C-H bond cleavage by bioinspired nonheme oxoiron(IV) complexes, including hydroxylation of n-butane

Article abstract of DOI:10.1021/ic502786y

The development of efficient and selective hydrocarbon oxidation processes with low environmental impact remains a major challenge of the 21st century because of the strong and apolar nature of the C-H bond. Naturally occurring iron-containing metalloenzymes can, however, selectively functionalize strong C-H bonds on substrates under mild and environmentally benign conditions. The key oxidant in a number of these transformations is postulated to possess an S = 2 Fe^IV - O unit in a nonheme ligand environment. This oxidant has been trapped and spectroscopically characterized and its reactivity toward C-H bonds demonstrated for several nonheme iron enzyme classes. In order to obtain insight into the structure-activity relationships of these reactive intermediates, over 60 synthetic nonheme Fe^IV(O) complexes have been prepared in various laboratories and their reactivities investigated. This Forum Article summarizes the current status of efforts in the characterization of the C-H bond cleavage reactivity of synthetic Fe^IV(O) complexes and provides a snapshot of the current understanding of factors that control this reactivity, such as the properties of the supporting ligands and the spin state of the iron center. In addition, new results on the oxidation of strong C-H bonds such as those of cyclohexane and n-butane by a putative S = 2 synthetic Fe^IV(O) species that is generated in situ using dioxygen at ambient conditions are presented.

Full text of DOI:10.1021/ic502786y

Forum Article
pubs.acs.org/IC
C−H Bond Cleavage by Bioinspired Nonheme Oxoiron(IV)
Complexes, Including Hydroxylation of n‑Butane
Scott T. Kleespies, Williamson N. Oloo, Anusree Mukherjee, and Lawrence Que, Jr.*
Department of Chemistry and Center for Metals in Biocatalysis, University of Minnesota, Minneapolis, Minnesota 55455, United
States
S
* Supporting Information
ABSTRACT: The development of efficient and selective hydrocarbon
oxidation processes with low environmental impact remains a major
challenge of the 21st century because of the strong and apolar nature of
the C−H bond. Naturally occurring iron-containing metalloenzymes can,
however, selectively functionalize strong C−H bonds on substrates under
mild and environmentally benign conditions. The key oxidant in a number
of these transformations is postulated to possess an S = 2 FeIVO unit in
a nonheme ligand environment. This oxidant has been trapped and
spectroscopically characterized and its reactivity toward C−H bonds demonstrated for several nonheme iron enzyme classes. In
order to obtain insight into the structure−activity relationships of these reactive intermediates, over 60 synthetic nonheme
FeIV(O) complexes have been prepared in various laboratories and their reactivities investigated. This Forum Article summarizes
the current status of efforts in the characterization of the C−H bond cleavage reactivity of synthetic FeIV(O) complexes and
provides a snapshot of the current understanding of factors that control this reactivity, such as the properties of the supporting
ligands and the spin state of the iron center. In addition, new results on the oxidation of strong C−H bonds such as those of
cyclohexane and n-butane by a putative S = 2 synthetic FeIV(O) species that is generated in situ using dioxygen at ambient
conditions are presented.
modeling, and computational investigations. Such studies
have produced candidates for the active oxidants in several
classes of important iron oxygenase enzymes, some of which
will be discussed in this Forum Article.
INTRODUCTION
■
The development of efficient and selective hydrocarbon
oxidation processes with low environmental impact remains a
major challenge of the 21st century.1 This is mainly due to a
number of factors, which include (i) a large kinetic barrier
associated with the C−H bond cleavage step that precedes the
functionalization step as a result of the strong (>96 kcal mol−1)
and apolar nature of the C−H bonds, (ii) the difficulty in
preventing further oxidation of the desired product because its
C−H bonds are more activated than those of the hydrocarbon
substrate, and (iii) the problem of controlling site selectivity
because most organic molecules contain different C−H bonds
of comparable strengths. Therefore, not only do new
methodologies to cleave the strong C−H bonds of hydro-
carbon substrates need to be identified, but these trans-
formations must also occur selectively and under mild, atom-
economical, and environmentally friendly conditions.1−3
Gaining an understanding of the structure−function relation-
ships in the metalloenzymes not only will be of great interest in
its own right but also may aid in the development of selective
and efficient synthetic transformations that operate via the
biological mechanism.7 Such “bioinspired” or “biomimetic”
catalysts3,8 may also possess added advantages over the
biological systems, such as an expanded substrate scope,
increased scale of production, and tunable selectivity and/or
specificity. Indeed, recent advances in the trapping and
characterization of high-valent oxoiron(IV) complexes have
shed light on the fundamental reaction steps and the nature of
reactive intermediates in C−H bond cleavage reactions
mediated by various metalloenzymes. In this Forum Article,
some highlights will be discussed.
