Hydrogen Atom vs. Hydride Transfer in Cytochrome P450 Oxidations: A Combined Mass Spectrometry and Computational Study

Article abstract of DOI:10.1002/ejic.201800273

Biomimetic models of short-lived enzymatic reaction intermediates can give useful insight into the properties and coordination chemistry of transition metal complexes. In this work, we investigate a high-valent iron(IV)-oxo porphyrin cation radical complex, namely [Fe^IV(O)(TPFPP^+·)]⁺, for which TPFPP is the dianion of 5,10,15,20-tetrakis(pentafluorophenyl) porphyrin. The [Fe^IV(O)(TPFPP^+·)]⁺ ion was studied by ion-molecule reactions in a Fourier transform ion cyclotron resonance mass spectrometer through reactivities with 1,3,5-cycloheptatriene, 1,3-cyclohexadiene and toluene. The different substrates give dramatic changes in reaction mechanism and efficiencies, whereby cycloheptatriene leads to hydride transfer, whereas cyclohexadiene and toluene react through hydrogen atom abstraction. Detailed computational studies point to major differences in ionization energy, as well as C–H bond energies of the substrates that influence the hydrogen atom abstraction vs. electron transfer pathways. The various variables that determine the pathways for hydride transfer vs. hydrogen atom transfer are elucidated and discussed.

Full text of DOI:10.1002/ejic.201800273

A Journal of
Accepted Article
Title: Hydrogen atom versus hydride transfer in cytochrome P450
oxidations: A combined mass spectrometry and computational
study
Authors: Samuel de Visser, Fabian Cantu Reinhard, Maria Elisa
Crestoni, and Simonetta Fornarini
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Eur. J. Inorg. Chem. 10.1002/ejic.201800273
Link to VoR: http://dx.doi.org/10.1002/ejic.201800273

10.1002/ejic.201800273
European Journal of Inorganic Chemistry
FULL PAPER
Hydrogen atom versus hydride transfer in cytochrome P450
oxidations: A combined mass spectrometry and computational
study
Fabián G. Cantú Reinhard,[a] Simonetta Fornarini,*[b] Maria Elisa Crestoni,*[b] and Sam P. de Visser*[a]
Abstract: Biomimetic models of short-lived enzymatic reaction
intermediates can give useful insight into the properties and
coordination chemistry of transition metal complexes. In this work we
hydroxylation has been the topic of many studies and
controversies have been posed whether it is initiated by a rate-
determining hydrogen atom abstraction (HAT) or a hydride
transfer (HT), as explained in Scheme 1.[5]
Thus, computational modelling on hydrogen atom abstraction
reactions by P450 enzymes with common aliphatic substrates,
investigate
complex, namely [FeIV(O)(TPFPP+•)]+ where TPFPP is the dianion of
5,10,15,20-tetrakis(pentafluorophenyl) porphyrin. The
a high-valent iron(IV)-oxo porphyrin cation radical
[FeIV(O)(TPFPP+•)]+ ion was studied by ion-molecule reactions in a
Fourier transform-ion cyclotron resonance mass spectrometer
through reactivities with 1,3,5-cycloheptatriene, 1,3-cyclohexadiene
and toluene. The different substrates give dramatic changes in
reaction mechanism and efficiencies, whereby cycloheptatriene
leads to hydride transfer, while cyclohexadiene and toluene react via
hydrogen atom abstraction. Detailed computational studies point to
major differences in ionization energy as well as C–H bond energies
of the substrates that influence the hydrogen atom abstraction
versus electron transfer pathways. The various variables that
determine the pathways for hydride transfer versus hydrogen atom
transfer are elucidated and discussed.
such as alkanes or alkylbenzenes, revealed a stepwise
mechanism with an initial hydrogen atom abstraction followed by
radical rebound to form alcohol product complexes.[6] However,
experimental studies of a synthetic iron(IV)-oxo porphyrin in a
reaction with the substrate 10-methyl-9,10-dihydroacridine
(AcrH2) gave hydride transfer instead.[7] Computational
modelling on these reaction mechanisms implicated that hydride
and hydrogen atom transfer pathways could both be feasible
under certain reaction conditions, but are determined by the
thermochemical properties of the oxidants, intermediates and
substrates.[8] To gain more insight into the mechanisms leading
to hydride transfer by CpdI of P450, we pursued a combined
mass spectrometry and computational study on the hydride
versus hydrogen atom abstraction patterns of model substrates.
In particular, we compare the reactivity patterns of these CpdI
models with substrates that are likely to react via either hydride
or hydrogen atom transfer. Thus, 1,3,5-cycloheptatriene (CHT)
is used as a model substrate for hydride transfer reactions and
1,3-cyclohexadiene (CHD) and toluene are considered for
aliphatic hydrogen atom transfer. We present here a detailed
joint mass spectrometric and computational study on substrate
activation by a synthetic CpdI model and establish the origin of
hydride versus hydrogen atom transfer pathways. Furthermore,
a detailed thermochemical and valence bond analysis explains
the intrinsic properties of substrate and oxidant that determine
hydride transfer versus hydrogen atom transfer by metal-oxo
oxidants.
Introduction
Cytochrome P450 enzymes are versatile catalysts in the human
body that detoxify the liver from foreign compounds, such as
drugs and some of their metabolites as well as other
xenobiotics.[1] Structurally, the P450s contain an iron(III)-heme
deeply embedded into a protein matrix, where the substrate
binds into a pocket of varying size depending on the P450
isozyme.[2] The P450 catalytic cycle uses molecular oxygen, two
reduction and two protonation equivalents in order to convert the
iron(III)-heme(water) resting state into its active form, namely the
high-valent iron(IV)-oxo porphyrin cation radical species called
Compound I (CpdI).[3] CpdI is known as one of the most efficient
and versatile oxidants in Nature and generally reacts with
substrates through oxygen atom transfer. For instance, it
converts aliphatic groups through a hydroxylation process into
alcohols, but also can activate aromatic C‒H bonds into phenols
and sulfides into sulfoxides.[1,4] In particular, substrate
Results
Mass spectrometry results.
The preparation of a high-valent iron(IV)–oxo porphyrin cation
radical complex, [FeIV(O)(TPFPP+•)]+ with TPFPP being the
dianion of 5,10,15,20-tetrakis(pentafluorophenyl) porphyrin, was
done by treatment of [FeIII(TPFPP)]Cl with iodosylbenzene as
oxygen atom donor, in a methanol/dichloromethane solution at –
40°C as previously reported.[9] The reaction mixture sampled by
electrospray ionization (ESI) and characterized by high
resolution Fourier transform-ion cyclotron resonance (FT-ICR)
mass spectrometry contains substantial amounts of an ion
cluster centered at m/z 1044.0116, revealing the incorporation of
one oxygen atom by the reactant species [FeIII(TPFPP)]+. When
trapped in the FT-ICR cell, mass-selected ions with m/z
1044.0116 are remarkably stable toward unimolecular
dissociation (kdiss ≤ 0.001 s–1) and their fragmentation patterns
and isotope patterns characterize them as [FeIV(O)(TPFPP+•)]+.
[a]
F.G. Cantú Reinhard, Dr S.P. de Visser
Manchester Institute of Biotechnology and School of Chemical
Engineering and Analytical Science
The University of Manchester, 131 Princess Street, Manchester M1
7DN, United Kingdom
sam.devisser@manchester.ac.uk
[b]
Prof. Dr. S. Fornarini, Prof. Dr. M.E. Crestoni
Dipartimento di Chimica e Tecnologie del Farmaco,
Università di Roma “La Sapienza”, P.le A. Moro 5, 00185, Roma,
Italy
mariaelisa.crestoni@uniroma1.it; simonetta.fornarini@uniroma1.it
Supporting information for this article is given via a link at the end of
the document.
1
This article is protected by copyright. All rights reserved.

