯ2;
Table 4, Entries 1–3), whereas for the bulky alcohols they
were quite different and much higher than those for the
linear ones (Table 4, Entries 4, 5). These results suggest that
the steric properties of the alcohols can have an important
role in determining the ratio of cis/trans-stilbene oxide.
Formation of hydrogen bonds between the alcohols and
the expected active oxidant(s)[8b] can cause significant steric
hindrance at the oxygenation site (Figure 4). In the course
of epoxidation of cis-stilbene, the formation of trans-stilb-
Table 3. Cyclooctene epoxidation yields (%) in various CH2Cl2/
CH3OH and CH2Cl2/t-C4H9OH solvents.[a]
Solvents
Alcohol (µL)/CH2Cl2 (1 mL) ratio
0
10
25
50
100
200
500
CH2Cl2/CH3OH
CH2Cl2/t-C4H9OH 78
78
80
83
85
89
85
95
75
100
62
100
45
100
[a] The molar ratio for [Mn(tpp)(OAc)]/im/cyclooctene/periodate
was 1:10:83:167 with concentration of [Mn(tpp)(OAc)] of
3ϫ10–3  in CH2Cl2 at 22Ϯ2 °C.
a
The initial increases observed in the epoxidation yields ene oxide requires free rotation about the alkene C=C
by adding small amounts of CH3OH (Ͻ25 µL) and t- bond. Such a rotation is expected to be hindered by the
C4H9OH (Ͻ100 µL) presumably reflect the solvation effects steric interactions of the hydrogen-bonded alcohols. It is
of the alcohols. At the low concentrations of CH3OH where apparent that the bulkier alcohols can generate a greater
its coordination to [Mn(tpp)(OAc)] is not so significant, ap- hindrance than the linear ones, and consequently a larger
parently the solvation effect is the predominant factor, and cis/trans-stilbene oxide ratio would result. Observation of
hence, the epoxidation yield is enhanced by increasing the the same cis/trans-stilbene oxide ratio of 12 for both neat
3220
www.eurjic.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2008, 3218–3224

Solvent Effects on Catalytic Epoxidation of Alkenes
Table 4. Epoxidation of cis-stilbene by nBu4NIO4 in the presence of [Mn(por)] catalysts in various solvents.[a,b]
Entry
Solvent
[Mn(tpp)(OAc)]
[Mn(tmp)(OAc)]
[Mn(tdcpp)(OAc)] [Mn(tpfpp)(OAc)]
Conv. / %
5 h
cis
oxide
trans
oxide
cis/trans
ratio
cis
oxide, 5 h
cis
cis
oxide, 24 h
oxide, 24 h
1
2
3
4
5
6
7
8
CH2Cl2/CH3OH[c]
CH2Cl2/C2H5OH[c]
CH2Cl2/n-C3H7OH[c]
CH2Cl2/i-C3H7OH[c]
CH2Cl2/t-C4H9OH[c]
70.5
85.5
88
92
95
65
80
83
88
92
74
5.5
5.5
5
4
3
12
14.5
16.5
22
≈31
12
25
27
31
36
39
23
28
31
42
55
14
17
18
31
37
[d]
CH2Cl2
80
6
48
35
26
CH2Cl2/CH3OH[e]
5(Ͻ3)[f]
Ͻ2
2 (0)[f]
0
3 (0)[f]
0
3 (0)[f]
0
[g]
CH2Cl2
[a] The molar ratio for [Mn(por)]/im/cis-stilbene/periodate was 1:10:83:167 with a concentration of [Mn(por)] of 3ϫ10–3 , at 22Ϯ2 °C.
[b] The epoxidation yields (%) were measured relative to the starting cis-stilbene. All the reactions were run at least in triplicate, and the
data represent an average of these reactions with Ϯ5–7%. [c] The reactions were carried out in CH2Cl2 (1 mL)/alcohol (0.5 mL). [d] The
reactions were performed in neat CH2Cl2 (1 mL) with no alcohol present. [e] The reactions were carried out in CH2Cl2 (1 mL)/CH3OH
(0.5 mL) with no imidazole. [f] The numbers in the parentheses are for CH2Cl2/t-C4H9OH under similar conditions. [g] The reactions
were carried out in neat CH2Cl2 (1 mL) with no imidazole or alcohol.
