Highly selective olefin epoxidation with the bicarbonate activation of hydrogen peroxide in the presence of manganese(III) meso-tetraphenylporphyrin complex: Optimization of effective parameters using the Taguchi method

Article abstract of DOI:10.1016/j.apcata.2009.10.033

A Mn^III-porphyrin-based catalytic system was explored for olefin epoxidations under mild reaction conditions using sodium bicarbonate-hydrogen peroxide as an oxidant. The Mn(TPP)OAc/imidazol/NaHCO₃ system efficiently catalyzed the epoxidation of olefins with H₂O₂. Cyclic olefins were transformed in excellent yield (80-100%) and selectivity (87-100%), the obtained selectivity and yields being much better than those observed in the absence of bicarbonate. In the presence of an excess of substrate, the turnover number 4286 was obtained with the Mn(TPP)OAc/Im/NaHCO₃/H₂O₂ system after 2 h. The bicarbonate-activated oxidation system is a simple, inexpensive, and relatively nontoxic alternative to other oxidants and peroxyacids, and it can be used in a variety of oxidations where a mild, neutral pH oxidant is required. Due to the various factors, such as solvent, reaction temperature, stoichiometric ratio of imidazole/NaHCO₃/H₂O₂, influencing the oxidation of olefins, the Taguchi method of system optimization was used to determine the percent of contribution (%P) of each factor. It was found that the solvent had the most influence on the oxidation (30.051%) and the imidazole amount stood in second place (22.286%).

Full text of DOI:10.1016/j.apcata.2009.10.033

Applied Catalysis A: General 372 (2010) 209–216
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Highly selective olefin epoxidation with the bicarbonate activation of hydrogen
peroxide in the presence of manganese(III) meso-tetraphenylporphyrin complex:
Optimization of effective parameters using the Taguchi method
a
a
b
Hassan Hosseini Monfared a, , Vahideh Aghapoor , Massomeh Ghorbanloo , Peter Mayer
*
a Department of Chemistry, Zanjan University 45195-313, Zanjan, Islamic Republic of Iran
b Fakulta¨t fu¨r Chemie und Pharmazie, Ludwig-Maximilians-Universita¨t, Mu¨nchen, Butenandtstr. 5-13, Haus D, D-81377 Mu¨nchen, Germany
A R T I C L E I N F O
A B S T R A C T
A MnIII-porphyrin-based catalytic system was explored for olefin epoxidations under mild reaction
conditions using sodium bicarbonate–hydrogen peroxide as an oxidant. The Mn(TPP)OAc/imidazol/
NaHCO3 system efficiently catalyzed the epoxidation of olefins with H2O2. Cyclic olefins were
transformed in excellent yield (80–100%) and selectivity (87–100%), the obtained selectivity and yields
being much better than those observed in the absence of bicarbonate. In the presence of an excess of
substrate, the turnover number 4286 was obtained with the Mn(TPP)OAc/Im/NaHCO3/H2O2 system after
2 h. The bicarbonate-activated oxidation system is a simple, inexpensive, and relatively nontoxic
alternative to other oxidants and peroxyacids, and it can be used in a variety of oxidations where a mild,
neutral pH oxidant is required.
Article history:
Received 3 June 2009
Received in revised form 18 October 2009
Accepted 21 October 2009
Available online 31 October 2009
Keywords:
Manganese porphyrin
Olefin oxidation
Hydrogen peroxide
Sodium bicarbonate
Taguchi
Due to the various factors, such as solvent, reaction temperature, stoichiometric ratio of imidazole/
NaHCO3/H2O2, influencing the oxidation of olefins, the Taguchi method of system optimization was used
to determine the percent of contribution (%P) of each factor. It was found that the solvent had the most
influence on the oxidation (30.051%) and the imidazole amount stood in second place (22.286%).
ß 2009 Elsevier B.V. All rights reserved.
1. Introduction
above oxygen donors, hydrogen peroxide provides a reasonably
cheap, readily available oxidant, giving only water as a by-product.
The catalytic epoxidation of olefins has been a subject of growing
interest in the production of chemicals and fine chemicals since
epoxides are the key starting materials fora wide variety of products
[1,2]. Much effort has beenmade todevelop new active and selective
epoxidation catalysts for those processes that require an elimination
of by-products. A wide variety of metal complexes have been
synthesized forthis purpose and their catalytic properties have been
extensively studied [3–10]. This transformation using environmen-
tally benign oxidants has aroused much interest, because the
chemical industry continues to require cleaner oxidation, which is
an advance over environmentally unfavored oxidations and a step
up from more costly organic peroxides.
Manganese meso-tetrakisaryl porphyrins and several other
metal porphyrin complexes have been studied intensively as
catalysts in epoxidation reactions of alkenes [11,12]. A variety of
oxidants such as iodosylarenes, alkylhydroperoxides, peracids and
hypochlorite in addition to H2O2 were employed [11,12]. Of the
However, it is necessary for efficient epoxidation that the hydrogen
peroxide should be cleaved heterolytically to give an ‘oxene’
(reaction (1a))
(1)
rather than homolytically to give hydroxyl radicals (reaction (1b)).
