Critical factors in determining the heterolytic versus homolytic bond cleavage of terminal oxidants by Iron(III) porphyrin complexes

Article abstract of DOI:10.1021/jacs.7b13037

Heterolytic versus homolytic cleavage of the metal-bound terminal oxidant is the key for determining the nature of reactive intermediates in metalloenzymes and metal catalyzed oxygenation reactions. Here, we study the bond cleavage process of hypochlorite

Full text of DOI:10.1021/jacs.7b13037

Subscriber access provided by TUFTS UNIV
Critical Factors in Determining the Heterolytic versus Homolytic
Bond Cleavage of Terminal Oxidants by Iron(III) Porphyrin Complexes
Sawako Yokota, and Hiroshi Fujii
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.7b13037 • Publication Date (Web): 25 Mar 2018
Downloaded from http://pubs.acs.org on March 25, 2018
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 13
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Critical Factors in Determining the Heterolytic versus Homolytic Bond
Cleavage of Terminal Oxidants by Iron(III) Porphyrin Complexes
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Sawako Yokota and Hiroshi Fujii*
Department of Chemistry, Graduate School of Humanities and Sciences, Nara Women’s University, Kitauoyanishi,
Nara 630-8506, Japan
ABSTRACT: Heterolytic versus homolytic cleavage of the metal bound terminal oxidant is the key for determining the
nature of reactive intermediates in metalloenzymes and metal catalyzed oxygenation reactions. Here, we study the bond
cleavage process of hypochlorite by iron(III) porphyrin complexes having 4-methoxy-2,6-dimethylphenyl (1), 2,4,6-
trimethylphenyl (2), 4-fluoro-2,6-dimethylphenyl (3), 2-chloro-6-methylphenyl (4), 2,6-dichlorophenyl (5), and 2,4,6-
trichlorophenyl (6) groups at the meso-position. Oxoiron(IV) porphyrin π-cation radical complexes (CompI) are charac-
terized from the reactions of 1 ~ 4 with tetra-n-butylammonium hypochlorite (TBA-OCl) in dichloromethane at –80 °C
while oxoiron(IV) porphyrin complexes (CompII) are characterized for 5 and 6 under the same conditions. For all of 1 ~
6, we find the formation of an epoxidation product in good yields from the catalytic reactions with TBA-OCl, suggesting
the heterolytic cleavages of the O-Cl bonds. CompI of 5 and 6 are reduced to the corresponding CompII by both chlo-
ride and hypochlorite while CompI of 1 ~ 4 are not. The reduction reactions with hypochlorite are much faster than
those with chloride. These results provide a mechanism where the O–Cl bond of the iron-bound hypochlorite is cleaved
heterolytically to form CompI for all of 1 ~ 6, but the subsequent reduction reaction with remaining hypochlorite affords
CompII for 5 and 6. The E(OCl⋅/OCl–) is the boundary to discriminate the identity of the final product: CompI or Com-
pII. The thermodynamic analysis based on the redox potential is successfully applied for explaining the bond cleavage
processes of the hypochlorite, hydroperoxide, and t-butyl peroxide complexes.
INTRODUCTION
The reactions of metal centers with terminal oxidants
(XO) are key steps to generate reactive high valent oxo-
intermediates in various metal catalyzed oxidation reac-
tions and heme enzymes such as peroxidases, catalases
Figure 1. Heterolytic and homolytic bond cleavage pro-
cesses of terminal-oxidant(O–X) adduct of iron(III) por-
phyrin complex.
and cytochromes P450.1-6 The reactions initially form
terminal-oxidant adducts of metal complexes, followed by
the cleavage of the O–X bonds of the metal bound termi-
nal oxidants to produce the high-valent metal-oxo inter-
mediates. Because of their significance, the cleavage pro-
cesses of the O–X bonds have been studied by the reac-
tions of various metal complexes with terminal oxidants
such as hydrogen peroxide, alkylperoxides, peracids, and
iodosylarenes.7-14 To date, there are two known types of
the bond cleavage fashion (Figure 1). One is the hetero-
lytic cleavage, in which the O–X bond is cleaved ionically
to form an atomic oxygen and X–. In this case, the initial
metal complex is oxidized to a high-valent metal oxo-
species having two-electron equivalent more oxidized
state. The other is the homolytic cleavage, in which the
O–X bond is cleaved radically to form O⋅ and X⋅. In this
case, the initial metal complex changes to a high-valent
metal oxo-species having one-electron equivalent more
oxideized state. For example, the heterolytic cleavage of
the hydroperoxide adduct (X = OH) of iron(III) porphyrin
is expected to form an oxoiron(IV) porphyrin π-cation
radical complex (compound-I, CompI) and hydroxide
(OH–), while the homolytic cleavage is anticipated to pro-
duce an oxoiron(IV) porphyrin complex (compound-II,
CompII) and a hydroxyl radical, ⋅OH. The heterolytic
cleavage would be more preferable than the homolytic
cleavage in catalytic reactions for avoiding undesirable
radical reactions.
Previously, Groves et al. reported that the heterolytic
cleavage of the O–O bond of the iron-bound m-
chloroperoxybenzoic acid (mCPBA) produces CompI in
dichloromethane while the homolytic cleavage forms the
iron(III) porphyrin N-oxide in toluene.10 The O–O bond
cleavage has been also studied in aqueous and aprotic
solvents. Traylor et al. proposed that the O–O bonds of
H2O2 and t-BuOOH are heterolytically cleaved in metha-
nol-dichloromethane mixture.7 On the other hand,
Bruice et al. provided an evidence that the O–O bond is
homolytically cleaved in aqueous and aprotic solvents and
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 13
the generation of the heterolytic cleaved product would bound hypochlorite spectroscopically. Here, we report
1
2
3
4
5
6
7
8
be due to the re-oxidation of the homolytic product by
another ROOH or the radical species formed from the
homolytic cleavage in solvent cage.8 The pH of the sol-
vent has been shown to be a key factor of the O–O bond
cleavage process.8,12 Nam et al. studied the bond cleavage
processes of the O–O bonds on the basis of the yields of
the catalytic epoxidation reactions.12 The main results are
that (1) the electron-deficient iron porphyrin complex
show a tendency to cleave the O–O bond more heterolyt-
ically than the less electron-deficient iron porphyrin
complex, (2) the heterolytic cleavage is more preferable as
the electron-withdrawing effect of the substituent of the
iron-bound O–O bond is higher (mCPBA > H2O2 > t-
BuOOH), (3) an electron-donating axial imidazole ligand
induces the homolytic cleavage more than a less electron-
donating one. The results of the porphyrin and axial lig-
and effects, (1) and (3), seems to contradict the push-pull
effect.4 On the other hand, as pointed out in their studies,
one should pay attention to determine the bond cleavage
fashion from the product yield of catalytic reaction be-
cause the catalytic product yield is affected not only by
the bond cleavage fashion, but also by various factors for
the reactive species produced in the catalytic reactions:
e.g. the formation and reaction rates, the lifetime, and
side reactions in the reaction mixture. This means that
the high-valent metal oxo-species in the reaction mixture
is not always identical to the one expected from the O–O
bond cleavage fashion. Nam et al. also studied the bond
cleavage process of the O–I bond of the iron-bound iodo-
sylarene and showed that the counter anion of iron por-
phyrin is an essential factor to determine the bond cleav-
age process.14(a) This result is reasonably understood from
a x-ray crystal structure of manganese(IV) salen iodosyl-
mesitylene complex, which shows that the counter anion
interacts with the manganese-bound O–I moiety and
changes the property of the I–O bond.15 While these
studies qualitatively clarified the factors affecting the
bond cleavage processes of these oxidants, it has not been
studied well for other terminal oxidants and the mecha-
nism for determining the fashion of the bond cleavage,
either the heterolytic or homolytic cleavage, has remained
unclear.
the heterolytic versus homolytic bond cleavage of hypo-
chlorite (OCl–) by iron(III) porphyrin complexes. The
electron-withdrawing effect of porphyrin ligand on the
bond cleavage process is studied by using iron(III) por-
phyrin complexes having various meso-phenyl groups, 1 ~
6, shown in Figure 2. This study clearly shows that the
O–Cl bond is cleaved heterolytically to produce CompI
for all of 1 ~ 6, but the identity of the final product is
changed to CompII by the subsequent reduction of the
formed CompI with hypochlorite for 5 and 6. This con-
clusion is supported by the comparison of the redox po-
tentials of CompI and hypochlorite. We also propose the
thermodynamic analysis based on the redox potentials of
CompI and a terminal oxidant (OX) to explain the bond
cleavage process of hypochlorite, hydrogen peroxide, and
t-butyl hydroperoxide. This analysis can be applicable to
predict the heterolytic versus homolytic bond cleavage
and the identity of high-valent metal oxo species charac-
terized as the final product complex in the reaction mix-
ture under various conditions.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. Structures of iron porphyrin complexes, 1 ~ 6,
used in this study
Results and Discussion
Electron-withdrawing Effect of meso-Phenyl
Group. To confirm the electron-withdrawing effect of
the meso-phenyl groups of 1 ~ 6, we measured the redox
potentials of CompI/CompII, E(I/II), at –40 °C.19 The
CompII of 1 ~ 6, 1-CompII ~ 6-CompII, were prepared
by published methods (see experimental section).20 The
cyclic voltammograms and differential pulse voltammo-
grams of 1-CompII ~ 6-CompII show reversible redox
peaks corresponding to CompI/CompII redox couples
(Figures S1 and S2) and the E(I/II) values are summarized
in Table 1. The E(I/II) values increases in numerical or-
der, 1 < 2 < 3 < 4 < 5 < 6.
