Mechanistic insight into peroxo-shunt formation of biomimetic models for compound II, Their reactivity toward organic substrates, and the influence of N-methylimidazole axial ligation

DOI: 10.1002/chem.201303694

Full Paper

&

Reaction Mechanisms

Mechanistic Insight into Peroxo-Shunt Formation of Biomimetic

Models for Compound II, Their Reactivity toward Organic

Substrates, and the Influence of N-Methylimidazole Axial

Ligation**

Maria Oszajca,^{[a, b]}Agnieszka Drzewiecka-Matuszek,^[c]Alicja Franke,^[b]Dorota Rutkowska-

Zbik,^[c]Małgorzata Brindell,^[a]Małgorzata Witko,^[c]Graz˙yna Stochel,*^[a]and Rudi van Eldik*^{[a, b]}

Abstract: High-valent iron-oxo species have been invoked

as reactive intermediates in catalytic cycles of heme and

nonheme enzymes. The studies presented herein are devot-

ed to the formation of compound II model complexes, with

the application of a water soluble (TMPS)Fe^III(OH) porphyrin

([meso-tetrakis(2,4,6-trimethyl-3-sulfonatophenyl)porphinato]-

iron(III) hydroxide) and hydrogen peroxide as oxidant, and

their reactivity toward selected organic substrates. The kinet-

ics of the reaction of H₂O₂with (TMPS)Fe^III(OH) was studied

as a function of temperature and pressure. The negative

values of the activation entropy and activation volume for

the formation of (TMPS)Fe^IV=O(OH) point to the overall asso-

ciative nature of the process. A pH-dependence study on

the formation of (TMPS)Fe^IV=O(OH) revealed a very high re-

activity of OOH^ꢀtoward (TMPS)Fe^III(OH) in comparison to

H₂O₂. The influence of N-methylimidazole (N-MeIm) ligation

on both the formation of iron(IV)-oxo species and their oxi-

dising properties in the reactions with 4-methoxybenzyl al-

cohol or 4-methoxybenzaldehyde, was investigated in detail.

Combined experimental and theoretical studies revealed

that among the studied complexes, (TMPS)Fe^III(H₂O)(N-MeIm)

is highly reactive toward H₂O₂to form the iron(IV)-oxo spe-

cies, (TMPS)Fe^IV=O(N-MeIm). The latter species can also be

formed in the reaction of (TMPS)Fe^III(N-MeIm)₂with H₂O₂or

in the direct reaction of (TMPS)Fe^IV=O(OH) with N-MeIm. In-

terestingly, the kinetic studies involving substrate oxidation

by (TMPS)Fe^IV=O(OH) and (TMPS)Fe^IV=O(N-MeIm) do not dis-

play a pronounced effect of the N-MeIm axial ligand on the

reactivity of the compound II mimic in comparison to the

OH^ꢀsubstituted analogue. Similarly, DFT computations re-

vealed that the presence of an axial ligand (OH^ꢀor N-MeIm)

in the trans position to the oxo group in the iron(IV)-oxo

species does not significantly affect the activation barriers

calculated for CꢀH dehydrogenation of the selected organic

substrates.

Introduction

voted to iron(IV)-oxo cation radical porphyrin species, termed

compound I (cpd. I). The very high reactivity of the latter was

confirmed in numerous biomimetic and enzyme studies with

application of synthetic iron(III) porphyrins and thermophilic

P450 isolated from Sulfolobus acidocaldarius, respectively.^[1–3]

However, the transient character and high instability of the re-

active intermediates in the catalytic cycle of P450 enzymes

hamper their trapping and direct characterisation under phys-

iological conditions. Therefore, there is still a debate on the

nature of the reactive intermediate(s) responsible for substrate

oxidation.^[4,5]Accordingly, in a number of enzyme studies,^[5,6]

cpd. 0 (ferric hydroperoxo intermediate) is proposed as

a second oxidant species apart from cpd. I. On the other hand,

computational results exclude cpd. 0 as potential oxidant due

to its significant basicity and proton affinity, which is reflected

in the relatively high energy barriers calculated for sulfoxida-

tion and epoxidation reactions.^[7]

In the catalytic cycles of heme enzymes, such as catalase, per-

oxidase and cytochrome P450, it is assumed that high-valent

iron(IV)-oxo species are key intermediates in oxygenation reac-

tions of organic substrates. Most of the attention has been de-

[a] M. Oszajca, Dr. M. Brindell, Prof. G. Stochel, Prof. R. van Eldik

Faculty of Chemistry, Jagiellonian University

Ingardena 3, 30-060 Krakow (Poland)

E-mail: stochel@chemia.uj.edu.pl

[b] M. Oszajca, Dr. A. Franke, Prof. R. van Eldik

Department of Chemistry and Pharmacy

University of Erlangen-Nꢀrnberg, Egerlandstrasse 1

91058 Erlangen (Germany)

E-mail: vaneldik@chemie.uni-erlangen.de

[c] A. Drzewiecka-Matuszek, Dr. D. Rutkowska-Zbik, Prof. M. Witko

Jerzy Haber Institute of Catalysis and Surface Chemistry

Polish Academy of Sciences, Niezapominajek 8

30-239 Krakow (Poland)

Another oxidant species, iron(IV)-oxo porphyrin (cpd. II), was

for a long time not considered to be responsible for CꢀH hy-

droxylation or other oxidation reactions. As a consequence,

little attention has been devoted to the study of models for

[**] Compound II=iron(IV)-oxo porphyrin

Supporting information for this article is available on the WWW under

http://dx.doi.org/10.1002/chem.201303694.

Chem. Eur. J. 2014, 20, 2328 – 2343

2328

ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

cpd. II.^[8,9]When Groves et al. reported that (TMP)Fe^IV=O is able

to oxidise olefins, cpd. II began to be perceived as an effective

oxidant playing a complementary role to cpd. I in oxygenation

reactions.^[10]More recent studies have demonstrated the reac-

tion of H₂O₂with iron(III) porphyrins is of particular interest

due to its biological relevance and environmental friendliness.

Therefore, one of the aims of our work focuses on the study of

the kinetics and mechanism of the reaction of hydrogen perox-

tivity of cpd. II in other reaction types, such as activation of Cꢀ ide with synthetic iron(III) porphyrins. Earlier insight into reac-

H bonds of alkanes and alkyl aromatics, as well as hydride

transfer of dihydronicotinamide adenine dinucleotide (NADH)

analogues.^[11]Furthermore, comparative studies on the reactivi-

ty of model complexes for the most discussed reactive inter-

mediates in P450 catalysis (namely, cpds. 0, I and II, where

cpd. 0 is the hydroperoxo iron(III) porphyrin intermediate) of

various types of reactions, such as OꢀH abstraction, CꢀH ab-

straction, epoxidation, sulfoxidation and hydride transfer, re-

vealed that cpd. II is the most efficient catalyst for hydride-

transfer reactions.^[12]

tions of water-soluble non-m-oxo dimer-forming porphyrins

with hydrogen peroxide was reported by Bruice et al. who

studied the pH and buffer dependence of the reaction of H₂O₂

with iron(III) porphyrins.^[20,21]

In this context, we present the results of a detailed mecha-

nistic study on the reaction of hydrogen peroxide with

a water-soluble iron(III) porphyrin, [meso-tetrakis(2,4,6-trimeth-

yl-3-sulfonatophenyl)porphinato]iron(III), Fe^III(TMPS), in alkaline

(pH 10) aqueous solution (Scheme 1a). In addition, the oxidis-

The active species in the catalytic cycle of various heme en-

zymes strongly differ in their function. The ability to catalyse

different types of reactions is connected with the structural dif-

ferences in the prosthetic group and its environment, which is

connected with the nature of an amino acid coordinated in

the proximal position to the iron(III) centre in the active site

(i.e., tyrosine for catalases, histidine for peroxidases and cys-

teine for P450). Thus, much effort has been devoted to the elu-

cidation of the effect of the axial ligands on the Fe^IV=O bond

strength and the reactivity of cpd. I models.^[3,13–15]To our

knowledge, there are no experimental studies on the influence

of axial ligation on the reactivity of iron(IV)-oxo species, the

model complex for cpd. II. Theoretical calculations showed

only insignificant axial ligand influence on the reaction barriers

of substrate oxidation by (Por)Fe^IV=O.^[9,16]Even a strong

s donor, such as hydrosulfide, has no significant effect on the

H-abstraction barrier on the ground-state triplet surface, how-

ever, it has a pronounced increasing effect on the energy barri-

er on the quintet surface.^[8]

Resonance Raman studies have shown that the presence of

N-MeIm or Im^ꢀas an axial ligand in iron(IV)-oxo complexes has

a significant influence on the v_Fefrequency.^[17]It is reasonable

=

O

to expect that the weaker Fe=O bond in iron(IV)-oxo species

should result in a higher reactivity of the latter toward organic

substrates. However, the first experimental results showed that

in the case of various substituted models for cpd. I, no correla-

tion between the Fe=O bond energy and oxidation capability

could be found.^[3,18]

Very recent results published by Takahashi et al.^[19]show

a correlation between the reactivity of (TMP^+·)Fe^IV=O(L) and

the redox potential of Fe^II/Fe^IIIin (TMP)Fe^III(L). Accordingly, this

study suggests that the reactivity of model cpd. I complexes

with various anionic axial ligands depends on the thermody-

namic stability of the final reaction product, that is,

(TMP)Fe^III(L).^[19]These findings lead to the question of what in-

fluence, if any, the nature of the axial ligand has on the reactiv-

ity of cpd. II model complexes toward organic substrates?

In biomimetic studies on the reactivity of high-valent

iron(IV)-oxo species, the synthetic iron(III) porphyrin complexes

are oxidised in so called “peroxo-shunt” reactions by the appli-

cation of various sacrificial oxidants, such as meso-chloroperox-

ybenzoic acid, iodosylbenzene or hydrogen peroxide. The reac-

Scheme 1. a) Structure of [meso-tetrakis(2,4,6-trimethyl-3-sulfonatophenyl)-

porphinato]-iron(III), (TMPS)Fe^III, and its reaction with H₂O₂to produce cpd. II.

b) Structures of the studied substrates.

ing capability of the hydroperoxide anion (HOO^ꢀ) towards

(TMPS)Fe^III(OH) was examined in kinetic experiments per-

formed at higher pH (up to 12.6). The studies were performed

in the absence and presence of various concentrations of N-

methylimidazole in order to gain more mechanistic informa-

tion on the role of axial ligands on the formation and reactivity

of Fe^IV=O in the studied system. Furthermore, we report the

complete catalytic cycle in which the cpd. II analogues with

different axial ligands (hydroxo or N-methylimidazole) oxidise

selected organic substrates (4-methoxybenzyl alcohol and 4-

methoxybenzaldehyde; Scheme 1b), which leads to almost

complete reformation of the catalyst. To the best of our knowl-

edge, this is the first time that the influence of N-methylimida-

zole (N-MeIm) as axial ligand on the reactivity of in situ gener-

ated iron(IV)-oxo species is examined experimentally.

Chem. Eur. J. 2014, 20, 2328 – 2343

www.chemeurj.org

2329

ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

The experimental results were supplemented by quantum

chemical calculations using density functional theory (DFT)

within the B3LYP functional. Such an approach has become

a standard tool for the clarification of catalytic reaction mecha-

nisms by iron-porphyrins.^[16,22]The combination of experimen-

tal and theoretical study enabled us to gain more detailed in-

sight into the role of the axial ligand on the reactivity of cpd. II

model complexes.