10.1002/ejic.201800273
European Journal of Inorganic Chemistry
FULL PAPER
Scheme 1. Products obtained from a reaction of CpdI with an aliphatic group.
Table 1. Thermodynamic and kinetic data and product distributions for the reaction of selected hydrocarbons with [FeIV(O)(TPFPP+•)]+.
[b]
[a]
[c]
Compound (A)
IE [a]
BDECH
IEradical
kexp
 (%)
%[A‒H]+ [d]
%[FeIII(TPFPP)]+
%ADD
Toluene
CHT
8.828
8.0‒8.3
8.25
87.9
72.9
72.9
7.242 [e]
6.28 [f]
0.36
2.67
1.58
3.9[h]
30
‒
80
20
‒
40
‒
60
CHD
6.82 [g]
17 [i]
100
‒
[a] In eV. [b] In kcal mol–1. [c] Second-order rate constant in units of 10‒10 cm3 molecule‒1 s‒1, at the temperature of the FT-ICR cell of 300K. [d]
product ion at m/z value for substrate minus H‒. [e] Data from Ref. 13. [f] The reaction of cycloheptatriene-7-[D1] displays a H/D kinetic isotope
effect of 2.5, resulting from the time independent ratio of the abundances of the C7H6D+ and C7H7+ product ions. [g] Data from Ref. 16. [h] Data
from Ref. 12. [i] Data from Ref. 11.
Scheme 2. Reaction channels observed for the reaction of [FeIV(O)(TPFPP+•)]+ with 1,3,5-cycloheptatriene in the FT-ICR cell.
2
This article is protected by copyright. All rights reserved.

10.1002/ejic.201800273
European Journal of Inorganic Chemistry
FULL PAPER
Moreover, when exposed to stationary concentrations of
the second-order rate constants for both reaction processes, as
displayed in Figure S1 in the Supporting Information.
candidate reductants, they react with
a wide variety of
compounds, including NO, NO2, aromatics, olefins, sulfides,
amines, and phosphites.[10]
The reaction of [FeIV(O)(TPFPP+•)]+ with cycloheptatriene-7-[D1]
gives a mixture of hydrogen and deuterium transfer and results
+
A
comprehensive combined mass spectrometric and
in the formation of both C7H6D+ and C7H7 ions in the product
+
computational approach recently succeeded in unveiling the full
details of multiple reaction channels and product distribution in
olefin epoxidation,[11] and aromatic hydroxylation,[12] in the gas-
phase, i.e. in a solvent-free environment. In this context, we
have become keenly interested in gaining further information on
the Compound I-like reactivity of [FeIV(O)(TPFPP+•)]+ with CHT,
that has been assessed and is described herein. CHT is an
isomer of toluene, although it lacks aromatic stabilization.
Therefore, the two compounds possess quite different physical
and chemical properties. In particular, CHT is characterized by a
markedly lower ionization energy (IE): 8.0 – 8.3 eV for CHT
versus 8.828 eV for toluene.[13] Also significantly smaller for CHT
versus toluene is its C–H bond dissociation energy (BDECH) for
removal of a hydrogen atom from the methylene group. Thus,
the BDECH of methylene C‒H bond of CHT is 72.9 kcal mol–1
whereas it is 87.9 kcal mol–1 for the cleavage of the H–CH2C6H5
bond in toluene.[14] However, the most notable reactivity feature
in the reactive behavior of CHT is the drive towards formation of
mixture. The abundance ratio of C7H6D+ and C7H7 product ions
is about 2.5 and is independent of the reaction time. Therefore,
the ratio of the abundances of C7H6D+ versus C7H7+ corresponds
to the H/D kinetic isotope effect for competing hydride versus
deuteride abstraction from cycloheptatriene-7-[D1]. The kinetic
plot for the reaction of [FeIV(O)(TPFPP+•)]+ with cycloheptatriene-
7-[D1] is given in Figure
1
showing the profiles of ion
abundances as a function of time.
100
m/z 91
Abundance
m/z 92
m/z 1028
m/z 1044
75
50
25
0
+
c-C7H7 , i.e. the tropylium ion, which has aromatic character and
hence is relatively stable. This species plays an important role in
the oxidation of CHT and CHT-derivatives and the process has
interesting synthetic applications, namely in anodic oxidation,
0
10
20
30
40
oxidation
chemoenzymatic studies.[15]
Because of the peculiar features of cycloheptatriene, we
envisaged detailed gas phase reactivity study towards
by
photoinduced
electron
transfer,
and
reaction time (s)
Figure 1. Ion abundances as a function of time after isolation of m/z 1044 ions,
i.e. [FeIV(O)(TPFPP+.)]+, in the FT-ICR cell in the presence of 5 × 10–8 mbar
cycloheptatriene-7-[D1] at room temperature.
a
[FeIV(O)(TPFPP+•)]+ would be interesting and could contribute to
elucidation of the intrinsic factors that determine the reactivity
behavior of this model of P450 Cpd I. An interesting comparison
may as well arise from the reaction of [FeIV(O)(TPFPP+•)]+ with
1,3-cyclohexadiene (CHD) as it has a comparable IE and also
similar C–H BDECH for the methylene group (IE and BDECH
values are equal to 8.25 eV and 72.9 kcal mol–1, respectively),
see Table 1.[13,14,16] However, ionization of the so-formed radical
It is interesting to note that the HT path is specific for CHT, while
formation of the reduced [FeIII(TPFPP)]+ ion is commonly
observed in the reactivity of both olefins and aromatic
compounds.[11,12] In order to find out how the reactivity of
[FeIV(O)(TPFPP+•)]+ with CHT differs to that with aliphatic
substrates, we decided to compare the reactivity of
[FeIV(O)(TPFPP+•)]+ with CHD and toluene. Thus, P450 Cpd I
models typically react with CHD through dehydrogenation and
the formation of benzene,[17] or alternatively by oxygen atom
transfer to form the corresponding epoxide. On the other hand,
using toluene as a substrate a mechanism was found starting
with hydrogen atom abstraction and followed by OH rebound to
form phenylmethanol products.[18] As follows, the aromatic C7H8
isomer of CHT, namely toluene, displays exclusive formation of
the reduced [FeIII(TPFPP)]+ ion with a markedly lower efficiency
of only 4%, which includes also a formal addition path leading to
a complex at m/z value corresponding to the sum of oxidant and
substrate (ADD pathway), Table 1. Also 1,3-cyclohexadiene,
with conjugated double bonds, undergoes exclusively an oxygen
atom transfer process (OAT) with 17% efficiency, which could
implicate substrate epoxidation.[11] Thus, in aromatic
hydroxylation and double bond epoxidation reactions studied for
analogous systems,[11,12,19] OAT is the common product ion, and
hence would implicate epoxidation of CHD.
•
•
is remarkably easier for c-C7H7 when compared to the c-C6H7
radical from 1,3-cyclohexadiene (6.28 vs. 6.82 eV, respectively).
Table 1 gives a summary of all cited thermochemical data.[13,14,16]
The thermal reactivity of [FeIV(O)(TPFPP+•)]+ towards CHT
occurs along two reaction channels, namely a process that
yields the reduced [FeIII(TPFPP)]+ ion and oxidized (neutral)
substrate, and a formal hydride transfer (HT) pathway, whereby
+
C7H7 as ionic product is released together with (neutral)
[FeIII(OH)(TPFPP)], Scheme 2. Thus, upon isolation of the ion at
m/z 1044 representing the [FeIV(O)(TPFPP+•)]+ complex, ion
abundances were measured as a function of time in the
presence of a stationary concentration of CHT. Indeed, the
abundance of ions at m/z 1044 decayed as a function of time
and two new peaks in the spectrum appeared at m/z 1028 and
m/z 91. These two peaks correspond to the products
+
[FeIII(TPFPP)]+ and C7H7 resulting from oxygen atom transfer
and hydride transfer from the reaction of [FeIV(O)(TPFPP+•)]+
with cycloheptatriene. The reaction is fast and characterized by
a reaction efficiency of  = 30% (see Table 1). Semi-logarithmic
plots of the reactant ion abundance as a function of time
obtained at varying CHT concentration enabled us to calculate
3
This article is protected by copyright. All rights reserved.