Hypochlorite (OCl–) has been used as a terminal oxi-
dant in many catalytic oxidation reactions. For example,
Jacobsen et al. showed highly enantioselective epoxida-
tion reactions by using a chiral manganese(III) salen
complex and sodium hypochlorite.16 The reactions have
been proposed to be catalyzed by an manganese(V) oxo
species generated by the heterolytic cleavage of the O–Cl
bond. Meunier et al. reported manganese porphyrin cata-
lyzed epoxidation and chlorination reactions by sodium
hypochlorite.17 While there have been many studies of
the catalytic oxidation reactions with hypochlorite, the
bond cleavage process of the metal-bound hypochlorite
has not been studied well. Recently, we succeeded in the
preparation and characterization of the hypochlorite ad-
ducts of iron(III) porphyrin complexes.18 This progress
allows us to study the bond cleavage process of the iron-
ACS Paragon Plus Environment

Page 3 of 13
Journal of the American Chemical Society
Table 1. Redox Potentials (vs. SCE) and yields (%) of
catalytic epoxidation reactions with TBA-OCl for 1 – 6
1
2
3
4
E(I/II) a
(V)
Yield (%) of cyclooctene ox-
ide b
5
at 4 °C (–20 °C)
6
7
8
9
1
0.845
0.853
0.924
0.999
1.117
53 (50)
51
2
3
4
5
6
53
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
55 (56)
55
1.167
55 (59)
a The E1/2 of ferrocene was 0.380 V in the present electro-
b
chemical cell at –40 °C. The yields are based on TBA-OCl.
The yields without catalyst are 7 % at 4 °C and 2 % at –20 °C.
The yields are average values of three different experiments.
Reaction of 1 ~ 6 with Hypochlorite. Figure 3 shows
absorption spectra of 1-Cl, 4-Cl, and 6-Cl and their spec-
tral change upon the addition of 10 equivalents of TBA-
OCl in dichloromethane at –80 °C. The results for 2-Cl, 3-
Cl, and 5-Cl are also shown in Figure S3. The absorption
spectra of the reaction products of 1-Cl ~ 3-Cl with TBA-
OCl, showing strong absorptions at around 670 nm, are
identical to those of 1-CompI ~ 3-CompI prepared by the
oxidation with ozone (Figure S4), respectively. These
results indicate that the heterolytic cleavages occur for 1 ~
3. Unique absorption spectral changes are observed for
4. With the addition of TBA-OCl, the absorption spec-
trum of 4-Cl initially changes to a new one with absorp-
tion peaks at 539 nm and 680 nm within a few minutes,
and then slowly changes to a final spectrum with the ab-
sorption peak at 536 nm with disappearance of the peak
at 680 nm over 1 hour (the red and green lines in Figure
3(b)). The spectrum of the intermediate is not identical to
the spectrum of either authentic 4-CompI or 4-CompII
(Figure S4 and S5). The absorption peak at 680 nm sug-
gests the formation of CompI, but the intensity is lower
than that of the authentic 4-CompI (Figure S4). When 50
equivalents of TBA-OCl were reacted with 4-Cl, the inten-
sity of the 680 nm peak increased (Figure S6). Moreover,
the spectrum of the final product was also obtained from
the reaction of 4-CompI with TBA-OCl at –80 °C (Figure
S7). These results indicate that 4-CompI is initially
formed from the reaction with TBA-OCl, but the formed
4-CompI reacts with remaining TBA-OCl to give a de-
composed complex. The decomposition process of 4 is
faster than those of 1 ~ 3 because 4-CompI is more reac-
tive than 1-CompI ~ 3-CompI due to the stronger elec-
tron-withdrawing meso-substituent.21 All of these results
indicate that the heterolytic cleavage occurs even for 4.
On the other hand, with addition of TBA-OCl, the ab-
sorption spectra of 5-Cl and 6-Cl initially change to new
ones showing the absorption peaks at 542 and 544 nm
(the blue lines in Figure 3(c) and S3(c)), followed by the
final spectra with 552 and 558 nm with clear isosbestic
Figure 3. Absorption spectral change of (a) 1-Cl, (b) 4-Cl,
and (c) 6-Cl after addition of 10 equivalent of TBA-OCl in
dichloromethane at –80 °C. Black solid line: before reac-
tion, pink line: immediately after addition of TBA-OCl,
gray line: after addition of TBA-OCl, interval: (a) 3 s, (b)
50 s, (c) 240 s, red line: the reaction product with TBA-
OCl. Green line in (b): the final reaction product after 1
hour. Blue line in (c): the intermediate (the bis-
hypochlorite complex) of 6 formed in the reaction. Inset:
EPR spectra (4 K) of the reaction intermediates and prod-
ucts shown by the red, green, and blue lines in the ab-
sorption spectra.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 13
alytic reactions at –20 °C, at which TBA-OCl itself cannot
points, respectively. The absorption spectra of the inter-
1
2
3
4
5
6
7
8
mediates are very close to that of iron(III) porphyrin bis-
hypochlorite complex, (ClO)2, reported previously.18 The
absorption spectra of the final products for 5 and 6 are
almost identical to those of the authentic 5-CompII and
6-CompII, respectively (Figure S5). These results suggest
that 5-CompII and 6-CompII are produced from the re-
actions of 5-Cl and 6-Cl with TBA-OCl via 5-(ClO)2 and 6-
(ClO)2.
oxidize cyclooctene to the epoxide (Table 1). These re-
sults suggest that the reactions of 1-Cl ~ 6-Cl with TBA-
OCl initially generate CompI by the heterolytic cleavages
of the O–Cl bonds. Since the epoxidation reactions of 5-
CompI and 6-CompI compete with their reduction reac-
tions to CompII and self-decomposition reactions in their
catalytic reactions, the moderate yields of cyclooctene
oxide suggest that the epoxidation reactions would be
faster than the reduction reactions and the decomposi-
tion reactions under the catalytic reaction conditions. In
addition, since 5-CompI and 6-CompI are effectively
stronger oxidants than other CompI, 5-CompI and 6-
CompI can transfer the oxygen atom to olefin more effec-
tively than other CompI, resulting in slightly higher
yields.
9
These assignments of the complexes produced from the
bond cleavage processes were further confirmed by EPR
spectroscopy (Figure 3 and Figure S8). The EPR spectra of
the reaction products with TBA-OCl for 1 ~ 3 show strong
EPR signals at g = 4.45, 3.55, and 1.98. These EPR spectra
are identical to those of compound I22 and supports the
heterolytic bond cleavages for 1 ~ 3. The compound-I like
EPR signals are also observed for 4, but the intensity of
the EPR signals is much lower than those of 1 ~ 3. In ad-
dition, the EPR signals disappear with progress in the
reaction. These results support the heterolytic cleavage
for 4 and the further decomposition of the formed 4-
CompI. The initial formations of 5-(ClO)2 and 6-(ClO)2
in the reactions of 5-Cl and 6-Cl with TBA-OCl are also
confirmed by EPR spectra with small g-anisotropy (g =
2.253, 2.139, and 1.965 for 5-(ClO)2 and 2.254, 2.137, and
1.963 for 6-(ClO)2).18 Disappearance of the EPR signals
with progress in the reactions is also consistent with the
formation of CompII from the O–Cl bond cleavage of 5-
(ClO)2 and 6-(ClO)2. The very small signals at g = 2.0
observed for 5 and 6 may be due to organic radical species
formed in the bond cleavage processes.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Reaction of CompI with Chloride. To further inves-
tigate the mechanism, we examined the reactions of 1-
CompI ~ 6-CompI, prepared by other oxidant, with chlo-
ride at –80 °C. We examined the preparation of 1-CompI
~ 6-CompI from the oxidation of 1-Cl ~ 6-Cl with ozone
gas at –80 °C. 1-Cl ~ 4-Cl were oxidized to 1-CompI ~ 4-
CompI, respectively, but 5-Cl and 6-Cl could not be oxi-
dized to the corresponding CompI species. These results
suggest the reactions of 5-CompI and 6-CompI with
chloride. Stable 5-CompI and 6-CompI were prepared by
the oxidation of the iron(III) porphyrin nitrate complexes
of 5 and 6 by ozone at –80 °C. When 50 equivalents of
tetra-n-butylammonium chloride (TBA-Cl) were added to
1-CompI ~ 6-CompI at –80 °C, no reactions are observed
for 1-CompI
~ 4-CompI for 1 hour except self-
decomposition reactions to 1-Cl ~ 4-Cl, respectively (Fig-
ure S9). However, 5-CompI and 6-CompI are reduced to
the mixture of the corresponding Cl and CompII during 1
hour (Figure S9). The reactions followed the first-order
kinetics in the presence of 50 equivalents of TBA-Cl, and
the rate constants depended on the concentration of
TBA-Cl (Figure S10). The second-order rate constants
were estimated to be 0.14 and 0.46 M-1s-1 at –80 °C for 5
and 6, respectively. These results indicate that 1-CompI ~
4-CompI do not react with chloride, but 5-CompI and 6-
CompI slowly react with chloride.