The formation of hydroxyl radicals leads to somewhat indiscrimi-
nate oxidation of organic substrates, with hydrogen abstraction
being important. In a reaction of type (1a), in which an oxene or its
equivalent is involved, oxidation is much more selective and, with
olefins, leads directly to epoxides. There have been investigations
concerning the hetero- and homolytic cleavage of hydroperoxides
and hydrogen peroxide in the presence of various metalloporphyr-
ins. The early porphyrin-based catalysts often showed rapid
deactivation due to the oxidative degradation of the ligand [13–
16]. Spectacular improvements in the robustness and activity of
catalysts for olefin epoxidation and hydroxylation of alkenes were
obtained after the introduction of halogen substituents into the
porphyrin ligands [17]. General disadvantages of manganese
* Corresponding author. Tel.: +98 241 5152576; fax: +98 241 2283203.
E-mail addresses: monfared@mail.znu.ac.ir, monfared@znu.ac.ir,
monfared_2@yahoo.com (H.H. Monfared).
0926-860X/$ – see front matter ß 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.apcata.2009.10.033

210
H.H. Monfared et al. / Applied Catalysis A: General 372 (2010) 209–216
porphyrin-basedepoxidationcatalystsarethedifficultyinpreparing
the ligands and the often tedious purification that is required.
Hydrogen peroxide has a high active oxygen content compared
to other stoichiometric oxidants, that is peroxycarboxylic acids,
periodate, and peroxymonosulfate [18]. Hydrogen peroxide is a
widely used oxidant with high active oxygen content [19–21], but
it is a rather slow oxidizing agent in the absence of activators.
Epoxidation of alkenes by HCO3ꢀ/H2O2 in CD3CN/D2O [22] has
been described but the reaction time is high (24 h) and the
conversion of
a-methylstyrene is only 50%. In another study
hydrogen peroxide was electrogenerated in situ in aqueous
bicarbonate solutions and has been used for water soluble
alkenes using Mn2+ as the catalyst [23]. MnSO4/NaHCO3/
H2O2/tBuOH system has been used for epoxidation of alkenes
[24] in 16 h as well. A number of groups have discussed the use of
catalysts containing species such Mo(CO)6 immobilized in a
polymer bound aliphatic amine [25], complexes of oxodiperoxo-
8-hydroxy-quinolinolato Mo(VI) over mesoporous silica [26],
oxodiperoxo molybdenum(VI) complex [MoO(O2)2(saloxH)] (sal-
oxH2 = salicylaldoxime) [27] and hexathiocyanatorhenate(IV)
(Ph4P)2[Re(NCS)6] [28] in the presence of the H2O2/NaHCO3
system as an oxidant in the oxidation of olefins. Simple metal salts
like Mn2+, Cr2+, Fe3+ and Mn-porphyrins with NaHCO3/H2O2 are
used for olefin epoxidation. Halogenated porphyrins were
required to resist oxidation of the catalysts [29]. Ammonium
bicarbonate has been used for diastereoselective epoxidation of
allylically substituted alkenes using sterically bulky metallopor-
phyrins as catalyst provides high trans-selectivities (i.e., trans:cis-
epoxide ratio) [30].
Fig. 1. Mn(TPP)OAc employed in this study.
30
m
m ꢁ 320
m
m ꢁ 0.25
mm) with flame-ionization detector and
gas chromatograph–mass spectrometry (Hewlett-Packard 5973
Series MS-HP gas chromatograph with a mass-selective detector).
1H NMR spectra were recorded by using Bruker 250 MHz
spectrometer. IR spectra were recorded using Perkin-Elmer 597
and Nicolet 510P spectrophotometers. The electronic transition
spectra were recorded on a Thermospectronic model
spectrometer with 1 cm quartz cells was used for recording and
storage of UV–vis absorbance spectra.
a, UV–vis
The catalysis of sulfide oxidation by bicarbonate ions,
bicarbonate ion catalysis of methionine oxidation to methionine
sulfoxide by H2O2 in water [31], and bicarbonate-catalyzed
hydrogen peroxide oxidation of cysteine and related thiols to
the corresponding disulfides [32] have also been reported.
Such systems can have one or more disadvantages, such as the
long reaction time [22,24], need to complex catalysts like the
sterically bulky metalloporphyrins that are not easy to make
[29,30] and low selectivity and by-products like diol [22]. Our
interest in the development of methods to accomplish selective
oxidation of olefins led us to consider another little-explored mode
of reactivity for the formation of epoxides through the use of
NaHCO3/H2O2/Imidazole/Mn(TPP)OAc catalysis. In this paper, we
describe our studies aimed at catalyzing efficient oxygen transfer
from hydrogen peroxide to olefins using the H2O2/NaHCO3 system
as the source of oxygen and using the statistical Taguchi design to
obtain the optimum oxidation condition.
2.2. Oxidation reactions
Reactions were carried out in a 25 ml round bottom flask.
Based on our previous work from catalytic oxidation of olefins
by the catalyst/imidazole/oxidant/solvent system [35] and
some preliminary experiments, in a typical experiment, H2O2
(1–10 mmol) was added to a flask containing the Mn(TPP)OAc
(1.4 ꢁ 10ꢀ3 mmol), cyclooctene (1 mmol) and NaHCO3 (0.26–
5.25 mmol) in solvent (3 ml). Reactions were carried out for 6 h,
under magnetic stirring at room temperature. At appropriate
intervals, aliquots were removed and quantified immediately by
GC. Imidazole (7–21
mmol) was added to some reactions. The
oxidation products were identified by comparing their retention
times with those of authentic samples or alternatively by 1H
NMR and GC–Mass analyses. Yields are based on the added olefin
and were determined by a calibration curve. Control reactions
were carried out in the absence of catalyst, under the same
conditions as the catalytic runs. No products were detected.