Results and Discussion

Before the basic solutions of (TMPS)Fe^III(OH) containing various

concentrations of N-methylimidazole were subjected to the ox-

idation reaction with hydrogen peroxide, systematic spectro-

scopic and kinetic studies on the substitution reaction of

(TMPS)Fe^III(OH) with N-MeIm were carried out for a detailed

speciation of the Fe(TMPS) complex under the selected reac-

tion conditions (for further details see the Supporting Informa-

tion). The Fe^III(TMPS) complex exists in aqueous solution as

monomeric diaqua-ligated [(TMPS)Fe^III(H₂O)₂]^3ꢀat low pH, and

as monohydroxo-ligated [(TMPS)Fe^III(OH)]^4ꢀat higher pH (pK_a=

6.9). [(TMPS)Fe^III(H₂O)₂]^3ꢀand [(TMPS)Fe^III(OH)]^4ꢀrepresent the

six-coordinate spin-admixed (S=3/2, 5/2, Int.% ~26%) and

predominantly five-coordinate high-spin (S=5/2) complexes,

respectively.^[23]

Figure 1. Spectral changes observed during formation of the (TMPS)Fe^IV=

O(OH) complex after addition of H₂O₂to (TMPS)Fe^III(OH). Experimental condi-

tions: [(TMPS)Fe^III(OH)]=1ꢁ10^ꢀ5m, [H₂O₂]=8ꢁ10^ꢀ5m, [borate buffer]=

0.05m, pH 10.0, T=58C.

observed when a subequivalent amount of meta-chloroperoxy-

benzoic acid (m-CPBA) was added to an aqueous solution of

(TMPS)Fe^III(OH) at pH 10.0. Thus, the application of these two

sacrificial oxidants leads to the formation of the same reactive

iron(IV)-oxo species, which can be regarded as a cpd. II ana-

logue. This observation is in line with previous reports on the

reaction of water-soluble Fe^IIIporphyrins with hydrogen perox-

ide and organic peroxides in alkaline aqueous solutions, as

well as spectro-electrochemical generation of (TMPS)Fe^IV=

O.^[25–27]Comparison of the electronic spectral changes obtained

in the reaction with both employed oxidants, indicated that

formation of the iron(IV)-oxo species in the reaction with hy-

drogen peroxide is considerably less efficient than in the reac-

tion with m-CPBA, and did not achieve completion even at

a relatively high excess of oxidant. In contrast, the addition of

even a subequivalent amount of m-CPBA appears to be suffi-

cient to achieve a relatively high reaction yield. This can arise

from the low reactivity of (TMPS)Fe^III(OH) toward hydrogen

peroxide combined with a relatively fast decomposition of the

generated high-valent iron(IV) porphyrin, which strongly de-

pends on the selected concentration of H₂O₂.

In the following we present results for the oxidation of

(TMPS)Fe^III(OH) by H₂O₂in the absence of N-methylimidazole at

pH 10, kinetic studies of the formation of iron(IV)-oxo as a func-

tion of pH, the oxidation of (TMPS)Fe^III(OH) by H₂O₂in the pres-

ence of N-methylimidazole at pH 10, and experimental and

theoretical (DFT) studies on the oxidation of 4-methoxybenzyl

alcohol and 4-methoxybenzaldehyde.

Oxidation of the (TMPS)Fe^III(OH) by H₂O₂in the absence of

N-MeIm at pH 10

Oxidation of (TMPS)Fe^III(OH) by H₂O₂was carried out in a buffer

at pH 10.0. It was shown previously that the acidity of the reac-

tion medium strongly affects the catalytic activity of water-

soluble iron(III) porphyrins, which was ascribed to pH-depen-

dent changes in iron(III) porphyrin speciation and its electro-

chemical properties.^[24,25]Accordingly, a gradual change in the

nature of the product formed upon oxidation of Fe^III(TMPS),

The iron(IV)-oxo species formed under such reaction condi-

tions can be considered as a six-coordinate intermediate with

either H₂O or OH^ꢀligands trans to the oxo group. In order to

resolve which of these two ligands binds more strongly to the

iron(IV)-oxo species, DFT calculations were performed for the

binding of L (where L=H₂O, OH^ꢀ) to (TMPS)Fe^IV=O (the triplet

state as the lower lying iron spin state in (TMPS)Fe^IV=O was

used for these calculations). The optimised structures and

bond lengths for (TMPS)Fe^IV=O(L) are presented in Figure S3 in

the Supporting Information. Based on the theoretical calcula-

tions, the most probable structure for [(TMPS)Fe^IV=O(L)]^4ꢀ/5ꢀin-

cludes the OH^ꢀligand in the trans position to the oxo group

since the binding energies for OH^ꢀand H₂O to [(TMPS)Fe^IV=

O]^4ꢀare ꢀ13.4 and 0.2 kcalmol^ꢀ1, respectively. In view of the

fact that the investigated complexes are anionic species, the

IV

from the oxo-iron(IV) porphyrin cation radical (TMPS⁺)Fe =O

C

at low pH to oxo-iron(IV) porphyrin (TMPS)Fe^IV=O at high pH

was observed.^[25]Therefore, to stabilise the cpd. II analogue as

a sole reactive intermediate, basic conditions were selected in

the present studies.

The time-resolved UV/Vis spectral changes indicated that

the oxidation of (TMPS)Fe^III(OH) by H₂O₂resulted in the accu-

mulation of a new species with a characteristic Soret band

maximum at 420 nm and Q band maximum at 524 nm, which

can be ascribed to the spectrum of the iron(IV)-oxo species

(Figure 1).^[25,26]

The formation of (TMPS)Fe^IV=O(OH) was followed by a slow

reformation of the starting (TMPS)Fe^III(OH) complex accompa-

nied by porphyrin damage. Analogous spectral changes were

Chem. Eur. J. 2014, 20, 2328 – 2343

www.chemeurj.org

2330

ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

effect of the solvent is expected

not to be negligible. In this con-

text, the above-presented values

and those provided later already

include a correction for the sol-

vent effect (e=80). Thus, the

presence of H₂O in the sixth co-

ordination position can be ex-

cluded since formation of

[(TMPS)Fe^IV=O(H₂O)]^4ꢀis not en-

ergetically favoured.

Scheme 2. Schematic representation of the formation of (TMPS)Fe^IV=O and (TMPS^+·)Fe^IV=O.

the formation of oxo-iron(IV) porphyrin or oxo-iron(IV) porphy-

IV

The reaction sequence for the formation of (TMPS)Fe^IV= rin cation radical ((Por⁺)Fe =O), respectively (Scheme 2).

C

O(OH) consists of H₂O₂binding, leading to the formation of

the hydroperoxo-iron(III) intermediate, followed by OꢀO bond

cleavage. In terms of formation of the hydroperoxo species,

the acidity of H₂O₂is expected to play an important role. Hy-

drogen peroxide (H₂O₂, pK_a=11.6) is known to be a very poor

ligand and it was shown that it does not displace H₂O and

Taking into account the low thermodynamic stability of

+

IV

C

(Por )Fe =O species at the studied pH, the subsequent very

fast process leading to the conversion of the latter species to

(Por)Fe^IV=O should be kept in mind.^[25]Therefore, it is reasona-

ble to assume that even if the heterolytic OꢀO bond cleavage

occurs, it should be followed by fast comproportionation of

ꢀ

IV

III

IV

ClO₄in the gallium(III) porphyrin centre.^[28]Since, the pK_a

(TMPS⁺)Fe =O and (TMPS)Fe (OH) to form (TMPS)Fe =O and/

or reduction by H₂O₂under condition of an excess of H₂O₂, ac-

cording to Scheme 2.

C

value of H₂O₂bound to the iron centre is expected to be

much lower (pK_a=3.2 to 4 for H₂O₂coordinated to Fe^IIIcentre

of peroxidase hemin)^[29]than that of bulk H₂O₂, the formation

of (TMPS)Fe^III(OOH) can proceed even at a much lower pH

than 11.6.^[30]Since under the studied reaction conditions

(pH 10) both forms of hydrogen peroxide, H₂O₂and OOH^ꢀ, are

present in solution (2.4% of hydrogen peroxide exists in the

deprotonated form at this pH), the formation of hydroperoxo

intermediate may in principle occur with the participation of

both these forms. The first reactivity scenario involves direct

protonation of the coordinated OH^ꢀligand by H₂O₂, leading to

the formation of H₂O and OOH^ꢀ, by which the formed OOH^ꢀ

can easily substitute the labile coordinated water molecule at

the iron(III) centre. By analogy for the monohydroxo complex,

it is reasonable to assume that hydroperoxo intermediate

exists as a five-coordinate species [(TMPS)Fe^III(OOH)]^4ꢀas pro-

posed previously.^[25]If a water molecule is present in the sixth

coordination site trans to OOH^ꢀ, it will be very weakly bound,

such that the hydroperoxo species can be better described as

a five-coordinate complex. This assumption is in agreement

with previous reports, which provide evidence for the exis-

tence of a hydroperoxo intermediate as a high-spin, five-coor-

dinate iron(III) species.^[31,32]The proposed mechanism is as-

sumed to operate at lower pH where H₂O₂exists in the reac-

tion medium as main reactive species. At higher pH values

where the OOH^ꢀform predominates, another reactivity pattern

involving direct coordination of OOH^ꢀto the iron(III) centre in

the trans position to OH^ꢀshould be taken into consideration

(see below for more details). Since under the studied reaction

conditions (pH 10) 97.6% of hydrogen peroxide exists in its

protonated form, further discussion of the results will concern

the first proposed mechanism involving H₂O₂as an oxidant.

Formation of high-valent iron(IV) porphyrin species is pre-

ceded by OꢀO bond cleavage in the hydroperoxo-iron(III) in-

termediate, which is followed by fast OH^ꢀbinding as present-

ed in Equation (1). In the case of the reaction with hydrogen

peroxide, the initial coordination of oxidant can be followed

by homolytic or heterolytic OꢀO bond cleavage resulting in

Detailed kinetic studies on the reaction of (TMPS)Fe^III(OH)

with hydrogen peroxide were performed by the application of

stopped-flow techniques and revealed the formation of

(TMPS)Fe^IV=O(OH) in a single kinetic step.^[33]Kinetic measure-

ments were carried out under pseudo-first- order conditions

(at least eightfold excess of H₂O₂over the porphyrin concentra-

tion). First-order rate constants (k_obs) were determined by fit-

ting a single-exponential function to the recorded kinetic

traces. The dependence of k_obson the hydrogen peroxide con-

centration showed a linear behaviour with an intercept at the

origin (Figure 2). The overall reaction scheme is presented in

Equation (1).

Due to decomposition of the porphyrin complex on increas-

ing [H₂O₂], it was not possible to study the oxidation reaction

Figure 2. Dependence of k_obson H₂O₂concentration for the reaction with

(TMPS)Fe^III(OH) in the absence of N-MeIm, and in the presence of 30- , 150-

and 500-fold excess of N-MeIm over the porphyrin complex. Experimental

conditions: [(TMPS)Fe^III(OH)]=1ꢁ10^ꢀ5m, [borate buffer]=0.05m, pH 10.0,

T=58C.

Chem. Eur. J. 2014, 20, 2328 – 2343

www.chemeurj.org

2331

ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

¼

DS and DV given in Table 1

are the mean values from two

independent measurements.

The overall negative activation

entropy and volume data deter-

mined for the formation of

under conditions of higher [H₂O₂]. Therefore, the identification

of the rate-determining step in the reaction sequence shown

in Equation (1) is not straightforward. If the formation of the

hydroperoxo intermediate is a fast pre-equilibrium with OꢀO

bond scission as the rate-determining step, then saturation ki-

netics should be observed at appropriate high oxidant concen-

tration. On the other hand, the observation that the efficiency

of the formation of the high-valent oxo species with H₂O₂is

relatively low in comparison with the reaction yield when m-

CPBA is used as an oxidant, along with the fact that the pres-

ence of the electron-rich TMPS porphyrin ring is supposed to

significantly accelerate the OꢀO bond scission in the hydroper-

oxo complex, support the assumption that in the studied

system the formation of Fe^III(TMPS)(OOH) should be the rate-

determining step. Thus, the observed first-order rate constant

for the oxidation reaction in Equation (1) can be expressed

with Equation (2) by application of the steady-state approxima-

tion to Fe^III(TMPS)(OOH):

(TMPS)Fe^IV=O(OH) from the reaction of H₂O₂with

(TMPS)Fe^III(OH) point to the overall associative nature of the

process, which can be expressed as a composite of three

terms in Equations (3) and (4). Formation of [(TMPS)Fe^III-

(OOH)]^4ꢀvia protonation of coordinated OH^ꢀby H₂O₂, followed

¼

by coordination of OOH^ꢀis expected to give negative DS (k₁)

¼

and DV (k₁) values. The second term involves OꢀO bond scis-

sion, oxidation of Fe^IIIto Fe^IVand the transition of a high-spin

five-coordinate [(TMPS)Fe^III(OOH)]^4ꢀspecies to a low spin

[(TMPS)Fe^IV=O]^4ꢀ.^[34]In terms of entropy and volume changes

the expected contributions of these processes can partially

¼

compensate each other such that DS (k₂) and DV (k₂) are ex-

pected to have small absolute values, most probably slightly

negative. Since k should be accompanied by positive activa-

ꢀ1

tion entropy and volume data, the sum (k_ꢀ1+k₂) should again

be accompanied by small absolute activation entropy and

volume data, most probably slightly positive depending on

the magnitudes of k and k₂. Thus, for Equation (1) it is rea-

ꢀ1

sonable to expect overall negative activation entropy and

volume data as observed in our experiments.

