Functional models of nonheme diiron enzymes: Kinetic and computational evidence for the formation of oxoiron(IV) species from peroxo-diiron(III) complexes, and their reactIVity towards phenols and H2O2

Article abstract of DOI:10.1039/c6dt01598k

The reactivity of the previously reported peroxo adducts [Fe₂(μ-O₂)(L¹)₄(CH₃CN)₂]²⁺, and [Fe₂(μ-O₂)(L²)₄(CH₃CN)₂]²⁺, (L¹ = 2-(2′-pyridyl)benzimidazole and L² = 2-(2′-pyridyl)-N-methylbenzimidazole) towards H₂O₂ as catalase mimics, and towards various phenols as functional RNR-R2 mimics, is described. Kinetic, mechanistic and computational studies gave direct evidence for the involvement of the (μ-1,2-peroxo)diiron(iii) intermediate in the O-H activation process via formation of low-spin oxoiron(iv) species.

Full text of DOI:10.1039/c6dt01598k

Dalton
Transactions
View Article Online
View Journal
PAPER
Functional models of nonheme diiron enzymes:
kinetic and computational evidence for the
formation of oxoiron(IV) species from peroxo-
diiron(III) complexes, and their reactivity towards
phenols and H O †
Cite this: DOI: 10.1039/c6dt01598k
2
2
a
b,c
a
a
Miklós István Szávuly, Mihai Surducan, Emőke Nagy, Mátyás Surányi,
a
b
a
Gábor Speier, Radu Silaghi-Dumitrescu* and József Kaizer*
1
2+
The reactivity of the previously reported peroxo adducts [Fe
2
(μ-O
2
)(L )
4
(CH
3
2
CN) ]
, and [Fe
2
2
(μ-O )
2
2+
1
2
Received 25th April 2016,
Accepted 5th June 2016
(L ) (CH CN) ]
,
(L
=
2-(2’-pyridyl)benzimidazole and
L
= 2-(2’-pyridyl)-N-methylbenzimidazole)
4
3
2
2 2
towards H O as catalase mimics, and towards various phenols as functional RNR-R2 mimics, is described.
DOI: 10.1039/c6dt01598k
Kinetic, mechanistic and computational studies gave direct evidence for the involvement of the (µ-1,2-
www.rsc.org/dalton
peroxo)diiron(III) intermediate in the O–H activation process via formation of low-spin oxoiron(IV) species.
species was fully characterized based on XAS (r(Fe–Fe) = 3.44/
Introduction
5
3.48 Å) and single-crystal X-ray diffraction data. Similar spectro-
Nonheme diiron enzymes ribonucleotide reductases (RNR-R2), scopic and kinetic evidence was found for a biomimetic dinuc-
9
9
6
steraoyl-ACP Δ -desaturase (Δ D), soluble methane monooxy- lear diiron peroxides, including the three crystallographically
genase (sMMO), membrane bound alkane monooxygenases, characterized derivatives however, the reactivity of these species
7
and deoxyhypusine hydroxylase (hDOHH) catalyse a large has not been thoroughly explored. We have previously reported
variety of dioxygen-dependent transformations such as oxi- the synthesis and structure of iron(II) precursor complexes
1
II
1
II
2
dation, hydroxylation and oxygenation. RNR-R2 is responsible [Fe (L )
3
](CF
bidentate ligands L (L = 2-(2′-pyridyl)benzimidazole) and L
2
3 3 2 3 3 3 2
SO ) (1) and [Fe (L ) ](CF SO ) (2) based on the
2
for the formation of an essential tyrosyl radical from tyrosine,
Δ D for the dehydrogenation of fatty acid alkyl chains, sMMO (L = 2-(2′-pyridyl)-N-methylbenzimidazole) (Scheme 1), and the
2
for the hydroxylation of methane and alkanes, and hDOHH for UV/Vis (νmax = 720 nm (ε = 1360 M cm ) and 685 nm (ε =
the hydroxylation of deoxyhypusine to form hypusine. For such 1400 M cm ), respectively), EPR, and resonance Raman ν(O–
enzymes, catalytic (µ-1,2-peroxo)diiron(III) (P) intermediates have O) = 876 cm ; (ν(Fe–O) = 463 cm ) spectroscopic characteri-
2
been postulated, which subsequently undergo O–O bond scis- zation of a transient green species with a Fe (µ-1,2-O )Fe (3,4)
sion leading to the formation of mixed valent iron(III)iron(IV) (X), core derived from the reaction of 1 and 2 with H O . Based on
1
1
9
3
4
−1
−1
5
−1
−1
−1
−1
III
III
8
2
2
or diiron(IV) (Q) intermediates to react via electrophilic hydrogen the similarities on the structural and spectroscopic behaviours
atom transfer (HAT) reaction. The formation of (µ-1,2-peroxo) of the enzymatic and synthetic (µ-1,2-peroxo)diiron(III) species,
diiron(III) species in these enzymes can be proved by their compound 3 and 4 are good structural models for the RNR-R2
characteristic ligand to metal charge transfer absorptions enzymes. Herein, we present the reactivity of 3 and 4 towards
between 600 and 750 nm, and the stretching mode of the (µ-1,2-
2 2
H O as catalase-mimics, and towards various phenols as func-
peroxo)diiron(III) core in their resonance Raman spectra tional RNR-R2 mimics, while also seeking direct evidence for
(Table 1). Furthermore, the generated and trapped hDOHHperoxo the involvement of the (µ-1,2-peroxo)diiron(III) intermediate in
the O–H activation process.