2. Experimental
3. Results and discussion
2.1. Materials and equipment
3.1. Oxidation of cyclooctene
Manganese(II) acetate tetrahydrate was purchased from Sigma.
Hydrogen peroxide (30% in water) was supplied by Merck and
stored at 5 8C, and was periodically titrated to confirm its purity.
The exact concentration of hydrogen peroxide was determined by
titration with potassium permanganate. Solvents of the highest
grade commercially available (Merck) were used without further
purification. The olefins and other reagents were purchased from
Merck and Fluka and used as received. The free base porphyrin
H2TPP was prepared and purified by the method reported
previously [33]. Manganese insertion into this free base porphyrin
was carried out using the method of Adler et al. [34]. Fig. 1 shows
the structure of the Mn-porphyrin complex used in this study.
The reaction products of oxidation were determined
and analyzed by an HP Agilent 6890 gas chromatograph
equipped with a HP-5 capillary column (phenyl methyl siloxane
In the first instance, the effect of various reagents was evaluated
in the catalytic oxidation of cyclooctene in the presence of
Mn(TPP)OAc with hydrogen peroxide as an oxidant. All reactions
were carried out with 1 mmol of cyclooctene in CH3CN and in the
presence of 1.4
mmol catalyst at room temperature for 5 h.
Cyclooctene oxide was the sole product. The results are
summarized in Table 1. The results of control experiments
revealed that the oxidation of cyclooctene in the absence H2O2
does not occur. As reported by Mansuy [17] due to oxidative
degradation of the Mn(TPP)OAc catalyst, the epoxide yield by
Mn(TPP)OAc/Im was very low (Table 1, run no. 1). Interestingly, an
improvement in the cyclooctene conversion was observed when
imidazole was replaced with sodium bicarbonate (epoxide yield
5%, Table 1, run no. 2). The mixture of imidazole/NaHCO3 catalyzed

H.H. Monfared et al. / Applied Catalysis A: General 372 (2010) 209–216
211
Table 1
Control experiments in oxidation of cyclooctene by H2O2
a
.
Run
Catalyst
Co-catalyst
Activator
Conversion (%)
Epoxide yield (%)
<1
1
2
3
4
5
Mn(TPP)OAc
Mn(TPP)OAc
None
Imidazole
None
None
<1
5.0
15
3
NaHCO3
NaHCO3
NaHCO3
NaHCO3
5.0
Imidazole
None
15
3
None
Mn(TPP)OAc
Imidazole
43
43
a
Reaction conditions: catalyst (1.4 mmol), cyclooctene (1.0 mmol), CH3CN (3 ml), Im (14 mmol), NaHCO3 (0.07 mmol), H2O2 (5.0 mmol), reaction time (5 h) at room
temperature. Conversions and yields are based on starting cyclooctene.
Table 2
the cyclooctene oxidation up to yield 15% with H2O2 in the absence
of Mn(TPP)OAc (Table 1, run no. 3). NaHCO3 by itself produced only
3% of the epoxide (Table 1, run no. 4). Whether the improvement
effect of NaHCO3 was a catalyst-stabilizing effect or from the
formation of peroxymonocarbonate ion was not clear. However,
the most striking catalytic enhancement giving a significant
quantity of cyclooctene oxide was with Mn(TPP)OAc/Im/NaHCO3
(yield 43%, Table 1, run no. 5). Of these three systems, the
Mn(TPP)OAc/Im and Mn(TPP)OAc/NaHCO3 catalysts were dis-
counted due to the low yield of the epoxide. Thus, the remaining
Mn(TPP)OAc/Im/NaHCO3 system was optimized.
Parameters and their values corresponding to their levels to be studied in the
experiments.
Parameter
Levels
1
2
3
4
A. Imidazol (
mmol)
0
7
14
0.53
c
21
5.25
d
B. NaHCO3 (mmol)
C. Solvent
0.26
0.35
b
a
1
1
D. H2O2 (mmol)
E. Time (h)
3
5
10
4
2
3
a, MeOH; b, CH2Cl2; c, CHCl3; d, CH3CN.
The results of the control experiments led us to conclude that by
optimization of H2O2, imidazole, solvent, NaHCO3 and reaction
time it is possible to obtain a high yield of epoxide in a short time
and to get insight into the relative importance of each factor. In the
search for suitable reaction conditions to achieve the maximum
oxidation of cyclooctene, the Taguchi method was used.
labor-intensive and time consuming. Hence, the analyses using
conventional experimental methods are sometimes inefficient.
Therefore, the Taguchi method [36] was introduced in this work.
The Taguchi method determines the experimental condition
having the least variability as the optimum condition. The
variability can be expressed by a signal to noise (S/N) ratio
[36]. The experimental condition having the maximum S/N ratio is
considered as the optimal condition.
Based on our previous work from catalytic oxidation of olefins
by the catalyst/imidazole/oxidant/solvent system [35] and the
results of Table 1, the main operational parameters and their levels
were selected and showed in Table 2.
The design of experiments is an approach for systematically
varying the controllable input factors and observing the effects
of these factors on the output product parameters. The
assumption used in this study was that the individual effects
of the input factors on the output product parameters are
separable. That means the effect of an independent variable on
the performance parameter does not depend on the different
level settings of any other independent variable. The orthogonal
3.2. Taguchi experiment design
In addition to the nature of catalyst and olefin concentration,
concentrations of axial ligand (imidazole), activator (sodium
hydrogencarbonate), solvent and oxidant (hydrogen peroxide)
influence the catalytic oxidation of olefins. The essential feature of
catalytic oxidation is to give a high epoxide yield and selectivity at
short time. Therefore, in order to achieve a high epoxide yield and
relatively fast oxidation, understanding the relationship between
the different system parameters and the oxidation time of the
olefins is essential. In the oxidation of olefins by catalytic
methods, however, there are many parameters that are important
to obtain high epoxide yields. The interrelationships between the
parameters are complex, and the conventional approach of
experimenting with one variable (or one factor) at a time is
Table 3
Experimental layout of cyclooctene oxidation using the L16 orthogonal array and experimental results for percent of epoxide yield.