k₁k₂½H₂O₂ꢁ

k₁k₂

_ꢀ1þ k₂

ð2Þ

k_obs

¼

;

k_on

¼

k

_ꢀ1þ k₂

¼

ð3Þ

ð4Þ

DS ¼ DS ðk₁Þ þ DS ðk₂ÞꢀDS ðk_ꢀ1þ k₂Þ

¼

DV ¼ DV ðk₁Þ þ DV ðk₂ÞꢀDV ðk_ꢀ1þ k₂Þ

The second-order rate constant, k_on, determined by fitting

a linear function to the obtained data at 58C (in the absence

of added N-MeIm) is equal to 536(ꢂ16)m^ꢀ1

s

^ꢀ1. Kinetic meas-

Oxidation of (TMPS)Fe^III(OH) by H₂O₂in the absence of N-

MeIm as function of pH

urements as a function of temperature and pressure allowed

¼

the determination of the activation parameters (DH , DS ,

DV ) for the formation of (TMPS)Fe^IV=O(OH). The influence of

¼

In order to examine the oxidising ability of the hydroperoxide

anion (HOO^ꢀ) toward (TMPS)Fe^III(OH), we extended our study

to higher pH values. Kinetic studies on the pH-dependent

(pH 9–12.6) formation of (TMPS)Fe^IV=O(OH) indicated a dramatic

increase in the rate constant with increasing pH (Figure 3).

temperature on the combined value of k₁k₂/(k +k₂) was mea-

ꢀ1

sured in the temperature range from 5 to 258C. The resulting

positive activation enthalpy and slightly negative activation en-

tropy were determined from the slope and the intercept of the

linear dependence of ln(k_onT^ꢀ1) versus T^ꢀ1, respectively (Fig-

ure S4a in the Supporting Information and Table 1). In order to

determine the influence of pressure on k_on, the oxidation of

(TMPS)Fe^III(OH) by H₂O₂was examined in the pressure range

10–130 MPa. The obtained linear dependence of ln(k_on) versus

pressure allowed the determination of the activation volume

¼

(Figure S4b in the Supporting Information). The values of DH ,

Table 1. Rate constants and activation parameters for oxidation of Fe^III-

(TMPS) by H₂O₂in the absence and presence of N-MeIm.

[N-MeIm]

0m

[N-MeIm]

3ꢁ10^ꢀ4m^[a]1.5ꢁ10^ꢀ3m^[b]5ꢁ10^ꢀ3m^[c]

k_on[m^ꢀ1s^ꢀ1

]

536ꢂ16

41ꢂ1

950ꢂ12

38ꢂ2

400ꢂ8

204ꢂ10

46ꢂ1

¼

DH [kJmol^ꢀ1

]

–

¼

DS [Jmol^ꢀ1K^ꢀ1

]

ꢀ43ꢂ3

ꢀ6.1ꢂ0.5

ꢀ50ꢂ8

ꢀ6.8ꢂ0.2

ꢀ35ꢂ4

ꢀ13ꢂ3

¼

DV [cm³mol^ꢀ1

Figure 3. Dependence of the observed rate constant for the formation of

(TMPS)Fe^IV=O(OH) on pH. Experimental conditions:

[(TMPS)Fe^III(OH)]=1ꢁ10^ꢀ5m, [H₂O₂]=2ꢁ10^ꢀ4m, T=258C.

[a] 30-fold excess; [b] 150-fold excess; [c] 500-fold excess.

Chem. Eur. J. 2014, 20, 2328 – 2343

www.chemeurj.org

2332

ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

This behaviour reflects the much higher reactivity of OOH^ꢀ

than H₂O₂, since no pH-dependent process related with

(TMPS)Fe^IIIis observed in this pH range.^[20,35]The pK_avalue de-

termined from the pH dependence shown in Figure 3 is

11.26ꢂ0.03, which is in good agreement with the acid dissoci-

ation constant of hydrogen peroxide (pK_a=11.6). The depend-

ence of the observed rate constant for the formation of

(TMPS)Fe^IV=O(OH) on pH allowed the determination of the lim-

iting rate constants for the reaction of (TMPS)Fe^III(OH) with

H₂O₂(k₁=8.9(ꢂ0.5)ꢁ10²m^ꢀ1

s

^ꢀ1) and OOH^ꢀ(k₂=4.8(ꢂ0.1)ꢁ

10⁴m^ꢀ1^ꢀ1). The second-order rate constants determined from

s

the linear dependence of k_obsversus total [H₂O₂] (Figure S5 in

the Supporting Information) at pH 10 and 11.6 and 258C are

1.66(ꢂ0.03)ꢁ10³m^ꢀ1s^ꢀ1and 3.15(ꢂ0.03)ꢁ10⁴m^ꢀ1

s

^ꢀ1, respec-

tively.

A comparison of these values reveals an over 50-fold in-

crease in reactivity on going from H₂O₂to OOH^ꢀas oxidant.

This indicates that at higher pH, OOH^ꢀshould be the sole reac-

tive species responsible for the generation of the cpd. II mimic.

To gain more insight on the interaction between OOH^ꢀand

[(TMPS)Fe^III(OH)]^4ꢀ, DFT calculations were performed. Two pos-

sible reaction pathways were considered in which OOH^ꢀat-

tacks either the Fe^IIIcentre in the trans position to OH^ꢀ, or the

coordinated OH^ꢀligand. A comparison of computed interac-

tion energies indicated that the former is favoured by 6.3 kcal

mol^ꢀ1. This suggests that the interaction between OOH^ꢀand

[(TMPS)Fe^III(OH)]^4ꢀmost probably involves attack of OOH^ꢀon

the Fe^IIIcentre trans to the OH^ꢀgroup. However, it is assumed

that the resulting product will be very unstable due to the

very fast homolytic cleavage of the OꢀO bond as a result of

the trans effect of the strong donor OH^ꢀligand.

Figure 4. Spectral changes observed during formation of the (TMPS)Fe^IV=

O(N-MeIm) complex: a) after addition of [H₂O₂]=8ꢁ10^ꢀ5m to (TMPS)Fe^IIIin

the presence of a 75-fold excess of N-MeIm (7.5ꢁ10^ꢀ4m), b) after addition of

a 100-fold excess of N-MeIm (1ꢁ10^ꢀ3m) to the solution of (TMPS)Fe^IV=

O(OH). Experimental conditions: [(TMPS)Fe^III(OH)]=1ꢁ10^ꢀ5m, [borate buf-

fer]=0.05m, pH 10.0, T=208C.

Oxidation of (TMPS)Fe^III(OH) by H₂O₂in the presence of N-

MeIm at pH 10

In order to mimic the catalytic active site in peroxidase en-

zymes which have histidine imidazole coordinated as axial

ligand, we explored the influence of N-methylimidazole on the

oxidation of (TMPS)Fe^IIIby H₂O₂. Since N-alkylimidazole deriva-

tives do not possess a free NꢀH moiety which could serve as

general acid catalyst, the use of N-MeIm in the present study

enables the examination of only the proximal effect of the ni-

trogen base (push effect) on the formation of high-valent oxo-

iron porphyrin species.^[36]

midazole substituted oxo-iron(IV) porphyrin, namely (TMP)Fe^IV=

O(N-MeIm).^[17,32,37]On the basis of these observations, it is sug-

gested that the new species formed in the reaction of

(TMPS)Fe^III(OH) with H₂O₂in the presence of N-MeIm can be at-

tributed to (TMPS)Fe^IV=O(N-MeIm).

These conclusions are further supported by theoretical cal-

culations of binding energies (DE_b), which showed that the

binding ability of N-MeIm to the [(TMPS)Fe^IV=O]^4ꢀcomplex

(DE_b=ꢀ94.1 kcalmol^ꢀ1) is much higher than that of the OH^ꢀ

anion (DE_b=ꢀ13.4 kcalmol^ꢀ1). The optimised structure and

bond lengths of (TMPS)Fe^IV=O(L), L=N-MeIm, OH^ꢀ, are pre-

sented in Figure S3 in the Supporting Information. This sug-

gests that N-MeIm binds much stronger than OH^ꢀsuch that

Rapid scan measurements showed that in the presence of

appropriately high N-MeIm concentrations, the reaction of the

iron(III) porphyrin with hydrogen peroxide leads to the forma-

tion of a new species characterised by the red shifted Soret

and Q bands maxima in comparison to the spectrum of

(TMPS)Fe^IV=O(OH) (compare Figures 1 and Figure 4a). More-

over, the addition of N-MeIm to the solution of (TMPS)Fe^IV= when N-MeIm is present in solution, it can easily coordinate to

O(OH) generated in situ in the reaction of (TMPS)Fe^III(OH) with

H₂O₂, resulted in spectral changes that suggest the formation

of the same species (Figure 4b). Similar shifts in the Soret and

Q bands to longer wavelengths have already been observed

following the addition of N-MeIm to (TMP)Fe^IV=O in organic

solvent, and have been ascribed to the formation of N-methyli-

the oxo-Fe^IVcentre and substitute the hydroxo ligand in

[(TMPS)Fe^IV=O(OH)]^5ꢀ. Clearly, the Fe^IV=O group must play an

important role in the stability of the mentioned complexes.

Moreover, it is assumed that the coordination of an anionic

ligand, such as OH^ꢀ, to the iron centre which is already ligated

to an O^2ꢀaxial group and the porphyrin ring possessing signifi-

Chem. Eur. J. 2014, 20, 2328 – 2343

www.chemeurj.org

2333

ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

We performed systematic kinetic studies on the reaction of

H₂O₂with (TMPS)Fe^IIIin the presence of three different N-MeIm

concentrations, corresponding to the highest, medium and

lowest values of k_obs(Figure 5). The linear dependence of k_obs

on [H₂O₂] allowed us to determine the second-order rate con-

stants at the selected [N-MeIm] (Figure 2) and the resulting

values of k_onare included in Table 1.

It is suggested that the increase in k_obswith increasing [N-

MeIm] observed in the low N-MeIm concentration range can

be ascribed to the formation of a small fraction of the mono-

N-methylimidazole substituted species, (TMPS)Fe^III(H₂O)(N-

MeIm) [see the Supporting Information for more details],

which is more reactive than (TMPS)Fe^III(OH) toward H₂O₂. A

similar behaviour was observed by Beck et al.^[38]for the reac-

tion of [5,10,15,20-tetrakis(2,6-dimethyl-3-sulfonatophenyl)-

porphinato]iron(III) with tBuOOH in the presence of nitrogen

bases in aqueous solution. Another important effect, which is

suggested to contribute to the acceleration of the oxidation

reaction in the presence of N-MeIm, is the “push effect” of the

N-MeIm axial ligand in the (TMPS)Fe^III(OOH)(N-MeIm) intermedi-

ate. As already shown, the presence of N-MeIm in the trans po-

sition to the OꢀO bond leads to the destabilisation of the

latter by pushing electron density into the antibonding orbitals

of the hydroperoxide group.^[39]As a consequence, the activa-

tion energies for heterolysis/homolysis of the OꢀO bond de-

crease significantly and result in the acceleration of the forma-

tion high-valent iron-oxo species.

Scheme 3. General reaction scheme for the formation of (TMPS)Fe^IV=O(N-

MeIm).

cant negative charge (TMPS^6ꢀ), would be a less favourable pro-

cess than the binding of a neutral ligand, such as N-MeIm. This

assumption is in line with computational calculations and with

the experimental finding showing that the formation of the di-

hydroxo derivative of (TMPS)Fe^III(high negatively charged spe-

cies) cannot be observed even under conditions of very high

pH.^[23]

Indeed, the (TMPS)Fe^IV=O(N-MeIm) complex can be ob-

tained, experimentally, in two different ways (Scheme 3). How-

ever, the formation of (TMPS)Fe^IV=O(N-MeIm) according to

route A was more effective than via route B. This arises from

the fact that the addition of N-MeIm to a solution of

(TMPS)Fe^IV=O(OH) causes depletion of the latter due to its par-

tial reduction.