Nature has used metalloenzymes for the selective function-
alization of strong C−H bonds on substrates such as alkanes,
fatty acids, and steroids in atom-economical transformations
that operate under mild and environmentally benign con-
ditions.4,5 These metal-containing enzymes reductively activate
molecular oxygen to generate a high-valent oxidant that attacks
substrates with excellent regio-, chemo-, and stereoselectivity.6
In the past 25 years, significant mechanistic insights into how
metalloenzymes work have been obtained via the character-
ization of intermediates by X-ray crystallography and various
spectroscopic methods, detailed kinetic studies, synthetic
C−H Bond Oxygenation by Nonheme Iron Metal-
loenzymes and Model Complexes. Various iron-containing
metalloenzymes reductively activate dioxygen to oxidize strong
aliphatic C−H bonds. Cytochrome P450 does so with good
regio- and stereoselectivity,4 and much is known about its
Special Issue: Small Molecule Activation: From Biological Principles
to Energy Applications
Received: November 24, 2014
© XXXX American Chemical Society
A
DOI: 10.1021/ic502786y
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Forum Article
Scheme 1. Reactive Iron Enzyme Intermediates That Have Been Spectroscopically Characterized
catalytic mechanism.9−11 This enzyme family binds dioxygen at
an active site consisting of an iron(II) porphyrin cofactor that is
attached to the protein backbone through an axial cysteine.12
The bound dioxygen is reduced in one-electron steps to
superoxo and peroxo forms and then converted with the help of
protons to a (Porph•+)FeIV(O) intermediate called Cpd I
(Scheme 1), which has been trapped, characterized spectro-
scopically, and shown to be responsible for substrate
oxidation.13
Nonheme diiron enzymes such as soluble methane
monooxygenase (MMO) and fatty acid desaturases, on the
other hand, employ an active site where two iron centers work
hand in hand to reductively activate dioxygen and carry out a
variety of C−H bond oxidation reactions.5 Soluble MMO,
which is the most studied nonheme diiron oxygenase to date,
activates dioxygen at a carboxylate-rich diiron center.14−16
Following the heme paradigm (Scheme 1), binding of dioxygen
onto the diiron(II) center affords a (μ-1,2-peroxo)diiron(III)
intermediate called P, which has been characterized and shown
to epoxidize electron-rich olefins.17 However, substrate C−H
bond cleavage requires a subsequent intermediate called Q,
which arises from O−O bond cleavage of P. Intermediate Q
has been characterized spectroscopically to have a [FeIV2(μ-
O)2] diamond core18,19 and demonstrated to be kinetically
competent at hydroxylating methane.20
Oxoiron(IV) species have also been found to be key
oxidizing intermediates in the mechanisms of mononuclear
nonheme iron enzymes that catalyze a variety of oxidative
transformations, including hydroxylation, halogenation, desatu-
ration, and heterocylic ring formation.5,21 Dioxygen activation
in these enzymes takes place at an iron center that is bound to a
2-His-1-carboxylate facial triad motif.22 The two electrons
required for dioxygen activation are supplied either by organic
cofactors like α-ketoglutarate (α-KG),21 tetrahydrobiopterin,23
and ascorbate24,25 or by the substrates themselves (e.g.,
isopenicillin N synthase).26−28 The reductive activation of the
bound dioxygen leads to the formation of an S = 2 oxoiron(IV)
species,29 which has been trapped and characterized by various
spectroscopic methods in several enzyme classes including
taurine:α-KG dioxygenase (TauD),30−32 prolyl-4-hydroxylase
(P4H),33 the halogenases CytC334,35 and SyrB2,36 and pterin-
dependent phenylalanine (PheH)37 and tyrosine hydroxylase
(TyrH).38 The kinetic competence of the oxoiron(IV)
intermediates toward C−H bond cleavage has also been
demonstrated via large substrate deuterium KIEs.31 In the cases
of TauD, P4H, CytC3, and SyrB2, the oxoiron(IV)
intermediates are generated upon oxidative decarboxylation of
the α-KG cofactor, as shown in Scheme 1.29
Given the identification of oxoiron(IV) species as key
intermediates in the catalytic cycles of nonheme iron
enzymes,29 synthetic FeIVO complexes have been prepared
and their reactivities toward C−H bond cleavage investigated.
These efforts have led to the characterization of over 60
synthetic nonheme oxoiron(IV) complexes over the past
decade.39−42 Besides the structural and spectroscopic character-
ization of these complexes, a wealth of reactivity data involving
hydrogen-atom abstraction has rapidly accumulated. This
Forum Article summarizes the current status of efforts in the
characterization of the hydrogen-atom-transfer (HAT) reac-
tivity of oxoiron(IV) complexes and what factors may control
this reactivity. In addition, new results are presented on a model
system that activates dioxygen under ambient conditions to
generate a putative S = 2 FeIV(O) species that can oxidize
strong C−H bonds such as those of cyclohexane (bond
dissociation energy BDE = 99 kcal mol−1) and light alkanes
such as n-butane (BDE = 98 and 101 kcal mol−1).43
Mechanistic Considerations. Following precedents from
studies of high-valent heme complexes,11,13,44,45 the mechanism
of C−H bond oxidation by nonheme FeIV(O) complexes can
be construed to involve at least two steps.46,47 The first step
involves HAT from the target C−H bond to (L)FeIV(O) to
afford an alkyl radical and FeIIIOH (Scheme 2). That HAT
contributes to the rate-determining step is usually deduced by
measuring the kinetic isotope effect (KIE) between protio and
deutero substrates and/or demonstrating a linear correlation of
Scheme 2. Proposed Mechanism of C−H Bond Oxidation by
a
FeIV(O) Complexes
a
Step a represents HAT, which may be followed by an intramolecular
oxygen rebound step b or intermolecular trapping of the alkyl radical
with traps such as dioxygen (step c).
B
DOI: 10.1021/ic502786y
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Forum Article
the reaction rate constants on substrate C−H BDEs, based
upon the Evans−Polanyi relationship of log(kH) = αΔHo +
constant. ΔHo is directly related to the strength of the C−H
bond, while the proportionality constant α is a measure of how
far the transition state has progressed along the reaction
coordinate.48,49
The higher reactivity of the S = 2 state has been exploited by
Shaik in the two-state-reactivity (TSR) model that was put
forward to rationalize the reactivity differences among various S
= 1 FeIV(O) complexes.53,56 This model postulates that the net
activation barrier for C−H bond cleavage by intermediate-spin
oxoiron(IV) complexes represents a weighted blend of the
barrier on the ground triplet and excited quintet surfaces. Given
that the transition state on the quintet surface lies lower than
that on the triplet ground state,51 decreasing the triplet−quintet
gap (ΔETQ) increases the accessibility of the quintet state,
which, in turn, lowers the net barrier for C−H cleavage (Figure
2). Thus, ligand-field effects that decrease ΔETQ are predicted
to increase the rate of C−H bond cleavage by S = 1 FeIV(O)
complexes.