Experimental number
A
B
C
D
E
Epoxide yield (%)
First
1
Second
2
Third
3
S/N
1
2
1
1
1
1
2
2
2
2
3
3
3
3
4
4
4
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
2
1
4
3
3
4
1
2
4
3
2
1
1
2
3
4
3
4
1
2
4
3
2
1
2
1
4
3
1
2
3
4
4
3
2
1
2
1
4
3
3
4
1
2
11.2
1
11
11.5
1.3
37
21.006
2.119
2
3
37.8
48.9
37
36
31.343
33.761
31.212
35.212
33.580
21.532
34.140
34.121
35.631
25.361
34.251
25.878
27.512
29.480
4
49.4
35.8
59
48
5
36.3
58
6
56
7
47.7
12.1
51
47
48.6
12
8
11.7
50.8
50
9
51
10
11
12
13
14
15
16
51.5
61.6
20
51
59
60.9
19
17
51.3
21.5
23
51
52.5
20
18
24
24.3
28
31.7
30

212
H.H. Monfared et al. / Applied Catalysis A: General 372 (2010) 209–216
Table 4
Table 5
Response for the Taguchi analysis of cyclooctene oxidation data.
Analysis of variance.
Parameter
Mean S/N ratio
Factor
DOF (f)
Sum of sqrs. (S)
Variance (V)
Percent (%P)
Level 1
Level 2
Level 3
Level 4
A
3
3
3
3
3
0
240.837
133.584
324.750
227.465
153.99
80.279
44.528
108.250
75.821
51.330
22.286
12.361
30.051
21.049
14.250
B
A. Imidazol (
mmol)
22.057
30.152
30.332
26.456
26.043
30.384
24.332
21.551
23.383
24.830
32.313
32.017
28.223
31.539
31.542
29.280
27.534
33.928
32.656
31.620
C
B. NaHCO3 (mmol)
C. Solvent
D
E
D. H2O2 (mmol)
E. Time (h)
Other/error
Total
15
1080.628
100.00%
array of L16 type was used, and is represented in Table 3. L and
16 mean Latin square and the replication number of the
experiment, respectively. Five–four level factors can be posi-
tioned in an L16 orthogonal array table. The number in table
indicates the levels of a factor [37]. The structure of Taguchi’s
L16 design and the results of measurement are shown in Table 3.
In the Taguchi method, the terms ‘signal’ and ‘noise’ represent
the desirable and undesirable values for the output character-
istic, respectively. The Taguchi method uses the S/N ratio to
measure the quality characteristic deviating from the desired
value. The S/N ratios are different according to the type of
characteristic. In the case where bigger characteristics are
Fig. 2. Average S/N ratio for different levels of the parameters in oxidation of cyclooctene by Mn(TPP)OAc/Im/NaHCO3 system: (a) imidazole, (b) NaHCO3, (c) solvent, (d) H2O2,
and (e) reaction time. See Table 2 for parameters and their values corresponding to their levels to be studied in the experiments.

H.H. Monfared et al. / Applied Catalysis A: General 372 (2010) 209–216
213
better, the S/N ratio is defined as [37]
ꢀ
ꢁ
1=y21 þ 1=y22 þ 1=y32 þ ꢂ ꢂ ꢂ þ 1=yn2
S=N ¼ ꢀ10 log
(2)
n
where yi is the characteristic property, n is the replication number
of the experiment. The unit of S/N ratio is decibel (dB), which is
frequently used in communication engineering. Table 3 shows the
S/N ratio for the oxidation of cyclooctene calculated using Eq. (2).
The mean S/N ratio for each level of the parameters was
summarized as S/N response, which was shown in Table 4.
Fig. 2 shows the average S/N ratio graph for oxidation of
cyclooctene. Therefore, the optimum condition is A3, B3, C4, D4
and E4. In other words, based on the S/N ratio, the optimal
parameters (conditions) for 1 mmol cyclooctene oxidation are A
(imidazole) at level
3 (14 mmol), B (NaHCO3) at level 3
(0.53 mmol), C (solvent) at level 4 (CH3CN), D (H2O2) at level 4
(10 mmol) and E (time) at level 4 (4 h). Finally, in this case 100%
epoxide yield can be obtained. This epoxide yield was confirmed
twice by oxidation of cyclooctene with the obtained optimum
conditions. In order to conduct an analysis of the relative
importance of each factor more systematically, an analysis of
variance (ANOVA) was applied to the data [36]. The main objective
Fig. 3. Effect of hydrogen peroxide concentration on the oxidation of cyclooctene
by Mn(TPP)OAc to the epoxide. Reaction conditions: cyclooctene (1 mmol),
Mn(TPP)OAc
(1.4 ꢁ 10ꢀ3 mmol),
imidazol
(1.4 ꢁ 10ꢀ2 mmol),
NaHCO3
(0.50 mmol), CH3CN (3 mmol) and room temperature (27 ꢃ 1 8C).