CH2Cl2 (Table 4, Entry 6) and CH2Cl2/CH3OH (Table 4, Active Oxidant
Entry 1) suggests analogous overall steric effects for the sol-
vation of the active oxidant(s) by CH3OH and CH2Cl2. Ab-
sence of any trans-stilbene oxide for the epoxidation of cis-
stilbene by the other [Mn(por)] catalysts indicates that the
steric hindrance at their oxygenation site is so great that no
rotation about the alkene bond is possible.
[Mn(tpp)(OAc)]
The UV/Vis spectra for the interaction of
[Mn(tpp)(OAc)] with nBu4NIO4 in CH2Cl2 in the absence
or presence of imidazole, and also in CH2Cl2/alcohol sol-
vents containing imidazole, are shown in Figure 5. The So-
ret band for [Mn(tpp)(OAc)] in CH2Cl2 (λmax = 478 nm)
displayed no changes upon the addition of an excess
amount of nBu4NIO4 {nBu4NIO4/[Mn(tpp)(OAc)], 200;
solution a; Figure 5a}. This suggests that the replacement
–
of OAc– by IO4 and the formation of [Mn(tpp)(IO4)] has
no effect on the UV/Vis spectrum. In contrast, addition of
imidazole into solution a {im/[Mn(tpp)(OAc)], 100; solu-
tion b} caused some broadening and intensity loss of the
corresponding Soret band (λmax = 474 nm) and also broad-
ening and intensification of the band ≈404 nm (Figure 5b).
However, contrary to solution a, the [Mn(por)] species was
unstable in solution b and degraded in ca. 30 min. This indi-
cates that coordination of imidazole to [Mn(tpp)(IO4)]
yields a highly active oxidant.[8b] When 50 µL of t-C4H9OH
was added into solution b (2 mL) to give solution c, the
UV/Vis spectrum reduced in intensity and broadened (Fig-
ure 5c). However, the presence of t-C4H9OH in solution c
greatly increased the lifetime of the [Mn(por)] species, and
the UV/Vis spectrum did not change much over a period of
2 h. For CH3OH (50 µL) in solution b (yielding solution d),
the UV/Vis spectrum dramatically changed and a new
broad band appeared rapidly (λmax = 416 nm; Figure 5d).
The other alcohols also gave similar UV/Vis spectra and
produced new broad bands (λmax = 414Ϯ2 nm). This band,
which is presumably due to [Mn(tpp)(O)(im)]+,[11] disap-
peared rapidly in the presence of cyclooctene.
Figure 4. Possible hydrogen bonding of ROH to the oxygen donor
sites of the presumed active oxidants: (a) [Mn(por)(IO4)(im)] and
(b) [Mn(por)(O)(im)]+. Some hydrogen-bonded ROH groups are
omitted for clarity and their likely positions are indicated by dashed
lines.
The epoxidation yields for alkenes are expected to be re-
lated to the stereoelectronic properties of the species in-
volved. cis-Stilbene is bulkier than cyclooctene, but it con-
tains a more electron-rich C=C bond because of π reso-
nance with the phenyl rings. The higher epoxidation yields
for cis-stilbene than cyclooctene in the presence of the least
hindered [Mn(tpp)(OAc)] catalyst in all the solvents, except
CH2Cl2/t-C4H9OH, clearly suggests that the electron-rich
C=C bond is more reactive than the electron-poor one.
However, the steric effects become the dominating factor
The UV/Vis spectra for
a 30 min interaction of
when bulky [Mn(por)] catalysts and/or bulky alcohols are [Mn(tpp)(OAc)] (10–5 ) with nBu4NIO4 in the presence of
employed.