The formation of the mono-N-MeIm complex with water in

the sixth coordination position was confirmed by kinetic and

theoretical studies. The rate constant for N-MeIm release from

(TMPS)Fe^III(N-MeIm)₂does not show a significant pH depend-

ence, which is in agreement with the presence of (TMPS)Fe^III-

(H₂O)(N-MeIm) as an intermediate. The existence of

(TMPS)Fe^III(OH)(N-MeIm) would result in a significant influence

of pH on the rate constant for the back reaction when OH^ꢀis

substituted by N-MeIm (see the Supporting Information for

more details).

In order to study the influence of N-MeIm on the hydrogen

peroxide oxidation of (TMPS)Fe^III(OH), the formation of the

oxo-iron(IV) porphyrin was followed as a function of N-MeIm

concentration under condition of constant [H₂O₂]. The oxida-

tion reaction was studied in the N-MeIm concentration range

0–5ꢁ10^ꢀ3m, which corresponds to 0–500-fold excess over the

porphyrin concentration. At relatively low [N-MeIm], up to

about 30-fold excess over the porphyrin, k_obsincreased with in-

creasing [N-MeIm], followed by a monotonic decrease reaching

the constant value, k=3.58(ꢂ0.07)ꢁ10^ꢀ2s^ꢀ1, at very high

[MeIm] (>400-fold excess over the porphyrin; Figure 5).

To further support the conclusions drawn from the experi-

mental study, the binding energies (DE_b) of the axial ligands

(H₂O, OH^ꢀ, N-MeIm) to the [(TMPS)Fe^III(N-MeIm)]^3ꢀcomplex

were determined by application of DFT calculations. Different

spin multiplicities of the (TMPS)Fe^III(N-MeIm) species were in-

vestigated and the quartet state, S=3/2 was chosen as the en-

ergetically lowest spin state. The optimised structures and

bond lengths of (TMPS)Fe^III(N-MeIm)(L), where L=H₂O, OH^ꢀ

and N-MeIm, are shown in Figure S6 in the Supporting Infor-

mation. According to the theoretical calculations when OH^ꢀis

present in the trans position, a higher stabilisation by the axial

ligand is achieved (DE_b=ꢀ76.8 kcalmol^ꢀ1) in contrast to a H₂O

ligand for which the binding energy is much more positive

(DE_b=ꢀ21.9 kcalmol^ꢀ1). The fact that the mono-N-methylimi-

dazole substituted intermediate did not really accumulate in

the reaction medium (only traces of this species could be ob-

served experimentally), implies that a relatively weak stable

complex is formed as an intermediate under the selected reac-

tion conditions. The value of DE_bfound for the formation of

the bis-N-methylimidazole substituted complex (DE_b=

Figure 5. Dependence of k_obson the N-MeIm concentration for the forma-

tion of iron(IV)-oxo species. Experimental conditions: [(TMPS)Fe^III(OH)]=

1ꢁ10^ꢀ5m, [H₂O₂]=2ꢁ10^ꢀ4m, [N-MeIm]=(0–5)ꢁ10^ꢀ3m, [borate buf-

fer]=0.05m, pH 10.0, T=58C.

Chem. Eur. J. 2014, 20, 2328 – 2343

www.chemeurj.org

2334

ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

ꢀ62.1 kcalmol^ꢀ1) is significantly more negative than for

[(TMPS)Fe^III(H₂O)(N-MeIm)]^3ꢀ, which accounts for the fact that

the bis-N-methylimidazole complex accumulates in the reac-

tion medium even under conditions of very low N-MeIm con-

centration. The very strong tendency of OH^ꢀto bind to

[(TMPS)Fe^III(N-MeIm)]^3ꢀexcludes the formation of [(TMPS)Fe^III-

(N-MeIm)(OH)]^4ꢀas an intermediate on the way to the forma-

tion of [(TMPS)Fe^III(N-MeIm)₂]^3ꢀ, since in such a case the mono-

N-methylimidazole complex should be easily observed experi-

mentally, because DE_bfor the second N-MeIm binding is much

more positive (DE_b=ꢀ62.1 kcalmol^ꢀ1).

[(TMPS)Fe^III(OH)]. The obtained results enabled the construc-

tion of Eyring plots and the calculation of the activation en-

thalpy and entropy from the slope and the intercept, respec-

tively (Figure S7a in the Supporting Information). The depend-

ence of ln(k_on) on pressure revealed a linear behaviour and was

used to determine the activation volume from the slope (Fig-

ure S7b). The obtained values of the activation parameters are

listed in Table 1. A comparison of the data determined for the

formation of the Fe^IV=O species in the absence and presence

of added N-MeIm does not show any significant difference in

the activation parameters except for the activation volume

(Table 1). This can be due to an effect of the composite nature

of these parameters as seen from Equation (6). The activation

volume for the formation of high-valent iron species in the

presence of a large excess of N-MeIm is twice as high as the

values found in the absence of N-MeIm or in the presence of

a low N-MeIm excess. The overall negative activation entropy

and volume data point to the overall associative nature of the

process. Such behaviour is predicted to result from a more

negative activation volume for the binding of H₂O₂to

On going to higher N-MeIm concentrations, the values of

k_obsfor the formation of the oxo-iron(IV) porphyrin species

become smaller. In the presence of a 500-fold excess of N-

MeIm over the porphyrin, (TMPS)Fe^III(N-MeIm)₂exists as a sole

reacting species, the oxidation of which leads to the formation

of an oxo-iron(IV) intermediate. Comparison of the first-order

rate constant, k=3.58(ꢂ0.07)ꢁ10^ꢀ2s^ꢀ1, obtained in the pres-

ence of the high N-MeIm concentration with the first-order

rate constant, 4.7(ꢂ0.8) s^ꢀ1, determined for the release of N-

MeIm from (TMPS)Fe^III(N-MeIm)₂complex (Figure S2 in the Sup-

porting Information), indicates that N-MeIm dissociation from

the bis-N-methylimidazole substituted species cannot be the

rate-determining step for the binding of H₂O₂to (TMPS)Fe^III(N-

MeIm)₂. Thus, it is reasonable to expect that the formation of

the high-valent oxo-iron species under conditions of high [N-

MeIm] does not involve the mono-N-MeIm substituted inter-

mediate and should occur through a direct reaction of the

(TMPS)Fe^III(N-MeIm)₂complex with the oxidant. On the other

hand, H₂O₂is a very weak ligand and does not seem to be

able to substitute N-MeIm in (TMPS)Fe^III(N-MeIm)₂. Taking to-

gether all these considerations, the rational explanation for the

observed reactivity behaviour is that the oxidation of

(TMPS)Fe^III(N-MeIm)₂to (TMPS)Fe^IV=O(N-MeIm) under condi-

tions of high N-MeIm concentrations must proceed through

the protonation of coordinate N-MeIm by H₂O₂and formation

of N-MeImH⁺and OOH^ꢀ. Protonated N-MeIm cannot coordi-

nate efficiently to the iron centre and can be easily substituted

by OOH^ꢀ.

¼

[(TMPS)Fe^III(MeIm)₂]^3ꢀ(DV (k₁’)) associated with charge genera-

tion on going to the transition state from the reactant

[(TMPS)Fe^III(MeIm)₂]^3ꢀto the product [(TMPS)Fe^III(MeIm)-

(OOH)]^4ꢀstate, and is connected with an increase in solvent

electrostriction.

When only a small excess of N-MeIm is present in solution,

the activation parameters for the formation of Fe^IV=O are in

good agreement within the experimental error limits with the

activation parameters obtained in the absence of N-MeIm.

Under such conditions there are three reactive species, namely

(TMPS)Fe^III(OH), (TMPS)Fe^III(H₂O)(MeIm) and (TMPS)Fe^III(MeIm)₂,

present in solution that can simultaneously undergo oxidation

reactions with H₂O₂. Therefore, the activation enthalpy, entropy

and volume measured under such conditions are only appar-

ent activation parameters since they represent composite tem-

perature and pressure effects that originate from these three

reactions. In this context, the interpretation of temperature

and pressure effects for the reaction of H₂O₂with the

[(TMPS)Fe^III(H₂O)(MeIm) intermediate seems to be difficult to

assign.

The general reaction scheme for the oxidation of (TMPS)Fe^III-

(N-MeIm)₂species under such conditions is presented by Equa-

tion (5). Accordingly, the observed rate constant can be ex-

pressed by Equation (6):

Substrate oxidation by iron(IV)-oxo complexes

We studied the oxidation of 4-methoxybenzyl alcohol (4-MB-

alc) and 4-methoxybenzaldehyde (4-MB-ald) by iron(IV)-oxo

species generated in situ in the reaction of (TMPS)Fe^III(OH) with

hydrogen peroxide in buffered aqueous solution at pH 10.0

and 208C.

Product analysis revealed that the major products in the re-

action mixture were 4-methoxybenzaldehyde (11(ꢂ2)%), based

on the amount of total [H₂O₂] added to the reaction mixture,

and 4-methoxybenzoic acid (16(ꢂ2)%) in the oxidation reac-

tion of 4-methoxybenzyl alcohol and 4-methoxybenzaldehyde,

respectively. Similar reaction products were identified in sub-

strate oxidation reactions with porphyrin cation radical

iron(IV)-oxo species.^[40,41]It is suggested that the relatively low

product yield can result from an inefficient formation of

k⁰k⁰½H₂O₂ꢁ

k₁⁰k₂⁰

k_o⁰_bs

¼

;

k_on

¼

1

2

ð6Þ

k_ꢀ⁰₁½N ꢀ MeImꢁ þ k₂⁰

In order to examine the influence of N-MeIm on the forma-

tion of the iron(IV)-oxo species at various temperatures and

pressures, we studied the oxidation of (TMPS)Fe^III(OH) in the

presence of 30- and 500-fold excess of [N-MeIm] over the

Chem. Eur. J. 2014, 20, 2328 – 2343

www.chemeurj.org

2335

ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

iron(IV)-oxo species in the reaction with H₂O₂, as well as the

ability of H₂O₂to act as a reducing agent for (TMPS)Fe^IV=O(OH)

which leads to a depletion of H₂O₂. The samples for product

analysis were obtained under inert conditions due to the ob-

served influence of dioxygen on the product yield. In the case

of 4-MB-alc oxidation, only a slight increase in product yield

was observed, whereas for the oxidation of 4-MB-ald per-

formed in the presence of O₂, a very high yield of 4-methoxy-

benzoic acid was obtained. This indicates that carbon radicals

of 4-MB-alc and 4-MB-ald formed via CꢀH abstraction may

escape from the solvent cage and be scavenged by O₂to yield

the corresponding peroxyl radicals which undergo further reac-

tions.^[42]

Scheme 4. General scheme for the production of iron(IV)-oxo species, and

the subsequent reaction with selected substrates. Note that the iron(III) por-

phyrin complex is denoted as (TMPS)Fe^III(X)_n, in which X=OH, n=1 or N-

MeIm, n=2.

Treatment of the iron(IV)-oxo complex, generated by mixing

(TMPS)Fe^III(OH) with H₂O₂in the concentration ratio of 1:1, with

100 equivalents of 4-MB-ald resulted in a very fast reformation

of the iron(III) species followed by a slow reaction indicating

reformation of the oxo-iron(IV) porphyrin species (Figure 6).

Figure 7. Decomposition of (TMPS)Fe^IV=O(OH) after addition of various con-

centrations of 4-methoxybenzyl alcohol. Experimental conditions:

[(TMPS)Fe^III(OH)]=1ꢁ10^ꢀ5m, [H₂O₂]=5ꢁ10^ꢀ6m, [borate buffer]=0.05m,

pH 10.0, T=58C.

Figure 6. Kinetic trace recorded at 428 nm for the formation of (TMPS)Fe^IV=

O(OH) and its decomposition after addition of 1ꢁ10^ꢀ3m 4-methoxybenzal-

dehyde. Experimental conditions: [(TMPS)Fe^III(OH)]=1ꢁ10^ꢀ5m, [H₂O₂]=

1ꢁ10^ꢀ5m, [borate buffer]=0.05m, pH 10.0, T=58C.

conditions of a constant substrate concentration and varying

concentrations of H₂O₂and vice versa (Figure S8 in the Sup-

porting Information).