Insights into HAT from the target C−H bond to an FeIV(O)
moiety have been obtained from the application of density
functional theory (DFT).50−55 In a tetragonal FeIV(O)
geometry, the key frontier molecular orbitals (FMOs) likely
to be involved in the HAT step are the singly occupied π* dxz/yz
2
and the unoccupied σ* dz orbitals (Figure 1). The transfer of a
Figure 2. Representation of the TSR scenario during the hydrogen-
atom-abstraction reactions of an FeIV(O) complex in a tetragonal
geometry. ΔETQ is the energy gap between the triplet and quintet
states, 3,5TSH are the respective transition states. ΔG⧧ is the free-
Figure 1. (a) C−H abstraction by S = 1 FeIVO complexes via the π-
attack pathway. C−H abstraction by S = 2 FeIVO complexes (b) via
the π-attack pathway and (c) via the σ-attack pathway. Adapted with
permission from ref 51. Copyright 2011 Nature Publishing Group.
3
5
energy barrier from the ground state R to TSH, and 3,5IH are the
corresponding hydrogen-atom-abstracted intermediate complexes.
Adapted with permission from ref 57. Copyright 2015 Amercian
Chemical Society.
hydrogen atom from the σ(C−H) orbital to the FeIV(O)
species results in protonation of the oxo atom and the
introduction of an electron into one of the singly occupied π*
dxz/yz orbitals of FeO. Because of the shape of the π* FMO,
this “π-attack” pathway requires a nearly perpendicular
approach of the substrate C−H bond relative to the FeO
bond with a Fe−O−HC angle of ∼120° for optimal overlap
with the oxo px/y orbital (Figure 1a,b).50,55 Such an approach
could result in substantial steric interactions between the
substrate and equatorial ligands that may raise the reaction
barrier for the π-attack pathway.
Table 1 compares HAT reactivity data for selected
oxoiron(IV) complexes characterized to date. A quick perusal
shows that the oxidation rates for 9,10-dihydroanthracene
(DHA) reported at −30 or −40 °C can differ by as much as 6
orders of magnitude (Figure 3a). Although the number of
examples is smaller, cyclohexane oxidation rates also span a
comparable range when the 65 °C temperature difference of
the reactions is taken into account (Figure 3b). It is clear from
Table 1 and Figure 3 that the HAT reactivity of nonheme
FeIV(O) complexes can be markedly affected by the properties
of the supporting ligands such as topology, denticity, and
sterics, as well as the nature of the variable cis or trans ligand
and the spin state of the iron center. These factors will be
discussed in the following sections.
S = 1 Oxoiron(IV) Complexes. Our discussion starts with
oxoiron(IV) complexes supported by an equatorial array of four
nitrogen donors, which leaves a site available trans to the oxo
ligand for coordination by a monodentate ligand such as the
solvent or added anions (Scheme 3). The prototypical example
for this group is the S = 1 [(TMC)FeIV(O)(NCMe)]2+
complex 1, which was prepared by the oxidation of its iron(II)
precursor with iodosylbenzene (PhIO) in CH3CN at −40 °C.59
Complex 1 was the first oxoiron(IV) complex to be isolated
An additional reaction pathway available only to S = 2 FeIV
2
O complexes involves the unoccupied FeO σ*dz orbital,
which requires a vertical approach of the C−H σ-bond collinear
with the FeO bond for an Fe−O−HC angle of 180° for
optimal orbital overlap (Figure 1c).50,55 Such a vertical
substrate attack would minimize steric interactions with the
equatorial ligands during the C−H abstraction process.
2
Furthermore, the transfer of an electron into the empty dz
orbital results in a large increase in the number of exchange
interactions, which lowers the quintet barrier relative to its
triplet counterpart.51 These two factors have been proposed to
make a high-spin S = 2 complex more reactive than the
corresponding intermediate-spin S = 1 species.51,52,55
C
DOI: 10.1021/ic502786y
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Forum Article
Table 1. C−H Bond Cleavage Reactivity of Various Oxoiron(IV) Complexes
(L)FeIV(O)
where L =
λmax/nm
δ
c-C6H12
(99*)
temp
entry
(ε/M−1 cm−1
)
(ΔEQ/mm s−1
)
DHA (77*)
CHD (78*)
PhEt (87*)
9.6 × 10−5
(°C)
ref
58
1
TMC
824 (400)
0.17 (1.24)
15
0
0.14
2.5 × 10−3
0.12
6.4 × 10−4
59, 60
61
−40
15
2
3
4
TBC
885 (360)
735 (240)
695 (400)
0.22 (0.97)
0.12 (1.98)
−0.04 (0.93)
0.015
58
13-TMC
N4Py
5.7
18
5.4
−40
25
61
4.0 × 10−3
1.3 × 10−3
5.5 × 10−5
3.9 × 10−4
4.0 × 10−4
46, 62
58
15
0.07
0.96
−40
25
62
5
Bn-TPEN
740 (400)
0.01 (0.87)
100
4.8
0.069
46, 62
62
−40
−30
25
a
6
TPA
724 (300)
750 (200)
0.01 (0.92)
0.05 (0.73)
5.4 × 10−3
64, 65
66
7a
PyTACN
b
5.7
4.2
0.016
15
66
0.81
7.4
−30
25
67
7b
Py2TACN
740 (340)
900 (200)
758 (120)
770 (180)
760 (130)
62
0.027
−40
0
8α
8β
9a
9b
cis-α-BQCN
cis-β-BQCN
BP1
0.16
68
68
69
62
0.014
0
8
−35
25
BP2
730 (400), 916 (sh)
1.1
0.37
−40
25
9c
BP3
730 (380), 896 (sh)
40
62
0.014
−40
−40
−30
11
15
Me3NTB
770 (200)
400 (9800)
825 (260)
400, 900
0.02 (1.53)
3.1 × 103
9.4 × 102
1.2
1.5
2.1
0.25
63
70
TMG3tren
0.09 (−0.29)
9.0 × 10−2
17
18
tpaPh
0.09 (0.51)
0.08 (0.58)
1.4
18
−30
−30
84
71
TMG2dien
380 (8200)
805 (270)
650 (300)
900 (75)
57
19
TQA
0.24 (−1.05)
0.37
−40
72
*
a
The value in parentheses reflects the C−H bond dissociation energy of this substrate in units of kcal mol−1. Although spectroscopic
characterization was conducted on [(TPA)FeIV(O)]2+, the kinetic data were obtained using [(5Me3TPA)FeIV(O)]2+. The electronic absorption
b
spectra for the (5Me3TPA)FeIV(O) and (TPA)FeIV(O) complexes at −40 °C are similar. The oxidation rate for PhEt is not available; listed instead
is that for 2,3-dimethylbutene (BDE = 84 kcal mol−1).