10.1002/ejic.201800273
European Journal of Inorganic Chemistry
FULL PAPER
Scheme 3. Mechanisms considered in this work for the reactions of 1 and 2 with CHT, CHD and toluene and the products obtained.
1.223
HO: 1.460
FeO: 1.649
174.4
i476 cm‒1
4TSHA1,TPFPP
CO: 2.053 (1.831)
FeO: 1.954 (1.868)
i145 (i281) cm‒1
4TS2TPFPP (2TS2TPFPP
)
Figure 2. Free energy landscape for oxidation reactions of cycloheptatriene by 4,22. Free energies use electronic energies at UB3LYP/BS2//UB3LYP/BS1 in kcal
mol–1 and contain zero-point, entropic and thermal corrections to 298K. Also given are optimized geometries of 4TSHA1 and 4,2TS2TPFPP with bond lengths in
angstroms, angles in degrees and the imaginary frequency in cm–1
.
4
This article is protected by copyright. All rights reserved.

10.1002/ejic.201800273
European Journal of Inorganic Chemistry
FULL PAPER
Computational modelling.
To support the experiments and gain insight into the reaction
the doublet and quartet spin states of the reactant complex to be
within 0.5 kcal mol‒1 for 4,2[FeIV(O)(Por+•)]+ as well as for
4,2[FeIV(O)(TPFPP+•)]+: 4,22.
pathways
for
1,3,5-cycloheptatriene
activation
by
[FeIV(O)(TPFPP+•)]+ we performed a computational study on the
mechanisms and pathways leading to the various products for
Figure
2
gives the potential energy landscape for
cycloheptatriene activation on one of the aliphatic C‒H bonds by
activation of CHT. As
a
comparison, we also included
4,22. The substrate and oxidant initially form a reactant complex
computational studies using CHD and toluene as substrates,
since the experimental studies with these substrates have
considerably different product distributions (Table 1). Initial
calculations using the CHT substrate utilized a TPFPP scaffold
(
4,2Re) and then react with a fast hydrogen atom abstraction
leading to
a very low-energy intermediate. Although the
4TSHA1,TPFPP barrier is lower in energy than isolated reactants, it
is slightly higher in energy than the reactants complex.
abbreviated to porphyrin (Por) without side chains, 2,41. However, Therefore, the hydrogen atom abstraction will be fast. On the
in a second set of calculations the full system (TPFPP, 2,42) was
also studied for the full reaction pathway of the predominant
hydroxylation reactions, on the quartet and doublet spin states
(see Scheme 3). In the case of CHT substrate technically,
oxygen atom transfer can lead to various isomeric epoxides,
namely at positions 3 or 5 (Scheme 3) or alternatively a
hydroxylation process can lead to alcohols from H1, H3, H4 and
H5. Hydrogen atom abstraction from position H2 was also
attempted but led to the same transition state and local minimum
as the one from H1. We decided to investigate all these
doublet spin state, we were unable to locate the structure of the
hydrogen atom transition states as all attempts led to the
product complexes instead. Similarly to previous calculations on
aliphatic hydroxylation by P450 CpdI, the doublet and quartet
spin states are close in energy and only at the products stage
they diverge due to differences in orbital occupation.[4b,6b,24]
4
The group spin densities of TSHA1,TPFPP give FeO = 2.16, TPFPP
= ‒0.06 and SubH = 0.90, which points to a hydrogen atom
abstraction transition state coupled to a triplet spin iron(IV)-oxo
group. Interestingly, the intermediate (4IR1TPFPP) that connects to
mechanisms
recently.[11,12]
using
established
procedures
reported
this transition state has spin densities corresponding to FeO =
2.96, TPFPP = 0.02 and Sub = 0.02, which implies that the
+
A detailed benchmark study aimed at reproducing experimental
reaction free energies of activation of substrate sulfoxidation
reactions by an iron(IV)-oxo species showed that PBE0 and
B3LYP were the preferred density functional theory methods to
reproduce experimental data and activation enthalpies to within
3 – 4 kcal mol–1 from experiment were obtained for a series of
substrates.[20] Furthermore, for several chemical systems we
calculated bifurcation pathways and predicted the correct
product distributions. For instance, we calculated the hydrogen
atom abstraction pathways from the six possible hydrogen
atoms of a proline residue in a peptide chain by prolyl-4-
hydroxylase and found a low-energy pathway for the formation
of R-4-hydroxyproline, in agreement with experimental
observation.[21] Furthermore, for the cytochrome P450
peroxygenase reaction pathways were calculated for fatty acid
decarboxylation and hydroxylation and the predicted product
distributions matched experiment very well.[22] As such we
utilized these methods in this work.
To gain further insight into hydride transfer versus hydrogen
atom transfer pathways, we included calculations for the full
mechanism of 1,3-cyclohexadiene and toluene, see Scheme 3.
The full set of results is documented in the Supporting
Information (Tables S1 – S16; Figures S2 – S10), while we
focus on the main trends here. Note that hydride transfer is not
always a feasible mechanism as will be discussed later;
however, molecular orbital swaps were attempted in all cases to
estimate the energy difference between hydrogen atom and
hydride transfer pathways. Further details on the possibilities of
hydride transfer follow in the thermochemical analysis of
structures in the Discussion section.
intermediate is an ion-molecule complex of C7H7 with a neutral
[Fe(OH)(TPFPP)]0 molecule. Therefore, the mechanism starts
with a fast hydrogen atom abstraction, but en route to the
intermediate another electron transfer takes place to form the
overall hydride transfer intermediate 4,2IR1. In the next stage of
the reaction the OH‒ rebound gives alcohol product complexes
(P) via the rebound transition state TS2. As can be seen from
Figure 2, the rebound encounters a barrier of 3.9 kcal mol‒1 in
the quartet spin state, but 9.9 kcal mol‒1 in the doublet spin state.
These barriers, particularly the one in the doublet spin state are
relatively high and may implicate dissociation of the complex into
+
individual C7H7
and [Fe(OH)(TPFPP)]0. Indeed, the
experimental measurements observed C7H7+ products alongside
the reduced complex [Fe(TPFPP)]+. It could very well be that the
latter is formed through the low-barrier high-spin reaction
mechanism, whereas the low-spin mechanism leads to
dissociation of the intermediate delivering free C7H7+ ions.
Apart from calculations on the full structure 4,22 and its
mechanism with substrates, we did a more comprehensive study
using the smaller model 4,21, whereby mechanisms for various
hydrogen atom abstractions and oxygen atom transfer
processes were investigated. In general, the hydrogen atom
abstraction mechanism is the same for 4,21 as compared to 4,22.
The 4,2TSHA1 barriers are relatively early on the potential energy
surface with short C–H and long O–H bonds (1.189 and 1.143 Å
and 1.560 and 1.762 Å, respectively) indicative of a fast reaction
process (see Figure 3). Although generally aliphatic
hydroxylation mechanisms are stepwise via
a
radical
intermediate; for the small model no stable radical and/or
cationic intermediates could be located, which implies a fast and
efficient rebound to alcohol products. Therefore, the small model
is not good enough to mimic the experimental system. Further
evidence that the small model fails to capture the chemical
properties of the meso-substituents follows from the calculated
bond dissociation energies (BDEOH) of the O‒H bond of the
[FeIV(OH)(Por)]+ versus [FeIV(OH)(TPFPP)]+ complexes.
Let us start with a description of a typical reaction mechanism of
oxygen atom transfer, namely of cycloheptatriene activation at
its methylene group by the iron(IV)-oxo model. In the following,
we will add the label for the reaction process as a subscript to
the molecule. Calculations were done starting from the
isoelectronic doublet and quartet spin state structures of
[FeIV(O)(Por+•)]+, 4,21. Similarly to our previous work on this
chemical system as well as reports on P450 CpdI,[4b,12,23] we find
5
This article is protected by copyright. All rights reserved.