Catalytic Reactions of 1 ~ 6 with Hypochlorite. The
above results provide the following two possible mecha-
nisms. One is that the bond cleavage process determines
the final product; the O–Cl bonds of the iron-bound hy-
pochlorites are cleaved heterolytically to form CompI for
1 ~ 4, but homolytically to generate CompII for 5 and 6.
The other is that the reactivity of CompI to chloride or
hypochlorite determines the final product; the O–Cl
bonds are cleaved heterolytically for all of 1 ~ 6 to produce
CompI, but the subsequent reactions of CompI with
chloride or hypochlorite produce CompII as a finally
characterized complexes for 5 and 6. To reveal the reac-
tion mechanism, we examined the catalytic epoxidation
reactions of cyclooctene with 1-Cl ~ 6-Cl and TBA-OCl.
Previously, the heterolytic versus homolytic bond cleav-
ages of the terminal oxidants have been studied by using
yields of the epoxidation products of the catalytic reac-
tions.7,8,12 These analyses are based on an assumption that
CompI formed from the heterolysis of the O–O bond is a
reactive compound to produce the epoxide from olefin,
but CompII formed from the homolysis is not so reactive
as to afford the epoxide.
Reaction of CompI with Hypochlorite. We further
examined the reactions of 1-CompI ~ 6-CompI with hy-
pochlorite at –80 °C. 1-CompI ~ 3-CompI do not react
with hypochlorite after the addition of 10 equivalents of
TBA-OCl (Figure 4 and S11). 4-CompI slowly reacted with
TBA-OCl and changed to a new compound having ab-
sorption peak at 536 nm for 10 min (blue line in Figure
S11(c)). The new compound was identical with the de-
composed compound formed from the reaction of 4-Cl
with TBA-OCl (Figure S7(b)). The similar results were
also obtained for the reactions of 1-CompI ~ 4-CompI
prepared from the iron(III) porphyrin nitrate complexes;
1-CompI ~ 3-CompI did not react with hypochlorite, but
4-CompI formed the new compound. The reaction of 4-
CompI from the nitrate complex is much slower than
that from the chloride complex. 4-CompI prepared from
the nitrate complex changed to the new compound for 1
The yields of cyclooctene oxide from the catalytic reac-
tions of 1-Cl ~ 6-Cl at 4 °C are listed in Table 1. For all of 1
~ 6, we found good yields of cyclooctene oxide. Interest-
ingly, the yields for 5 and 6, which afford CompII in the
absence of substrate, are slightly higher than those of 1 ~
4. Furthermore, similar results were obtained for the cat-
ACS Paragon Plus Environment

Page 5 of 13
Journal of the American Chemical Society
tions with hypochlorite clearly support the formation of
hour. On the other hand, 5-CompI and 6-CompI, pre-
1
2
3
4
5
6
7
8
pared from the iron(III) porphyrin nitrate complexes,
produce the corresponding CompII immediately after the
addition of 10 equivalents of TBA-OCl (Figure 4 and S11).
5-CompII and 6-CompII further change to new com-
pounds having absorption peak at 538 nm (blue lines in
Figures 4(c) and S11(d)), which are very close to the final
product obtained from the reaction of 4-CompI with
TBA-OCl. The reaction rate constants for the formations
of 5-CompII and 6-CompII could not be measured with
an absorption spectrometer because of extremely fast
reactions, but are expected to be larger than 103 M-1s-1
from the reactions completed within a few second.
Therefore, the reduction reactions with TBA-OCl are
much faster than those with TBA-Cl. Overall, 1-CompI ~
4-CompI are not reduced to the corresponding CompII
by hypochlorite, but 5-CompI and 6-CompI are immedi-
ately reduced to the corresponding CompII. Although
the absorption spectral features of the final compounds
observed for 4 ~ 6 are close to that of iron(IV) bis-
methoxide complex, we need further spectroscopic study
to identify them and to clarify the formation mechanism.
CompI for all of 1 ~ 6. The detection of 5-CompII and 6-
CompII is due to the subsequent reduction reactions
with hypochlorite because the reduction reactions with
hypochlorite are much faster than those with chloride for
5 and 6. These results provide a possible mechanism that
the reactions of 1-Cl ~ 6-Cl with TBA-OCl initially pro-
duce CompI by the heterolytic cleavages of the O–Cl
bonds, but the formed 5-CompI and 6-CompI are imme-
diately reduced to the corresponding CompII by hypo-
chlorite remaining in the reaction mixture when the sub-
strate is not present (Scheme 1). In the presence of the
substrate (e.g. cyclooctene), the reactions of 5-CompI and
6-CompI, formed from the heterolysis of the O–Cl bond,
with cyclooctene compete with the reduction reactions
with hypochlorite, resulting in the formation of cy-
clooctene oxide in moderate yields (Scheme 1).
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 1. A proposed mechanism for the reactions
of iron(III) porphyrin complexes with hypo-
chlorite.
The formation of CompII for 5 and 6 can be explained
by the redox potential of hypochlorite. The standard po-
tential of one-electron oxidation process of hypochlorite,
E°(OCl⋅/OCl–), has been reported to be 1.39 V (1.15 V vs.
SCE).23 This value is close to the E(I/II) values of 5-
CompI and 6-CompI and is altered by the solvent. The
E(OCl⋅/OCl–) would be lower than the E(I/II) values of 5-
CompI and 6-CompI in dichloromethane at –80 °C, re-
sulting in the rapid reduction to the corresponding Com-
pII. The E(OCl⋅/OCl–) value is an essential factor in de-
termining the identity of the finally characterized com-
plex from the reaction of iron(III) porphyrin with hypo-
chlorite in the absence of substrate; CompI is detected
for the iron porphyrin complex with the E(I/II) lower than
the E(OCl⋅/OCl–) while CompII is characterized for the
iron porphyrin complex with the E(I/II) higher than the
E(OCl⋅/OCl–).
Figure 4. Absorption spectral change for the reactions of
(a) 1-CompI and (b) 6-CompI with TBA-OCl in dichloro-
methane at –80 °C. Green line: CompI, red line: immedi-
ately after addition of 10 equiv of TBA-OCl, blue line:
after 1000 s.
The standard potential of one-electron oxidation pro-
cess of chloride, E°(Cl⋅/Cl–), has been reported to be 2.432
V (2.190 V vs. SCE).23 Since the E°(Cl⋅/Cl–) is much higher
than E(I/II) in dichloromethane, chloride cannot reduce
CompI to produce CompII and chlorine radical (Cl⋅) by
the one-electron redox process. Therefore, another re-
duction reaction by chloride should be considered. One
of the possible reactions is that chloride reduces CompI
to yield CompII and chlorine molecule (Cl2) because the
Mechanism of O–Cl Bond Cleavage. Although 5-
CompII and 6-CompII are detected as the final products
from the reactions of 5-Cl and 6-Cl with TBA-OCl, the
good yields of cyclooctene oxide from the catalytic reac-
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 13
be changed by the solvent, the E°(Cl⋅/Cl–) is much higher
standard potential of two-electron redox process of chlo-
ride, E°(Cl2/Cl–), has been known to be 1.358 V (1.116 V vs.