imidazole in various CH2Cl2/alcohol solvents are presented
Eur. J. Inorg. Chem. 2008, 3218–3224
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
3221

D. Mohajer, M. Abbasi
FULL PAPER
It is noteworthy that the steric strains at the oxygenation
sites of the two possible active oxidants (Figure 4) must be
different. Thus, it is expected that the oxygenation of cis-
stilbene by bulky [Mn(tpp)(IO4)(im)] would result in a
higher cis/trans-stilbene oxide ratio than by the less-hin-
dered [Mn(tpp)(O)(im)]+.[8b] In contrast, on the basis of the
UV/Vis spectra, different ratios of these active oxidants
might be available in the epoxidation of cis-stilbene by
nBu4NIO4 and [Mn(tpp)(OAc)] in the presence of imidazole
in various CH2Cl2/alcohols solvents. Indeed, the lower
cis/trans-stilbene oxide ratios obtained for the smaller
alcohols relative to those of the bulkier ones (Table 4) might
be partly due to the higher [Mn(tpp)(O)(im)]+/
[Mn(tpp)(IO4)(im)] ratios in the presence of the smaller
alcohols.
Figure 5. UV/Vis spectra for the interaction of [Mn(tpp)(OAc)]
(10–5 ) with an excess amount of nBu4NIO4: (a) in neat CH2Cl2;
(b) in CH2Cl2 containing imidazole; (c) in CH2Cl2 (2 mL)/t-
C4H9OH (50 µL) containing imidazole; (d) in CH2Cl2 (2 mL)/
CH3OH (50 µL) containing imidazole. All the spectra were re-
corded 5 min after adding nBu4NIO4, as the last component, at
room temperature.
[Mn(tmp)(OAc)]
The UV/Vis spectra for various solutions of
[Mn(tmp)(OAc)] in CH2Cl2 containing only nBu4NIO4
{nBu4NIO4/[Mn(tmp)(OAc)], 200; solution a} or imidazole
{im/[Mn(tmp)(OAc)], 50; solution b} in the presence of t-
C4H9OH (solution c) or CH3OH (solution d) are given in
Figure 7. Similar to [Mn(tpp)(OAc)], the UV/Vis spectra for
[Mn(tmp)(OAc)] in neat CH2Cl2 (λmax = 478 nm), with or
without an excess amount of nBu4NIO4, were exactly the
same (Figure 7a). However, in contrast to [Mn(tpp)(OAc)],
addition of imidazole into solution a generated a new dis-
tinct band (λmax = 414 nm), which was presumably related
to the [Mn(tmp)(O)(im)]+ species (Figure 7b).[11c] When t-
C4H9OH (50 µL) was added into solution b (2 mL), the
in Figure 6. It is shown that the conversion of the
[Mn(tpp)(IO4)(im)] complex (λmax
=
478 nm) into
[Mn(tpp)(O)(im)]+ (λmax = 414 nm) is virtually complete in
CH2Cl2/CH3OH and the relative intensities of the Mn=O
bands for different alcohols increases in accord with their
acid strengths, that is, CH3OH Ͼ C2H5OH Ͼ n-C3H7OH
Ͼ i-C3H7OH Ͼ t-C4H9OH. In the case of bulky i-C3H7OH
and t-C4H9OH, the formation of Mn=O species is far from
completion and their corresponding UV/Vis spectra did not
display many changes over a period of several hours. In
fact, [Mn(tpp)(O)(im)]+ showed unusually high stability in
all CH2Cl2/alcohol solvents, particularly in the presence of
the bulkier alcohols.
solution c obtained also displayed a similar band (λmax
=
410 nm; Figure 7c). For CH3OH in solution b (producing
solution d), the spectral changes due to the formation of
the [Mn(tmp)(O)(im)]+ complex were very dramatic (Fig-
ure 7d).
Figure 7. UV/Vis spectra for the interaction of [Mn(tmp)(OAc)]
(10–5 ) with an excess amount of nBu4NIO4: (a) in neat CH2Cl2;
(b) in CH2Cl2 containing imidazole; (c) in CH2Cl2 (2 mL)/
Figure 6. The UV/Vis spectra of [Mn(tpp)(OAc)] (2ϫ10–5 mmol)/
im (2ϫ10–3 mmol)/nBu4NIO4 (5ϫ10–3 mmol) in CH2Cl2 (2 mL)/
alcohol (50 µL) solvents [alcohol = CH3OH (a); C2H5OH (b); t-C4H9OH (50 µL) containing imidazole; (d) in CH2Cl2 (2 mL)/
nC3H7OH (c); i-C3H7OH (d); t-C4H9OH (e)]. All the spectra were
recorded 30 min after the final addition of nBu4NIO4, at room tem-
perature.