10.1002/ejic.201800273
European Journal of Inorganic Chemistry
FULL PAPER
2.259 (2.408)
1.631 (1.623)
i143.5 (i86.4) cm‒1
4TSOAT1 (2TSOAT1
)
1.189 (1.143)
2.133 (2.361)
1.648 (1.625)
1.560 (1.762)
175.9 (174.8 )
1.645 (1.621)
i168.1 (i74.2) cm‒1
4TSOAT3 (2TSOAT3
i262.2 (i67.9) cm‒1
4TSHA1 (2TSHA1
)
)
Figure 3. Free energy landscape for HAT and OAT reactions from cycloheptatriene by 4,21. Free energies use electronic energies at UB3LYP/BS2//UB3LYP/BS1
in kcal mol–1 and contain zero-point, entropic and thermal corrections. Also given are optimized geometries of 4,2TSHAT1 4,2TSOAT1 and 4,2TSOAT3 with bond lengths
in angstroms, angles in degrees and the imaginary frequency in cm–1
,
.
In particular, we calculate an energy difference between the
iron(IV)-hydroxo and the sum of the iron(IV)-oxo porphyrin cation
radical model and a hydrogen atom of 42.8 kcal mol‒1 for
[FeIV(OH)(TPFPP)]+ and 83.6 kcal mol‒1 for [FeIV(OH)(Por)]+.
This difference results from a change of 8.0 kcal mol‒1 in the
electron affinity of the iron(IV)-oxo species and 32.2 kcal mol‒1
drop in pKa value of the iron(IV)-hydroxo species. As a
consequence of this, the small model reacts with lesser
exothermicity and higher hydrogen atom abstraction barriers
with substrates with aliphatic groups than the large system.
The barriers for hydrogen atom abstraction for both the small
and large models (4,21 and 4,22) are relatively small. Previously, a
series of hydrogen atom abstraction barriers with a range of
aliphatic substrates showed early transition states to correspond
to lower reaction barriers.[25] Indeed, 4,2TSHA1 are low in energy
and only G‡ = 6.0/6.3 kcal mol–1 above the energy of reactants
in the doublet/quartet spin state. Note also that the imaginary
frequency in the transition state is relatively low (i262 cm‒1 for
4TSHA1 and i476 cm‒1 for 4TSHA1,TPFPP). Typical values of the
imaginary frequency for hydrogen atom abstraction barriers are
well over i1500 cm‒1, which usually leads to a major kinetic
isotope effect (KIE) for the replacement of the transferring
hydrogen atom by deuterium.[26] We cal
stable radical intermediates could be located and hence a
concerted hydroxylation process is predicted. In previous work
on P450 model complexes,[4b,6,24,25] most HAT pathways showed
a shallow intermediate that was separated with a small barrier to
products (typically less than 1 kcal mol–1 on the doublet spin
state surface and less than 5 kcal mol–1 on the high-spin
surface); however, no such local minimum could be located here
and all attempts converged to the product complexes instead.
Therefore, the absence of a radical intermediate may have to do
with the lack of meso-substituents in the model that affects the
electron affinity and hydrogen atom abstraction ability of the
complex as will be discussed in the thermochemical section later.
The group-spin densities for the doublet and quartet
intermediates give most radical character on the FeO group and
virtually no spin density is seen on the substrate. Consequently,
these intermediates correspond to a formal hydride transfer from
substrate to oxidant. The reaction is highly exothermic and 4,2IR1
are below reactants by G = 55.5 (quartet) and 47.7 (doublet)
kcal mol‒1. Clearly, the small model does not capture the
properties of the intermediate state well even though the
hydrogen atom abstraction barriers are similar in structure and
electronic configuration.
In contrast to the intermediates 4,2IR1, the hydrogen atom
abstraction transition states have significant radical character on
the substrate. To be specific, a spin density of 0.88 (‒0.73) is
found on the C7H8 unit in 4,2TSHA1. Therefore, during the
transition state a hydrogen atom is transferred, which is quickly
followed by another electron transfer before reaching the
intermediate state.
culate a value of KIE = 2.7, which matches the experimental rate
constant ratio perfectly. Nevertheless, typical KIE values of
hydrogen atom abstraction barriers are well above 10 for
analogous systems.[27]
For all the smaller models dealing with 4,2[FeIV(O)(Por+•)]+, after
passing the hydrogen atom abstraction transition states, the
system directly collapses to the alcohol product complex and no
6
This article is protected by copyright. All rights reserved.

10.1002/ejic.201800273
European Journal of Inorganic Chemistry
FULL PAPER
Sub = 0.34 (‒0.42)
Sub = 0.87 (‒0.76)
Sub = 0.90 (‒0.73)
Sub = 0.91 (‒0.78)
Fe = 1.49 (1.55)
O = 0.62 (0.43)
Fe = 1.66 (1.50)
O = 0.52 (0.58)
Por = ‒0.10 (‒0.30)
Fe = 1.22 (1.25)
O = 0.84 (0.80)
Fe = 1.59 (1.38)
O = 0.61 (0.61)
Por = ‒0.10 (‒0.26)
Por = 0.02 (‒0.21)

Por = 0.61 (‒0.63)

4TSOAT3 (2TSOAT3
)
4Re (2Re)
4TSHA1 (2TSHA1
)
4TSOAT1 (2TSOAT1
)
Figure 4. Group spin densities of reactant complexes and low-energy transition states for the reaction of [FeIV(O)(Por+•)]+ with CHT as calculated at
UB3LYP/BS2//UB3LYP/BS1.
The last step, then, refers to an OH‒ transfer from
[FeIII(OH)( TPFPP)] to C7H7+ to form the alcohol product complex. reactions have very small barriers of less than 1.2 kcal mol–1 for
hydroxylation or dehydrogenation products. Interestingly, all
Subsequently, we calculated hydrogen atom abstraction by 4,21
at position H3, H4 and H5 and the doublet and quartet barrier
heights are given in Figure 3 as well. As follows, all these
hydrogen atom abstraction barriers are well over 20 kcal mol–1
and will not be competitive with hydrogen atom abstraction from
H1 or H2. This is not surprising as hydrogen atom abstraction
from aromatic centers (or olefins) generally requires a lot of
energy and does not give a stable radical.[28]
We then studied C–O activation through an electrophilic attack
of the oxo group onto the carbon atom at position 3 and 5.
Similarly to the aliphatic hydroxylation reported in Figure 3, the
reaction is concerted with a C–O bond formation barrier that
directly leads to epoxide product complexes without the
formation of a radical intermediate. The two pathways are
competitive and only a small energy difference on the C‒O bond
formation barrier is seen between attacks on position 3 versus 5.
Therefore, the oxygen atom transfer on position 3 and 5 will be
competitive, although their barriers are higher in energy than
that for aliphatic hydrogen atom abstraction. Similarly to the
aliphatic hydroxylation barriers, also the oxygen atom transfer
barriers are early on the potential energy surface with long C‒O
and short Fe‒O distances (Figure 3).
concerted reaction mechanisms leading to products directly. As
such, the reaction of 4,2[FeIV(O)(TPFPP+•)]+ with CHD should
lead to a mixture of products. This is surprising as CHD as a
substrate with iron(IV)-oxo or manganese(V)-oxo complexes in
solution typically gives dehydrogenation to benzene as the sole
product.[29] Our gas-phase model, therefore, has properties
significantly deviating from solution-based oxidants and
consequently gives different reactivity patterns, which we
analyze in detail below. Most probably the lack of an axial ligand
in our [FeIV(O)(TPFPP+•)]+ model gives it significantly different
chemical properties and results in reactivity differences.
Discussion
In order to understand the mechanistic details of the reaction
pathways and find the origin of hydride versus hydrogen atom
abstraction processes, we did a detailed analysis of the
electronic and thermochemical properties of reactants and
intermediates. Let us start with a look at the spin density
distribution of reactant complexes and rate determining
transition states, see Figure 4. As can be seen from the spin
densities in the reactant complexes of [FeIV(O)(Por+•)]+ with CHT,
designated 2,4Re, there is considerable spin density found on the
Thereafter, we investigated the mechanistic landscapes for
oxygen atom transfer of [FeIV(O)(TPFPP+•)]+ with toluene and
1,3-cyclohexadiene (CHD) on both spin states. In particular, we
focused on aliphatic hydroxylation of toluene and CHD, and
hydroxylation/dehydrogenation of CHD and epoxidation of CHD.
Toluene undergoes the expected HAT with a free energy of
activation of 13.4 kcal mol‒1 on the doublet spin state and 16.3
kcal mol‒1 on the quartet states (Figure S9, Supporting
Information). In both cases a radical intermediate is formed
substrate moiety, i.e. Sub
=
0.34 (‒0.42) in 4Re (2Re),
respectively. The systems with TPFPP as equatorial ligand give
analogous spin density distributions, see Tables S11 and S12
(Supporting Information). Similar spin density distributions are
also seen for the reactant complexes containing cyclohexadiene
and toluene. Therefore, upon approach of the substrate to the
iron(IV)-oxo species, a considerable charge transfer happens
from the substrate to the porphyrin group, which thereby loses
radical character. This type of charge transfer in the reactant
complexes is not seen in similar iron(IV)-oxo complexes bearing
an axial ligand.[18,25]
•
representing [FeIV(OH)(TPFPP)]+-C7H7 ]. However, it is expected
its lifetime will be short as the reaction is completed with an OH•
rebound to form the final alcohol products with negligible barrier.
Finally, the reaction of 4,2[FeIV(O)(TPFPP+•)]+ with CHD was
investigated computationally leading to either epoxidation,
7
This article is protected by copyright. All rights reserved.