a
Department of Chemistry, University of Pannonia, 8201 Veszprém,
Wartha Vince u. 1., Hungary. E-mail: kaizer@almos.vein.hu
b
Department of Chemistry, Babes-Bolyai University, RO-400024 Cluj-Napoca,
Experimental
Materials and methods
Romania. E-mail: rsilaghi@chem.ubbcluj.ro
c
National Institute for Research and Development of Isotopic and Molecular
Technologies, Donath Street, No. 65-103, RO-400293 Cluj-Napoca, Romania
All manipulations were performed under a pure argon atmo-
sphere using standard Schlenk-type techniques unless other-
†
Electronic supplementary information (ESI) available: Experimental details of
catalyses, kinetic, and computational data. See DOI: 10.1039/c6dt01598k
This journal is © The Royal Society of Chemistry 2016
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Table 1 Properties of (µ-1,2-peroxo)diiron(III) units in enzymes and their models
UV-vis, λmax/nm (ε, M− cm−1
1
)
Raman ν(O–O)/cm−1
Ref.
Complex
hDOHH
sMMO
Δ9D
630 (2800)
725 (1800)
700 (1100)
700 (1800)
490 (1100), 640 (1100)
480 (1000), 620 (1000)
480 (1000), 620 (1000)
505 (1300), 650 (1300)
730 (2400)
690 (1500)
720 (1360)
685 (1400)
694 (1725)
588 (1500)
644 (3000)
577 (1500)
700 (br) (1700)
855
—
5b
4b,c
3a,b
2d
6k
6i
6i
6i,h
6i
7d
8
8
6f
6b,d
6l
898
868
847
844
853
854
925
874
876
876
885
900
908
847
884
RNR-R2
2
+
[Fe
[Fe
[Fe
[Fe
[Fe
[Fe
[Fe
[Fe
[Fe
[Fe
[Fe
[Fe
[Fe
2
2
2
2
2
2
2
2
2
2
2
2
2
(µ-O)(µ-O
(µ-O)(µ-O
(µ-O)(µ-O
(µ-O)(µ-O
(µ-OH)(µ-O
(µ-O)(µ-O )(IndH)]
2
2
2
2
)(6-Me
)(BQPA)]
)(6-Me-BQPA)]
)(BnBQA)(CH
3
-TPA)]
2
+
2
+
2
+
3
CN)
2
]
3
+
2
)(BnBQA)(CH
3
CN) ]
2
2
+
2
1
2+
2+
(µ-O
(µ-O
2
)(L )
)(L )
4
(CH
(CH
3
CN)
CN)
2
]
]
2
2
4
3
2
(HB(3,5-iPr
(N-Et-HPTB)(µ-O
(6-Me
(6-Me
2
pz)
3 2
) )(µ-O
2
)(µ-O
2
CCH
2 6 5 2
C H ) ]
2
)(OPPh
)(µ-OH)]
)(µ-O)]
3
)
2
]
+
2
-BPP)
-BPP)
2
(µ-O
(µ-O
2
2
2
2
6l
6e
2
+
(Ph-bimp)(µ-O
2
)(µ-O
2
CC
6
H
5
)]
6
-Me
methylpyridyl-2-methyl)amine; BnBQA = N-benzyl-N,N-bis(2-quinolylmethyl)amine; IndH = l,3-bis(2′-pyridilimino)isoindoline); pz = pyrazole;
N-Et-HPTB = anion of N,N,N′,N′-tetrakis(1′-ethylbenzimidazolyl-2′-methyl)-2-hydroxy-1,3-diaminopropane); 6-Me -BPP = N,N-bis(6-methyl-2-pyri-
3
-TPA = tris(6-methyl-2-pyridylmethyl)amine; BQPA = bis(2-quinolylmethyl)(2-pyridylmethyl)amine; 6-Me-BQPA = bis(2-quinolylmethyl)(6-
2
dylmethyl)-3-aminopropionate); Ph-bimp = 2,6-bis[bis{2-(1-methyl-4,5-diphenylimidazolyl)-methyl}aminomethyl]-4-methylphenolate).
intermediates 3 and 4 were generated in situ with stoichio-
metric amounts of hydrogen peroxide before the addition of
the corresponding substrate. The oxidation reactions were fol-
lowed with UV-Vis spectroscopy between 400 and 1000 nm.
The reaction rates were calculated using the initial rate
0
.5
method (Vox = V
rate (V ) were extracted from plots of concentration [(µ-1,2-
peroxo)diiron(III)] intermediate versus time, and were corrected
with the decomposition reaction rate (V ). The krel values for
the Hammett plot were calculated as k /k for different para-
and meta-substituted substrates. For the determination of the
KIE (SIE), separate reactions were carried out for 2,6-di-tert-
butylphenol and 2-chloro-1,4-hydroquinone in the presence of
i
0
− V = kox[S][3 or 4] ). The oxidation reaction
i
Scheme 1
0
X
H
wise stated. Solvents used for the reactions were purified by
standard methods and stored under argon. Iron(II) triflate was
purchased from commercial sources. The ligands L and L ,
1 2
and complexes 1 and 2 were prepared according to published
procedures. Infrared spectra were recorded with an Avatar 330
FTIR Thermo Nicolet instrument. UV/Vis spectra were recorded
with an Agilent 8453 diode-array spectrophotometer with quartz
cells. Microanalyses were performed by the Microanalytical
Service of the University of Pannonia and Atlantic Microlab.
1
00 equivalents of D O or H O (based on 3 and 4 concen-
2 2
8
tration), and the kinetic isotope effect was calculated from
k /k For product analysis, the samples were filtered through
H
D.
alumina to remove the complex. The products were identified
by GC–MS and confirmed by comparison with authentic
samples. The products were quantified by GC relative to biphe-
nyl as an internal standard.