and crystallographically characterized. All four methyl groups of
the TMC ligand were found to be on the same side of the
macrocycle and opposite to the position of the oxo group, and
MeCN occupied the coordination site trans to the oxo group.
This complex exhibits a t1/2 of 10 h at 25 °C. Two other closely
related complexes have also been crystallized, namely,
[(TMCPY)FeIV(O)]2+ and [(TMCDMA)FeIV(O)]2+ (Scheme
3),73 where one of the hydrogen atoms on a methyl group in
each complex was respectively replaced by a pyridyl group and
a −C(O)NMe2 group, providing a fifth ligand that coordinated
to the iron center trans to the oxo group. The pyridine
substitution decreased t1/2 of the complex to 7 h, while the
amide substitution increased t1/2 of the complex 12-fold to 120
h, demonstrating the importance of the axial ligand in
modulating the stability of the FeIVO complex.
observation of nonclassical KIE values of 10−20 for
dihydroanthracene and xanthene oxidation.60,61
Substitution of the axially bound MeCN of [(TMC)FeIV(O)-
(NCMe)]2+ with anions to obtain an [(TMC)FeIV(O)(X)]+
(1X) series shortened their t1/2 values and accelerated the
oxidation of DHA (BDE = 77 kcal mol−1) and cyclohexadiene
(CHD; BDE = 78 kcal mol−1). The HAT rates increased with
the basicity of the anionic ligand X− (X = NCMe < O2CCF3 <
N3 < SR),60 spanning a 40-fold difference in the HAT reactivity
and exhibiting a linear correlation when log k2 was plotted
against the redox potential of 1X (see the black line in Figure 4:
black crosses for DHA and filled diamonds for CHD). This
observed trend was surprising because the oxidizing power of
the oxoiron(IV) center would be expected to diminish as the
axial ligand became more electron-donating. The counter-
intuitive trend was rationalized by invoking the TSR model
discussed in the Introduction (Figure 2).56,60 DFT calculations
showed that the energy gap between the ground triplet and
excited quintet states (ΔETQ) decreased as the axial ligand
became more electron-donating. The decreased separation
between these two states increased the accessibility to the more
reactive quintet state as the hydrogen-atom abstraction
progressed along the reaction coordinate, thereby resulting in
the experimentally observed “antielectrophilic” trend. In
The high thermal stability of these complexes translated into
a relatively poor C−H abstraction reactivity,59 and only
substrates with relatively weak C−H bonds could be
oxidized.60,61 Within this narrow range of substrates, a linear
relationship could be discerned in the plot of log k2′ values
versus substrate C−H BDE values for [(TMC)FeIV(O)-
(NCMe)]2+, with faster rates associated with weaker C−H
bonds. These results indicated the involvement of a rate-
determining C−H abstraction step, which was supported by the
D
DOI: 10.1021/ic502786y
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Forum Article
Figure 4. Plot of log k2 determined in the oxidation of CHD (red
diamonds), DHA (black crosses), and PPh3 (green circles) at 0 °C
against Ep,c values of [FeIV(O)(TMC)(X)]n+ complexes in CH3CN or
1:1 CH3CN/CH3OH. The filled circles, filled diamonds, and crosses
represent data reported in ref 60, while the open circles and open
diamonds correspond to more recent data reported in ref 73. The red
line is the linear correlation for the OAT rates represented by the filled
and open circles, and the black line is the linear correlation for the
HAT rates represented by the filled diamonds and crosses. The letters
at the bottom of the figure correspond to the axial ligands of 1, with a
−
= −CHC(O−)(NMe2), b = RS−, c = −CH2C(O)NMe2, d = N3 , e
= CF3COO−, f = −CH2Py, and g = MeCN. See Scheme 3 for
structures of the FeIV(O) complexes.
Figure 3. Reactivity comparisons of high-valent oxoiron complexes in
the oxidation of (a) DHA at −30 °C (checkered red), −35 °C (striped
purple), and −40 °C (solid blue) and (b) cyclohexane at 25 °C
(striped green) and −40 °C (solid blue). See Schemes 3−6 for
structures of the respective (L)FeIV(O) complexes.
the newer complexes could be rationalized by including
tunneling contributions.57
Variations on the TMC macrocycle have also been
investigated. Replacing the N-methyl substituents of the
TMC ligand with benzyl groups to give the S = 1
[(TBC)FeIV(O)]2+ complex 2 resulted in a 150-fold increase
in the HAT reactivity (ethylbenzene oxidation),58 which could
be rationalized by a TSR argument. The weaker ligand field, as
indicated by the red shift of its near-IR band to 885 nm,
presumably increased accessibility to the more reactive quintet
spin state. Shrinking the TMC macrocycle by one carbon atom
to obtain the smaller 13-TMC analogue 3 also increased the
HAT reactivity (DHA oxidation) but by 3 orders of
magnitude.61 However, in this case, the near-IR band blue-
shifted to 735 nm, so the TSR argument cannot be applied to
this case. These contradictory observations demonstrate that
understanding the HAT reactivity is not so straightforward.