SCE) and close to E(I/II).23 In fact, we observed an irre-
versible broad redox peak at 1.02 V vs. SCE in the CV of
TBA-Cl in acetonitrile (Figure S1 and S2). In addition, we
previously reported the formation of Cl2 from the reaction
of CompI having pentafluorophenyl group with chlo-
ride.24 The reaction mechanism of CompI with chloride
is not clear at present, but Cl2 may be produced via the
coordination of chloride to the heme-iron center of
CompI, followed by the intermolecular coupling of the
iron-bound Cl to form CompII and Cl2. In the previous
report, we found the coordination of chloride to the iron-
center before the formation of CompII in the absorption
spectral change of the reduction reaction.24 The reaction
of CompI with chloride hardly occurs in the presence of
excess hypochlorite and substrate because this reaction is
much slower than the reactions with hypochlorite and
substrate. In fact, the second-order reaction rate constant
of 6-CompI having nitrate axial ligand with cyclooctene
at –80 °C is estimated to be 5.5 × 10 M-1s-1 (Figure S12) and
that of 6-CompI having chloride axial ligand is expected
to be 1-order magnitude greater than this rate.25 Under
the catalytic conditions where the concentration of sub-
strate is 1 ~ 3-order magnitude greater than that of hypo-
chlorite, the reaction rate of the epoxidation reaction
would be comparable to that of the reduction reaction of
hypochlorite, providing the epoxidation product from the
catalytic reactions.
1
2
3
4
5
6
7
8
than the E(I/II) of 1 ~ 6, listed in Table 1. Thus, the ther-
modynamic analysis of the iron-bound hypochlorite pre-
dicts that the state formed by the heterolytic cleavage is
more stable than that by the homolytic cleavage, resulting
in the formation of CompI for all of the iron porphyrin
complexes in this study. This is consistent to the results
of catalytic epoxidation reactions in this study.
9
Scheme 2. Thermodynamic analysis of the bond
cleavage process of the iron-bound terminal oxi-
dant (O–X).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The present analysis would be applicable to the bond
cleavage processes of other terminal oxidants. For exam-
ple, the heterolytic bond cleavage of the O–O bonds of
iron(III) hydroperoxide complex (X = OH) affords CompI
and hydroxide while the homolytic cleavage yields Com-
pII and hydroxyl radical. The standard potential of one
electron redox process of hydroxide, E°(HO⋅/HO–), is cal-
Thermodynamic Analysis of Bond Cleavage Pro-
cesses. To gain further insight into the bond cleavage
process, we considered the bond cleavage reactions from
thermodynamic analysis. Scheme 2 shows thermodynam-
ic analysis of the bond cleavage processes of the iron-
bound terminal oxidant (OX). The bond cleavage fashion
of the iron-bound terminal oxidant should be determined
by the relative stability between two states formed by the
heterolytic and homolytic bond cleavages. The relative
stability should depend on the free energies of reactions
for the heterolytic and homolytic processes. Moreover,
according to the Hess’s law, the relative stability can be
estimated from the difference of the ionization potentials
between CompII and X–, which are rationalized by their
redox potentials, E(I/II) and E(X⋅/X–), using the relation-
ship ꢀG° = –nFE°, where n is the number of electron in
the redox process, F is Faraday constant. When E(I/II) is
lower than E(X⋅/X–), the state formed by the heterolytic
cleavage is more stable than that by the homolytic cleav-
age and the O–X bond should be cleaved heterolytically.
However, when E(I/II) is higher than E(X⋅/X–), the O–X
bond should be cleaved homolytically.
culated to be +1.800
V (+1.558 V vs. SCE) from
E°(HO⋅/H2O) = +2.730 V and the pKa value of water.23
Although this value would be changed by the solvent, the
E°(HO⋅/HO–) is much higher than the E(I/II) values of
most iron porphyrin complexes (Figure 5). Therefore, the
heterolytic cleavage of the O–O bond occurs more prefer-
entially than the homolytic cleavage and CompI is gener-
ated from the iron(III) hydroperoxide complex even in the
absence of proton (Scheme 3).
As observed in the reactions of hypochlorite, the subse-
quent reduction reaction of CompI by hydrogen peroxide
Scheme 3. Reaction Scheme for the bond cleavage
of hydroperoxide complex.
For an iron(III) porphyrin hypochlorite adduct (X = Cl),
the heterolytic cleavage forms CompI and chloride while
the homolytic cleavage produces CompII and chlorine
radical. According to the above discussion, the bond
cleavage fashion should be determined by E(I/II) and
E(Cl⋅/Cl–). The standard redox potential of one-electron
redox process of chloride is reported to be E°(Cl⋅/Cl–) =
+2.432 V (+2.190 V vs. SCE). 23 Although this value would
ACS Paragon Plus Environment

Page 7 of 13
Journal of the American Chemical Society
has to be considered for some iron porphyrins and/or
under some conditions. The standard potential of one-
1
2
3
4
5
6
7
8
9
electron process of hydrogen peroxide, E°(HO2⋅/H2O2),
has been reported to be +1.46 V (+1.218 V vs. SCE) and in
the range of the E(I/II).23 Therefore, hydrogen peroxide
would become a possible reductant for CompI having
high E(I/II) value, such as CompI of meso-
pentafluorophenylporphyrin (TPFPP): E(I/II) = 1.60 V
(+1.358 V vs. SCE).19 The reaction of iron(III) complex of
TPFPP would initially form CompI by the heterolytic
cleavage of hydroperoxide, but the subsequent reduction
reaction of CompI with hydrogen peroxide may occur to
produce CompII in the absence of substrate. Further-
more, this reduction reaction would compete with the
oxygenation reaction in the presence of substrate
(Scheme 3). Even when the E(I/II) is lower than the
E°(HO2⋅/H2O2), the one electron reduction of CompI
with hydrogen peroxide would become a possible reaction
in neutral and basic aqueous solutions. As shown in Fig-
ure 5, the E°(HO2⋅/H2O2) and E(I/II) shift to lower poten-
tial as the pH becomes higher. In addition, since the pKa
of CompII is smaller than that of hydrogen peroxide, the
E(I/II) value becomes higher than the E°(HO2⋅/H2O2) val-
ue in neutral and basic aqueous solutions. In this case,
although the O–O bond is heterolytically cleaved, Com-
pII may be identified from the reactions of most iron(III)
porphyrin complexes with hydrogen peroxide. This is
consistent to the previous report that CompII is pro-
duced from hydrogen peroxide in basic conditions.7,8
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5. Dependence of E(HO⋅/H2O), E(HO2⋅/H2O2),
and E(I/II) on the pH of aqueous solution. The following
parameters are used for the preparation; E(I/II) = 1.0 V,
pKa(CompII) = 5.0, pKa(CompI) < 0, pKa(H2O2) = 11.7, T =
298 K. The heterolytic cleavage is expected at pH from 0
to 14 (the area drawn by pale blue). The regions drawn
by green and red indicate the pH regions expected to
form CompI and CompII as final products, respectively.
tion. Proton should be taken into account in the bond
cleavage reaction. The bond cleavage fashion is sensitive
to the pKa value of t-butanol in solvent used for the reac-
tion. E°(tBuO•/tBuOH) can be calculated to be +2.59V
(+2.35 V vs. SCE) in acetonitrile from the pKa value of 44.8
and much higher than E(I/II).26 Therefore, the state
formed by the heterolytic cleavage becomes more stable
than that by the homolytic cleavage, resulting in the for-
mation of CompI from the reaction of an iron(III) por-
phyrin with t-butyl hydroperoxide in acetonitrile. How-
ever, E°(tBuO•/tBuOH) can be calculated to be +1.07 V
(+0.828 V vs. SCE) in water from the pKa value of 19 if we
assume that E°(tBuO•/tBuO–) in water is the same as that
in acetonitrile.26 This value is smaller than E(I/II) of most
iron porphyrins, resulting in the homolytic cleavage in
water. In a water-organic solvent mixture, the pKa value
of t-butanol, and thus E°(tBuO•/tBuOH), becomes higher,
as the content of organic solvent is higher. When
E(tBuO•/tBuOH) becomes higher than E(I/II) in a water-
organic solvent mixture, the heterolytic cleavage occurs.
Moreover, the E(tBuO•/tBuOH) and E(I/II) are changed
by the pH of the solvent and the pKa value of CompII,
respectively, the bond cleavage fashion is also altered by
these factors. These predictions deduced from the pro-
posed mechanism are consistent with the previous exper-
imental results.7,8
In heme enzymes, the terminal oxygen atom of the
iron-bound hydroperoxide makes hydrogen bonding in-
teraction with amino acid residues and water molecules
in the distal side.4 This interaction is called the pull-effect
and has been proposed to assist the heterolytic cleavage
of the O–O bond to generate CompI and water. The pre-
sent analysis also provides reasonable explanation for the
pull effect.
As mentioned in above paragraph,
E°(HO⋅/H2O) is about 900 mV higher than E°(HO⋅/HO–).
This means that the state formed by the heterolytic cleav-
age is stabilized much more than the state formed by the
homolytic cleavage by the distal hydrogen bonding inter-
action: the pull effect. According to the linear free energy
relation, this drastic stabilization would lower the activa-
tion energy of the heterolytic bond cleavage process more
than the homolytic cleavage process, accelerating the het-
erolytic O–O bond cleavage reaction, i.e. the formation of
CompI.