Catalase-like activities
All reactions were carried out in a 50 mL reactor containing a Computational details
stirring bar under air. Acetonitrile (30 mL) was added to the
All calculations were started from the crystal structure of Fe(II)
complex and the flask was closed with a rubber septum.
Hydrogen peroxide was injected through the septum with a
Hamilton syringe. The reactor was connected to a graduated
burette filled with mercury and dioxygen evolution was
measured volumetrically at time intervals of 30 or 60 s. The
reaction rates were calculated using the initial rate method.
2
8
(
3
L ) (CCDC 955599), where one ligand was substituted with
acetonitrile and oxygenic ligands so as to obtain [Fe(IV)
2+
(
L
2
)
2
(MeCN)O] (given as Fe(IV)O from now) and [(L
2
)
2
(MeCN)
4+
Fe(III)–O–O–Fe(III)(L ) (MeCN)] (given as Fe O from now).
2 2 2 2
The barriers for oxidations and O–O bond breaking were
obtained by determining the transition states (TS) for the pro-
cesses. All calculations were carried out with the Gaussian 09
Oxidation reactions
9
suite. To determine the different broken symmetry solutions
All reactions were carried out under thermostated conditions for the dimeric peroxo complexes, the fragment editor from
at different temperatures indicated in text, in 1 cm quartz cuv- Gaussview 5 was employed: fragments were defined to force
ettes with stirring under argon. The (µ-1,2-peroxo)diiron(III) the desired spin solution, then a guess for the resulting wave-
Dalton Trans.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Dalton Transactions
Paper
function was made followed by the determination of the accu-
rate wavefunction and subsequent geometry optimisation.
A total of 12 different broken symmetry solutions were com-
puted and they are denoted with the following notation: S =
n(X, Y), where n is the spin multiplicity, X the number of
unpaired electrons on the first iron atom, Y the number of
unpaired electrons on the second iron atom, and positive/
negative values signify spin up/down electrons. All geometry
optimisations were carried out with the B3LYP functional and
the 6-31G(d,p) basis set. Final energies are Gibbs free energies
and contain solvent corrections (acetonitrile, conductor-like
polarized continuum model), dispersion energies computed
1
0,11
with Grimme’s D3 method with BJ damping,
and single
point calculations using the higher 6-311++G(d,p) basis set for
all atoms except for iron for which one extra set of d and f
functions was added. All wavefunctions were confirmed to be
stable. Intrinsic reaction coordinate calculations were used for
confirming the TS of the monoiron complexes, but not for the
diiron complexes due the very high computational cost.
0
Fig. 2 Formation and temperature-dependent stability of 4. [2] = 2 ×
−3
−3
10
2 2 0
M, [H O ] = 8 × 10 M in MeCN at 685 nm.
Table 2 Temperature-dependent stability of (µ-1,2-peroxo)diiron(III)
complexes (3, 4)
T (°C)
3 t1/2 (s)
4 t1/2 (s)
Results and discussion
Catalase-like activity of 3 and 4
1
1
0
5
1650
1200
645
390
n.a.
n.a.
4740
3210
1770
870
The reactivity of complex 1 and 2 towards hydrogen peroxide 20
2
3
5
03
was investigated in MeCN by measuring changes in UV-Vis,
and by volumetric determination of evolved dioxygen. Upon
formation, the peroxo-diiron(III) complexes 3 and 4 decompose
to form dioxygen and regenerate the precursor Fe(II) com-
plexes, suggesting a catalase-like reactivity (Fig. 1 and 2).
The thermal stability of the (µ-peroxo)-bridged species
shows considerable dependence on both the substitution (NH
versus NMe) in the ligand as well as on the temperature
2 2
H O and O was examined in MeCN at 25 °C using initial rates
method monitoring the increase of the evolved dioxygen, and
it was found that both complexes have such activity. Fig. 3
shows typical dioxygen evolution versus time curves for the dis-
proportionation process catalyzed by 3 and 4. Based on the
observed initial rates, compound 4 was more reactive than 3.
(Table 2). The half-life (t1/2’s) for complex 3 is 11 min, while
for complex 4 it is 54 min at 20 °C, demonstrating that
complex 3 is thermally much less stable compared to complex
The estimated initial rates (under same conditions) were V
0
=
M
−5
−1
−1
−5
2
s
.95 × 10 M s (TOF = 0.65 min ), and V
0
= 5.10 × 10
4
. The catalase-like activity of the in situ formed peroxo-
−
1
−1
(TOF = 1.43 min ), respectively to the compound 3 and 4
diiron(III) complexes 3 and 4 to disproportionate H O into
2
2
Fig. 1 Formation and temperature-dependent stability of 3. [1]
0
= 2 ×
Fig. 3 Disproportionation of H
2
O
2
catalyzed by the in situ formed 3
= 0.2 M at 25 °C in MeCN.
−3
−3
−3
1
0
M, [H
2
O
2
]
0
= 8 × 10 M in MeCN at 720 nm.
and 4. [1 or 2]
0
= 1 × 10 M, [H
2
O
2
]
0
This journal is © The Royal Society of Chemistry 2016
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Scheme 2 (µ-1,2-Peroxo)diiron(III)-mediated oxidation of phenols and
hydroquinones.