Table 6
a
Oxidation of various olefins catalyzed by Mn(TPP)OAc/Im/NaHCO3/H2O2
.
No.
1
Alkene
Product(s)b
Epoxide yield (%)c
85
Conversion (%)c
85
2
3
4
100
86
100
98.5
100
100
5
6
80
93
80
93
7
40d
31d
71
8
18d
18

214
H.H. Monfared et al. / Applied Catalysis A: General 372 (2010) 209–216
Table 6 (Continued )
No.
9
Alkene
Product(s)b
Epoxide yield (%)c
Conversion (%)c
26
26
10
11
1-Octene
1-Decene
1-Octene oxide
1-Decene oxide
26
40
26
40
a
Reaction conditions: Mn(TPP)OAc (1 mg, 1.376 mmol), Imidazole (14 mmol), NaHCO3 (0.53 mmol) olefin (1.0 mmol), acetonitrile (3 ml), H2O2 (10 mmol), reaction time (4 h)
and reaction temperature (25 ꢃ 1 8C). Conversions and yields are based on the starting substrate.
b
All products were identified by comparison of their physical and spectral data with those of authentic samples.
Epoxide yield and conversion with respect to olefin.
Yields determined by 1H NMR.
c
d
of ANOVA is to extract from the results how much variation each
cis- and trans-epoxides, respectively [41]. Typical epoxidizing
reagents such as peroxyacids have been shown to react with cis
olefins about two times faster than the corresponding trans olefin
[42]. For m-chloroperoxybenzoic acid in methylene chloride at
room temperature, cis- and trans-stilbenes are epoxidized at equal
rates. This unequal reactivity was inferred to result from the
nonbonded interactions between the phenyl groups of trans-
stibene and the phenyl groups of Mn(TPP)OAc. A comparison of
this catalytic system with previously reported [Mn(DCPP)]Cl/Im/
benzoic acid/H2O2 systems shows that the conversion and the
selectivity of trans-stilbene are higher than in the other systems
[43]. The only product of the norbornene oxidation was endo-
norbornene oxide. This result is in spite of the exo-epoxide from the
oxidation of norbornene by HCO3ꢀ/H2O2 in CD3CN/D2O [22] and
oxodiperoxo molybdenum(VI) complex [MoO(O2)2(saloxH)] (sal-
oxH2 = salicylaldoxime)/NaHCO3/H2O2/CH3CN [27] systems.
The Mn(TPP)OAc/Im/NaHCO3/H2O2 system is active enough to
oxidize even olefins with lowest activity like 1-octene and 1-
decene (Table 6, nos. 10 and 11, with 100% selectivity) and
cyclooctane to cyclooctanol (6%) and cyclooctanone (31%), not
shown in Table 6. The activity of Mn(TPP)OAc/Im/NaHCO3/H2O2
system is much higher than HCO3ꢀ/H2O2/CD3CN/D2O [22] and
MnSO4/NaHCO3/H2O2/tBuOH [24]
factor causes relative to the total variation observed in the result.
From the results of ANOVA in Table 5, the solvent had the largest
percent (30.051%). The axial ligand imidazole (22.286%) and
hydrogen peroxide (21.049%) indicated the second and third,
respectively. Consequently, it can be concluded that the most
influential factor was in the order of the solvent. NaHCO3 had the
least influence on the oxidation reaction. On the other hand, the
degree of freedom (DOF) for each factor was 3 and total DOF was
15, so the DOF for error term was 0, and finally the variance for the
error term (Ve), obtained by calculating error sum of squares and
dividing by error degrees of freedom, could not be calculated.
Henceforth, it was impossible to calculate the F-ratio, defined as
the variance of each factor dividing by Ve. In order to eliminate the
zero DOF from the error term, a pooled ANOVA was applied. The
process of ignoring a factor once it was deemed insignificant was
called pooling.
As confirming the results of the optimization of effective
parameters using the Taguchi method, one factor, namely, the
influence of reaction time on cyclooctene conversion at different
[H2O2]/[cyclooctene] molar ratios was studied with the continuous
variation method. The highest conversion of cyclooctene was
achieved when the [H2O2]/[cyclooctene] molar ratios was 10:1
(Fig. 3). This finding is in complete agreement with the result of the
Taguchi method and confirmed it.
3.4. The catalyst anion and substrate concentration effects
3.3. Oxidation of various olefins
To get insight into the role of the effects of catalyst anion and
cyclooctene concentration on oxidation with two different catalytic
systems, Mn(TPP)OAc/Im/NaHCO3/H2O2 and Mn(TPP)Cl/Im/
NaHCO3/H2O2 were examined. The results of these studies
(Table 7) revealed that the catalytic activity of Mn(TPP)OAc is
better than that of Mn(TPP)Cl and with the Mn(TPP)OAc/Im/
NaHCO3/H2O2 system,theturnovernumber4286wasobtainedafter
2 h. Theless reactivityofMn(TPP)Cl relativetothatofMn(TPP)OAcis
The applicability of the catalyst system was demonstrated in
the epoxidation of a number of representative olefins under the
same reaction conditions which proved to be the best for
cyclooctene (Table 6). Cyclic olefins were transformed in excellent
yield (80–100%) and selectivity (87–100%) (Table 6, run nos. 1–5).
Electron-rich olefins displayed a greater reactivity than electron
deficient olefins. The relatively electron-poor 1,3-cyclooctadiene
shows lower reactivity than cyclooctene. This reflects the
electrophilic nature of the oxygen transfer from the manganese–
oxo [38] intermediate to the olefinic double bond. Allylic oxidation
or formation of by-products, which is seen in the metalloporphyrin
systems [39,40] was not observed in the epoxidation of alkenes
Table 7
Effects of catalyst anion nature and cyclooctene concentration on the reactivity of
catalyst/imidazole/NaHCO3/H2O2 systema.