CH3OH (50 µL) containing imidazole. All the spectra were re-
corded 2 h after adding nBu4NIO4, as the last component, at room
temperature.
3222
www.eurjic.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2008, 3218–3224

Solvent Effects on Catalytic Epoxidation of Alkenes
n-propanol (99.8%), and isopropyl alcohol (99.5%) were obtained
from Merck or Fluka, and tert-butyl alcohol (99.4%) was pur-
chased from Aldrich. Deionized, double-glass distilled water was
used for all experiments. Alkenes were obtained from Fluka and
Merck. The purity of the starting alkenes was checked by gas
chromatography or NMR spectroscopy to ensure that no oxidation
products were present before the reactions. A constant temperature
bath was used to maintain the temperature at 22Ϯ2 °C.
Similar to [Mn(tpp)(OAc)], the formation of Mn=O spe-
cies in the presence of alcohols of smaller molecular size
was faster than that for the larger ones (Figure S1). Also,
the stability of the Mn=O species was considerably greater
in the presence of the bulkier alcohols. However, in every
case the lifetime of [Mn(tmp)(O)(im)]+ was much longer
than that of [Mn(tpp)(O)(im)]+.
These results clearly indicate that the formation rates of
Mn=O species and their relative stabilities strongly depend
upon the nature of the alcohols. Apparently, hydrogen
bonding between the alcohols and the coordinated per-
iodate (Figure 4) can facilitate the heterolytic cleavage of
MnO–IO3 bond[3,4] and the formation of a [Mn(por)-
Materials: Porphyrin ligands were prepared by previously reported
procedures,[13] and manganese was inserted by standard route.[14]
Tetra-n-butylammonium periodate was synthesized by the method
of Santaniello[15] with some modifications. A solution of NaIO4
(2.14 g, 100 mmol) in water (10 mL) was added with stirring to a
solution of tetra-n-butylammonium hydrogen sulfate (nBu4NHSO4;
3.4 g, 100 mmol) in water (20 mL). The precipitated nBu4NIO4 was
dissolved in CH2Cl2 (50 mL), and the organic phase was dried with
magnesium sulfate (MgSO4) or sodium sulfate (Na2SO4) and then
filtered. After evaporation of the solvent, the remaining paste was
washed with hexane (10 mL) and dried under vacuum.
–
(O)(im)]+ species. Furthermore, the obtained IO3 and
[Mn(por)(O)(im)]+ will be stabilized by hydrogen bonding
to the alcohols. Accordingly, small alcohols with stronger
acidities and hydrogen-bonding capabilities than those of
the larger ones can yield Mn=O species more efficiently, as
observed. In contrast, the resulting Mn=O species
are stabilized more effectively by the larger alcohols than
the smaller ones, as the former can cause greater hindrance
and shielding of the active oxidant(s) through hydrogen
bonding.
Instrumentation: Electronic absorption spectra were recorded with
a MultiSpect-1501 Shimadzu spectrophotometer. 1H NMR spectra
were recorded with a Bruker Avance DPX 250 MHz spectrometer.
GC analyses were performed with a Shimadzu 14A series gas chro-
matograph capillary column and a flame-ionization detector. The
separation of products was achieved by on-column injection on a
glass column (length 1.5 m, internal diameter 3 mm) packed with
silicon DC-200; mesh 60–80. Nitrogen was used as the carrier gas.
The UV/Vis spectra for the interaction of the electron-
poor [Mn(tpfpp)(OAc)] (Figure S2) and [Mn(tdcpp)(OAc)]
(not shown) with nBu4NIO4 in various CH2Cl2/alcohol sol-
vents showed no sign of formation of Mn=O species, which
indicates that the occurrence of the Mn=O complex re-
quires an electron-rich porphyrin. Accordingly, the primary
active oxidizing species for the electron-poor [Mn(por)] is
presumably the [Mn(por)(IO4)(im)] complex. However,
[Mn(por)(O)(im)]+ may occur transiently.