The (µ-1,2-peroxo)diiron(III) complexes 3 and 4, which are
well-characterized iron-based oxidants with an absorbance
band in the visible region (νmax = 720 nm (1360 M cm ) and
νmax = 685 nm (1400 M−1 cm−1), respectively) were generated
Fig. 4 Dependence of the initial reaction rate (V
0
) on the catalyst (3 or
catalyzed by the
2 0
O ] = 0.2 M at 25 °C in MeCN.
−1
−1
4) concentration for the disproportionation of H
2 2
O
in situ formed 3 and 4. [H
2
by reaction of 1 and 2 with H O in MeCN, and the rate of the
2
2
decay of the absorption band at 720 nm and 685 nm was
measured as a function of the concentration of added sub-
strates (Fig. 5 and 6).
The reaction of 3 and 4 was investigated in detail. To gen-
erate complexes 3 and 4, 1 and 2 (0.5 mM) were dissolved in
at 25 °C. In order to determine the rate dependence on the
various reactants, disproportionation runs were performed at
different substrate and catalyst concentrations (Table S1, ESI†).
At constant [H
2 2 0 2 2
O ] , the initial rate of H O disproportionation
0
.5
MeCN, and treated with 4 equivalent of aqueous H O , which
2 2
varies linearly with the in situ-formed [catalyst 3 or 4]
,
led to the maximum generation of 3 and 4 (<320 and 700 s), as
judged by UV-vis spectroscopy, followed by a slower decompo-
sition of the green species, and the formation of 3,3′,5,5′-tetra-
tert-butyl-4,4′-diphenoquinone. No shifts have been observed
for the 2,6-di-tert-butylphenols, excluding their complexation
meaning that both reactions are half-order in catalyst, and
suggesting a dissociation process via homolytic cleavage of the
O–O bond (Fig. 4). At low H
2 2
O concentrations, the reactions
are first-order in peroxide concentration as shown in Fig. S1
(
4
1
ESI†), to establish a rate law of –d[H
2
O
2
]/dt = kcat[H
2 2
O ][3 or
0
.5
−3
−1/2 −1
with the oxidant. Plots of reaction rates (Vox = V
i
− V = kox[S]
0
]
with kcat = 2.81 × 10
M
s
for 3 and kcat = 5.06 ×
0
.5
−
3
−1/2
s−1 for 4 at 25 °C.
[3 or 4] ) versus 2,6-di-tert-butylphenol concentrations, as
shown for 3 and 4 in Fig. 7, state that the reaction is first order
with respect to the substrate concentration in both cases
0
M
(
Tables S2 and S3, ESI†), which in turn indicates that the rate-
Substrate oxidation studies
determining step of the stoichiometric reaction involves sub-
Stoichiometric O–H activation of 2,6-di-tert-butylphenols
mediated by complex 3 and 4. Firstly, we have investigated the
properties of 3 and 4 as oxidants against various organic com-
pounds, in order to obtain an initial assessment of their oxi-
dizing power. To get direct evidence for the involvement of the
III
III
2
Fe (µ-1,2-O )Fe species in the O–H activation processes,
reactions of 3 and 4 with various substrates, such as 2,6-di-tert-
butylphenol, and 1,4-hydroquinones were investigated. The
distinction between nucleophilic versus radical peroxo charac-
ter can be proved with phenols since deprotonation leads to
phenoxide coordination, whereas HAT provides a phenoxyl
radical, which can lead to an easily-identifiable 2,2′-biphenol-
coupled product in case of 2,6-di-tert-butylphenols or 1,4-
benzoquinones after the disproportionation of the forming
semiquinones (Scheme 2). Peroxo O–O bond cleavage resulting
in oxoiron(IV) intermediate may also occur prior to HAT in
these reactions. Complexes 3 and 4 are capable of performing
HAT reactions with 2,6-di-tert-butylphenol 3,3′,5,5′-tetra-tert-
butyl-4,4′-diphenoquinone (∼80% based on 1 and 2) formed
via the phenoxyl intermediate, and 1,4-hydroquinone trans-
formed to 1,4-benzoquinone (∼90%). These results raise the
question of which active species is involved in the mechanism.
Fig. 5 UV-Vis spectral change during the 3-mediated oxidation of
−3
phenols monitored at 720 nm. Reaction conditions: [3] = 0.5 × 10 M,
2.6-DTBPh] = 0.1 M, in MeCN at 20 °C. Inset shows time course of the
[
decay of 3, without substrate (a), and with 0.1 M 2.6-DTBPh (b), Δt =
45 s.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Dalton Transactions
Paper
Fig. 8 Dependence of the reaction rate (Vox) on the (µ-1,2-peroxo)diiron(III)
(3 or 4) concentrations for the oxidation of 2.6-di-tert-butylphenol
Fig. 6 UV-Vis spectral change during the 4-mediated oxidation of
−3
phenols monitored at 685 nm. Reaction conditions: [4] = 0.5 × 10 M,
2.6-DTBPh] = 0.06 M, in MeCN, at 20 °C Inset shows time course of the
decay of 4, without substrate (a), and with 0.06 M 2.6-DTBPh (b), Δt =
50 s.
mediated by 3 (□) or 4 (■) at 20 °C in MeCN. [2.6-DTBPh]
= 0.1 M.
0
[
1
Scheme 3
Fig. 7 Dependence of the reaction rate (Vox) on the substrate concen-
trations for the oxidation of 2.6-di-tert-butylphenol mediated by 3 (□) or
= 0.5 × 10− M.
3
4
(■) at 20 °C in MeCN. [3 or 4]
0
0 i
strate oxidation. At constant [substrate] , the reaction rates (V
and Vox) of the decay of 3 and 4 vary linearly with the in situ
0
.5
formed [catalyst 3 or 4] , meaning that both reactions show a
half-order dependence in catalyst, suggesting a dissociation
process via a homolytic cleavage of the O–O bond in a fast
equilibrium (Fig. 8 and Scheme 3).