Cyclooctene
Catalyst
(mmol)
Epoxide
yield (%)
Reaction
time (h)
TONb
such as cyclohexene.
a-Methylstyrene was converted to the
epoxide with 93% conversion in 4 h. This conversion is about two
times faster than of its oxidation (50%) by HCO3ꢀ/H2O2 system in
24 h in acetonitrile [22]. In the epoxidation of cis- and trans-
stilbene the steric effects are distinct. The sterically demanding
trans-stilbene is less reactive than the cis isomer. Epoxidation of
trans-stilbene proceeds with absolute stereospecificity. The
oxygenation of cis-stilbene results in a cis- to trans-stilbene oxide
ratio of 31:40. It was previously reported that the oxidation of cis-
stilbene by Mn(TPP)Cl–PhIO, Mn(TPP)Cl–PhIO–Im and Mn(TPP)Cl–
PhIO–pyridine systems giving 39:61, 93:7 and 57:43 mixture of
Mn(TPP)OAc
Mn(TPP)OAc
Mn(TPP)OAc
Mn(TPP)OAc
Mn(TPP)Cl
1
2
100
3
714
86
5
1229
5
85
5
3036
10
1
60 (65)c
93 (100)c
87
2 (5)
3 (5)
5
4286 (4643)c
664
Mn(TPP)Cl
2
1243
Mn(TPP)Cl
10
67 (67)b
2 (5)
4786
a
Reaction conditions: catalyst (1.4
m
mol), CH3CN (3 ml), Im (14 mmol), NaHCO3
(0.53 mmol), H2O2 (5.0 mmol), reaction time (5 h) at room temperature.
b
Turnover number = [epoxide]/[catalyst]. Yields based on starting cyclooctene.
c
The results in parenthesis correspond to the results after 5 h.

H.H. Monfared et al. / Applied Catalysis A: General 372 (2010) 209–216
215
Fig. 4. The visible spectrum of the intermediate formed upon addition of H2O2
(0.21 M) to a solution of Mn(TPP)OAc (2.8 ꢁ 10ꢀ5 M), imidazole (6.99 ꢁ 10ꢀ6 M)
and NaHCO3 (1.0 ꢁ 10ꢀ2 M) in the UV–vis cell. Blue, Mn(TPP)OAc/Im (1); green,
Mn(TPP)OAc/Im/H2O2 after 1 min; red, Mn(TPP)OAc/Im/NaHCO3/H2O2 after 1 min.
Fig. 5. The visible spectrum of Mn(TPP)OAc/Im, Mn(TPP)OAc/Im/H2O2 and
Mn(TPP)OAc/Im/NaHCO3/H2O2 after 1 and 5 min. Conditions as in Fig. 4.
probably due to the weak basicity of the chloride ion relative to that
of the acetate counter ion.
3.5. Evidence for the formation of reactive intermediate
Clearly, the Mn(TPP)OAc complex generates active species upon
reaction with H2O2/Im/NaHCO3, because Mn(TPP)OAc/Im/NaHCO3
can participate in efficient epoxidation as described above. The
characteristics of stilbene oxidation by the Mn(TPP)OAc/Im/
NaHCO3/H2O2 system are very similar to those of the PhIO–
Mn(TPP)Cl–imidazole system [41]: cis-stilbene is more reactive
than trans-stilbene and gives the cis-epoxide predominantly. The
only product of the oxidation of norbornene is endo-epoxide. The
epoxidation of norbornene by m-chloroperbenzoic acid proceeds
almost exclusively exo [44], whereas the formation of exo-
norbornene oxide and endo-norbornene oxide in low exo/endo
ratio has been reported with iodosylarenes using several
substituted tetraphenylporphyrin and teterapyridylporphyrin
iron(III) salts as catalysts [45]. It is concluded that the active
oxygenizing species does not occur directly by the species resulted
from H2O2–NaHCO3.
The addition of H2O2 to a CH3CN solution of Mn(TPP)OAc/Im/
NaHCO3/cyclooctene leads to the immediate appearance of a new
species, characterized by peaks at 424 and 523 nm in the visible
spectrum (Fig. 4). The spectrum is identical to that of H2O2/
Mn(TPP)OAc/Im/cyclooctene. Although, the oxidation results show
that NaHCO3 dramatically affects the activation of H2O2 by
Mn(TPP)OAc/Im, the similar characteristics of the Mn(TPP)OAc–
H2O2–NaHCO3–Im and Mn(TPP)Cl–PhIO-systems suggest a com-
mon active species. This species is very similar to that formed upon
the reaction of PhIO with Mn(TPP)Cl which has been described as a
high-valent Mn–oxo complex [38].
The lack of similar oxidizing ability, however, in the
oxidation of cyclooctene with the Mn(TPP)OAc/Im/H2O2 system
suggests that the effect of NaHCO3 is at least in part due to its
stabilizing effect on H2O2. The effect of sodium bicarbonate on
the oxidation with hydrogen peroxide is like the effects of
organic acids. As known, enhancement of the activity by acetic
acid in the oxidation of cyclohexane has been reported in the
literature [46,47]. In the medium of acetic acid, hydrogen
peroxide is more stable and diminishes the self-decomposition
by forming peroxyacetic acid. Activation of hydrogen peroxide
by citric acid in acidic solution has been reported in [48,49]. It is
proposed that the formation of peroxycitric acid is just like the
formation of peroxyacetic acid.