The second FeIVO complex to be crystallized was
supported by the pentadentate N4Py ligand.46,74 [(N4Py)-
FeIV(O)]2+ (4) is another S = 1 complex and has t1/2 of 60 h
(Scheme 4). Despite the fact that this complex was 6-fold more
stable than 1 at 25 °C, 4 was found to attack a much broader
range of hydrocarbon substrates. Indeed, it was the first
nonheme FeIVO complex shown to cleave the C−H bonds of
cyclohexane (BDE = 99 kcal mol−1), albeit quite slowly. The
substitution of N4Py with Bn-TPEN, a pentadentate ligand
with a different topology, gave rise to the less stable FeIVO
complex 5 (t1/2 = 3 h at 25 °C), which was more reactive and
oxidized cyclohexane at a 10-fold faster rate.
Scheme 3. S = 1 FeIV(O)(L) Complexes Supported by
Cyclam and Related Ligands
contrast, the oxygen-atom-transfer (OAT) rates for the series
(represented by green filled and open circles in Figure 4)
followed the opposite but expected trend, where more
electrophilic FeIVO complexes reacted more rapidly with
PPh3 (Figure 4, red line).
When the 1X series was expanded to include subsequently
characterized [(TMCPY)FeIV(O)]2+, [(TMCDMA)FeIV(O)]2+,
and the conjugate base of [(TMCDMA)FeIV(O)]2+,73 one linear
correlation held up but the other did not. The additional PPh3
data points (Figure 4, green open circles) conformed to the red
OAT line, supporting the correlation between the OAT rates
and the electrophilicity of the FeO unit. However, the
corresponding data points for CHD oxidation (Figure 4, red
open diamonds) did not fall at all near the black
antielectrophilic line, suggesting that there are other factors
besides the redox potential that control the HAT reactivity.
More recent DFT calculations suggested that the HAT data for
Additional complexes of pentadentate ligands with various
combinations of tertiary amine and pyridine donors, namely,
Py2TACN (7b), BP2 (9b), and BP3 (9c) (Scheme 4) behaved
similarly, with higher reactivity associated with complexes
exhibiting higher redox potentials and having pyridine rings
perpendicular to the FeO unit.62 An excellent positive
correlation was observed for the plot of log k2(OAT) values
E
DOI: 10.1021/ic502786y
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Forum Article
in the FeIV/III redox potential, with that of 8α (0.72 V vs SCE)
being 0.11 V higher than that of 8β (0.61 V vs SCE).
A study was also carried out with a series of [(L8py2)-
FeIV(O)(L)]2+ (10; Scheme 4) complexes, where the
supporting L8py2 ligand provided an equatorial array of four
nitrogen donors and L represented an axial pyridine N-oxide
donor with various 4-substituents.76 In the set of HAT
experiments with benzyl alcohol as the substrate, the oxidation
rates accelerated with more electron-donating substituents,
corresponding to a Hammett ρ value of −1.4, consistent with
the antielectrophilic trend found for HAT reactions of the 1X
series (Figure 4, black line). Surprisingly, the corresponding
OAT reactions also exhibited an antielectrophilic trend with a
Hammett ρ value of −1.3, which was quite a puzzling
observation.
Scheme 4. S = 1 FeIV(O)(L) Complexes Supported by
Pentadentate and Tetradentate Ligands
The complex that stands out among its peers in the S = 1
FeIVO subset is [(Me3NTB)FeIV(O)]2+ (11; Scheme 5)
Scheme 5. Highly Reactive High-Valent Oxoiron Complexes
versus FeIV/III potentials, where the OAT rates increased with
higher FeO electrophilicity. In contrast to the antielec-
trophilic trend observed for the 1X series (Figure 4, black line),
the HAT rates parallel the electrophilic trend observed in the
plot of log k2(OAT) versus FeIV/III potential, although the
correlation is not as good as that found for the OAT rates. It
was suggested that the downshift in the rates seen for the BP2/
BP3 subset may be associated with constraints imposed by the
bicyclic framework of the BP ligands.
For 4 and 5, large nonclassical KIE values of 30 and 50,
respectively, were observed at 25 °C in the oxidation of
ethylbenzene versus ethylbenzene-d10; in fact, for 5, the KIE
value was temperature-dependent and increased from 40 at 40
°C to 400 at −40 °C, which was ascribed to hydrogen-atom
tunneling effects.75 Further investigation of the KIE value as a
function of the substrate revealed a surprising pattern of
reactivity, where nonclassical values were found for substrates
with C−H BDE values of up to 93 kcal mol−1, but classical
values were associated with substrates with stronger C−H
bonds. For example, a KIE value of 6 was obtained for
cyclohexane oxidation by 5 at 40 °C. This unexpected behavior
was rationalized via the TSR model, where the tunneling
mechanism depends on the C−H BDE.
because it exhibits the highest oxidation rate of cyclohexane
(BDE = 99 kcal mol−1) to date for any S = 1 FeIV=O complex
(Figure 3).63 The cyclohexane oxidation ability of 11 even
exceeds those of the two [(TAML)FeV(O)]− complexes 12a
and 12b77,78 and [(TDCPP•)FeIV(O)]+ (13),63 which is a
synthetic analogue of Cpd 1, the highly reactive intermediate
involved in the oxidation chemistry of heme enzymes. The
HAT reactivity of 11 is at least 3 orders of magnitude higher
than that of 6, and the main structural difference between the
two is that 11 has three benzimidazole donors instead of
pyridines. The bar graphs in Figure 3 emphasize that the S = 1
complexes shown have DHA oxidation rates covering a range of
6 orders of magnitude, while the subset that can oxidize
cyclohexane exhibits rates that span a similar range after
adjustment for temperature differences.