10.1002/ejic.201800273
European Journal of Inorganic Chemistry
FULL PAPER
Figure 5. Valence bond curve crossing diagrams for a stepwise hydroxylation (a) and concerted hydroxylation (b) pathways. Dots represent electrons and a line
in between two dots is a bonding orbital with two electrons.
In one previous study[20c,30] of a nonheme iron(IV)-tosylimido
complex with CHD also electron transfer in the reactant
complexes was observed due to the large electron affinity of the
iron(IV)-tosylimido group as compared to P450 CpdI, where this
electron transfer is not seen. The group spin densities of
TSHA,CHT and TSOAT,CHT are also given in Figure 4 and display
even more charge-transfer from substrate to porphyrin as
compared to the reactant complexes. In particular, in 4,2TSHA,CHT
a spin density of Sub = 0.90 (–0.73) is found for the quartet
(doublet) spin state and hence almost a full electron transfer has
taken place at this stage, so that the transition state corresponds
to hydrogen atom abstraction.
However, the subsequent intermediates IR1 for either the Por or
TPFPP ligand systems have no radical on the substrate group,
hence are cationic. Therefore, along the pathway from TSHA to
IR1 an extra electron transfer has occurred to give an overall
hydride transfer leading to the local minimum IR1. As such, even
though IR1 mimics a hydride transfer local minimum, in fact the
reaction proceeds via an initial hydrogen atom transfer in the
transition state followed by a fast electron transfer en route to
the intermediate IR1. Clearly, the hydride transfer is split into a
hydrogen atom transfer followed by an electron transfer and
these two processes are not simultaneous, namely the hydrogen
atom transfer happens first, i.e. in the TS, and the subsequent
electron transfer happens after the TS. As a result of this, the
subsequent OH rebound will refer to an OH– transfer to the
substrate cation. The early second electron transfer will lower
the energy of the iron-hydroxo intermediate in the mechanism.
As a matter of fact the energy of 4,2IR1 is lower than that of the
alcohol product complexes with a substantial rebound barrier in
+
the low-spin state. It is, therefore, likely that C7H7 will be
released from IR1 as is also detected in the mass spectrum. The
calculated mechanism and the obtained charge and spin
distributions clearly support the experimental observations and
explain the various product distributions.
To understand the obtained potential energy profiles and
reaction mechanisms of the reaction of [FeIV(O)(TPFPP+•)]+ with
CHT substrate, we set up a valence bond (VB) curve crossing
diagram to gain insight into the electronic changes during the
reaction. In the past, we used these valence bond diagrams to
understand the electronic features of reactants that determine
the transition states. Thus, we showed that in hydrogen atom
abstraction transition states by metal(IV)-oxo oxidants, the rate
determining barrier correlates with either the strength of the C‒H
bond that is broken or the O‒H bond that is formed.[18,25,31]
Furthermore, these VB diagrams explained the electronic origin
of regioselectivities and bifurcation pathways.[20b,21,22,24b,32] Note
that in the following all calculations represent the large model as
its thermochemistry was seen to be slightly different from the
small model (compare Figures 2 and 3).
8
This article is protected by copyright. All rights reserved.

10.1002/ejic.201800273
European Journal of Inorganic Chemistry
FULL PAPER
Figure 6. Thermochemical reaction scheme for individual electron, proton and hydride transfer from iron(IV)-oxo and iron(IV)-hydroxo complexes. Data represent
reaction enthalpies (in kcal mol–1) as calculated at UB3LYP/BS2//UB3LYP/BS1 with ZPE corrections included. Quartet spin data in parenthesis, doublet spin data
out of parenthesis.
The VB diagram explain the difference between stepwise and
concerted reaction mechanisms and gives the various electron
transfer processes and, in particular, highlights how hydrogen
atom transfer and hydride transfer differ (Figure 5). Previously, a
stepwise mechanism for substrate hydroxylation for aliphatic
hydroxylation by CpdI of P450 was reported with a VB
landscape similar to the one shown on the left-hand-side of
Figure 5.[25,31‒33] Thus, in VB theory the reactant configuration
has wave function r and connects to an excited state in the
product geometry with wave function r*. The product wave
function is P in the product geometry and connects to an
excited state in the reactant geometry with wave function P*.
The point where these two wave functions cross should lead to a
transition state for the reaction. However, another wave function
for the radical intermediate bifurcates these two curves and
gives a local minimum with wave function Rad. As it happens,
the first crossing point is between Rad and r is well lower in
energy than the one between r and P, hence the reaction
proceeds via a preferred stepwise mechanism via a radical
intermediate. Therefore, our analysis will be focused on the
electronic configuration of reactants and products and how the
electronic configuration of reactants needs to change to initiate
the chemical reaction.
substrate, a hydrogen atom abstraction transition state results in
breaking of the C‒H orbital in the substrate into atomic orbitals
2
(2pC and 1sH). Furthermore, the xz *xz1 pair of orbitals revert to
2
atomic orbitals 2pO 3dxz1, whereby one electron from the 2pO
orbital is promoted to the a1u orbital. Finally, the remaining 2pO
electron pairs up with the 1sH electron to form the O‒H orbital.
Consequently, the hydrogen atom abstraction barrier (ETS(H,Rad)
)
is determined by the strength of the C‒H bond that is broken
(BDECH), the strength of the O‒H bond that is formed (BDEOH),
the strength of the xz/*xz orbitals that are broken (E/*xz) and
the excitation energy from 2pO to the porphyrin group (Eexc), Eq
1.[24c,32]
ETS(H,Rad)  BDECH,CHT – BDEOH + E/*xz + Eexc
(1)
In the concerted mechanism (Figure 5b) these same bond
breaking and bond forming reactions occur. However, at the
same time, an electron is promoted from the 2pC orbital of the
CHT group into the metal-type orbitals, i.e. into the *z2 orbital in
the quartet spin state and into 3dxz in the doublet spin state.
Therefore, the barrier (ETS(H,Cat)) for the concerted mechanism
will be dependent on the energy to break the C‒H bond (BDECH),
the energy to form the O‒H bond (BDEOH), the energies to split
the xz/*xz pair of orbitals into atomic orbitals (E/*xz), the
electron transfer from 2pO,x to a1u (Eexc) and from 2pC to *z2/3dxz.
In our model the reaction starts from [FeIV(O)(TPFPP+•)]+, which
is an iron(IV)-oxo porphyrin cation radical species with orbital
2
2
1
1
1 [11]
occupation xz yz *xz *yz a1u
.
Occupation of the xz and
The latter excitation refers to the ionization energy of CHT (IESub)
*xz orbitals with three electrons means there is a two-center
three-electron bond located in the xz-plane along the Fe‒O axis.
In addition, there is another two-center three-electron bond in
the yz-plane due to occupation of yz and *yz with three
electrons (shown in red in Figure 5). In addition,
[FeIV(O)(TPFPP+•)]+ has a singly occupied orbital with a1u
symmetry on the porphyrin ligand. Upon activation of the
and the electron affinity of the iron(IV)-oxo complex (EAFeO), Eq
2.
ETS(H,Cat)  BDECH,CHT – BDEOH + E/*xz + Eexc + IESub + EAFeO
(2)
9
This article is protected by copyright. All rights reserved.