We further apply this analysis to the O–O bond cleav-
age process of the iron-bound t-butyl peroxide (Scheme
4). The redox potential of one-electron redox process of
t-butoxide has been calculated to be E°(tBuO•/tBuO–) = –
0.058 V (–0.30 V vs. SCE) in acetonitrile and much lower
than E(I/II).25 Therefore, the state formed by the homo-
lytic cleavage is more stable than that by the heterolytic
cleavage, and CompII would be generated from the t-
butyl peroxide complex in aprotic solvent. However, the
bond cleavage fashion would be altered in the presence of
proton, such as the reactions of iron(III) porphyrins with
t-butyl hydroperoxide and the reactions in aqueous solu-
The subsequent reduction reaction of CompI to Com-
pII with excess t-butyl hydroperoxide is also a possible
reaction. The redox potential of one-electron redox pro-
cess of t-butyl hydroperoxide has been estimated to be
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 13
the line of the dependence of the E(I/II) intersects that of
Scheme 4. Reaction Scheme for the bond cleavage
of t-butyl peroxide complex.
the E(tBuO•/tBuOH) at pH = ~ 5.5. Therefore, the hetero-
lytic cleavage is expected at pH < 5.5 (the area shown by
pale purple) because the E(tBuO•/tBuOH) is larger than
E(I/II) while the homolytic cleavage is expected at pH >
5.5 (the area shown by pale yellow) because the E(I/II) is
larger than the E(tBuO•/tBuOH). Moreover, the identity
of the final compound in the absence of substrate may be
switched from CompI to CompII at pH = ~5.0 because
the line of the dependence of the E(I/II) intersects that of
the E(tBuO2•/tBuO2H) at pH = ~ 5.0. In the pH range
from 5.0 to 5.5, the O–O bond of t-butyl hydroperoxide is
heterolytically cleaved to form CompI, but CompII may
be characterized from the reaction because of the subse-
quent reduction reaction of CompI with t-butyl hydrop-
eroxide. In addition, under the catalytic reaction condi-
tions where large excess of t-butyl hydroperoxide and
substrate are present, the oxygenation reaction of CompI
with substrate would compete with the reduction reac-
tion in this range. The parameters used here are reasona-
ble values for an iron porphyrin complex having the o-
difluorophenyl or o-dichlorophenyl group at the meso
position.19,28 These analyses are consistent with the previ-
ous result that the high valent oxo-species generated by
the bond cleavage of the O–O bond of 2-methyl-1-
phenylpropan-2-yl hydroperoxide is switched from Com-
pI to CompII at pH = 4 ~ 5 in a solvent mixture of water-
methanol-acetonitrile (5:2:3).12
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
E(tBuO2•/tBuO2H) = +1.47 V vs NHE (+1.228 V vs. SCE) in
water at pH = 0.27 E(tBuO2•/tBuO2H) would also be
changed by the content of organic solvent and the pH of
solvent, as discussed for E°(tBuO•/tBuOH). The reduction
of CompI to CompII with t-butyl hydroperoxide would
be a possible reaction for CompI having high E(I/II) value
and in neutral and basic aqueous solution.7,8,12 After all,
the bond cleavage fashion of t-butyl hydroperoxide and
the identity of the final product are changed by the sol-
vent and pH conditions.
We show one example of the present thermodynamic
analysis for the prediction of the bond cleavage process of
t-butyl hydroperoxide in a water-organic solvent mixture
(Figure 6). Under the conditions described in Figure 6,
To further confirm these discussions, we investigated
the reactions of iron(III) TPFPP complex with t-butyl
hydroperoxide in organic solvents. From the thermody-
namic analysis, we expect that the reaction of iron(III)
TPFPP complex with t-butyl hydroperoxide leads to the
homolytic cleavage of the O–O bond to form CompII in
the presence of a base (in the absence of proton), but the
heterolytic cleavage to form CompI, followed by the re-
duction to CompII by the subsequent reaction with t-
butyl hydroperoxide in the presence of proton (methanol).
Moreover, we expect that the catalytic epoxidation reac-
tion with cyclooctene affords cyclooctene oxide in the
presence of proton because of the heterolysis of the O–O
bond, but does not produce the epoxide in the presence
of base because of the hemolysis of the O–O bond. The
absorption spectra of the reactions in the presence of tet-
ra-n-butylammonium hydroxide (TBA-OH) and in meth-
anol indicate the immediate formation of CompII at –40
°C (Figure S13). The catalytic reaction with cyclooctene in
the presence of TBA-OH at room temperature does not
afford cyclooctene oxide (yield: 0 % based on t-butyl hy-
droperoxide), but that in the presence of methanol pro-
duces cyclooctene oxide in good yield (yield: 51 % based
on t-butyl hydroperoxide). These results are consistent
with the predictions and support our proposed mecha-
nism.
Figure
6.
Dependence of E(tBuO•/tBuOH),
E(tBuO2•/tBuO2H), and E(I/II) on the pH in a water-
organic solvent mixture. The following parameters are
used for the preparation; E°(tBuO•/tBuOH) is 1.50 V,
E(tBuO2•/tBuO2H) = +1.47 V, E(I/II) = 1.40 V, pKa(CompII)
= 4.0, pKa(CompI) < 0, pKa(tBuO2H) > 14, T = 298 K. The
area drawn by pale blue and pale yellow indicate the pH
ranges expected the heterolytic and homolytic cleavages,
respectively. The regions drawn by green and red indi-
cate the pH regions expected to form CompI and CompII
as final products, respectively.
It should be pointed out here that the observation of
CompII from the reaction of iron porphyrin with a ter-
minal oxidant does not necessarily indicate the homolytic
cleavage of the O-X bond because it can be produced by
the subsequent reduction reaction of CompI formed from
ACS Paragon Plus Environment

Page 9 of 13
Journal of the American Chemical Society
Furthermore, these assignments were confirmed by the
the heterolytic cleavage of the O–X bond. In addition, the
1
2
3
4
5
6
7
8
subsequent reduction reaction of CompI usually com-
petes with the oxygenation reaction in the presence of
substrate when the E(I/II) is higher than the redox poten-
tial of the terminal oxidant. Therefore, while the high
yield of the reaction product suggests the heterolytic
cleavage of the O-X bond (the formation of CompI), a
low yield of the reaction product does not necessarily
indicate the homolytic cleavage.
dependence of the rate constant on the concentration of
TBA-OCl; the rate from 6-Cl to 6-(ClO)2 depends on the
concentration of TBA-OCl, but that from 6-(ClO)2 to 6-
CompII does not (Figure S15).
Based on the present results, we discuss the reaction
mechanism (Scheme 5). The reaction of ferric chloride
complex with hypochlorite initially forms an undetectable
mono-hypochlorite complex (OCl). The absorption spec-
tral change with clear isosbestic points and the pseudo-
first order kinetics for 1 and 2 indicate that the rate-
limiting step is the binding of hypochlorite to the iron
center: Cl → OCl. Therefore, 1-CompI and 2-CompI
would be formed via 1-OCl and 2-OCl, respectively. With
an increase in the electron-withdrawing effect of the me-
so-substituent, the reaction pathway via the bis-
hypochlorite complex would be more dominant. 3-
CompI would be produced via both 3-OCl and 3-(OCl)2,
resulting in the disappearance of the isosbestic points and
biphasic kinetics, as shown in Figure S3(b) and S14(c).
The fast reaction process would be assignable to the pro-
cess via 3-(OCl)2 because the strong axial ligand effect has
been know to accelerate the formation of CompI.11 For
more electron-deficient iron porphyrin (4 ~ 6), the reac-
tions via the bis-hypochlorite complexes are main reac-
tion pathways. 5-(OCl)2 and 6-(OCl)2 are observed as
metastable intermediates in the reactions. Although 4-
(OCl)2 cannot be detected in the reaction of 4-Cl with
TBA-OCl, but 4-(OCl)2 is characterized as an initial in-
termediate in the reaction of iron(III) porphyrin complex
having a weakly binding axial anion such as nitrate (Fig-
ure S16). The heterolytic cleavage occurs for 4-(OCl)2 ~ 6-
(OCl)2, but 5-CompII and 6-CompII are characterized as
the final products. Since 5-CompI and 6-CompI are not
detected in their absorption spectral change, the bond
cleavage step of 5-(OCl)2 and 6-(OCl)2 are the rate-
limiting steps. The estimated reaction rate constants are
summarized in Table 2. As shown in Table 2, both of the
binding step of hypochlorite to the iron-center and the
heterolytic bond cleavage step become slower as the elec-
tron-withdrawing effect of the meso-substituent is
stronger.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 2. Apparent Rate Constants for the Reactions
of 1 ~ 6 with TBA-OCl (1 mM) at –80 °C.