The reaction kinetic parameters for complexes 3 and 4 are
−
5
−1/2 −1
#
−1
#
k
−
6
ox = 20.2 × 10
M
s
, ΔH = 43 kJ mol , ΔS
=
=
−
1
−1
−5
−1/2
s−1, ΔH
at 293 K, respectively
#
171 J mol
K
, and k = 5.75 × 10
M
ox
4 kJ mol− , ΔS = −108 J mol−1
1
#
K
−1
(Fig. 9). As shown in Fig. 10, there is a linear correlation
between log kox and O–H bond dissociation energy (BDE) for 3,
which suggests that the oxidizing species involved is selective
in nature. This may be taken as evidence for a HAT including a
concerted hydrogen abstraction (Table S6, ESI†).
Fig. 9 Eyring plots for the oxidation of 2.6-di-tert-butyl-phenols
mediated by 3 (□) or 4 (■) in MeCN.
This journal is © The Royal Society of Chemistry 2016
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Fig. 11 Hammett plot for the oxidation of para-substituted 2.6-di-tert-
Fig. 10 Plot of rate constants versus O–H BDE for the oxidation of
para-substituted 2.6-di-tert-butylphenols (0.06 M) by 3 (□) or 4 (■) in
MeCN at 20 °C.
butylphenols mediated by 3 (□) or 4 (■) at 20 °C. Reaction conditions:
−3
[
0 0
3 or 4] = 0.5 × 10 M; [4R-2.6-DTBPh] = 0.06 M.
This is supported by a kinetic isotope effect (KIE) of 1.78
for the 3-mediated oxidation of 2.6-di-tert-butylphenol. A
similar plot is observed for 4, but with slower rates, in line
with its observed stability. The introduction of substituents in
the para position of the phenyl ring of phenol affected the rate
appreciably, the electron-releasing substituents accelerated the
rate significantly, suggesting the electrophilic character of the
key oxidant in both cases (ρ = −0.67 and −0.71 for 3 and 4,
respectively) (Table 3, and Fig. 11).
Stoichiometric O–H activation of 1,4-hydroquinones
mediated by complex 3 and 4. The reactivities of complex 3
and 4 with 1,4-hydroquinones in HAT show that both com-
plexes exhibit enhanced reactivities when compared to the 2.6-
di-tert-butylphenol oxidation. Kinetic experiments, similarly to
the latter substrate, revealed 1st-order dependence on the
Fig. 12 Dependence of the oxidation reaction rates (Vox) on the sub-
2
(
-chloro-1,4-hydroquinone, and half-order dependence on the strates concentrations for the oxidation of 2-chloro-1,4-hydroquinone
−
1/2 −1
#
□
■
with 3 ( ) or 4 ( ) in MeCN at 5 °C.
µ-1,2-peroxo)diiron(III) 3 with kox = 0.38 M
s
, ΔH
=
6 kJ mol− and ΔS = −121 J mol
1
#
−1
K
−1
at 5 °C (Table 3,
3
Fig. 12 and 13, and Tables S4 and S5, Fig. S2, ESI†). A slower
rate, kox = 0.06 M−
1/2
s , was observed with respect to complex
−1
Table 3 Comparison of kinetic data obtained by 3 and 4 complexes at
a
b
2
0 °C and 5 °C
Substrate
Complex
—
3
4
t
2
k
1/2 (s)
.6-DTBPh
645
3210
a
−
5
−1/2
s−1
)
ox (10
M
20.22
43
−171
−0.67
1.78
5.75
64
−108
−0.71
ΔH (kJ mol−1)
#
#
−1
K
−1)
ΔS (J mol
ρ
KIE
2
k
b
2
-Cl-H Q
−
1/2 −1
ox (M
s
)
−1
0.62
36
−121
2.9
0.059
51
−83
2.3
#
Fig. 13 Dependence of the reaction rates on the (µ-1.2-peroxo)
diiron(III) intermediate concentrations for the oxidation of 2-chloro-1,4-
hydroquinone with 3 (□) or 4 (■) in MeCN at 5 °C.
ΔH (kJ mol
)−1)
#
−1
ΔS (J mol
K
KIE
Dalton Trans.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Dalton Transactions
Paper
4
. The different reactivities can be explained by the introduc-
tion of the methyl-substituent on the NH group of the benz-
imidazole which increases the electron density on the iron
centre, and decreases its electrophilic character. Upon using
meta-substituted 1,4-hydoquinones with electron donating
groups the rate of the decay processes were increased remark-
ably (Fig. 14 and 15, and Table S7, ESI†), suggesting that the
metal-based oxidant is electrophilic. The non-linear nature of
the Hammett plot can be explained by the two available O–H
groups in different positions. A similar trend was reported for
IV
the [(TMC)Fe (O)]-mediated (TMC
= 1,4,8,11-tetramethyl-
1,4,8,11-tetraazacyclotetradecane) oxidation of substituted
hydroquinones. A kinetic isotope effect (KIE) value of 2.9 was
obtained by determining the rate constants (k ) corresponding
ox
to separate reactions of 2-chloro-1,4-hydroquinone with 3,
Fig. 14 Plot of rate constants versus O–H BDE for the oxidation of sub-
stituted hydroquinones (0.01 M) by 3 (□) (−0.147) or 4 (■) (−0.253) in using the same conditions as described above, demonstrating
MeCN at 5 °C.
that the H-atom abstraction is involved in the rate-determining
step. This value is comparable to that was observed for the
IV
12
[(TMC)Fe (O)]-mediated oxidation above (2.7).