10.1002/ejic.201800273
European Journal of Inorganic Chemistry
FULL PAPER
Consequently, the VB analysis highlights the fundamental
differences between hydrogen atom and hydride transfer
processes. In particular, it shows that hydride transfer is only
hydrogen atom transfer with a subsequent electron transfer. As
such, hydride transfer processes should be seen as consecutive
hydrogen atom and electron transfer processes, where an initial
possible if the second electron transfer is an exothermic reaction. hydrogen atom abstraction is followed by a quick electron
Therefore, the hydrogen atom abstraction barrier is always lower
in energy than the hydride transfer barrier unless the energy
transfer as also reported previously for the reactivity of
analogous heme and nonheme iron(IV)-oxo oxidants. Our
thermochemical modelling shows that the original reactant
cannot react via electron transfer as it is endothermic; however,
the subsequent intermediate, namely the iron(IV)-hydroxo
species, has a sufficiently large electron affinity to abstract
electrons from the substrate. This mechanism may have
relevance to enzymatic and biomimetic reactivity work where
often hydride transfer processes are proposed and explains the
intrinsic properties of substrate and oxidant that determines the
reaction mechanism.
difference between the ionization energy of the substrate (IESub
)
and the electron affinity of the oxidant EAFeO is negative. To find
out if that is the case for substrates CHT, CHD and toluene, we
estimated values for all these individual contributions from
adiabatic electron and hydrogen atom transfer energies or the
molecular orbital differences in the reactant complexes.
Figure 6 gives the enthalpic differences for individual electron,
hydrogen atom and hydride transfer processes from CHT to
iron(IV)-oxo porphyrin cation radical. Thus, the reactions from
left to right represent electron transfer to form [FeIV(O)(TPFPP)]0
+•
and C7H8 and has a small endothermic driving force of HET
=
6.6 (6.2) kcal mol–1 in the doublet (quartet) spin states.
Therefore, a long-range electron transfer between oxidant and
substrate will not happen. Furthermore, the difference between
radical and cationic pathways displayed in Figure 5 implicated
that the radical pathway would be lower for an endothermic
electron transfer. As this is the case, the combination of VB and
thermochemical modelling predicts a rate-determining hydrogen
atom abstraction first.
Experimental Section
Experiment.
All chemicals and solvents were research grade products purchased
from commercial sources and used as received. For the synthesis of the
iron(IV)-oxo porphyrin cation radical complex, a reaction of (5,10,15,20-
tetrakis(pentafluorophenyl)porphinato) iron(III) chloride, [FeIII(TPFPP)]Cl
was applied with iodosylbenzene (C6H5IO), which was synthesized
according to a literature procedure[34] and stored at –20 °C. The
cycloheptatriene-7-[D1] was prepared by the reduction of 7-
acetoxynorbornadiene with LiAlD4 in tetrahydrofuran as described in the
literature.[35] Subsequently, it was purified by preparative GLC using a 3
m column filled with Chromosorb 80/100 W-AW coated with a base
deactivated polyethyleneglycol stationary phase, mounted on a Carlo
Erba FRACTOVAP Mod ATC/f series 410 gas chromatograph. The
identification, purity and deuterium content was obtained by GLC-MS
analyses on a Hewlett-Packard 5890 gas chromatograph coupled with a
model 5989B quadrupole mass spectrometer, by using a 50 m long, 0.2
mm i.d. fused silica capillary column, coated with cross-linked
methylsilicone film. The extent of D-incorporation in cycloheptatriene-7-
[D1] was found equal to 98.0 atom %D.
By contrast, the enthalpy of reaction for hydrogen atom
abstraction is calculated as HHAT = –22.2 (–22.5) kcal mol–1 in
the doublet (quartet) spin states. Consequently, approach of
substrate on [FeIV(O)(TPFPP+•)]+ will lead to a hydrogen atom
•
abstraction and the formation of [FeIV(OH)(TPFPP)]+ and C7H7
rather than electron transfer. However, the electron transfer
•
between [FeIV(OH)(TPFPP)]+ and C7H7 has an exothermic
driving force of HET = –19.3 (–24.8) kcal mol–1 for the doublet
(quartet) pathways. This means that although no electron
transfer will take place between reactants, as soon as the
hydrogen atom transfer has taken place it becomes energetically
feasible and will happen quickly.
Indeed, the DFT calculations reported above give a mechanism
whereby an initial hydrogen atom abstraction took place followed
by a fast electron transfer in an overall hydride transfer process.
If we now calculate the reaction enthalpies for Figure 6 but with
cyclohexadiene or toluene as a substrate, we find an analogous
result, whereby the long-range electron transfer is energetically
unfavorable over hydrogen atom transfer but by an even larger
amount. In the case of cyclohexadiene, the electron transfer
Instrumental.
All procedures and methods follow those from our previous studies on
these chemical systems.[11,12] Mass spectrometric studies were
performed on
a Bruker BioApex Fourier transform-ion cyclotron
resonance (FT-ICR) mass spectrometer, which is implemented with a
cylindrical infinity cell, a 4.7 T superconducting magnet and an Apollo I
electrospray ionization (ESI) source. Analyte solutions were infused into
the mass spectrometer at a continuous flow rate of 120 L h–1 by a
syringe pump through a 50 m i.d. fused-silica capillary. Subsequently,
ions were desolvated by applying an N2 counter current of drying gas
heated at 400 K, trapped and accumulated in a radiofrequency-only
hexapole ion guide for 0.8 seconds, and then pulsed into the ICR cell
(held at room temperature, 300 K). An ion ejection procedure was used
to select the ions of interest, namely [FeIV(O)(TPFPP+•)]+ with m/z 1044,
and their reactivity and fragmentation patterns were studied through ion-
molecule reactions by inserting neutral collision gases to the ICR cell at
stationary pressures (in the range 1.0 – 15 × 10–8 mbar) by a needle
valve. The product ion abundances were monitored as a function of time
and analyzed. The pressure readings, obtained from a cold-cathode
sensor (IKR Pfeiffer Balzers S.p.A., Milan, Italy), were calibrated against
the rate constant of proton transfer from methane cation radical to
methane, Eq 3, for which the rate constant has been accurately
determined at k = 1.1  10–9 cm3 s–1 and weighted by using individual
response factors.[36]
•
from C6H7 to [FeIV(OH)(TPFPP)]+ is still exothermic by 6.0 kcal
mol–1, while it is endothermic for toluene. Therefore,
cyclohexadiene is expected to also react via an overall hydride
transfer with an initial hydrogen atom abstraction followed by
electron transfer, whereas no such electron transfer will be
expected for toluene.
Conclusions
In this work a combined mass spectrometry and density
functional theory study is presented on the reactivity of
[FeIV(O)(TPFPP+•)]+ with CHT and selected substrates. We find a
reaction mechanism whereby electron transfer is determined by
the intermediate structure. Thus, although a formal hydride
transfer is thermochemically favorable, actually the initial step is
10
This article is protected by copyright. All rights reserved.