Cl → CompI
(s-1)
Cl → (OCl)2
(s-1)
(OCl)2 → CompI
(s-1)
(1.14 ± 0.05) × 10-1
(1.11 ± 0.04) × 10-1
1
2
3
(3.76 ± 0.28) ×
(1.92 ± 0.07) × 10-
10-2
1
(9.51 ± 0.42) × 10-
4
5
6
3
(7.66 ± 0.37) ×10- (3.26 ± 0.11) × 10-3
2
(1.46 ± 0.09) ×10- (1.40 ± 0.09) × 10-
2
3
Kinetics of the O–Cl Bond Cleavage. We examined
the kinetics of the bond cleavage processes of hypo-
chlorite by iron(III) porphyrins. The time courses of the
absorption for the bond cleavage processes are summa-
rized in Figure S14. The absorption spectral changes with
clear isosbestic points indicate that no intermediates,
such as mono-hypochlorite complex, OCl, and bis-
hypochlorite complex, (OCl)2, are accumulated in the
reactions of 1-Cl and 2-Cl with TBA-OCl (Figures 3 and
S3). The time courses of the absorbance for the reactions
can be fitted with a single exponential functions, provid-
ing apparent reaction rate constants from 1-Cl and 2-Cl to
1-CompI and 2-CompI, respectively. The reaction of 3-Cl
with TBA-OCl also affords 3-CompI, but the spectral
change does not afford clear isosbestic points (Figure
S3(b)). The time course of the absorbance cannot be sim-
ulated well with a single exponential function, but with a
double exponential function, suggesting two pathways
from 3-Cl to 3-CompI. The time of the absorbance at 680
nm for 4 can be simulated with a single-exponential func-
tion. The absorption spectral change for the reactions of
5-Cl and 6-Cl with TBA-OCl indicate the formation of 5-
(ClO)2 and 6-(ClO)2, followed by the formation of 5-
CompII and 6-CompII, respectively (Figures 3(c). and
S3(c)). The time courses of the absorbance for the con-
secutive reactions can be simulated well with a double
exponential function, providing the reaction rates from 5-
Cl and 6-Cl to 5-(ClO)2 and 6-(ClO)2, and from 5-(ClO)2
and 6-(ClO)2 to 5-CompII and 6-CompII, respectively.
Scheme 5. Proposed reaction mechanism for the
reaction of chloroiron(III) Porphyrin complex
with hypochlorite
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 13
without
further
purification.
4-Methoxy-2,6-
In summary, we studied here the bond cleavage pro-
cesses of hypochlorite by iron(III) porphyrin complexes of
1 ~ 6. This study showed that the O–Cl bonds of the iron-
bound hypochlorites are cleaved heterolytically to gener-
ate CompI for all of 1 ~ 6, but CompII is produced as a
final product by the subsequent reduction of CompI by
hypochlorite for 5 and 6. The essential factor in deter-
mining the final reaction compound, CompI or CompII,
is whether the E(I/II) is larger or smaller than
E(OCl⋅/OCl–). We propose a new thermodynamic analy-
sis based on the comparison between the redox potentials
of CompI and the terminal oxidant for predicting of the
bond cleavage of terminal oxidant. The bond cleavage
processes of hypochlorite, hydroperoxide, and t-butyl
peroxide are successfully explained by the thermodynam-
ic analysis. The heterolytic versus homolytic cleavage of
the metal-bound terminal oxidant (XO) can be predicted
by the comparison of E(I/II) with E(X⋅/X). In addition,
this analysis predicts that, in spite of the heterolyic cleav-
age, CompII may be identified as the final product, due
to the reduction of the formed CompI with a terminal
oxidant, when the E(I/II) is higher than the one-electron
oxidation potential of the terminal oxidant, E(OX⋅/OX).
dimethylbenzaldehyde was synthesized by the Vilsmeier reac-
tion of 3,5-dimethylanisole, followed by the purification by the
recrystallization
dimethylbenzaldehyde was prepared by the bromination of 5-
fluoro-1,3-dimethylbenzene, followed by the formylation with
dimethylamide and n-butyl litium.30
TBA-OCl in dichloromethane. Excess amount of solid NaOCl
･5H2O was added to 0.1 M dichloromethane solution of tetra-n-
butylammonium chloride at room temperature and the mixture
was stirred vigorously. The progress of the exchange reaction
was monitored by the absorption spectroscopy. TBA-OCl
formed from the anion exchange reaction showed the absorption
peak at 300 nm. Concentration of TBA-OCl was calculated from
the intensity of the absorbance at 300 nm with ε = 300 M-1cm-1.
After decantation, the TBA-OCl solution was cooled with ice.
The TBA-OCl solution was prepared every time, just before use,
because it gradually decomposed in dichloromethane.
1
2
3
4
5
6
7
8
in
hexane.29
4-fluoro-2,6-
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
meso-Tetraarylporphyrins.
meso-Tetraarylporphyrins (free
base of iron porphyrins 1 ~ 6) were prepared from pyrrole and
the corresponding benzaldehydes according to a previously pub-
lished procedure.31 There porphyrins were purified by silica gel
column with dichloromethane as an eluent. Spectroscopic data.
UV-vis (nm) in dichloromethane. 1: 419, 515, 548, 591, 647. 2: 418,
515, 547, 592, 650. 3: 417, 513, 545, 590, 646. 4: 417, 513, 544, 589,
1
645. 5: 418, 513, 589, 650, 704. 6: 419, 513, 589, 651, 705. H NMR
(400 MHz, ppm from TMS) in CDCl3 at 298 K: 1: 8.62 (py-H),
1.85 (o-CH3), 6.99 (m-H), 4.04 (p-OCH3), -2.54 (NH). 2: 8.38 (py-
H), 1.85 (o-CH3), 7.30 (m-H), 2.60 (p-CH3), -2.51 (NH). 3: 8.60
(py-H), 1.85 (o-CH3), 7.18, 7.15 (m-H), -2.56 (NH). 4: 8.62 (py-H),
7.68, 7.66, 7.63, 7.62 and 7.60 (m-H), 7.53 and 7.51 (p-H), 1.94 and
1.92 (o-CH3), -2.52 (NH). 5: 8.64 (py-H), 7.76 (m-H), 7.78 (p-H), -
2.56 (NH). 6: 8.66 (py-H), 7.82 (m-H), -2.64 (NH).
Ferric chloride complexes. Ferric chloride complexes of 1 ~ 6
were prepared by insertion of iron into free base porphyrins of 1
~ 6 with FeCl2 and sodium acetate in acetic acid, respectively,
and purified with a silica gel column.21 Spectroscopic data. UV-
vis (nm) in dichloromethane. 1: 377, 422, 508, 578, 662, 696. 2:
376, 419, 509, 577, 661, 696. 3: 374, 417, 506, 579, 655, 690. 4: 372,
418, 506, 581, 653, 683. 5: 374, 419, 508, 582, 648. 6: 360, 418, 507,
580, 645. 1H NMR (400 MHz, ppm from TMS) in CDCl3 at 298 K:
1: 79.8 (py-H), 15.2 and 13.7 (m-H), 4.86 (p-OCH3), 6.45 and 3.66
(o-C H3). 2: 79.5 (py-H), 15.9 and 14.3 (m-CH3), 4.14 (p- CH3),
6.46 and 3.91 (o- CH3). 3: 80.4 (py-H), 14.9 and 13.5 (m-H). 4: 81.0
(py-H), 15.2, 14.3, 13.7 and 13.0 (m-H), 8.22 (p-H). 5: 79.7 (py-H),
13.9 and 12.6 (m-H), 8.17 (p-H). 6: 80.1 (py-H), 13.8 and 12.6 (m-
H).
Experimental Section
Instrumentation. UV-visible absorption spectra were rec-
orded on an Agilent 8453 spectrometer (Agilent Technologies)
equipped with
a
USP-203 low-temperature chamber
1
(UNISOKU). H NMR spectra were measured on a Lambda-400
spectrometer (JEOL). The chemical shifts were referenced to the
residual peaks of the deutrrated solvents: dichloromethane (5.32
ppm) and chloroform (7.24 ppm). The concentrations of NMR
samples were 1 ~ 3 mM. EPR spectra were measured on EMX
Plus continuous-wave X-band spectrometer (Bruker). For EPR
measurements at 4 K, an ESR910 helium-flow cryostat (Oxford
Instruments) was used. The following parameters were com-
monly used for EPR measurements, microwave frequency: 9.46
GHz, modulation frequency: 100 kHz, modulation amplitude: 10
gauss, microwave power: 1 mW, time constant: 81.9 ms, receiver
gain: 30 dB, sample concentration: ~1mM. The cyclic voltammo-
grams and differential pulse voltammograms were measured
with an ALS612A electrochemical analyzer (BAS) in degassed
acetonitrile containing 0.1M tetra-n-butylammonium perchlo-
rate (TBAP) as a supporting electrolyte. A platinum electrode
was used as the working electrode and a platinum-wire electrode
was employed as the counter-electrode. The potentials were
recorded with respect to a saturated calomel electrode (SCE) as
the reference electrode. The reaction product analyses were
performed using a gas chromatograph mass-spectrometer GC-
MS QP-2010 SE (Shimadzu). Ozone gas was generated by the
UV irradiation of oxygen gas (99.999%) with an ozone generator
PR-1300 (Clear Water) and used without further purification.