Scheme 1. Two possible reaction pathways for the epoxidation of olefins by
Mn(TPP)OAc/Im/NaHCO3/H2O2 system: (a) direct oxidation of olefin by HCO4ꢀ, or
(b) oxidation of olefin by the active species [Mn–oxo]# which is formed in situ by the
ꢀ
reaction of HCO4 and Mn(TPP)OAc/Im.
The reactivity of Mn(TPP)OAc/Im/NaHCO3/H2O2 system is more
than that of Mn(TPP)OAc/Im/H2O2 (Table 1). This finding was also
confirmed by the faster reduction of Mn(TPP)OAc/Im Soret band in
the presence of NaHCO3/H2O2 relative to that of Mn(TPP)OAc/Im/
H2O2 after 1 min (Fig. 5). Carboxylic acids and nitrogen-containing
additives are considered to facilitate the heterolytic cleavage of the
O–O bond in the manganese hydroperoxy intermediate, resulting
in a catalytically active manganese(V)–oxo intermediate [43].
Drago et al. described the activating H2O2 by using the bicarbonate
ion in their study of the oxidation of chemical warfare agents and
sulfides in mixed tert-butyl alcohol/water solvent [50]. They
concluded on the basis of NMR evidence that the active oxidant ꢀin
the catalytic pathway is peroxymonocarbonate ion (HCO4
)
formed by a labile preequilibrium reaction between bicarbonate
ion and H2O2 (Eq. (3)).
HCOꢀ3 þ H2O2 Ð HCO4ꢀ þ H2O
(3)
We may summarize our findings as Scheme 1. The active
species in the olefin oxidation with Mn(TPP)OAc/Im/NaHCO3/H2O2
system is [Mn–oxo]# which is formed by the reaction of HCO4ꢀ and
Mn(TPP)OAc. This active species oxidizes olefins to the corre-
sponding epoxides. Since the oxidation of norbonene was endo-
epoxide, so the oxidation proceeds mainly by route b in Scheme 1.
4. Conclusion
The activation of C55C bonds of olefins and C–H bonds of alkanes
can be achieved by the Mn(TPP)OAc/Im/NaHCO3 system and
hydrogen peroxide as the clean oxidant. The Taguchi method of
system optimization was used to determine the optimized
conditions and the percent of contribution (%P) of each factor. It
was found that the solvent had the most influence on the oxidation
and the imidazole amount stood in second place (22.286%). The

216
H.H. Monfared et al. / Applied Catalysis A: General 372 (2010) 209–216
MnIII-porphyrin in the presence of bicarbonate gives remarkably
excellent selectivity and yield for the epoxidation of cyclic and
linear olefins with H2O2. Sodium bicarbonate by stabilizing H2O2
prevents its fast decomposition and promotes its heterolytic
cleavage for selective epoxidation. The bicarbonate-activated
oxidation system is a simple, inexpensive, and relatively nontoxic
alternative to other oxidants and peroxyacids, and it can be used in
a variety of oxidations where a mild, neutral pH oxidant is
required. The components of the system are inexpensive and
environmentally friendly. The oxidation with Mn(TPP)OAc/Im/
NaHCO3/H2O2 proceeds by the mediation of Mn–oxo species.
[16] S. Banfi, F. Montanari, G. Pozzi, S. Quici, Tetrahedron 50 (1994) 9025–9036.
[17] D. Mansuy, Coord. Chem. Rev. 125 (1993) 129–141.
[18] R.A. Sheldon, Top. Curr. Chem. 164 (1993) 21–34.
[19] V.N. Kislenko, A.A. Berlin, Russ. Chem. Rev. 60 (1991) 470–488.
[20] W.C. Schumb, C.N. Satterfield, R.L. Wentworth, Hydrogen Peroxide, Reinhold,
New York, 1955.
[21] J.P. Farr, W.L. Smith, D.S. Steichen, in: J.I. Kroschwitz (Ed.), Kirk-Othmer Ency-
clopedia of Chemical Technology, vol. 4, Wiley-Interscience, New York, 1997, pp.
271–300.
[22] H. Yao, D.E. Richardson, J. Am. Chem. Soc. 122 (2000) 3220–3221.
[23] K.-P. Ho, T.H. Chan, K.-Y. Wong, Green Chem. 8 (2006) 900–905.
[24] B.S. Lane, K. Burgess, J. Am. Chem. Soc. 123 (2001) 2933–2934.
[25] S. Tangestaninejad, V. Mirkhani, M. Moghadam, G. Grivani, Catal. Commun. 8
(2007) 839–844.
[26] S.K. Maiti, S. Dinda, M. Nandi, A. Bhaumik, R. Bhattacharyya, J. Mol. Catal. A: Chem.
287 (2008) 135–141.
[27] N. Gharah, S. Chakraborty, A.K. Mukherjee, R. Bhattacharyya, Chem. Commun.
(2004) 2630–2632.
Acknowledgement
[28] S. Dinda, S.R. Chowdhury, K.M. Abdul Malik, R. Bhattacharyya, Tetrahedron Lett.
46 (2005) 339–341.
[29] B.S. Lane, K. Burgess, Chem. Rev. 103 (2003) 2457–2473.
[30] C. Chi-ming,W. Man-kin, US Patent 7,482,478 (2009).
[31] D.E. Richardson, H. Yao, K.M. Frank, D.A. Bennett, J. Am. Chem. Soc. 122 (2000)
1729–1739.