10.1002/ejic.201800273
European Journal of Inorganic Chemistry
FULL PAPER
+
CH4 • + CH4 → CH5+ + CH3
(3)
•
FGCR thanks the Conacyt Mexico for a studentship. The COST
Action (CM1305 ECOSTBio, Explicit Control Over Spin-States in
Technology and Biochemistry) is acknowledged for support.
Rate constants for the reaction processes were derived from the ion
abundance of the reactant ion, [FeIV(O)(TPFPP+•)]+, monitored as a
function of time at each selected pressure. The slope of these semi-
logarithmic plots gave us the pseudo first-order rate constants for the
disappearance of [FeIV(O)(TPFPP+•)]+, then divided by the substrate
concentration to obtain second-order rate constants (kexp) at 300 K. All
measurements were done at least in triplicate and averaged. Whereas
the reproducibility of kexp values is within 10%, the estimated error in the
absolute rate constants is estimated to be 30%. The ratio of the second-
order rate constant and the collision rate constant (kADO), as described in
Eq 4, enabled us to estimate the reaction efficiencies (). Values for kADO
were calculated using the parametrized trajectory theory.[37]
Keywords: High-valent iron(IV)-oxo • Porphyrin • density
functional theory • FT-ICR MS • reaction mechanisms
[1]
a) M. Sono, M. P. Roach, E. D. Coulter, J. H. Dawson, Chem. Rev.
1996, 96, 2841‒2888; b) J. T. Groves, Proc. Natl. Acad. Sci. USA, 2003,
100, 3569–3574; c) Ed.: P. R. Ortiz de Montellano, Cytochrome P450:
Structure, Mechanism and Biochemistry, 3rd ed., Kluwer Academic /
Plenum Publishers, New York, 2005; d) P. R. Ortiz de Montellano,
Chem. Rev. 2010, 110, 932–948; e) Eds.: K. M. Kadish, K. M. Smith, R.
Guilard, Handbook of Porphyrin Science, World Scientific Publishing
Co., New Jersey, 2010.
 = kexp/kADO  100%
(4)
[2]
[3]
a) D. R. Nelson, Human Genomics 2009, 4, 59–65; b) F. P. Guengerich,
Chem. Res. Toxicol. 2001, 14, 611–650; c) T. L. Poulos, Chem. Rev.
2014, 114, 3919–3962.
Sample preparation.
a) B. Meunier, S. P. de Visser, S. Shaik, Chem. Rev. 2004, 104, 3947–
3980; b) I. G. Denisov, T. M. Makris, S. G. Sligar, I. Schlichting, Chem.
Rev. 2005, 105, 2253‒2278; c) J. Rittle, M. T. Green, Science 2010,
330, 933–937.
The [FeIV(O)(TPFPP+•)]+ ion was synthesized by adding iodosylbenzene
(0.5 mM) to 10 µM of [FeIII(TPFPP)]Cl in a methanol/dichloromethane
(1:1) mixture and was stable for about 1 h, if kept cooled at –40°C. The
high-resolution ESI FT-ICR mass analysis of the reaction mixture gave a
prominent peak centered at m/z 1044 with isotopic pattern conforming to
an iron(IV)-oxo complex, [FeIV(O)(TPFPP+•)]+, as well as a signal for the
reduced form [FeIII(TPFPP)]+ at m/z 1028. As already described in
previous contributions,[10a,11,12] the synthetic procedure leads to the
formation of an additional fraction of isomeric species, most likely
corresponding to a four-coordinate iron(III) complex oxidized on the
porphyrin ring and unable to perform any oxidation reactions. This portion,
quantified by its complete trapping by NO gives the [FeIII(TPFPP-
O)(NO)]+ adduct, similarly to the reduced form, [FeIII(TPFPP)]+, and was
discarded from the kinetic analysis.
[4]
[5]
a) W. Nam, Acc. Chem. Res. 2007, 40, 522–531; b) Y. Watanabe, H.
Nakajima, T. Ueno, Acc. Chem. Res. 2007, 40, 554–562.
a) W. Nam, Y.-M. Lee, S. Fukuzumi, Acc. Chem. Res. 2014, 47, 1146–
1154; b) M. Asaka, H. Fujii, J. Am. Chem. Soc. 2016, 138, 8048–8051;
c) L. Ji, A. Franke, M. Brindell, M. Oszajca, A. Zahl, R. van Eldik, Chem.
Eur. J. 2014, 20, 14437–14450.
[6]
[7]
a) F. Ogliaro, N. Harris, S. Cohen, M. Filatov, S. P. de Visser, S. Shaik,
J. Am. Chem. Soc. 2000, 122, 8977–8989; b) S. Shaik, D. Kumar, S. P.
de Visser, A. Altun, W. Thiel, Chem. Rev. 2005, 105, 2279–2328; c) D.
Li, Y. Wang, K. Han, Coord. Chem. Rev. 2012, 256, 1137–1150; d) M.
R. A. Blomberg, T. Borowski, F. Himo, R.-Z. Liao, P. E. M. Siegbahn,
Chem. Rev. 2014, 114, 3601–3658.
a) Y. J. Jeong, Y. Kang, A. R. Han, Y. M. Lee, H. Kotani, S. Fukuzumi,
W. Nam, Angew. Chem. Int. Ed. 2008, 47, 7321–7324; b) S. Fukuzumi,
H. Kotani, Y.-M. Lee, W. Nam, J. Am. Chem. Soc. 2008, 130, 15134–
15142.
Density functional theory modelling.
[8]
[9]
L. Tahsini, M. Bagherzadeh, W. Nam, S. P. de Visser, Inorg. Chem.
2009, 48, 6661–6666.
Calculations were done using density functional theory methods as
implemented in Gaussian-09,[38] and follow methods and procedures as
reported and tested previously on analogous complexes and
reactions.[20,39] In general, the unrestricted B3LYP hybrid density
functional theory is used for all geometry optimizations and
frequencies.[40] All structures were optimized in the gas phase with an
LACVP basis set on iron (with core potential) and 6-31G on the rest of
the atoms (H, C, N, O): basis set BS1.[41] Single points using a triple-
quality basis set on iron (with core potential), i.e. LACV3P+, and 6-
311+G* on the rest of the atoms were done to correct the energies: basis
set BS2. All stationary points had real frequencies and transition states
were characterized with a single imaginary frequency for the correct
mode. Energies reported here are UB3LYP/BS2//UB3LYP/BS1 with zero-
point energy included. Previous studies showed little differences in
optimized geometries and reaction kinetics between geometry
optimizations at UB3LYP/BS2 and UB3LYP/BS1 level of theory,[11,12]
hence the latter was used here.
M. E. Crestoni, S. Fornarini, F. Lanucara, Chem. Eur. J. 2009, 15,
7863–7866.
[10] a) R. Chiavarino, R. Cipollini, M. E. Crestoni, S. Fornarini, F. Lanucara,
A. Lapi, J. Am. Chem. Soc. 2008, 130, 3208–3217; b) M. E. Crestoni, S.
Fornarini, Inorg. Chem. 2007, 46, 9018–9020; c) M. E. Crestoni, S.
Fornarini, Inorg. Chem. 2005, 44, 5379–5387.
[11] M. A. Sainna, S. Kumar, D. Kumar, S. Fornarini, M. E. Crestoni, S. P.
de Visser, Chem. Sci. 2015, 6, 1516–1529.
[12] F. G. Cantú Reinhard, M. A. Sainna, P. Upadhyay, G. A. Balan, D.
Kumar, S. Fornarini, M. E. Crestoni, S. P. de Visser, Chem. Eur. J.
2016, 22, 18608‒18619.
[13] E. P. Hunter, S. G. Lias, NIST Chemistry Webbook, NIST Standard
Reference Database, Number 69; Eds.: P. J. Linstrom, W. G. Mallard,
National Institute of Standards and Technology: Gaithersburg MD,
20899, (retrieved June 7, 2017), http://webbook.nist.gov.
[14] Ed.: D. R. Lide, CRC Handbook of Chemistry and Physics, 90th Ed.,
Internet Version 2010.
Kinetic isotope effects are calculated from the free energy of activation
difference between hydrogen atom abstraction transition state and
isolated reactants (G‡HA) between the substrate with all hydrogen atoms
and the one with one or more hydrogen atoms replaced by deuterium,
according to Eq, 5, with R being the gas constant and T the estimated
temperature (300K).[42]
[15] a) T. Shono, T. Nozoe, H. Maekawa, Y. Yamaguchi, S. Kanetaka, H.
Masuda, T. Okada, S. Kashimura, Tetrahedron 1991, 47, 593–603; b)
D. Jacobi, W. Abraham, U. Pischel, L. Gubert, R. Stösser, W. Schnabel,
J. Chem. Soc. Perkin Trans. 2 1999, 1695–1702; c) Y. Xu, C. R.
Johnson, Tetrahedron Lett. 1997, 38, 1117–1120.
[16] O. Krechkivska, C. Wilcox, G. D. O’Connor, K. Nauta, S. H. Kable, T. W.
Schmidt, J. Phys. Chem. A 2014, 118, 10252−10258.
KIE = exp((G‡HA,D – G‡HA,H)/RT)
(5)
[17] a) D. Kumar, L. Tahsini, S. P. de Visser, H. Y. Kang, S. J. Kim, W. Nam,
J. Phys. Chem. A 2009, 113, 11713–11722; b) R. Latifi, M. A. Sainna, E.
V. Rybak-Akimova, S.. P. de Visser, Chem. Eur. J. 2013, 19, 4058–
4068.
Acknowledgements
11
This article is protected by copyright. All rights reserved.