Epoxidation of Cyclooctene: Stock CH2Cl2 solutions of [Mn(por)]
(3 mmol), imidazole (30 mmol), neat cyclooctene (34 µL,
250 mmol), and neat dodecane (2 µL) as an internal standard, were
put into a 5-mL round-bottomed flask containing CH2Cl2 (1 mL)
and alcohol (0.5 mL) and then tetra-n-butylammonium periodate
(0.217 g, 500 mmol) was added to initiate the reaction. The reaction
solution was stirred for the required time, and aliquots (2 µL) were
then removed from the reaction solution and injected into the GC
column to monitor the formation of cyclooctene epoxide. The con-
centration of the epoxide was determined by comparison of its
peak area with that of the internal standard. Both peak heights
and peak areas were used to measure the concentration of various
species; averaging of the two methods usually gave the most consis-
tent results. The temperature program for the analysis was 1 min at
80 °C, and the rate program was set at 20 °Cmin–1 up to 175 °C
and 1.5 min at 175 °C. Data were obtained in triplicate to ensure
precision, and the repeatability of the data was within Ϯ10%.
Conclusions
The effects of alcohols in CH2Cl2/alcohol solvents upon
the epoxidation of alkenes by nBu4NIO4 and [Mn(por)] in
the presence of imidazole are apparently related to their
steric bulk and/or acidities. Coordination of the alcohols to
[Mn(por)], which is more efficient for smaller alcohols than
for larger ones, would lead to reduction of the epoxidation
yields. In contrast, solvation of the active oxidants by the
bulkier alcohols causes a longer lifetime of the active oxi-
dant(s) and greater steric hindrance at the oxygenation
site(s) than that achieved by the smaller ones. Finally, the
smaller alcohols with stronger acidities and hydrogen-bond-
ing capabilities than those of the larger ones produce
Mn=O complexes more rapidly.
Epoxidation of cis-Stilbene: Stock CH2Cl2 solutions of [Mn(por)]
(3 mmol), imidazole (30 mmol), and cis-stilbene (45 mg, 250 mmol)
were introduced into a 5-mL round-bottomed flask containing
CH2Cl2 (1 mL), with or without alcohol (0.5 mL), and then nBu4-
NIO4 (0.217 g, 500 mmol) was added to initiate the reaction. The
reaction solution was stirred for the required time, and then the
solvent was removed under reduced pressure. n-Pentane (5 mL) was
added to precipitate [Mn(por)] and nBu4NX (X=OAc–, IO4 , IO3 ).
The resulting mixture was filtered, and n-pentane was removed un-
der vacuum. The composition of the products was determined by
1H NMR spectroscopy (CCl4, TMS). cis-Stilbene epoxide was
characterized by a singlet at δ = 4.35–4.45 ppm for the epoxidic
proton and the trans isomer by a singlet at δ 3.80–3.90 ppm.[1b]
–
–
Experimental Section
General: All the solvents were distilled under an atmosphere of
nitrogen prior to use. Dichloromethane was purified by distillation
over P2O5 or CaH2 before use and stored over molecular sieves
(Linde 4A). Reagent-grade carbon tetrachloride was dried by pass-
ing it through activity I neutral alumina and stored over molecular
sieves (Linde 4A). n-Pentane, methanol (99.8%), ethanol (99.6%),
UV/Vis Spectra: Stock CH2Cl2 solutions of [Mn(por)] (10–5 ,
2 mL), imidazole (0.5 , 10–25 µL), and nBu4NIO4 (0.25 , 50 µL)
were added into a 1-cm UV cell containing CH2Cl2 (2 mL) and
alcohol (50 µL), if needed. The spectral changes were monitored at
different times.
Eur. J. Inorg. Chem. 2008, 3218–3224
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
3223

D. Mohajer, M. Abbasi
FULL PAPER
esue, B.-H. Ye, F. Tani, Y. Naruta, Angew. Chem. Int. Ed. 2004,
43, 98–100; k) W. J. Song, M. S. Seo, S. D. George, T. Ohta, R.
Song, M.-J. Kang, T. Tosha, T. Kitagawa, E. I. Solomon, W.
Nam, J. Am. Chem. Soc. 2007, 129, 1268–1277.
Supporting Information (see footnote on the first page of this arti-
cle): UV/Vis spectra of [Mn(tmp)(OAc)]/im/nBu4NIO4 in various
CH2Cl2/alcohol solvents; UV/Vis spectra for interaction of
[Mn(tpfpp)(OAc)] with nBu4NIO4 in various solvents.