The observed reactivity behaviour arises from an inefficient

formation of (TMPS)Fe^IV=O, which did not reach completion

even at a relatively high excess of H₂O₂, accounting for the

presence of unconsumed hydrogen peroxide in solution which

caused the re-initiation of the catalytic cycle (Scheme 4). More-

over, even if a subequivalent amount of H₂O₂was used in the

presence of a small excess of the substrate, the existence of

unconsumed oxidant suppressed the reformation of the iron-

(III) complex by the re-initiation of the catalytic cycle

(Scheme 4, reaction 1). The kinetic traces obtained under such

conditions deviated markedly from a single exponential char-

acter (Figure 7).

In order to gain further insight into the reactivity behaviour,

which is controlled by the operation of the catalytic cycle, ad-

ditional studies were performed at pH 11.6. Since under such

conditions the second-order rate constant for the formation of

(TMPS)Fe^IV=O(OH) is much larger than that at pH 10, and the

yield of the formed iron(IV)-oxo species is also much higher,

the observed reactivity behaviour is totally different from that

observed at lower pH (compare Figure S9a and b in the Sup-

porting Information). Namely, the observed reduction of

iron(IV)-oxo species and the reformation of the starting

(TMPS)Fe^III(OH) proceeds already under catalytic conditions,

which in turn does not allow the direct determination of the

second-order rate constant for the reaction between

(TMPS)Fe^IV=O(OH) and substrate, as can be done for example

at lower pH where the slower and inefficient formation of

iron(IV)-oxo species does not interfere with the first direct reac-

tion. Thus, at pH 11.6, the observed rate constant is slower

than at pH 10 due the operation of the catalytic cycle shown

in Scheme 4 that will continue until all the H₂O₂has been used

Detailed kinetic studies revealed that the complication de-

scribed above depends strongly on the concentration ratio of

free oxidant and substrate present in the reaction medium, as

well as the ratio of the second-order rate constants for oxo-

iron(IV) porphyrin formation and substrate oxidation. This as-

sumption is nicely supported by an experiment in which the

degree of (TMPS)Fe^IV=O(OH) reduction was examined under

Chem. Eur. J. 2014, 20, 2328 – 2343

www.chemeurj.org

2336

ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

up. Studies on substrate oxidation in the presence of various

concentrations of H₂O₂nicely show the influence of uncon-

sumed oxidant on the reaction rate for the reformation of

(TMPS)Fe^III(OH) that slows down with increasing [H₂O₂] (Fig-

ure S10 in the Supporting Information). Under the higher pH

conditions where the second-order rate constant for the for-

mation of the iron(IV)-oxo species is significantly larger than its

reduction by substrate, the sequential addition of hydrogen

peroxide (even in excess over the porphyrin) to the solution

containing a significant excess of the substrate, leads to re-ini-

tiation of the catalytic cycle (Scheme 4 and Figure 8).

Figure 8. Kinetic trace recorded at 428 nm for the formation of (TMPS)Fe^IV=

O(OH) and its decomposition after several additions of [H₂O₂]=5ꢁ10^ꢀ5m,

and 6ꢁ10^ꢀ3m 4-methoxybenzaldehyde. Experimental conditions:

[(TMPS)Fe^III(OH)]=1ꢁ10^ꢀ5m, pH 11.6, NaOH, I=0.05m NaNO₃, T=58C.

Figure 9. a) Kinetic trace recorded at 428 nm for the formation of

(TMPS)Fe^IV=O(OH) and its decomposition after addition of:

1) [H₂O₂]=1ꢁ10^ꢀ5m, and 2) 2ꢁ10^ꢀ3m 4-methoxybenzaldehyde. b) Typical

kinetic trace (black line) with single exponential fit (grey line) registered for

reduction of (TMPS)Fe^IV=O(OH) back to (TMPS)Fe^III(OH) upon addition of

6ꢁ10^ꢀ3m 4-methoxybenzyl alcohol. Experimental conditions:

[(TMPS)Fe^III(OH)]=1ꢁ10^ꢀ5m, [borate buffer]=0.05m, [H₂O₂]=5ꢁ10^ꢀ6m,

pH 10.0, T=58C.

Based on these observations, it was reasonable to study the

reactivity of the iron(IV)-oxo species toward selected substrates

under conditions of a subequivalent concentration of hydro-

gen peroxide and relatively high concentration of alcohol or al-

dehyde (>1 mm). Indeed, selection of such conditions allowed

us to observe complete reduction of the oxo-iron(IV) poprhyrin

to the starting iron(III) porphyrin species with kinetic traces dis-

playing single exponential behaviour (Figure 9).

rate constant determined for the formation of the oxo-iron(IV)

poprhyrin, respectively (Figure 5). The determined k_obsvalues

plotted against the substrate concentration showed a linear

dependence without a meaningful intercept (Figure 10). The

second-order rate constants determined from a linear fit of the

obtained data are listed in Table 2.

Influence of axial ligand on the reactivity of the cpd. II

model-kinetic studies

In order to examine the influence of N-MeIm ligation on the

catalytic activity of (TMPS)Fe^III, the oxidation reactions of the

selected substrates were also performed in the presence of N-

MeIm. The kinetics of the direct reduction of oxo-iron(IV) popr-

hyrin species by the selected substrates was studied by the ap-

plication of a stopped-flow technique including a premix

mode. The iron(IV)-oxo species was generated by mixing equal

volumes of porphyrin and H₂O₂in a concentration ratio of

1:0.5. Subsequently, substrate oxidation was examined by fol-

lowing depletion of the oxo-iron(IV) porphyrin complex on the

addition of substrate. Kinetic studies for substrate oxidation

were performed in the absence and presence of 30- and 500-

fold excess of N-MeIm over the porphyrin concentration, which

corresponds with the highest and the lowest second-order

Analogous measurements were performed for m-CPBA as

oxidant. Mixing of the iron(III) porphyrin species with m-CPBA

in a concentration ratio 1:0.5 led to the formation of the oxo-

iron(IV) poprhyrin, which was in turn reduced to the iron(III)

porphyrin after mixing with the substrate. The rate constants

found for the oxidation of 4-MB-alc and 4-MB-ald by the oxo-

iron(IV) porphyrin species generated by using the two different

oxidants are in very good agreement (Table 2). This finding

clearly indicates that the same reactive species is responsible

for the oxidation of the selected substrates independent of the

applied sacrificial oxidant.

The fact that the addition of the organic substrates to the

solution of (TMPS)Fe^IV=O(OH) resulted in the decomposition of

Chem. Eur. J. 2014, 20, 2328 – 2343

www.chemeurj.org

2337

ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

(III) porphyrin species should result in the suppression of the

oxidation reaction leading to a decrease in the k_obsvalues for

the substrate oxidation reaction. However, the performed ex-

periment revealed that increasing concentrations of added

(TMPS)Fe^III(OH) did not have a meaningful effect on the result-

ing rate constants (the observed changes in k_obswere within

the experimental error limits; see Figure S11 in the Supporting

Information). Thus, the obtained results allow us to conclude

that (TMPS)Fe^IV=O(OH) is the sole active oxidant in the investi-

gated system.

The performed studies further showed that there is no sig-

nificant effect of N-MeIm present in the axial position trans to

oxygen in iron(IV)-oxo species (model compound for cpd. II) on

the second-order rate constants for the oxidation of the select-

ed alcohol and aldehyde. The obtained results are rather sur-

prising in terms of numerous reports showing the influence of

axial ligands on the reactivity and selectivity of cpd. I. Gross

et al. have shown in biomimetic studies that the rate of styrene

epoxidation was strongly affected by various anionic axial li-

gands.^[3]Moreover, results reported by Fujii et al. showed ex-

treme enhancement of the reactivity of cpd. I by axially coordi-

nated imidazole and phenolate.^[13]A recent theoretical study

dealing with the axial ligand effect (Cl^ꢀvs. CH₃CN) on the ep-

oxidation reaction by iron(IV)-oxo porphyrin cation radical

models, revealed that neutral ligands influence an orbital inter-

action between the metal and porphyrin ring in a different

way to the anionic ligands. As a consequence of this, the dif-

ferent electron affinity of iron(IV)-oxo species possessing neu-

tral and anionic ligands leads to their dissimilar reactivity to-

wards inert organic substrates.^[43]

&

Figure 10. Plots of k_obsversus substrate concentration: , 4-methoxybenzyl

*

alcohol; , 4-methoxybenzaldehyde. Experimental conditions:

[(TMPS)Fe^III(OH)]=1ꢁ10^ꢀ5m, [H₂O₂]=5ꢁ10^ꢀ6m, [borate buffer]=0.05m,

pH 10.0, T=208C.

Table 2. Second-order rate constants for the reaction of iron(IV)-oxo spe-

cies generated by application of H₂O₂or m-CPBA as an oxidant in water

at pH 10.0 and 208C.

N-MeIm

excess

4-Methoxybenzyl alcohol

10^ꢀ1ꢁk [m^ꢀ1s^ꢀ1

4-Methoxybenzaldehyde

10^ꢀ2ꢁk [m^ꢀ1s^ꢀ1

]

H₂O₂

m-CPBA

H₂O₂

m-CPBA

0-fold

30-fold

500-fold

2.72ꢂ0.09

2.95ꢂ0.06

3.26ꢂ0.05

2.82ꢂ0.05

2.90ꢂ0.10

3.10ꢂ0.03

3.67ꢂ0.07

3.61ꢂ0.05

3.82ꢂ0.02

3.28ꢂ0.06

3.62ꢂ0.07

3.82ꢂ0.05

However, studies on the effect of an axial ligand on the elec-

tronic structure and reactivity of cpd. II are limited. Recently,

a pronounced effect of N-MeIm and Im^ꢀligands on the v_Fe

the latter back to the starting iron(III) porphyrin, may not be

sufficient evidence that the oxo-iron(IV) porphyrin species is

really the reactive species responsible for the oxidation reac-

tions. According to studies by Pan and Newcomb,^[40]the true

form of the oxidant must not always be identical to the spe-

cies the depletion of which is actually followed by UV/Vis spec-

troscopy. Indeed, they have nicely demonstrated that in the

oxidation of alkenes and alcohols by the selected oxo-iron(IV)

porphyrins, an oxo-iron(IV) porphyrin cation radical rather than

the directly observed iron(IV)-oxo species was the true oxidant

in the studied reactions. The explanation for such reactivity be-

haviour involves the existence of a mixture of reactive inter-

mediates in the reaction medium, that is, oxo-iron(IV) poprhyr-

in species with traces of a much more reactive oxo-iron(IV)

porphyrin cation radical (not directly observed by UV/Vis spec-

troscopy), both coupled via a comproportionation/dispropor-

tionation equilibrium.

=

O

frequency was found, indicating that the presence of these ni-

trogen bases in the axial position resulted in weakening of the

Fe=O bond.^[17]This may suggest that such effects should be re-

flected in the reactivity of cpd. II. However, in the case of cpd. I

no correlation between the Fe=O bond strength and reactivity

for the various enzymes and porphyrin model complexes was

found.^[3,44]An explanation for this observation was provided by

the multistep oxidation mechanism proposed by Gross et al.,^[3]

according to which the first step, a one-electron transfer from

substrate to Fe^IV=O, does not depend on the axial ligand,

whereas the second step involving intramolecular electron

III

transfer from iron to the porphyrin radical, (Por⁺)Fe (O)!

C

(Por)Fe^IV(O), is mainly affected by the axial ligand the nature of

which can lower the energy barrier.^[3]In the case of a substrate

oxidation process by cpd. II, intramolecular electron transfer

does not occur. This can, therefore, account for the observed

lack of influence of N-MeIm ligation on the oxidation of 4-MB-

alc and 4-MB-ald.