We are grateful to the Research Council of Zanjan University for
financial support of this study.
References
[32] C.A.S. Regino, D.E. Richardson, Inorg. Chim. Acta 360 (2007) 3971–3977.
[33] A.D. Adler, F.R. Longo, J.D. Finarelli, J. Goldmacher, J. Assour, L. Korsakof, J. Org.
Chem. 32 (1967) 476.
[34] A.D. Adler, F.R. Longo, F. Kampas, J. Kim, J. Inorg. Nucl. Chem. 32 (1970) 2443–
2445.
[35] H. Hosseini-Monfared, S. Sadighian, M.-A. Kamyabi, P. Mayer, J. Mol. Catal. A:
Chem. 304 (2009) 139–146.
[36] A. Kamyabi-Gol, S.M. Zebarjad, S.A. Sajjadi, Colloid Surface A: Physicochem. Eng.
Aspects 336 (2009) 69–74.
[37] S.T. Kim, M.S. Park, H.M. Kim, Sens. Actuator B 102 (2004) 253–260.
[38] J.T. Groves, W.J. Kruper, R.C. Haushalter, J. Am. Chem. Soc. 102 (1980) 6375–6377.
[39] D. Mohajer, H. Hosseini-Monfared, J. Chem. Res. (1998) 772–773.
[40] A.J. Appleton, S. Evans, J.R.L. Smith, J. Chem. Soc., Perkin Trans. 2 (1995) 281–284.
[41] D. Mansuy, P. Battioni, J.-P. Renaud, J. Chem. Soc., Chem. Commun. (1984) 1255–
1257.
[42] B.M. Lynch, K.H. Pausacker, J. Chem. Soc. (1955) 1525–1531.
[43] J.T. Groves, Y. Watanabe, T.J. McMurry, J. Am. Chem. Soc. 105 (1983) 4489–4490.
[44] H.C. Brown, J.H. Kawakami, S. Ikegami, J. Am. Chem. Soc. 92 (1970) 6914–6917.
[45] D. Ostovic, T.C. Bruice, Acc. Chem. Res. 25 (1992) 314–320.
[46] T. Sooknoi, J. Limtrakul, Appl. Catal. A: Gen. 233 (2002) 227–237.
[47] G.B. Shul’pin, G. S-Fink, J.R.L. Smith, Tetrahedron 55 (1999) 5345–5358.
[48] A.W. Karunditu, C.M. Carr, K. Dodd, Text. Res. J. 64 (1994) 570.
[49] Y. Du, Y. Xiong, J. Li, X. Yang, J. Mol. Catal. A: Chem. 298 (2009) 12–16.
[50] R.S. Drago, K.M. Frank, Y.-C. Yang, G.W. Wagner, The Proceedings of the 1997
ERDEC Scientific Conference on Chemical and Biological Defense Research, U.S.
Army Edgewood Research, Development, and Engineering Center, 1998.
[1] G. Franz, R.A. Sheldon, in: B. Elvers, S. Hawkins, G. Shulz (Eds.), 5th ed., UIImann’s
Encyclopedia of Industrial Chemistry, vol. A, no. 18, VCH, Weinheim, 1991, pp.
261–311.
[2] J.T. Lutz, in: K. Othmer, M. Grayson, D. Eckroth, G.J. Bushey, C.I. Eastman, A.
Klingsberg, L. Spiro (Eds.), 3rd ed., Encyclopedia of Chemical Technology, vol. 9,
Wiley, New York, 1980, p. 251.
[3] R.A. Sheldon, J.K. Kochi (Eds.), Metal-catalyzed Oxidations of Organic Compounds,
Academic, New York, 1981.
[4] P.R. Ortiz de Montellano (Ed.), Cytochrome P-450: Structure, Mechanism and
Biochemistry, Plenum, New York, 1976.
[5] H. Hosseini Monfared, M. Ghorbani, Monatsh. Chem. 132 (2001) 989–992.
[6] H. Hosseini Monfared, M. Ghadimi, J. Chem. Res. (5) (2003) 313–314.
[7] R. De Paula, M.M.Q. Simo˜es, M. Grac¸a, P.M.S. Neves, J.A.S. Cavaleiro, Catal. Com-
mun. 10 (2008) 57–60.
[8] G.B. Shul’pin, Y.N. Kozlov, S.N. Kholuiskaya, M.I. Plieva, J. Mol. Catal. A: Chem. 299
(2009) 77–87.
[9] S.T. Castaman, S. Nakagaki, R.R. Ribeiro, K.J. Ciuffi, S.M. Drechsel, J. Mol. Catal. A:
Chem. 300 (2009) 89–97.
[10] K.C. Gupta, A.K. Sutar, Coord. Chem. Rev. 252 (2008) 1420–1450.
[11] T. Katsuki, Coord. Chem. Rev. 140 (1995) 189–214.
[12] B. Meunier, Chem. Rev. 92 (1992) 1411–1456.
[13] W.A. Lee, T. Bruice, J. Am. Chem. Soc. 107 (1985) 513–514.
[14] P. Battioni, J.-P. Renaud, J.E. Bartoli, M. Reina-Artiles, M. Fort, D. Mansuy, J. Am.
Chem. Soc. 110 (1988) 8462–8470.
[15] P.L. Anelli, S. Banfi, F. Lengramandi, F. Montanari, G. Pozzi, S. Quici, J. Chem. Soc.,
Perkin Trans. 1 (1993) 1345–1357.