S = 2 Oxoiron(IV) Complexes. The majority of the FeIV
O units found in synthetic nonheme complexes described thus
far have S = 1 ground states, while those associated with
nonheme iron enzyme intermediates have S = 2 ground states.
Calculations have predicted S = 2 FeIVO units to be much
more reactive toward C−H abstraction than their S = 1
counterparts.50−56,79 However, the paucity of synthetic
examples of S = 2 oxoiron(IV) complexes makes it difficult
to ascertain experimentally this computational prediction.
The first example of a synthetic S = 2 FeIVO complex was
reported by Bakac and co-workers,80,81 which was generated by
the ozonolysis of an aqueous [FeII(OH2)6]2+ ion at 25 °C and
formulated as [(H2O)5FeIV(O)]2+ (14). This complex exhibited
a half-life of only 20 s at room temperature and was found to
attack substrates like CH3CN with very high rate constants.
Closely related FeIVO complexes supported by tetraden-
tate ligands have also been found to have S = 1 ground states.
There are fewer systematic studies that allow convenient
comparison of the HAT reactivity, but it seems likely that
oxoiron(IV) complexes of tetradentate ligands are more
reactive than their pentadentate counterparts (Figure 3 and
Table 1). The two [(BQCN)FeIV(O)(NCMe)]2+ complexes
8α and 8β represent an interesting pair because they are
topological isomers (Scheme 4).68 8α was found to oxidize
ethylbenzene by a factor of 10 faster than 8β, demonstrating
the role that ligand topology can play in modulating the HAT
reactivity. This reactivity difference may stem from a difference
F
DOI: 10.1021/ic502786y
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Forum Article
However, its high HAT reactivity, together with the fact that it
could only be observed in aqueous media, made this complex a
challenge to characterize with the spectroscopic detail achieved
for other FeIVO complexes.
An even more reactive S = 2 complex, [(TQA)FeIV(O)-
(NCMe)]2+ (19), has just been obtained by substituting the
pyridine moieties of the TPA ligand of 6 with weaker-field
quinoline ligands (Scheme 6).72 This complex has a half-life of
15 min at −40 °C and is formulated as a six-coordinate
complex, unlike the previously discussed S = 2 complexes 15−
18. Significantly, 19 exhibits the highest cyclohexane oxidation
rate (k2 = 0.37 M−1 s−1 at −40 °C) found to date for this class
of complexes (Figure 3b). For comparison, the closely related S
= 1 [FeIV(O)(TPA)(NCMe)]2+ complex 6 does not react at all
with cyclohexane at −40 °C, while the cyclohexane oxidation
rates for the S = 1 complexes [FeIV(O)(N4Py)]2+ (4) and
[FeIV(O)(BnTPEN)]2+ (5) at 25 °C are 3−4 orders of
magnitude slower than that for 19 at −40 °C even without
correcting for the 65 °C temperature difference.46 The
observed reactivity of 19 thus supports the DFT-based
expectations of a more reactive S = 2 FeIVO center.
The first S = 2 FeIV(O) complex to be isolated was the
trigonal-bipyramidal [(TMG3tren)FeIV(O)]2+ complex 15
(Scheme 6).70 The C3 symmetry about the iron(IV) center
Scheme 6. Crystallographically and/or Spectroscopically
Characterized S = 2 FeIV(O)(L) Complexes
An exception to this general trend is the S = 1 complex
[FeIV(O)(Me3NTB)]2+ (11), which is almost as reactive as 19,
with k2(cyclohexane) = 0.23 M−1 s−1 at −40 °C.63 This
surprising similarity suggests that FeIV(O) complexes, despite
having different ground spin states, can achieve comparably
high HAT reactivity. DFT calculations on 11 suggest that its
high reactivity derives from a highly reactive S = 2 excited-state
surface that lies in close proximity of the S = 1 ground state. If
this postulate can be experimentally verified, these results
would suggest that oxoiron(IV) complexes in the S = 1 state
can be just as reactive as their S = 2 congeners as long as the
excited S = 2 spin state is easily accessible. Clearly, more effort
is needed to assess the relative energies of the S = 1 and 2 spin
states in oxoiron(IV) complexes experimentally and correlate
these ΔETQ values with their observed HAT reactivity.
gave rise to a d-orbital splitting pattern with a
(dxz,dyz)2(dxy,dx −y ) (dz ) configuration. Although this complex
had a t1/2 value at 25 °C of 30 s, just slightly longer than that of
14, the possibility of cooling the MeCN solvent to −40 °C
allowed its half-life to be extended significantly. In addition, it
was determined that its self-decay occurred by intramolecular
attack of a ligand methyl C−H bond by the FeO unit,
showing that this FeO moiety was capable of attacking a
relatively strong C−H bond (BDE ∼ 93 kcal mol−1). On the
basis of this result, the 12 ligand methyl groups were
perdeuterated in order to further extend the lifetime of 15,
2
0
2
2
2
A challenge in analyzing the above reactivity comparisons is
how to factor in the effect of the supporting ligand. In the cases
discussed thus far, different supporting ligands with distinct
topologies were required to obtain a different spin state for the
FeIVO unit. Que and co-workers found a way to address this
challenge by preparing a series of oxo-bridged diiron complexes
(20−25; Figure 5) supported by the same TPA* ligand [TPA*
= tris(3,5-dimethyl-4-methoxypyridyl-2-methyl)amine].64,85
This series consisted of FeIIIFeIV and FeIVFeIV complexes with
either an Fe2(μ-O)2 diamond core or an Fe−O−FeO open
core, and for the latter subset, the FeIVO unit could have an S
= 1 or 2 spin state depending on the oxidation state of the
adjacent iron center (Figure 5). Because the reactive members
of this series could only be prepared and studied at −80 °C,
DHA was the substrate of choice for comparing the C−H bond
cleavage ability of this series of complexes. This comparison
clearly demonstrates that a terminal oxo is at least 100-fold
better at HAT than a bridging one and an S = 2 FeIVO unit is
1000−10000-fold faster at HAT than its S = 1 analogue. The
most reactive members of this series have measured rates of
DHA oxidation of 320−360 M−1 s−1 at −80 °C,64,86 which
exceeds values obtained at −30 or −40 °C for all complexes
shown in Figure 3a except for the S = 1 Me3NTB complex
(11). When adjusted for the 40 °C temperature difference, the
DHA oxidation rates for 24 and 25 are likely to exceed that of
the Me3NTB complex.63 These results provide the strongest
support to date for the consensus DFT predictions of a more
reactive S = 2 FeIVO moiety.