10.1002/ejic.201800273
European Journal of Inorganic Chemistry
FULL PAPER
[18] a) S. P. de Visser, D. Kumar, S. Cohen, R. Shacham, S. Shaik, J. Am.
Chem. Soc. 2004, 126, 8362–8363; b) R. Latifi, J. S. Valentine, W.
Nam, S. P. de Visser, Chem. Commun. 2012, 48, 3491–3493.
[19] a) B. Chiavarino, M. E. Crestoni, S. Fornarini, F. Lanucara, Eur. J.
Mass Spectrom. 2010, 7, 407–414. b) F. Lanucara, M. E. Crestoni,
Chem. Eur. J. 2011, 17, 12092–12100.
[29] a) W. J. Song, Y. J. Sun, S. K. Choi, W. Nam, Chem. Eur. J. 2006, 12,
130–137; b) K. A. Prokop, S. P. de Visser, D. P. Goldberg, Angew.
Chem. Int. Ed. 2010, 49, 5091–5095.
[30] S. Kumar, A. S. Faponle, P. Barman, A. K. Vardhaman, C. V. Sastri, D.
Kumar, S. P. de Visser, J. Am. Chem. Soc. 2014, 136, 17102–17115.
[31] S. P. de Visser, J. Am. Chem. Soc. 2010, 132, 1087–1097
[32] X.-X. Li, V. Postils, W. Sun, A. S. Faponle, M. Solà, Y. Wang, W. Nam,
S. P. de Visser, Chem. Eur. J. 2017, 23, 6406–6418.
[20] a) F. G. Cantú Reinhard, A. S. Faponle, S. P. de Visser, J. Phys. Chem.
A 2016, 120, 9805‒9814; b) T. Yang, M. G. Quesne, H. M. Neu, F. G.
Cantú Reinhard, D. P. Goldberg, S. P. de Visser, J. Am. Chem. Soc.
2016, 138, 12375‒12386; c) A. K. Vardhaman, P. Barman, S. Kumar, C.
V. Sastri, D. Kumar, S. P. de Visser, Angew. Chem. Int. Ed. 2013, 52,
12288–12292.
[33] A. Hernández-Ortega, M. G. Quesne, S. Bui, D. J. Heyes, R. A. Steiner,
N. S. Scrutton, S. P. de Visser, J. Am. Chem. Soc. 2015, 137, 7474–
7487.
[34] H. Saltzman, J. G. Sharefkin, Organic Syntheses, Wiley, New York,
1973, Collect. Vol. 5, p 658.
[21] a) A. Timmins, M. Saint-André, S. P. de Visser, J. Am. Chem. Soc.
2017, 139, 9855–9866; b) A. Timmins, S. P. de Visser, Frontiers Chem.
2017, 5, Article 94, pp 1–13.
[35] B. Franzus, E. I. Snyder, J. Am. Chem. Soc. 1963, 85, 3902‒3903.
[36] a) M. Meot-Ner, In Gas Phase Ion Chemistry, Ed.: M. T. Bowers,
Academic Press, New York, 1979; Vol. 1; b) J. E. Bartmess, R. M.
Georgiadis, Vacuum 1983, 33, 149–154.
[22] A. S. Faponle, M. G. Quesne, S. P. de Visser, Chem. Eur. J. 2016, 22,
5478–5483.
[23] a) M. T. Green, J. Am. Chem. Soc. 1999, 121, 7939–7940; b) S. P. de
Visser, S. Shaik, P. K. Sharma, D. Kumar, W. Thiel, J. Am. Chem. Soc.
2003, 125, 15779–15788; c) S. P. de Visser, L. S. Tan, J. Am. Chem.
Soc. 2008, 130, 12961–12974; d) F. G. Cantú Reinhard, S. P. de
Visser, Inorganics 2017, 5, Article 77, pp 1–18.
[37] T. Su, W. J. Chesnavich, J. Chem. Phys. 1982, 76, 5183–5185.
[38] Gaussian-09, Revision C.01, 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, X. Li, H. P.
Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, 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, J. Normand, K.
Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M.
Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V.
Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O.
Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, 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, D. J. Fox, Gaussian, Inc., Wallingford CT, 2010.
[39] F. G. Cantú Reinhard, P. Barman, G. Mukherjee, J. Kumar, D. Kumar,
D. Kumar, C. V. Sastri, S. P. de Visser, J. Am. Chem. Soc. 2017, 139,
18328–18338.
[24] a) Y. Wang, D. Kumar, C. L. Yang, K. Han, S. Shaik, J. Phys. Chem. B
2007, 111, 7700–7710; b) L. Ji, A. S. Faponle, M. G. Quesne, M. A.
Sainna, J. Zhang, A. Franke, D. Kumar, R. van Eldik, W. Liu, S. P. de
Visser, Chem. Eur. J. 2015, 21, 9083–9092; c) F. G. Cantú Reinhard, S.
P. de Visser, Chem. Eur. J. 2017, 23, 2935‒2944.
[25] a) S. Shaik, D. Kumar, S. P. de Visser, J. Am. Chem. Soc. 2008, 130,
10128–10140; b) R. Latifi, M. Bagherzadeh, S. P. de Visser, Chem. Eur.
J. 2009, 15, 6651–6662.
[26] a) M. G. Quesne, D. Senthilnathan, D. Singh, D. Kumar, P. Maldivi, A.
B. Sorokin, S. P. de Visser, ACS Catal. 2016, 6, 2230–2243; b) P.
Barman, P. Upadhyay, A. S. Faponle, J. Kumar, S. S. Nag, D. Kumar,
C. V. Sastri, S. P. de Visser, Angew. Chem. Int. Ed. 2016, 55, 11091‒
11095.
[27] a) M. M. Abu-Omar, A. Loaiza, N. Hontzeas, Chem. Rev. 2005, 105,
2227–2252; b) M. Atanasov, P. Comba, S. Hausberg, B. Martin, Coord.
Chem. Rev. 2009, 253, 2306–2314; c) P. C. A. Bruijnincx, G. van Koten,
R. J. M. Klein Gebbink, Chem. Soc. Rev. 2008, 37, 2716–2744; d) M.
Costas, M. P. Mehn, M. P. Jensen, L. Que, Jr., Chem. Rev. 2004, 104,
939–986.
[40] a) A. D. Becke, J. Chem. Phys. 1993, 98, 5648–5652; b) C. Lee, W.
Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785–789.
[41] P. J. Hay, W. R. Wadt, J. Chem. Phys. 1985, 82, 270–283.
[42] a) D. Kumar, S. P. de Visser, P. K. Sharma, S. Cohen, S. Shaik, J. Am.
Chem. Soc. 2004, 126, 1907–1920; b) S. P. de Visser, Chem. Eur. J.
2006, 12, 8168–8177.
[28] S. P. de Visser, S. Shaik, J. Am. Chem. Soc. 2003, 125, 7413–7424.
12
This article is protected by copyright. All rights reserved.

10.1002/ejic.201800273
European Journal of Inorganic Chemistry
FULL PAPER
Entry for the Table of Contents
FULL PAPER
Fabián G. Cantú Reinhard, Simonetta
Fornarini,* Maria Elisa Crestoni,* and
Sam P. de Visser*
Page No. – Page No.
Hydrogen atom versus hydride
transfer in cytochrome P450
oxidations: A combined mass
spectrometry and computational
study
A combined mass spectrometry and computational study highlights the substrate
and oxidant specificities that determine hydride transfer and hydrogen atom transfer
mechanisms to cytochrome P450 Compound I models. The bifurcation pathways
are further elucidated through thermochemical and valence bond modelling.
13
This article is protected by copyright. All rights reserved.