In light of these results, the nature of the active oxidant in

the present system was carefully examined by an experiment

which is commonly used to prove the existence of a mixture

of reactive intermediates.^[40]The oxidation reactions of the se-

lected substrates were carried out with varying amounts of

(TMPS)Fe^III(OH) added to the prior generated iron(IV)-oxo spe-

cies. If (TMPS)Fe^IV=O(OH) really reacts through a disproportiona-

tion pathway as outlined above, then the addition of the iron-

The obtained results are supported by a recent theoretical

study on the influence of neutral nitrogen axial ligands on

model compounds for cpd. I and cpd. II.^[16]The conclusions

drawn from this study suggest that axial ligands enlarge the

Fe=O bond distance and decrease the polarity of the Fe=O

bond for both cpd. I and cpd. II, even though they cannot

Chem. Eur. J. 2014, 20, 2328 – 2343

www.chemeurj.org

2338

ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

switch the reactivity of these two oxidants. Moreover, the theo-

retical calculations revealed that although axial ligands have

a meaningful effect on the reaction selectivity of cpd. I, they

cannot switch the selectivity of oxidation reactions for cpd. II

but only decrease the difference in the energy barrier for the

selected type of oxidation reaction (dehydrogenation versus

epoxidation). On one hand, these results are in line with previ-

ous experimental studies that axial ligands in iron porphyrin

catalyst models of cpd. I have a prominent influence on the

distribution of epoxidation and hydroxylation products.^[2,10]On

the other hand, the results of the combined experimental and

theoretical studies performed by Kang et al. show that neutral

axial ligands can affect the reactivity of cpd. I, even though

they influence to a smaller degree the rate constant of sub-

strate oxidation in comparison with the effect observed for

anionic axial ligands.^[15]

it was shown that the oxidation occurs via a dual-hydrogen ab-

straction process, not a gem-diol dehydration, thus the re-

bound step does not operate in this case.^[45]Based on this

mechanistic investigation, our computational studies were lim-

ited to H abstraction in the case of alcohol oxidation. The

values of the transition-state barrier on the triplet surface

(³TS_H) for dehydrogenation of both 4-MB-alc and 4-MB-ald cat-

alysed by (Por)Fe^IV=O and (Por)Fe^IV=O(L), L=OH^ꢀor Im, are

presented in Table 3. The optimised geometries of the hydro-

gen abstraction transition states (³TS_H) for dehydrogenation of

4-MB-alc and 4-MB-alc by (Por)Fe^IV=O(L) are presented in Fig-

ure S13 in the Supporting Information.

Table 3. Energy barriers calculated for the rate-determining step (hydro-

gen abstraction) for 4-MB-alc and 4-MB-ald oxidation catalysed by (Por)-

Fe^IV=O(L) on the triplet spin state surface.

In the present study, the experimental results clearly demon-

strated that there is no significant difference in the reactivity

of the cpd. II mimics coordinated by OH^ꢀor N-MeIm ligands.

On the other hand, to quantify the effect of N-MeIm on the re-

activity of the cpd. II mimic, a kinetic study involving an axially

unligated cpd. II mimic seems to be a reasonable suggestion.

However, in aqueous solution, oxo-iron(IV) species can only be

stabilised at higher pH^[25]through the favourable coordination

of OH^ꢀto the oxidised metal centre, which in turn does not

permit the stabilisation of the unligated cpd. II mimic. There-

fore, in view of these experimental difficulties, theoretical cal-

culations for the oxidation of selected substrates by the unli-

gated and ligated oxo-iron(IV) species were performed in order

to evaluate the effect of N-MeIm on the reactivity of the cpd. II

mimic.

[a]

L

Substrate

DE [kcalmol^ꢀ1

]

–

4-MB-alc

4-MB-ald

20.2

19.8

18.6

35.2

30.1

29.4

OH

Im

–

OH

Im

[a] Calculations were performed in the gas phase.

In the case of aldehyde oxidation, the rebound step was

also considered. Therefore, in order to confirm the location of

the rate-determining step, an exemplary (Por)Fe^IV=O(Im) spe-

cies was chosen and the energy barrier for the oxygen re-

bound step in the aldehyde oxidation pathway was computed.

The initial hydrogen abstraction leads to a stable intermediate

(³I) via TS_H. Starting from I, the radical rebound via TS_rebgives

the hydroxylated product. A stable geometry for the (Por)-

Fe(Im)-OH–4-MB-ald adduct (³I) was found to be 7.3 kcalmol^ꢀ1

higher than the level of the substrate on the potential energy

Influence of axial ligand on the reactivity of the cpd. II

model-theoretical (DFT) calculations

3

In order to gain more insight into the obtained experimental

data, we performed DFT studies on the oxygenation of 4-MB-

alc and 4-MB-ald by (Por)Fe^IV=O and (Por)Fe^IV=O(L) with L=OH^ꢀ

and N-MeIm. To simplify the systems, the 2,4,6-trimethyl-3-sul-

fonatophenyl groups of the TMPS porphyrin were replaced by

hydrogen atoms and the imidazole (Im) ligand instead of N-

methylimidazole was used in the calculations. The optimised

structures for (Por)Fe^IVO and (Por)Fe^IVO(L), where L=OH^ꢀand

Im, are presented in Figure S12 in the Supporting Information.

Calculations were performed on triplet states for all considered

substrates and catalyst forms as it is the energetically lowest

spin state of (Por)Fe^IVO. Recent studies on methane oxidation

by an axially unligated (Por)Fe^IV=O revealed that the reaction

involves spin-crossing on the potential energy surface.^[13]In

order to check whether such reactivity scenario can take place

in our system, we also extended our calculations by examining

the quintet pathway for 4-MB-ald oxidation by (Por)Fe^IV=O and

(Por)Fe^IV=O(Im); see further calculations.

3

surface. It was found that the radical rebound barrier via TS_reb

3

is lower than that for H abstraction via TS_H(12.8 vs. 29.4 kcal

mol^ꢀ1in the gas-phase, respectively), confirming that the cal-

culation of the CꢀH abstraction barrier is sufficient to reveal

the axial ligand effect on the reactivity of the studied systems.

The data summarised in Table 3 do not show a significant in-

fluence of the nature of the axial ligand on the reactivity of

cpd. II on the triplet surface. The only meaningful difference in

3

the TS_Hbarriers was observed in the case of CꢀH abstraction

from aldehyde by substituted and non-substituted (Por)Fe^IV=O.

However, it is important to emphasise that the energies of the

³TS_Hbarriers for particular substrate oxidation by both (Por)-

Fe^IV=O(OH) and (Por)Fe^IV=O(Im) remain in good agreement.

A comparison of the obtained results for CꢀH abstraction

occurring on the ground triplet and quintet surface showed

that in the case of (Por)Fe^IV=O(Im), the reaction proceeds only

on the triplet energy surface, such as for the hydroxylation of

methane by the hydrosulfide-ligated iron(IV)-oxo complex.^[8]

The calculated difference in the TS energies amounts to 14 kcal

mol^ꢀ1in favour of the triplet pathway. In contrast, when the re-

As shown previously, the initial hydrogen abstraction is the

rate-determining step of the studied oxidation reactions for

both alcohol and aldehyde, which was proven by studying the

kinetic isotope effect (KIE).^[45,46]Moreover, in the case of alcohol

Chem. Eur. J. 2014, 20, 2328 – 2343

www.chemeurj.org

2339

ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

5

action is catalysed by (Por)Fe^IV=O, the TS_Henergy is compara-

formed in the TS can undergo relaxation in the transition

state?

ble to that of TS_H, the energy difference being 0.4 kcalmol^ꢀ1

,

3

which is below the accuracy of the employed DFT method and

thus does not allow to clearly indicate the prevalent pathway.

This finding is in agreement with results presented by Rosa

and Ricciardi^[8]showing that the presence of even strong

s donor axial ligand, such as hydrosulfide in cpd. II mimics, has

no influence on the ³TS_Hbarrier, however, pronouncedly in-

creases the transition-state energy on the quintet pathway

(⁵TS_H).

To evaluate the effect of substrate-radical relaxation on the

height of TS barriers, a DFT calculation for CꢀH hydrogen atom

abstraction from the studied substrates was performed but

without geometrical optimisation of the transition states. Ac-

cording to the above considerations, the obtained values for

3

the TS_Hbarriers should reflect the more reasonable trend ob-

served experimentally. Indeed, the results of the calculations,

yielding barrier heights augmented by 11.8 ((Por)Fe^IV=O(OH)),

13.3 ((Por)Fe^IV=O) and 15.4 kcalmol^ꢀ1((Por)Fe^IV=O(Im)) for 4-

MB-alc oxidation, clearly show that the tendency obtained

from experimental data is preserved if the calculations are per-

formed in the described manner.

Since the results of theoretical calculations on the ground

triplet state nicely correlate with the experimentally obtained

data showing the lack of influence of the N-MeIm axial ligand

on the reactivity of the cpd. II model toward 4-MB-alc and 4-

MB-ald oxidation, we can exclude the participation of the quin-

tet reactivity pathway in the studied oxidation reactions. Sur-

prisingly, the calculated DE_alcvalues are meaningfully lower

than for DE_ald, which does not correspond to the experimental-

ly measured second-order rate constants. It is known from the

literature that the hydrogen abstraction barrier, or rate con-

stant for CꢀH hydroxylation by oxo-metal complexes, corre-

lates with the bond-dissociation energy (BDE_CꢀH) of the sub-

strate, as well as with the energy of the newly formed OꢀH

bond (BDE_OꢀH). The BDE_OꢀHvalue characterises a particular

type of catalyst and does not depend on the substrate nature,

which implies that variation in the energy barrier for the oxida-

tion of alcohol and aldehyde should merely depend on

BDE_CꢀH. However, a recent theoretical study also underlines

a correlation between the bond strength (D_CꢀH) defined as

Although we are aware that none of the DFT approaches

(applying or completely neglecting the structure optimisation

of the TS) can accurately model the transition state, the com-

parison of the results obtained from theoretical and experi-

mental studies supports the statement that the substrate radi-

cal partially formed in the TS does not undergo full relaxation

in the transition state. This finding does not seem to be sur-

prising in view of the fact that the process of hydrogen ab-

straction should be significantly faster in comparison with

a radical-relaxation step due to the difference in atomic mass

between hydrogen and the rest of the substrate.

Another aspect that cannot be omitted in the consideration

of theoretical and experimental results is the fact that DFT cal-

culations of the TS for substrate oxygenation reactions were

performed for the simplified porphyrin models. Thus, the pre-

diction of the TS geometry and TS barrier does not involve the

effect of steric hindrance from the 2,4,6-trimethyl-3-sulfonato-

phenyl groups on the substrate accessibility to the iron(IV)-oxo

centre. It is reasonable to expect that the tetrahedral geometry

of alcohol (sp³a-C) can to some extent hamper the access of

this substrate to the active iron(IV)-oxo centre, leading in con-

sequence to the deceleration of the hydrogen-abstraction re-

action. In contrast, the planar aldehyde (sp²) structure can

much easier reach the iron(IV)-oxo centre.

C

D

_CꢀH=BDE_CꢀH+ jRE_Rj and the activation barrier, where RE_Ris

the relaxation energy of the substrate radical.^[47]Accordingly,

C

a noticeable difference in jRE_Rj for the substrates characterised

by comparable BDE_CꢀHvalues may have a crucial effect on the

energy barrier. Taking into consideration the geometry of an al-

dehyde (sp²CHO), it is reasonable to expect that structural re-

organisation in the transition state during hydrogen abstrac-

tion from aldehyde is only minor and the related benefit from

this relaxation energy is small. In contrast, alcohol radical for-

mation requires alteration of the tetrahedral arrangement (sp³

a-C) to a planar arrangement, such that in this case a meaning-

ful decrease in energy by relaxation is expected. Indeed, these

Conclusion

C

speculations are reflected in the calculated RE_Rvalues, namely

Although much work has been done on various aspects of

heme enzymes, modelled by different complexes involved in

a number of catalytic oxidation reactions, little attention has

been devoted to the study of mimics of cpd. II. The experimen-

tal study presented herein, covers the complete picture from

the formation, stability and further reactivity of water-soluble

cpd. II models. Importantly, we could follow the complete cata-

lytic cycle spectrophotometrically, in which the cpd. II ana-

logues with different axial ligands (hydroxo or N-methylimida-

zole) oxidise selected organic substrates, leading to almost

complete reformation of the catalyst. The kinetic and mecha-

nistic studies revealed that: 1) among the studied complexes,

six-coordinate (TMPS)Fe^III(H₂O)(N-MeIm) is the most reactive

species toward hydrogen peroxide to form the N-MeIm substi-

tuted oxo-iron(IV) species, that is, a biomimic for cpd. II perox-

idase enzyme; 2) the change in the nature of the axially coordi-

ꢀ0.88 and ꢀ15.03 kcalmol^ꢀ1for 4-MB-ald and 4-MB-alc, re-

C

spectively. Application of the obtained RE_Rvalues and the

bond-dissociation energies, which were calculated for the stud-

ied substrates to estimate D_CꢀH, revealed that in spite of the

ald

higher BDE value, the bond strength for 4-MB-ald is lower

CH

(BDE =96.3 and D_CꢀH=97.1 kcalmol^ꢀ1) than for 4-MB-alc

ald

CH

alc

CH

(BDE =87.7 and D_CꢀH=102.7 kcalmol^ꢀ1). The calculated D_CꢀH

values are in agreement with a higher energy barrier for the

oxidation of 4-MB-alc than for 4-MB-ald, and correlate with the

experimental rate constants found for substrate oxidation. In

the light of this finding, and taking into account the opposite

3

trend observed in the TS_Hbarriers calculated with complete

structure optimisation of the transition states, a relevant ques-

tion to ask is to what extent substrate radicals that are partially

Chem. Eur. J. 2014, 20, 2328 – 2343

www.chemeurj.org

2340

ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

nated ligand (N-MeIm vs. OH^ꢀ) trans to the oxo-iron(IV) bond

does not significantly affect the oxidation capability of the

cpd. II mimic.

a quartz dip-in detector (Spectralytics, Aalen, Germany) coupled to

a J&M TIDAS 16/300-1100 diode array spectrophotometer (J&M,

Aalen, Germany). High pressure stopped-flow experiments were

performed in the pressure range of 10 to 130 MPa on a custom-

built apparatus.^[48]OLIS KINFIT software (Bogart, GA, 1989) was

used for the analysis of kinetic traces.