which enabled its characterization by X-ray crystallography.82
A
similar C3-symmetric ligand design strategy was employed by
Borovik et al.83 and Chang et al.84 to obtain additional examples
of S = 2 FeIVO complexes, namely, [(H3buea)FeIV(O)]−
(16) and [(tpaPh)FeIV(O)]− (17) (Scheme 6). An X-ray
structure was also obtained for the Borovik complex.
The intermolecular HAT reactivity of 15 was, however, quite
sluggish because of the near-encapsulation of the FeO unit
by the bulky tetramethylguanidine groups of the TMG3tren
supporting ligand that hindered substrate access.82 Its rate of
DHA oxidation at −30 °C was, in fact, 20-fold slower than that
of 4 (Figure 3), but the corresponding rates of oxidation of the
smaller CHD (77 kcal mol−1) were essentially identical,
corroborating the steric access argument.70 The higher
reactivity predicted for an S = 2 FeIVO moiety was realized
when one arm of the TMG3tren supporting ligand was replaced
by a methyl group to obtain a less sterically hindered
TMG2dien ligand (Scheme 6), and the corresponding S = 2
[(TMG2dien)FeIV(O)]2+ complex (18) was found to have a
DHA oxidation rate 600-fold faster than that of 15 (Figure 3).71
G
DOI: 10.1021/ic502786y
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Forum Article
first-order in the iron complex and first-order in dioxygen,
indicating that the rate-determining step corresponds to an
early phase of the reaction. By analogy to the α-KG-dependent
enzymes, it is proposed that the oxidative decarboxylation step
forms a yet unobserved FeIVO oxidant 28a, which DFT
calculations suggest to have an S = 2 ground state when the
carboxylate is coordinated in a monodentate fashion.74,88
A
space-filling model of the putative oxidant is shown in Figure 6,
right.
We previously showed that it was possible to intercept the
high-valent species 28a by introducing external substrates that
could be oxidized intermolecularly in competition with the
intramolecular ligand hydroxylation.88 The lower yield of the
green chromophoric product from the latter reaction was used
as a convenient probe for assessing the relative rates of
intermolecular and intramolecular oxidation. Indeed, the
addition of 10 equiv of PhSMe into the reaction of 26a with
dioxygen completely prevented intramolecular self-hydroxyla-
tion. Instead, a colorless [FeII(TpPh2)(O2CPh)(OSR2)] adduct
was observed based on electrospray ionization mass spectrom-
etry, NMR, and Mossbauer evidence. The sulfoxide adduct
̈
formed in 70% yield, a value comparable to the amount of self-
hydroxylated product obtained in the absence of thioanisole.
Probing the thioanisole interception further revealed some
interesting but expected insights. The addition of stoichio-
metric PhSMe suppressed formation of the self-hydroxylated
product by 60%, demonstrating the high oxo-transfer ability of
the putative FeIVO oxidant 28a. Substituting PhSMe with the
more electron-rich tetrahydrothiophene decreased by 90% the
amount of self-hydroxylated product, while the less electron-
rich Ph2S had the opposite effect (40%). So, 28a is sensitive to
the basicity of the substrate. It should be noted, however, that
the reaction time was unchanged for these three experiments
because the rate-determining step is associated with the
formation of 28a rather than its decay.
In the 2009 paper on this system, limited success was
reported in intercepting 28a with hydrocarbons.88 While the
addition of substrates with weak C−H bonds such as DHA,
fluorene, or cyclohexene significantly decreased the yield of the
self-hydroxylated product, potential substrates with stronger
C−H bonds were much less effective. Indeed, the addition of
0.1 M ethylbenzene (BDE = 87 kcal mol−1), toluene (BDE =
90 kcal mol−1), or cyclohexane (BDE = 99 kcal mol−1) did not
affect the yield of 27a, raising the possibility that 28a may not
be a powerful enough oxidant to be able to cleave these
stronger C−H bonds. However, the observation that the use of
0.1 M cyclooctane (BDE = 95 kcal mol−1) as a substrate slightly
Figure 5. (a) Interconversions among oxo-bridged diiron-TPA*
complexes 20−25. Red and blue fonts respectively represent high-
and intermediate-spin iron centers. (b) Comparison of DHA oxidation
rates of the various diiron-TPA* complexes at −80 °C.
RESULTS AND DISCUSSION
■
C−H Bond Cleavage by a Fleeting FeIVO Species
Produced by Dioxygen Activation. All the FeIVO
complexes described above have been generated by an oxidant
other than dioxygen, which has allowed them to accumulate
and be characterized with various spectroscopic methods and
sometimes by X-ray crystallography. An exception to this
pattern is the putative oxoiron(IV) species that is produced by
the oxygenation of [FeII(TpPh2)(O2CC(O)Ph)] (26a), an
iron(II) complex that serves as a functional model for α-KG-
dependent enzymes like TauD (Scheme 1). Upon exposure to
dioxygen in a benzene solvent at room temperature, 26a
undergoes oxidative decarboxylation of its benzoylformate
ligand (the α-KG analogue) to form a green chromophore
designated as [FeIII(TpPh2*)(O2CPh)] (27a).89 The latter has
been identified as an iron(III) phenolate complex, resulting
from self-hydroxylation of one of the ligand phenyl rings
(Figure 6, left). Kinetic studies show that the reaction rate is
Figure 6. Left: Some reactions of 26a/26b with dioxygen in benzene at 25 °C. Right: DFT-calculated model for the putative oxidant. The peach-
colored atom in front is an oxygen atom from the benzoate ligand, and the benzoyl moiety was excised to give a clearer view of the iron center.