[7] a) T. Takata, W. Ando, Tetrahedron Lett. 1983, 24, 3631–3634;
b) K. S. Suslick, F. V. Acholla, B. R. Cook, J. Am. Chem. Soc.
1987, 109, 2818–2819; c) D. Mohajer, S. Tangestaninejad, Tet-
rahedron Lett. 1994, 35, 945–948.
Acknowledgments
[8] a) F. G. M. Gunter, P. Turner, J. Mol. Catal. 1991, 66, 121–141;
b) D. Mohajer, G. Karimipour, M. Bagherzadeh, New J. Chem.
2004, 28, 740–747.
The authors are thankful for financial supports from the Shiraz
University Research Council.
[9] a) X. H. Mu, F. A. Schultz, Inorg. Chem. 1992, 31, 3351–3357;
b) M. W. Nee, J. Lindsay Smith, J. Chem. Soc., Dalton Trans.
1999, 3373–3377.
[10] a) L. C. Yuan, T. C. Bruice, J. Am. Chem. Soc. 1986, 108, 1643–
1650; b) S. Campestrini, F. D. Furia, P. Ghiotti, F. Novello, C.
Travaglini, J. Mol. Catal. A 1996, 105, 17–23; c) D. Mohajer,
L. Sadeghian, J. Mol. Catal. A 2007, 272, 191–197.
[1] a) J. T. Groves, R. C. Haushalter, M. Nakamura, T. E. Nemo,
B. J. Evans, J. Am. Chem. Soc. 1981, 103, 2884–2886; b) J. T.
Groves, T. E. Nemo, J. Am. Chem. Soc. 1983, 105, 5786–5791;
c) J. P. Collman, J. I. Brauman, T. Hayashi, T. Kodadek, S. A.
Raybuck, J. Am. Chem. Soc. 1985, 107, 2000–2005; d) K. S.
Suslick, B. R. Cook, J. Chem. Chem. Commun. 1987, 200–202;
e) B. Meunier, Chem. Rev. 1992, 92, 1411.
[2] a) A. J. Castellino, T. C. Bruice, J. Am. Chem. Soc. 1988, 110,
158–162; b) R. D. Arasasingham, G. X. He, T. C. Bruice, J.
Am. Chem. Soc. 1993, 115, 7985–7991; c) E. Baciocchi, T. Bos-
chi, L. Cassioli, C. Galli, L. Jaquinod, A. Lapi, R. Paolesse,
K. M. Smith, P. Tagliatesta, Eur. J. Org. Chem. 1999, 3281–
3286.
[3] a) W. Nam, S. Y. Oh, Y. J. Sun, J. Kim, W. K. Kim, S. K. Woo,
W. Shin, J. Org. Chem. 2003, 68, 7903–7906; b) N. A. Stephen-
son, A. T. Bell, Inorg. Chem. 2006, 45, 2758–2766; c) N. A. Ste-
phenson, A. T. Bell, Inorg. Chem. 2006, 45, 5591–5599; d) N. A.
Stephenson, A. T. Bell, J. Mol. Catal. A 2007, 275, 54–62; e)
N. A. Stephenson, A. T. Bell, J. Mol. Catal. A 2007, 272, 108–
117.
[4] a) T. G. Traylor, R. Popovitz-Biro, J. Am. Chem. Soc. 1988,
110, 239–243; b) T. G. Traylor, F. Xu, J. Am. Chem. Soc. 1990,
112, 178–186; c) K. Yamaguchi, Y. Watanabe, I. Morishima, J.
Am. Chem. Soc. 1993, 115, 4058–4065.
[11] a) C. L. Hill, B. C. Schardt, J. Am. Chem. Soc. 1980, 102, 6374–
6375; b) J. T. Groves, W. J. Kruper, R. Haushalter, J. Am.
Chem. Soc. 1980, 102, 6375–6377; c) J. T. Groves, Y. Watanabe,
Inorg. Chem. 1986, 25, 4808–4810; d) S. Banfi, M. Cavazzini,
F. Coppa, S. V. Barkanova, O. L. Kaliya, J. Chem. Soc. Perkin
Trans. 2 1997, 1577–1583; e) Y. Iamamoto, M. D. Assis, K. J.