Computational characterisation of peroxo-diiron(III) complexes
and O–O homolysis
Our first computational step was the geometry optimisation of
the ground state structures of the 12 possible broken symmetry
Fig. 15 Hammett plot for the oxidation of hydroquinones (H
H, Br, Cl, Me, Bu). Reaction conditions: 0.5 mM 3 (□) or 4 (■), 0.01 M
2
Q–X, X:
t
Fig. 16 Potential energy surfaces for hemolytic bond breaking (the 6
substituted H
2
Q in MeCN at 5 °C.
lowest energy processes are shown).
Table 4 Energies (kcal mol− ), selected bond lengths (Å), spin densities for the 12 optimised spin solutions
1
Selected bond lengths
Spin densities
Fe1 Fe2
0.81
Spin solution
ΔG (vaccum)
ΔG (MeCN)
Fe1–O1
Fe2–O2
O–O
Fe–Fe
O1
0.15
0.20
0.17
0.23
0.35
0.44
0.25
−0.17
−0.03
0.38
O2
S = 0(1,−1)
S = 1(1,1)
S = 2(3,1)
S = 3(5,1)
S = 4(5,3)
S = 5(5,5)
S = 0(5,−5)
S = 1(3,−1)
S = 2(5,−1)
S = 1(5,−3)
S = 0(3,−3)
S = 3(3,3)
0.0
0.4
0.0
0.7
0.6
1.80
1.80
1.80
1.81
1.86
1.89
1.86
1.80
1.80
1.87
1.81
1.81
1.80
1.80
1.80
1.87
1.82
1.89
1.86
1.80
1.87
1.81
1.80
1.79
1.41
1.42
1.42
1.40
1.39
1.39
1.37
1.41
1.39
1.38
1.41
1.41
4.40
4.40
4.39
4.55
4.56
4.71
4.72
4.40
4.55
4.57
4.44
4.35
−0.81
0.84
−0.15
0.20
0.07
0.39
0.08
0.44
−0.25
0.03
0.35
0.08
−0.07
−0.02
0.84
0.86
0.90
−2.0
−4.4
−8.0
−10.9
−11.3
−2.5
−7.2
−8.2
−0.1
−1.9
2.84
4.18
0.7
−0.2
−0.8
−0.8
0.3
−2.2
−0.3
5.6
4.17
4.19
2.92
4.19
4.17
−4.17
−0.83
−0.86
4.16
2.81
2.85
2.82
4.15
−2.85
−2.80
2.91
0.05
0.02
2.6
This journal is © The Royal Society of Chemistry 2016
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Table 5 Energies (kcal mol− ), selected bond lengths (Å), spin densities for the transition states of the O–O homolytic bond breaking in Fe
1
O
2
2
models (the 6 lowest energy processes are shown)
Selected bond lengths
Spin densities
Fe1 Fe2
0.95
Spin solution
ΔG (vaccum)
ΔG (MeCN)
Fe1–O1
Fe2–O2
O–O
Fe–Fe
O1
0.71
−0.36
0.59
0.09
0.16
−0.69
O2
S = 0(1,−1)
S = 2(3,1)
S = 3(5,1)
S = 4(5,3)
S = 0(5,−5)
S = 1(3,−1)
15.3
17.5
12.6
10.9
12.2
13.1
17.5
22.3
20.0
20.8
25.8
17.5
1.69
1.71
1.66
1.71
1.69
1.70
1.69
1.68
1.72
1.67
1.69
1.69
1.86
1.91
1.84
1.82
1.78
1.85
4.57
4.61
4.82
4.82
5.14
4.56
−0.95
2.93
−0.71
0.58
0.09
0.45
−0.16
0.56
0.73
1.10
3.95
3.98
3.03
3.81
−0.96
−3.81
2.90
2 2
spin solutions for the Fe O complexes. In vaccuum, the lowest
energy solution was the antiferromagnetically coupled singlet
S = 0(5,−5) (containing 5 spin-up electrons on the first iron
atom and 5 spin-down electrons on the second), closely fol-
lowed by the antiferromagnetically coupled triplet S = 1(5,−3)
and the S = 5(5,5) solutions. After inclusion of solvation ener-
gies using acetonitrile as solvent, all 12 spin solutions were
found to be degenerate in energy, with approximately 6.5 kcal
−1
mol between the lowest energy solution S = 0(5,−5) and
highest energy S = 0(3,−3). Selected geometrical parameters are
described in Table 4. Fe–O bonds vary between 1.79–1.89 Å and
are typical of Fe(III)-peroxo ligands, while Fe–Fe distances are
8
similar to those derived from resonance Raman spectroscopy.
The O–O homolytic bond cleavage processes were then explored:
Fig. 16 (see Table 5 and S9 for all spin states, ESI†) show the
that the potential energy surfaces (PESs) intersect for different
spin solutions, suggesting that spin cross-over is possible (L cropped to N–C–C–N, for full models see coordinates in ESI†).
during homolysis. The TSs occur early along the O–O PES and
Fig. 17 Reaction barriers for the catalase reaction (kcal mol− in aceto-
1
nitrile), together with relevant bond lengths (in Å), spin densities and
partial charges (in parenthesis). Models are simplified for ease of view
2
are degenerate in energy, with values suggesting a feasible
process. All these data indicate that O–O homolysis is facile and
will occur on multiple spin PESs simultaneously.