The experimental results were further supported by DFT cal-

culations as implemented in the Turbomole package. This al-

lowed us to compare the reactivity of axially ligated cpd. II

models with non-ligated species since the generation of an ax-

ially non-ligated cpd. II mimic in aqueous medium is practically

impossible. Since the dehydrogenation process is the rate-de-

termining step in the studied substrate oxidation reactions, its

energy barrier was computed as a function of the axial ligand

(OH^ꢀ, N-MeIm) and compared with the nonligated catalyst.

The calculations revealed that for both substrates, the transi-

tion-state barrier on the triplet surface did not depend on the

type of the axial ligand, which is in good agreement with the

kinetic results, and only for 4-MB-ald the presence of the axial

Product analysis

In order to analyse the oxidation products, hydrogen peroxide

([H₂O₂]_total=2ꢁ10^ꢀ4m) was added to 2ꢁ10^ꢀ4m porphyrin complex

in borate buffer, pH 10.0, in four aliquots (5ꢁ10^ꢀ5m each) in order

to prevent porphyrin decomposition. After addition of the first ali-

quot and generation of the iron(IV)-oxo species, the selected sub-

strate was added to the reaction mixture. Subsequently, the next

three aliquots of H₂O₂were added to the reaction mixture at inter-

vals of several minutes. Since it was observed that the presence of

oxygen can influence the oxidation reactions on a longer time

scale, samples for the product analysis were prepared under inert

conditions. After completion of the oxidation reactions, the ob-

tained reaction mixtures were analysed by HPLC by the use of au-

thentic samples. Product analysis was performed with a Perkin–

Elmer HPLC Chromera system. A Brownlee Validated AQ C18 5 mm,

250ꢁ4.6 mm column was employed for the HPLC separation and

H₂O (50%) and CH₃CN (50%) or 0.01% TFA (trifluoroacetic acid) in

H₂O (50%) and CH₃CN (50%) were used as a mobile phase at

3

ligand slightly decreased the TS_Henergy barrier. Notably, the

transition state barriers calculated by the DFT method for hy-

drogen abstraction from the substrates, the radicals of which

differ remarkably in relaxation energy, should be interpreted

with caution since hydrogen abstraction processes in such sys-

tems appear to be much faster than the radical relaxation

steps. Therefore, the combination of experimental and theoret-

ical studies seems to be the best approach to improve our un-

derstanding of the mechanistic picture that describes the for-

mation and reactivity of the oxo-iron(IV) biomimics and to

avoid misinterpretation of obtained results.

a flow-rate of 1 mLmin^ꢀ1

.

Quantum chemical calculations

A quantum chemical method based on density functional theory

(DFT) with nonlocal B3LYP (Becke, three-parameter, Lee–Yang–Parr)

functional^[49]was applied to study the influence of the nature of

the axial ligands (OH^ꢀ, H₂O, N-MeIm) on the stability of various

possible coordinative forms of the iron porphyrins. A literature

survey revealed that the most popular DFT functional applied to

study iron complexes, especially tetrapyrroles, is B3LYP. The calcula-

tion consisted of geometry optimisations of the studied structures

that were further confirmed with vibrational analysis. All-electron

Gaussian-type orbitals of def-TZVP quality were used to define

atomic orbitals.^[50]The nature of various coordinated forms of the

catalyst in solution was confirmed by theoretical calculation of the

binding energies of possible ligands to the iron porphyrin, namely

Experimental Section

Materials

All chemicals used in this study were of analytical reagent grade.

The

iron(III)

porphyrin,

[meso-tetrakis(2,4,6-trimethyl-3-

sulfonatophenyl)porphinato]iron(III) hydroxide (tetrasodium salt),

Fe^III(TMPS)(OH), was purchased from Frontier Scientific (Inc. Utah,

USA) and used as received. Hydrogen peroxide (35% aqueous solu-

tion) was purchased from Sigma–Aldrich. m-Chloroperoxybenzoic

acid (m-CPBA) was purchased from Acros Organic (75%) and puri-

fied by recrystallisation from hexane. Boric acid, 4-methoxybenzyl

alcohol (98%; 4-MB-alc), 4-methoxybenzaldehyde (>99%; 4-MB-

ald) and N-methylimidazole (N-MeIm) were purchased from Sigma–

Aldrich. All solutions were prepared in deionised water.

[(TMPS)Fe^III(N-MeIm)]^3ꢀ+L^ꢀ/0![(TMPS)Fe^III(N-MeIm)(L)]^4ꢀ/3ꢀ

,

where

L=OH^ꢀ,

N-MeIm,

H₂O;

DE_b=E[(TMPS)Fe^III(N-MeIm)(L)]ꢀE-

[(TMPS)Fe^III(N-MeIm)]ꢀE[L], and to the cpd. II model complex,

namely [(TMPS)Fe^IV=O]^4ꢀ+L^ꢀ/0![(TMPS)Fe^IV=O(L)]^5ꢀ/4ꢀ, where L=

OH^ꢀ, N-MeIm, H₂O; DE_b=E[(TMPS)Fe^IV=O(L)]ꢀE[(TMPS)Fe^IV=

O]ꢀE[L]. The calculations were carried out for different orientations

of the meso-substituents with respect to the macrocyclic ring. For

the exemplary case of OH^ꢀbinding to (TMPS)Fe^IV=O we confirmed

that the ligand binding energies do not differ by more than

3.6 kcalmol^ꢀ1when different conformations of the complex are

considered. Here, we report the values obtained for the global

minimum structure, assuming that there is no conformation

change of meso-substituents upon axial ligand binding. A similar

problem has already been discussed by Zimmermann et al.,^[51]and

the additional error on ligand binding energy arising from the mul-

titude of conformers, the energy of which differ by not more than

3 kcalmol^ꢀ1, was approximated to 0.5 kcalmol^ꢀ1. The solvation was

accounted for by the COSMO model^[52]with default radii for the el-

ements (H=1.30, C=2.00, N=1.83, O=1.72, S=2.16) and 2.00 ꢂ

for iron. In order to reflect the environmental effect, the value of

e=80 representing an aqueous environment, was used in the cal-

Kinetic measurements

Ambient pressure stopped-flow measurements were performed on

an SX18.MV Applied Photophysics instrument equipped with a ther-

mostat (ꢂ0.18C). The kinetics of the reaction of H₂O₂with the iron-

(III) porphyrin in borate buffer solution (0.05m, pH 10.0) were stud-

ied under pseudo-first-order conditions. In a typical experiment,

buffer solutions of the porphyrin complex were rapidly mixed with

appropriately concentrated H₂O₂solutions in the volume ratio 1:1.

Reactions in the presence of N-methylimidazole were initiated

using pre-equilibrated solutions of Fe^III(TMPS)(OH) with N-MeIm. In

the case of oxidation of selected substrates, formation of high-

valent oxo-iron porphyrin species and their reduction by organic

substrates were followed using the Applied Photophysics stopped-

flow apparatus SX20 equipped with a sequential mixing mode.

Time-resolved UV/Vis spectra were recorded with the use of

Chem. Eur. J. 2014, 20, 2328 – 2343

www.chemeurj.org

2341

ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

227 9 –2 328; P. R. Ortiz de Montellano, J. J. De Voss, Nat. Prod. Rep. 2002 ,

19 , 47 7 –4 93.

culations. For the determination of binding energies, the complete

geometry of the TMPS ligand was considered. For the transition-

state determination the theoretical model was limited to the por-

phyrin without the substituents and N-MeIm was replaced by Im.

The influence of axial ligation on the catalytic oxidation of 4-MB-

alc and 4-MB-ald was examined knowing that in this type of oxida-

tion reactions, the rate-limiting step consists of hydrogen-atom ab-

straction and formation of the substrate radical. The energy barrier

accompanying this step was found and confirmed by the existence

of one imaginary frequency, corresponding to H abstraction in the

transition state. The barriers were computed with respect to the

separate reactants (so called “apparent reaction barriers”). Experi-

mental observations regarding the relative rates of 4-MB-alc and 4-

MB-ald oxidations were further supported based on the reorganisa-

tion energies of the 4-MB-alc and 4-MB-ald radicals. These energies

were calculated as an energy difference between their geometries

in the free substrate (no relaxation) and in the fully relaxed geome-

try of the radicals, following the concepts introduced by Shaik and

co-workers.^[47]In addition, in the case of aldehyde oxidation the

energy barrier for the rebound reaction step was calculated with

the use of (Por)Fe^IV=O(Im) as selected oxidant. The presented re-

sults were obtained with Turbomole v.6.3.^[53]

[5] J. P. Collman, L. Zeng, L. A. Decrꢃau, Chem. Commun. 2003 , 297 4 –2 975;

J. P. Collman, A. S. Chien, T. A. Eberspacher, J. I. Brauman, J. Am. Chem.

Soc. 2000 , 122 , 1109 8 –1 110 0; S. Jin, T. A. Bryson, J. H. Dawson, J. Biol.

Inorg. Chem. 2004, 9, 644–653; A. D. N. Vaz, D. F. McGinnity, M. J. Coon,

Proc. Natl. Acad. Sci. USA 1998 , 95 , 355 5 –3 560; A. D. N. Vaz, S. J. Per-

necky, G. M. Raner, M. J. Coon, Proc. Natl. Acad. Sci. USA 1996 , 93 , 464 4 –

4648; M. J. Cryle, J. J. De Voss, Angew. Chem. 2006 , 118 , 840 1 –8 403;

Angew. Chem. Int. Ed. 2006 , 45 , 822 1 –8 223; M. Imai, H. Shimada, Y. Wa-

tanabe, Y. Matsushima-Hibiya, R. Makino, H. Koga, T. Horiuchi, Y. Ishi-

mura, Proc. Natl. Acad. Sci. USA 1989 , 86 , 782 3 –7 827; S. A. Martinis,

W. M. Atkins, P. S. Stayton, S. G. Sligar, J. Am. Chem. Soc. 1989 , 111 ,

925 2 –9 253.

[6] M. Newcomb, D. Aebisher, R. Shen, R. E. P. Chandrasena, P. F. Hollenberg,

M. J. Coon, J. Am. Chem. Soc. 2003 , 125 , 606 4 –6 06 5; M. Newcomb, R.

Shen, S.-Y. Choi, P. H. Toy, P. F. Hollenberg, A. D. N. Vaz, M. J. Coon, J. Am.

Chem. Soc. 2000 , 122 , 267 7 –2 686; M. Newcomb, P. H. Toy, Acc. Chem.

Res. 2000 , 33 , 44 9 –4 55.

[7] F. Ogliaro, S. P. De Visser, S. Cohen, P. K. Sharma, S. Shaik, J. Am. Chem.

Soc. 2002 , 124 , 280 6 –2 817; P. K. Sharma, S. P. De Visser, S. Shaik, J. Am.

Chem. Soc. 2003 , 125 , 869 8 –8 699; T. Kamachi, Y. Shiota, T. Ohta, K.

Yoshizawa, Bull. Chem. Soc. Jpn. 2003 , 76 , 72 1 –7 32.

[8] A. Rosa, G. Ricciardi, Inorg. Chem. 2012 , 51 , 983 3 –9 845.

[9] L. Tahsini, M. Bagherzadeh, W. Nam, S. P. de Visser, Inorg. Chem. 2009 ,

48 , 666 1 –6 669.

[10] J. T. Groves, Z. Gross, M. K. Stern, Inorg. Chem. 1994 , 33 , 506 5 –5 072.