Materials. Anhydrous organic solvents were obtained com-
mercially and stored in a glove box. Dichloromethane was puri-
fied by passing through alumina column just before use in the
glove box. Sodium hypochlorite pentahydrate (NaOCl･5H2O)
was purchased from Wako (Japan) and used without further
purification. Tetra-n-butylammonium chloride (≥ 99.0 %) was
obtained from Aldrich. t-Buthyl hydroperoxide (5 M solution in
decane) was purchased from Aldrich. The concentration of t-
Buthyl hydroperoxide was determined by the iodometric titra-
tion.7a Other chemicals were purchased commercially and used
Ferric nitrate complexes. Ferric nitrate complexes. Ferric ni-
trate complexes of 1 ~ 6 were synthesized from reaction of ferric
chloride complexes with 2 equivalent of silver(I) nitrate, respec-
tively, and purified by recrystallization from dichlorometane-
hexane.32 Spectroscopic data. UV-vis (nm) in dichloromethane. 1:
333(sh), 414, 515, 578, 660. 2: 333(sh), 412, 514, 578, 656. 3: 338(sh),
512, 579, 655. 4: 339(sh), 412, 512, 578, 650. 5: 339(sh), 412, 512,
1
580, 645. 6: 340(sh), 411, 510, 575, 642. H NMR (400 MHz, ppm
from TMS) in CD2Cl2 at 298 K: 1: 72.6 (py-H), 15.5 and 14.5 (m-
H), 4.92 (p-OCH3), 6.13 and 3.95 (o-CH3). 2: 73.0 (py-H), 16.4 and
15.2 (m-H), 4.28 (p-CH3), 6.14 and 3.99 (o-CH3). 3: 74.6 (py-H),
15.3 and 14.3 (m-H), 6.04 and 3.95(o-CH3). 4: 76.8, 75.4, 74.0 and
72.5 (py-H), 15.5, 14.6 and 13.5 (m-H), 8.24 (p-H). 5: 76.7 (py-H),
14.3 and 13.4 (m-H), 8.14 (p-H). 6: 77.1 (py-H), 14.1 and 13.3 (m-H).
Reactions of Ferric Porphyrin with Hypochlorite.
Chloroiron(III) porphyrin complex (100 μM) in dichloromethane
was placed in a 1 cm quartz cuvette in a low-temperature cham-
ber set on a UV-visible absorption spectrometer. After stabiliza-
tion of solution temperature at –80 °C, 10 equivalent TBA-OCl
ACS Paragon Plus Environment

Page 11 of 13
Journal of the American Chemical Society
was then added to the solution with stirring, and the reactions
on t-butyl hydroperoxide, were averages of three independent
experiments and hardly changed by the reaction time (30 ~ 120
min).
were monitored by the absorption spectral change at constant
time intervals. The reaction rate constants were estimated from
the simulations of time courses of the absorbance at 670 nm for
1, 666 nm for 2, 670 nm for 3, 680 nm for 4, 552 nm for 5, and
558 nm for 6 by commercial software, Igor Pro version 6.37
(WaveMetrics), with single- or double-exponential function.
Catalytic Epoxidation Reactions of Cyclooctene. Di-
chloromethane solution containing iron(III) porphyrin chloride
complex (1-Cl ~ 6-Cl) was cooled to 4 °C . Cyclooctene was dded
to the solution. The catalytic reaction was initiated by addition
of TBA-OCl. The initial concentrations of iron porphyrin, cy-
clooctene, and TBA-OCl were 10 ꢁM, 1.0 M, and 3.0 mM, respec-
tively. The reaction mixture was stirred for 10 min. The reaction
was quitted by addition of tetra-n-butylammonium iodide (30
mM). The reaction mixture was passed through silica gel short
column. After addition of standard sample (n-dodecane) to the
eluent, the products and their yields were analyzed by GC-MS.
The catalytic reactions at –20 °C were carried out in a tempera-
ture controlled cooling bath under the following conditions, iron
porphyrin: 10 ꢁM, TBA-OCl: 3.0 mM, cyclooctene: 10 mM.
1
2
3
4
5
6
7
8
Electrochemistry. Oxoiron(IV) porphyrin complexes of 1 ~ 3
were prepared by published method4 and electrochemical meas-
urements were carried out in acetonitrile containing 0.1 M tetra-
n-butylammonium perchlorate at -40 ºC. Oxoiron(IV) porphyrin
complexes of 4 ~ 6 were prepared by ozone oxidation of the cor-
responding ferric hydroxide complexes of 4 ~ 6 in acetonitrile
containing 0.1 M tetra-n-butylammonium perchlortate at –40 °C.
Excess ozone gas was removed by bubbling Ar gas. The working
and counter electrodes were directly immersed in the sample
solution and the reference electrode (SCE) was immersed in the
solution connecting the sample solution through the ion perme-
ability porous glass. The electrochemical measurements were
carried out before and after the ozone oxidation at –40 °C.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ASSOCIATED CONTENT
Supporting Information
Reactions of CompI with Chloride. Iron(III) porphyrin
chloride or nitrate complex (0.1 mM) in dichloromethane was
placed in a 1 cm quartz cell. The cell was set on the low-
temperature chamber and cooled to –80 °C. After stabilization
of temperature, ozone gas was bubbled in the solution. The
formation of CompI was monitored by the absorption spectros-
copy. After formation of CompI, excess ozone gas was removed
by bubbling argon gas. After reconfirming the absorption spec-
trum of CompI, 50 equivalent of TBA-Cl in dichloromethane was
added to the solution. After addition, the absorption spectral
change was monitored at constant interval.
Figure S1 – S16 and experimental details. This material
is available free of charge via the Internet at
http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
fujii@cc.nara-wu.ac.jp
Notes
The authors declare no competing financial interest.
Reactions of CompI with Hypochlorite. Iron(III) porphy-
rin chloride (for 1 ~ 4) or nitrate complex (for 5 and 6) (0.1 mM)
in dichloromethane was placed in a 1 cm quartz cell. The cell
was set on the low-temperature chamber and cooled to –80 °C.
After stabilization of temperature, ozone gas was bubbled in the
solution. The formation of CompI was monitored by the ab-
sorption spectroscopy. After formation of CompI, excess ozone
gas was removed by bubbling argon gas. After reconfirming the
absorption spectrum of CompI, 10 equivalent of TBA-OCl in
dichloromethane was added to the solution. After addition, the
absorption spectral change was monitored at constant interval.
Catalytic Epoxidation Reactions of Cyclooctene with t-
butyl hydroperoxide. Iron(III) TPFPP trifluoromethanesul-
fonate complex, TPFPP-Tf, was prepared by the reaction of
iron(III) TPFPP chloride complex with silver trifluoro-
methanesulfonate in dichloromethane and purified by recrystal-
lization.33 The catalytic reaction in the presence of TBA-OH was
carried out in acetonitrile-dichloromethane mixture (3:1). Cy-
clooctene was dissolved in dichloromthane. TPFPP-Tf, TBA-
OH•30H2O, and t-butyl hydroperoxide were dissolved in ace-
tonitrile. The total volume of the reaction solution was 2.0 mL
and the final concentration of heme, cyclooctene, TBA-OH, and
t-butyl hydroperoxide are 0.5 mM, 500 mM, 50 mM, and 50 mM,
respectively. The reaction mixture was stirred for 30 ~ 120 min at
room temperature. The products were analyzed by GC-MS after
addition of a standard sample (n-tetradecane) to the reaction
solution. The catalytic reaction in the presence of methanol was
carried out in dichloromthane-methanol mixture (7:3). The total
volume of the reaction solution was 2.0 mL and the final concen-
tration of heme, cyclooctene, and t-butyl hydroperoxide are 0.5
mM, 500 mM, and 50 mM, respectively. The reaction mixture
was stirred for 30 ~ 120 min at room temperature. The products
were analyzed by GC-MS after addition of a standard sample (n-
tetradecane) to the reaction solution. The product yields, based
ACKNOWLEDGMENT
This study was supported by grants from JSPS (Grant
Nos. 25288032 & 17H03032), CREST, Nara Women’s
University, The Naito Foundation, and The Uehara
Memorial Foundation. We thank IMS for assistance of
EPR measurements.
REFERENCES
(1) Liu, W.; Groves, J. T. Acc. Chem. Res. 2015, 48, 1727-1735.