Ciuffi, C. M. C. Prado, B. Z. Prellwitz, M. Moraes, O. R. Nas-
cimento, H. C. Sacco, J. Mol. Catal. A 1997, 116, 365–374.
[12] a) P. H. Howard, W. M. Meylan, Handbook of Physical Proper-
ties of Organic Chemicals, CRC Lewis Publishers, Boca Raton,
FL, 1989; b) Y. Marcus, The Properties of Solvents, Wiley Series
in Solution Chemistry, 1998, vol. 4, ch. 4, pp. 165–170.
[13] a) G. M. Badger, R. A. Jones, R. L. Laslett, Aust. J. Chem.
1964, 17, 1028; b) A. D. Adler, F. R. Longo, V. Varadi, Inorg.
Synth. 1976, 16, 213; c) A. D. Adler, F. R. Longo, J. D. Finar-
elli, J. Goldmacher, J. Assour, L. Korsakoff, J. Org. Chem.
1967, 32, 476; d) P. Hoffmann, A. Robert, B. Meunier, Bull.
Soc. Chim. Fr. 1986, 129, 85–97; e) A. Bettelheim, B. A. White,
S. A. Raybuck, R. W. Murray, Inorg. Chem. 1987, 26, 1009; f)
J. S. Lindsey, I. C. Schreiman, H. C. Hsu, P. C. Kearney, A. M.
Marguerettaz, J. Org. Chem. 1987, 52, 827; g) J. L. Lindsey,
R. W. Wagner, J. Org. Chem. 1989, 54, 828–836; h) P.
Hoffmann, A. Robert, B. Meunier, Bull. Soc. Chim. Fr. 1992,
129, 85.
[5] a) W. Nam, M. H. Lim, H. J. Lee, C. Kim, J. Am. Chem. Soc.
2000, 122, 6641–6647; b) W. Nam, Y. O. Ryu, W. J. Song, J.
Biol. Inorg. Chem. 2004, 9, 654–660.
[6] a) J. T. Groves, M. K. Stern, J. Am. Chem. Soc. 1987, 109,
3812–3814; b) R. S. Czernuszewicz, O. Su, M. K. Stern, K. A.
Macor, D. Kim, J. T. Groves, T. G. Spiro, J. Am. Chem. Soc. [14] a) A. D. Adler, F. R. Longo, F. Kampas, J. Kim, J. Inorg. Nucl.
1988, 110, 4158–4165; c) J. T. Groves, M. K. Stern, J. Am.
Chem. Soc. 1988, 110, 8628–8638; d) J. T. Groves, J. Lee, S. S.
Marla, J. Am. Chem. Soc. 1997, 119, 6269–6273; e) N. Jin, J. T.
Groves, J. Am. Chem. Soc. 1999, 121, 2923–2924; f) N. Jin, J. L.
Bourassa, S. C. Tizio, J. T. Groves, Angew. Chem. Int. Ed. 2000,
39, 3849–3851; g) W. Nam, I. Kim, M. H. Lim, H. J. Choi, J. S.
Lee, H. G. Jang, Chem. Eur. J. 2002, 8, 2067–2071; h) R.
Zhang, M. Newcomb, J. Am. Chem. Soc. 2003, 125, 12418–
12419; i) R. Zhang, J. H. Horner, M. Newcomb, J. Am. Chem.
Soc. 2005, 127, 6573–6582; j) Y. Shimazaki, T. Nagano, H. Tak-
Chem. 1970, 32, 2443–2445; b) P. N. Balasubramanian, A.
Sinha, T. C. Bruice, J. Am. Chem. Soc. 1987, 109, 1456; c)
K. M. Kadish, C. Araullo-McAdams, B. C. Han, M. M.
Franzen, J. Am. Chem. Soc. 1990, 112, 8364; d) A. Robert, M.
Momenteau, B. Loock, B. Meunier, Inorg. Chem. 1991, 30,
706–711.
[15] E. Santaniello, A. Manzocchi, C. Farachi, Synthesis 1980, 563–
565.
Received: March 11, 2008
Published Online: June 3, 2008
3224
www.eurjic.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2008, 3218–3224