Computational characterisation of Fe(IV)O catalase and phenol
reactivity
The DFT estimates on the catalase reaction and the HAT reac-
tion of DTBPh are shown in Fig. 17 and 18 respectively; both
processes are predicted to occur through a concerted HAT reac-
tion, similarly to what was determined experimentally.
3
5
Although Fe(IV)O and Fe(IV)O are essentially degenerate in
energy, the reaction is preferred along the lower spin state
surface for both cases. For the catalase reaction, the S = 2 tran-
−
1
sition state is 9 kcal mol higher in energy that the S = 1 tran-
sition state, indicating that the preferred reaction path is along
S = 1, although there is a clear similarity in geometrical para-
meters for Fe–O and FeO–H and substrate-H (Fig. 17). Both plified for ease of view (L cropped to N–C–C–N and 2.6-DTBPh cut,
processes occur at a late stage along the O–H PES (Fig. S6,
Fig. 18 Reaction barriers for the hydrogen atom abstraction from 2.6-
DTBPh (kcal mol in acetonitrile), together with relevant bond lengths
−1
(
in Å), spin densities and partial charges (in parenthesis). Models are sim-
2
for full models see coordinates in ESI†).
ESI†), with only the S = 1 reaction being exergonic (Fig. 17). In
the case of the HAT reaction for DTBPh, no hydrogen bound
adducts could be obtained between the substrate and the
oxidant for any of the spin states, and no TS could be opti-
Conclusions
mised for the S = 2 spin state (consistent with the B3LYP PES The reactivity of two peroxo-diiron(III) complexes (3 and 4) has
given in Fig. S7, ESI†). All energies shown in Fig. 18 indicate a been investigated in O–H activation processes as a structural
preferred S = 1 spin state for this reaction.
and functional model of RNR-R2 enzyme. The decay of the
Dalton Trans.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Dalton Transactions
Paper
peroxo species was affected by 2,6-di-tert-butylphenol and 1,4-
hydroquinone, leading to a 2,2′-biphenol-coupled product and
4 (a) S.-K. Lee, B. G. Fox, W. A. Froland, J. D. Lipscomb and
E. Münck, J. Am. Chem. Soc., 1993, 115, 6450; (b) K. E. Liu,
A. M. Valentine, D. Wang, B. H. Huynh, D. E. Edmondson,
A. Salifoglou and S. J. Lippard, J. Am. Chem. Soc., 1995, 117,
10174; (c) A. M. Valentine, S. S. Stahl and S. J. Lippard,
J. Am. Chem. Soc., 1999, 121, 3876.
5 (a) V. V. Vu, J. P. Emerson, M. Martinho, Y. S. Kim,
E. Münck, M. H. Park and L. Que Jr., Proc. Natl. Acad.
Sci. U. S. A., 2009, 106, 14814; (b) Z. Han, N. Sakai,
L. H. Böttger, S. Klinke, J. Hauber, A. X. Trautwein and
R. Hilgenfeld, Structure, 2015, 23, 11.
6 (a) E. Y. Tshuva and S. J. Lippard, Chem. Rev., 2004, 104,
987; (b) Y. Dong, S. Menage, B. A. Brennan, T. E. Elgren,
H. G. Jang, L. L. Pearce and L. Que Jr., J. Am. Chem. Soc.,
1993, 115, 1851; (c) N. Kitajima, N. Tamura, H. Amagai,
H. Fukui, Y. Moro-oka, Y. Mizutani, T. Kitagawa, R. Mathur
and K. Heerwegh, J. Am. Chem. Soc., 1994, 116, 9071;
1
,4-benzoquinones, respectively. Based on detailed kinetic
(half-order dependence for 3 and 4), mechanistic (KIE = 1.78,
ρ = −0.7 for 2,6-DTBPh), and computational studies, an elec-
trophilic oxoiron(IV) species with S = 1 spin state, was
suggested as reactive species responsible for the HAT pro-
cesses. Based on our preliminary results complex 3 is capable
of oxidizing both oxygen-atom transfer and hydrogen-atom
abstraction, that is a further evidence for the presence of elec-
trophilic oxoiron(IV) species as key oxidant (Scheme 3). From
benzyl alcohol benzaldehyde (34% based on 3) was formed
and triphenylphosphine gave triphenylphosphine oxide (93%
based on 3) at 20 °C in MeCN (Fig. S3–S5, ESI†). Detailed
kinetic studies on these two systems are in progress.
(
d) Y. Dong, S. P. Yan, V. G. Young and L. Que Jr., Angew.
Acknowledgements
Chem., 1996, 108, 673, (Angew. Chem., Int. Ed. Engl., 1996,
35, 618); (e) T. Ookubo, H. Sugimoto, T. Nagayama,
H. Masuda, T. Sato, K. Tanaka, Y. Maeda, H. Okawa,
Y. Hayashi, A. Uehara and M. Suzuki, J. Am. Chem. Soc.,
1996, 118, 701; (f) K. Kim and S. J. Lippard, J. Am. Chem.
Soc., 1996, 118, 4914; (g) M. Kodera, M. Itoh, K. Kano,
T. Funabiki and M. Reglier, Angew. Chem., 2005, 117, 7266,
(Angew. Chem., Int. Ed., 2005, 44, 7104); (h) S. V. Kryatov,
S. Taktak, I. V. Korendovych, E. V. Rybak-Akimova,
J. Kaizer, S. Torelli, X. P. Shan, S. Mandal, V. L. Mac-Murdo,
A. Mairata i Payeras and L. Que Jr., Inorg. Chem., 2005, 44,
This work was supported by a grant from the Hungarian
Research Fund (OTKA) K108489. and COST Actions CM1205,
CM1201, and CM1003. The authors would like to acknowledge
funding from the Romanian Ministry for Education and
Research (grant PN-II-ID-PCE-2012-4-0488). The authors are
also indebted to the Data Center from NIRDIMT (Cluj-Napoca,
Romania) for providing the computational infrastructure and
technical assistance.