(2) Nam, W. Acc. Chem. Res. 2007, 40, 522-531.
(3) Meunier, B.; Bernadou, J. In Metal-Oxo and Metal-Peroxo
Species in Catalytic Oxidations; Meunier, B. Ed.; Structure
and Bonding; Springer: Berlin, 2000, 97, pp 1−35.
(4) Poulos, T. L. Chem. Rev. 2014, 114, 3919-3962.
(5) Fujii, H. In Heme Peroxidases; Raven, E. and Dunford, B.
Eds.; RSC Metallobiology Series; The Royal Society of
Chemistry: London, 2016, pp 183-217.
(6) Groves, J. T. J. Inorg. Biochem. 2006, 100, 434-447.
(7) (a) Traylor, T. G.; Xu, F. J. Am. Chem. Soc. 1990, 112, 178-186.
(b) Traylor, T. G.; Xu, F. J. Am. Chem. Soc. 1987, 109, 6201-
6203. (c) Traylor, T. G.; Fann, W.-P.; Bandyopadhay, D. J.
Am. Chem. Soc. 1989, 111, 8009-8010. (d) Traylor T. G.; Cic-
cone, J. P. J. Am. Chem. Soc. 1989, 111, 8413-8420. (e) Traylor
T. G.; Tsuchoya, S.; Byun, Y.-S.; Kim, C. J. Am. Chem. Soc.
1993, 115, 2775-2781. (f) Traylor, T. G.; Kim, C.; Richards, J. L.;
Xu, F.; Perrin, C. L. J. Am. Chem. Soc. 1995, 117, 3468-3474.
(8) (a) Bruice, T. A. Acc. Chem. Res. 1991, 24, 243-249. (b)
Almarsson, Ö; Bruice, T. C. J. Am. Chem. Soc. 1995, 117, 4533-
4544. (c) He, G.-X.; Bruice, T. C. J. Am. Chem. Soc. 1991, 113,
2747-2753.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 13
(9) Allentoff, A. L.; Bolton, J. L.; Wilks, A.; Thompson, J. A.;
(30) (a) Hu, Y.; Ishikawa, Y.; Hirai, K.; Tomioka, H. Bull. Chem.
Oritz de Montellano, P. R. J. Am. Chem. Soc. 1992, 114, 9744-
9749.
(10) (a) Groves, J. T.; Watanabe, Y. J. Am. Chem. Soc. 1986, 108,
7834-7836. (b) Groves, J. T.; Watanabe, Y. J. Am. Chem. Soc.
1986, 108, 7836-7837.
Soc. Jpn. 2001, 74, 2207-2218. (b) Gruenberg, M. F.; Goos-
sen, L. J. Chem. Eur. J. 2013, 19, 7334-7337.
(31) Lindsey, J. S.; Wagner, R. W. J. Org. Chem. 1989, 54, 828-
836.
(32) Wyllie, G. R. A.; Munro, O. Q.; Schulz, C. E.; Sheidet, W. R.
Polyhedron, 2007, 26, 4664-4672.
(33) Gismelseed, A.; Bominaar, E. L.; Bill, E.; Trautwein, A. X.;
Winkler, H.; Nasri, H.; Doppelt, P.; Mandon, D.; Fischer, J.;
Weiss, R. Inorg. Chem. 1990, 29, 2741.
1
2
3
4
5
6
7
8
(11) Yamaguchi, K.; Watanabe, Y.; Morishima, I. J. Am. Chem.
Soc. 1993, 115, 4058-4065.
(12) (a) Nam, W.; Choi, H. J.; Han, H. J.; Cho, S. H.; Lee, H. J.;
Han, S.-Y. Chem. Commun., 1999, 387-388. (b) Nam, W.;
Han, H. J.; Oh, S.-Y.; Lee, Y. J.; Choi, M.-H.; Han, S.-Y.; Kim,
C.; Woo, S. K.; Shin, W. J. Am. Chem. Soc. 2000, 122, 8677-
8684.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(13) Van Heuvelen, K. M.; Fielder, A. T.; Shan, X.; DeHont, R. F.;
Meier, K. K.; Bominaar, E. L.; Münck, E.; Que, Jr. L. Proc.
Natl. Acad. Sci. 2012, 109, 11933-11938.
(14) (a) Nam, W.; Lim, M. H.; Oh, S.-Y. Inorg. Chem. 2000, 39,
5572-5575. (b) Guo, M.; Dong, H.; Li, J.; Cheng, B.; Huang, Y.
Q.; Feng, Y. Q.; Lei, A. Nat. Commun. 2012, 3, 1190. (c)
Collman, J. P.; Zeng, L.; Brauman, J. I. Inorg. Chem. 2004, 43,
2672-2679.
(15) (a) Wang, C.; Kurahashi, T.; Fujii, H. Angew. Chem. Int. Ed.
2012, 51, 7809-7811. (b) Wang, C.; Kurahashi, T.; Inomata,
K.; Hada, M.; Fujii, H. Inorg. Chem. 2013, 52, 9557-9566.
(16) (a) Jacobsen, E. N.; Zhang, W.; Muci, A. R.; Ecker, J. R.;
Deng, L. J. Am. Chem. Soc. 1991, 113, 7063-7064. (b) Palucki,
M.; Finney, N. S.; Pospisil, P. J.; Güler, M. L.; Ishida, T.; Ja-
cobsen, E. N. J. Am. Chem. Soc. 1998, 120,948-954.
(17) Meunier, B.; Guilmet, E.; Caralho, M. E. D.; Poilblanc, R. J.
Am. Chem. Soc. 1984, 106, 6668-6676.
(18) Cong, Z.; Yanagisawa, S.; Kurahashi, T.; Ogura, T.;
Nakashima, S.; Fujii, H. J. Am. Chem. Soc. 2012, 134, 20617-
20620.
(19) Takahashi, A.; Kurahashi, T.; Fujii, H. Inorg. Chem. 2011, 50,
6922-6928.
(20) (a) Groves, J. T.; Gross, Z.; Stern, M. K. Inorg. Chem. 1994, 33,
5065-5072. (b) Jeong, Y. J.; Kang, Y.; Han, A. R.; Lee, Y. M.;
Kotani, H.; Fukuzumi, S.; Nam, W. Angew. Chem. Int. Ed.
2008, 47, 7321-7324. (c) Pan, Z.; Newcomb, M. Inorg. Chem.
2007, 46, 6767-6774.
(21) Fujii, H. J. Am. Chem. Soc. 1993, 115, 4641-4648.
(22) Fujii, H.; Yoshimura, T.; Kamada, H. Inorg. Chem. 1996, 35,
2373-2377.
(23) Armstrong, D. A.; Huie, R. E.; Lymar, S.; Koppenol, W. H.;
Merényi, G.; Neta, P.; Stanbury, D. M.; Steenken, S.; Ward-
man, P. Bioinorg. React. Mech. 2013, 9, 59-61.
(24) Cong, Z.; Kurahashi, T.; Fujii, H. Angew. Chem. Int. Ed. 2011,
50, 9935-9939.
(25) Takahashi, A.; Kurahashi, T.; Fujii, H. Inorg. Chem. 2009, 48,
2614-2625.
(26) (a) Workentin, M. S.; Maran, F.; Wayner, D. D. M. J. Am.
Chem. Soc. 1995, 117, 2120-2121. (b) Bietti, M.; DiLabio, G. A.;
Lanzalunga, O.; Salamone, M.; J. Org. Chem. 2010, 75, 5875-
5881. (c) Donkers, R. L.; Maran, F.; Wayner, D. D. M.; Wor-
kentin, M. S. J. Am. Chem. Soc. 1999, 121, 7239-7248. (d)
Wayner, D, D. M.; Parker, V. D. Acc. Chem. Res. 1993,
26,287-294.
(27) Das, T. N.; Dhanasekaran, T.; Alfassi, Z. B.; Naeta, P. J. Phys.
Chem. 1998, 102, 280-284.
(28) Boaz, N. C.; Bell, S. R.; Groves, J. T. J. Am. Chem. Soc. 2015,
137, 2875-2885.
(29) Kang, Y.-F.; Yan, W.-J.; Zhou, T.-W.; Dai, F.; Li, X.-Z.; Bao,
X.-Z.; Du, Y.-T.; Yuan, C.-H.; Wang, H.-B.; Re, X.-R.; Liu, Q.;
Jin, X.-L. Zhou, B.; Zhang, J. Chem. Eur. J. 2014, 20, 8904-
8908.
ACS Paragon Plus Environment

Page 13 of 13
Journal of the American Chemical Society
Critical Factors in Determining the Heterolytic versus Homolytic Bond Cleavage of Terminal Oxidants by
Iron(III) Porphyrin Complexes
1
2
3
4
Sawako Yokota and Hiroshi Fujii*
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
13
ACS Paragon Plus Environment