[11] Y. J. Jeong, Y. Kang, A.-R. Han, Y.-M. Lee, H. Kotani, S. Fukuzumi, W. Nam,

Angew. Chem. 2008 , 120 , 743 1 –7 434; Angew. Chem. Int. Ed. 2008 , 47 ,

732 1 –7 32 4; W. Nam, S.-E. Park, I. K. Lim, M. H. Lim, J. Hong, J. Kim, J.

Am. Chem. Soc. 2003 , 125 , 1467 4 –1 4675.

[12] C. Fertinger, N. Hessenauer-Ilicheva, A. Franke, R. van Eldik, Chem. Eur. J.

2009 , 15 , 1343 5 –1 3440.

[13] A. Takahashi, T. Kurahashi, H. Fujii, Inorg. Chem. 2009 , 48 , 261 4 –2 625.

[14] S. P. de Visser, L. Tahsini, W. Nam, Chem. Eur. J. 2009, 15, 5577–5587.

[15] Y. Kang, H. Chen, Y. J. Jeong, W. Lai, E. H. Bae, S. Shaik, W. Nam, Chem.

Eur. J. 2009 , 15 , 1003 9 –1 0046.

[16] Y. Sun, X. Hu, H. Li, A. F. Jalbout, Comput. Theor. Chem. 2011 , 966 , 6 2 –

69.

[17] H. Ishimaru, H. Fujii, T. Ogura, Chem. Lett. 2010 , 39 , 33 2 –3 33.

[18] M. T. Green, J. Am. Chem. Soc. 2006 , 128 , 190 2 –1 906.

[19] A. Takahashi, D. Yamaki, K. Ikemura, T. Kurahashi, T. Ogura, M. Hada, H.

Fujii, Inorg. Chem. 2012 , 51 , 729 6 –7 305.

Acknowledgements

The support from the National Science Centre (grant no. DEC-

2012/05/B/STS/00389) is gratefully acknowledged. M.O. thanks

the financial support from the International PhD Study Pro-

gram at the Faculty of Chemistry, Jagiellonian University,

within the Foundation for Polish Science MPD Program co-fi-

nanced by the European Regional Development Fund. A.F. and

R.v.E. gratefully acknowledge continued financial support from

the Deutsche Forschungsgemeinschaft. This work was carried

out with the equipment purchased through financial support

from the European Regional Development Fund in the frame-

work of the Polish Innovation Economy Operational Program

(contract no. POIG.02.01.00-12-023/08). D.R.Z. and A. D. M. re-

ceived funding from the Marian Smoluchowski Krakow Re-

search Consortium-a Leading National Research Centre KNOW

supported by the Ministry of Science and Higher Education.

[20] M. F. Zipplies, W. A. Lee, T. C. Bruice, J. Am. Chem. Soc. 1986 , 108 , 443 3 –

4445.

[21] R. Panicucci, T. C. Bruice, J. Am. Chem. Soc. 1990 , 112 , 606 3 –6 071.

[22] M. Wirstam, M. R. A. Blomberg, P. E. M. Siegbahn, J. Am. Chem. Soc.

1999 , 121 , 1017 8 –1 018 5; P. Rydberg, L. Olsen, U. Ryde, Curr. Inorg.

Chem. 2012, 2, 292–315; S. Shaik, S. Cohen, Y. Wang, H. Chen, D.

Kumar, W. Thiel, Chem. Rev. 2010 , 110 , 94 9 –1 017.

[23] M. Wolak, R. van Eldik, J. Am. Chem. Soc. 2005 , 127 , 1331 2 –1 3315.

[24] S. J. Yang, W. Nam, Inorg. Chem. 1998 , 37 , 60 6 –6 07; W. Nam, H. J. Choi,

H. J. Han, S. H. Cho, H. J. Lee, S.-Y. Han, Chem. Commun. 1999 , 38 7 –3 88.

[25] M. Wolak, R. van Eldik, Chem. Eur. J. 2007 , 13 , 487 3 –4 883.

[26] M. Liu, Y. O. J. Su, Electroanal. Chem. 1998, 452, 113–125.

[27] K. Murata, R. Panicucci, G. Enona, T. C. Bruice, J. Am. Chem. Soc. 1990 ,

112 , 607 2 –6 083.

[28] A. G. DiPasquale, J. M. Mayer, J. Am. Chem. Soc. 2008 , 130 , 181 2 –1 813.

[29] P. Jones, H. B. Dunford, J. Inorg. Biochem. 2005 , 99 , 229 2 –2 298.

[30] A. Brausam, J. Maigut, R. Meier, P. ꢄ. Szilꢅgyi, H.-J. Buschmann, W.

Massa, Z. Homonnay, R. van Eldik, Inorg. Chem. 2004 , 43 , 535 1 –5 359.

[31] J. T. Groves, Y. Watanabe, Inorg. Chem. 1987 , 26 , 78 5 –7 86; R. D. Arasa-

singham, A. L. Balch, L. Latos-Grazynski, J. Am. Chem. Soc. 1987 , 109 ,

584 6 –5 847.

Keywords: density functional calculations · enzyme models ·

hydrogen peroxide · oxidation · reaction mechanisms

[1] R. van Eldik, Coord. Chem. Rev. 2007 , 251 , 164 9 –1 662; I. G. Denisov,

T. M. Makris, S. G. Sligar, I. Schlichting, Chem. Rev. 2005 , 105 , 225 3 –

2277; B. Meunier, S. P. de Visser, S. Shaik, Chem. Rev. 2004 , 104 , 394 7 –

3980; M. Newcomb, P. F. Hollenberg, M. J. Coon, Arch. Biochem. Biophys.

2003 , 409 , 7 2 –7 9; Z. Pan, R. Zhang, M. Newcomb, J. Inorg. Biochem.

2006 , 100 , 52 4 –5 32; S. R. Bell, J. T. Groves, J. Am. Chem. Soc. 2009 , 131 ,

964 0 –9 641; W. Nam, Y. M. Goh, Y. J. Lee, M. H. Lim, C. Kim, Inorg. Chem.

1999 , 38 , 323 8 –3 240; Y. M. Goh, W. Nam, Inorg. Chem. 1999 , 38 , 91 4 –

920; S. G. Sligar, Science 2010 , 330 , 92 4 –9 25; J. Rittle, J. M. Younker,

M. T. Green, Inorg. Chem. 2010 , 49 , 361 0 –3 617; J. Rittle, M. T. Green, Sci-

ence 2010 , 330 , 93 3 –9 37.

[2] W. J. Song, Y. O. Ryu, R. Song, W. Nam, J. Biol. Inorg. Chem. 2005 , 10 ,

29 4 –3 04.

[32] R. D. Arasasingham, C. R. Cornman, A. L. Balch, J. Am. Chem. Soc. 1989 ,

111 , 780 0 –7 805.

[3] Z. Gross, S. Nimri, C. M. Barzilay, L. Simkhovich, J. Biol. Inorg. Chem.

1997 , 2 , 49 2 –5 05; Z. Gross, J. Biol. Inorg. Chem. 1996 , 1 , 36 8 –3 71; Z.

Gross, S. Nimri, Inorg. Chem. 1994 , 33 , 173 1 –1 732.

[4] S. Shaik, H. Hirao, D. Kumar, Acc. Chem. Res. 2007 , 40 , 53 2 –5 42; S.

Shaik, D. Kumar, S. P. de Visser, A. Altun, W. Thiel, Chem. Rev. 2005 , 105 ,

[33] P. N. Balasubramanian, R. W. Lee, T. C. Bruice, J. Am. Chem. Soc. 1989 ,

111 , 871 4 –8 721; P. N. Balasubramanian, J. R. Lindsay Smith, M. J. Davies,

T. W. Kaaret, T. C. Bruice, J. Am. Chem. Soc. 1989 , 111 , 147 7 –1 483; J. R.

Lindsay Smith, P. N. Balasubramanian, T. C. Bruice, J. Am. Chem. Soc.

1988 , 110 , 741 1 –7 418.

Chem. Eur. J. 2014, 20, 2328 – 2343

www.chemeurj.org

2342

ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

[34] A. Ikezaki, Y. Ohgo, M. Nakamura, Coord. Chem. Rev. 2009 , 253 , 205 6 –

2069; and references therein.

[47] S. Shaik, D. Kumar, S. P. de Visser, J. Am. Chem. Soc. 2008 , 130 , 1012 8 –

10140.

[35] A. Brausam, S. Eigler, N. Jux, R. van Eldik, Inorg. Chem. 2009 , 48 , 766 7 –

7678.

[36] T. G. Traylor, W. A. Lee, D. V. J. Stynes, J. Am. Chem. Soc. 1984 , 106 , 75 5 –

764.

[37] A. L. Balch, Y.-W. Chan, R.-J. Cheng, G. N. La Mar, L. Latos-Grazynski,

M. W. Renner, J. Am. Chem. Soc. 1984 , 106 , 777 9 –7 785.

[38] M. J. Beck, E. Gopinath, T. G. Bruice, J. Am. Chem. Soc. 1993 , 115 , 2 1 –2 9.

[39] K. Yamaguchi, Y. Watanabe, I. Morishima, J. Am. Chem. Soc. 1993 , 115 ,

405 8 –4 065.

[48] R. van Eldik, D. A. Palmer, R. Schmidt, H. Kelm, Inorg. Chim. Acta 1981 ,

50 , 13 1 –1 35; R. van Eldik, W. Gaede, S. Wieland, J. Kraft, M. Spitzer,

D. A. Palmer, Rev. Sci. Instrum. 1993 , 64 , 135 5 –1 357.

[49] P. A. M. Dirac, Proc. Royal Soc. Lond. 1929 , 123 , 71 4 –7 33; J. C. Slater,

Phys. Rev. 1951 , 81 , 38 5 –3 90; S. H. Vosko, L. Wilk, M. Nusair, Can. J.

Phys. 1980 , 58 , 120 0 –1 211; A. D. Becke, Phys. Rev. A 1988 , 38 , 309 8 –

3100; C. Lee, W. Yang, R. G. Parr, Phys. Rev. B 1988 , 37 , 78 5 –7 89; A. D.

Becke, J. Chem. Phys. 1993 , 98 , 564 8 –5 652.

[50] A. Schꢆfer, C. Huber, R. Ahlrichs, J. Chem. Phys. 1994, 100, 5829–5835.

[51] T. Zimmermann, Z. Chval, J. V. Burda, J. Phys. Chem. B 2009 , 113 , 313 9 –

3150.

[40] Z. Pan, M. Newcomb, Inorg. Chem. 2007 , 46 , 676 7 –6 774.

[41] B. Meunier, Chem. Rev. 1992 , 92 , 141 1 –1 456.

[42] O. Pestovsky, A. Bakac, J. Am. Chem. Soc. 2004 , 126 , 1375 7 –1 3764.

[43] D. Kumar, R. Latifi, S. Kumar, E. V. Rybak-Akimova, M. A. Sainna, S. P. De

Visser, Inorg. Chem. 2013 , 52 , 796 8 –7 979.

[52] A. Klamt, G. Schꢇꢇrmann, J. Chem. Soc. Perkin Trans. 2 1993 , 79 9 –8 05.

[53] V6.3 ed., TURBOMOLE GmbH saf, Forschungszentrum Karlsruhe, 2011.

[44] L. A. Andersson, J. H. Dawson, Struct. Bonding (Berlin) 1991, 74, 1–40.

[45] N. Y. Oh, Y. Suh, M. J. Park, M. S. Seo, J. Kim, W. Nam, Angew. Chem.

2005 , 117 , 430 7 –4 311; Angew. Chem. Int. Ed. 2005 , 44 , 423 5 –4 239.

[46] W. K. Seok, T. J. Meyer, Inorg. Chem. 2005 , 44 , 393 1 –3 941.

Received: September 20, 2013

Published online on January 17, 2014

Chem. Eur. J. 2014, 20, 2328 – 2343

www.chemeurj.org

2343

ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Article Doi

Mechanistic insight into peroxo-shunt formation of biomimetic models for compound II, Their reactivity toward organic substrates, and the influence of N-methylimidazole axial ligation

DOI: 10.1002/chem.201303694

Source and publish data:

Authors:

Article abstract of DOI:10.1002/chem.201303694

Full text of DOI:10.1002/chem.201303694

Products guided by the article

R&D Labs maybe for 100-09-4

Relevant to this article

Hot Product