8
5; (i) A. T. Fiedler, X. Shan, M. P. Mehn, J. Kaizer,
Notes and references
S. Torelli, J. R. Frisch, M. Kodera and L. Que Jr., J. Phys.
Chem. A, 2008, 112, 13037; ( j) M. A. Cranswick, K. K. Meier,
X. Shan, A. Stubna, J. Kaizer, M. P. Mehn, E. Münck and
L. Que, Jr., Inorg. Chem., 2012, 51, 10417; (k) Y. H. Zong,
Y. Zang, L. J. Shu, E. C. Wilkinson, L. Que, Jr.,
K. Kauffmann and E. Münck, J. Am. Chem. Soc., 1986, 108,
5027; (l) X. Zhang, H. Furutachi, S. Fujinami, S. Nagatomo,
Y. Maeda, Y. Watanabe, T. Kitagawa and M. Suzuki, J. Am.
Chem. Soc., 2005, 127, 826.
1
(a) J. Du Bois, T. J. Mizoguchi and S. J. Lippard, Coord.
Chem. Rev., 2000, 200–202, 443; (b) M. Fontecave,
S. Ménage and C. Duboc-Toia, Coord. Chem. Rev., 1998,
1
78–180, 1555; (c) S. V. Kryatov and E. V. Rybak-Akimova,
Chem. Rev., 2005, 105, 2175; (d) A. R. McDonald and
L. Que, Jr., Coord. Chem. Rev., 2013, 257, 414; (e) M. Costas,
M. P. Mehn, M. P. Jensen and L. Que, Jr., Chem. Rev., 2004,
1
04, 939; (f) E. Y. Tshuva and S. J. Lippard, Chem. Rev.,
2
004, 104, 987.
7 (a) C. Duboc-Toia, S. Menage, R. Y. N. Ho, L. Que Jr.,
C. Lambeaux and M. Fontecave, Inorg. Chem., 1999, 38,
1261; (b) Y. Mekmouche, H. Hummel, R. Y. N. Ho, L. Que
Jr., V. Schünemann, F. Thomas, A. X. Trautwein, C. Lebrun,
K. Gorgy, J.-C. Leprete, M.-N. Collomb, A. Deronzier,
M. Fontecave and S. Menage, Chem. – Eur. J., 2002, 8, 1196;
(c) M. Costas, C. W. Cady, S. V. Kryatov, M. Ray, M. J. Ryan,
E. V. Rybak-Akimova and L. Que Jr., Inorg. Chem., 2003, 42,
7519; (d) J. S. Pap, M. A. Cranswigh-Hergovich, G. Baráth,
M. Giorgi, G. T. Rohde, J. Kaizer, G. Speier and L. Que, Jr.,
Eur. J. Inorg. Chem., 2013, 3858.
8 J. S. Pap, A. Draksharapu, M. Giorgi, W. R. Browne,
J. Kaizer and G. Speier, Chem. Commun., 2014, 50,
1326.
9 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone,
B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato,
2
(a) J. M. Bollinger, C. Krebs, A. Vicol, S. Chen, B. A. Ley,
D. E. Edmondson and B. H. Huynh, J. Am. Chem. Soc.,
1
998, 120, 1094; (b) P. Moënne-Loccoz, J. Baldwin,
B. A. Ley, T. M. Loehr and J. M. Bollinger, Biochemistry,
998, 37, 14659; (c) J. Baldwin, C. Krebs, L. Saleh,
1
M. Stelling, B. H. Huynh, J. M. Bollinger and P. Riggs-
Gelasco, Biochemistry, 2003, 42, 13269; (d) A. J. Skulan,
T. C. Brunold, J. Baldwin, L. Saleh, J. M. Baldwin and
E. I. Solomon, J. Am. Chem. Soc., 2004, 126, 8842;
(
e) D. Yun, R. Garcia-Serres, B. M. Chicalese, Y. H. An,
B. H. Huynh and J. M. Bollinger, Biochemistry, 2007, 46,
925.
(a) J. A. Broadwater, J. Ai, T. M. Loehr, L. Sanders and
B. G. Fox, Biochemistry, 1998, 37, 14664;
b) J. A. Broadwater, C. Achim, E. Münck and B. G. Fox, Bio-
chemistry, 1999, 38, 12197.
1
3
(
This journal is © The Royal Society of Chemistry 2016
Dalton Trans.

View Article Online
Paper
Dalton Transactions
X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng,
R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth,
P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels,
O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski and
D. J. Fox, Gaussian 09, Revision D.01, Gaussian, Inc., Wall-
ingford CT, 2013.
J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda,
J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,
H. Nakai, T. Vreven, J. A. Montgomery, Jr., J. E. Peralta,
F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers,
K. N. Kudin, V. N. Staroverov, T. Keith, R. Kobayashi, 10 S. Grimme, J. Antony, S. Ehrlich and H. Krieg, J. Chem.
J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, Phys., 2010, 132, 154104.
S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, 11 S. Grimme, S. Ehrlic and L. Goerick, J. Comput. Chem.,
M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, 2011, 32, 1456.
J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, 12 K. Nehru, Y. Jang, S. Oh, F. Dallemer, W. Nam and J. Kim,
A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, Inorg. Chim. Acta, 2008, 361, 2557.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2016