Adapted with permission from ref 88. Copyright 2009 Wiley-VCH Verlag GmbH & Co. KGaA.
H
DOI: 10.1021/ic502786y
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Forum Article
decreased the yield of 27a provided some incentive to follow up
on these earlier observations and obtain the more recent results
described below.
In these more recent experiments, we have used higher
substrate concentrations in order to increase the intermolecular
oxidation rate and enhance the likelihood of intercepting the
nascent FeIVO oxidant 28a. Indeed, this expectation has been
borne out in the reactions of 26a with dioxygen in the presence
of up to 2 M toluene or cyclohexane (Figure 7, top, and Tables
2 and S1 in the Supporting Information, SI). For example, with
1 M toluene, the yield of 27a decreased from 64% to 48%
concomitant with the formation of PhCHO in 33% yield. Even
more significant were the experiments with 1 M cyclohexane,
where even less 27a was formed (34%) and cyclohexane
oxidation products were obtained in 44% yield with a 1:3 ratio
of cyclohexanol to cyclohexanone. Parallel experiments carried
out on the corresponding pyruvate complex [FeII(TpPh2)-
(O2CC(O)CH3)] (26b) to serve as a check on the generality
of the results obtained for 26a gave comparable results (Tables
2 and S1 in the SI). In previous work, 26b was found to react
with dioxygen 4-fold more rapidly than 26a at 25 °C, achieving
completion within 15 min (vs 1 h for 26a), but the amount of
self-hydroxylated 27b obtained was lower.87 The lower extent
of self-hydroxylation for 26b may reflect a less efficient
formation of 28b relative to that of 28a or the decay of a
higher fraction of 28b via an alternative unproductive pathway
in competition with self-hydroxylation. However, as seen in
Figure 7, bottom, and Table 2, oxygenation of 26b gave rise to
intermolecular oxidation products either equal to or higher than
26a.
The major (or only) oxidation products observed in the
toluene and cyclohexane oxidation are not the alcohols but the
corresponding aldehyde or ketone. These outcomes suggest
two mechanistic possibilities. The first scenario entails
subsequent oxidation of the initially formed alcohol product
by a second 1 equiv of 28. This notion is not implausible
because oxygenation of the iron(II) precursor, being rate-
determining, makes this step slower than the substrate
oxidation step, and alcohol oxidation would be expected to
be much more facile than oxidation of the hydrocarbon. The
second scenario involves diffusion of the alkyl radical formed in
the initial hydrogen-atom abstraction into solution and reaction
with the excess dioxygen present. At this point, we are unable
to assess the likelihood of the nascent alkyl radical undergoing
oxygen rebound. What is clear, however, is that the nascent
alkyl radical is formed by the action of a metal-based oxidant,
based on results from competitive experiments between a
substrate and its perdeuterated analogue in their respective
oxidations by 26a and 26b. Product analysis showed product
KIEs (PKIEs) of 12 for cyclohexane oxidation and 15 for
toluene oxidation (Tables 2 and S3 in the SI). Such nonclassical
KIE values have been observed for the reactions of C−H bonds
with a number of nonheme oxoiron(IV) complexes61−63,66,72
and support the postulate that 28 is also a member of this class
of high-valent iron complexes.
Figure 7. Top: Comparison of the product yields in the reactions of
26a with dioxygen as a function of the cyclohexane (blue) or toluene
(red) concentration. Bottom: Comparison of the product yields in the
reactions of 26a (blue) and 26b (green) with dioxygen as a function of
the cyclohexane concentration. Solid bars: yield of oxidized substrate.
Dashed bars: yield of 27a/27b. See Figure 6 for structures of 26a and
26b.
Table 2. Product Yields in the Reactions of Dioxygen with
Having demonstrated the ability of 28 to oxidize cyclo-
hexane, we shifted our attention to the light alkane substrate n-
butane, which has a high solubility in benzene (approximately 4
M at 10 °C by NMR). Indeed, bubbling n-butane into a
benzene solution of 26a/26b, followed by dioxygen, resulted in
a lower yield of 27a/27b, suggesting the interception of 28a/
a
FeII(TpPh2)(BF/PRV) (26a/26b)
self-hydroxylation
yield
substrate oxidation yield
26a 26b
concn
(M)
substrate
none
26a
26b
1
28b by n-butane. H NMR analysis of the product solution
64(1)
48(2)
40(2)
0
0
showed the characteristic multiplets of the C2 proton of 2-
butanol at 3.7 ppm and the C3 protons of 2-butanone at 1.8
ppm (Figure 8). Because the region below 1.8 ppm contained
strong signals from residual n-butane, 1D-TOCSY experiments
were carried out to identify protons coupled to these multiplets
as a means of corroborating our assignments. Indeed, features
were observed at 1.5, 1.2, and 0.8 ppm upon irradiation of the
3.7 ppm multiplet belonging to 2-butanol (Figure S3 in the SI)
and at 1.6 and 0.8 ppm in the corresponding experiment for the
2-butanone multiplet at 1.8 ppm (Figure S4 in the SI). It
toluene
1
1
27(3) 33(5)
PKIE = 15
33(3)
PKIE = 15
cyclohexane
34(3)
39(2)
22(2) A + K 44(5), A + K 51(2),
A/K 11/33,
PKIE = 12
A/K 2/49,
PKIE = 12
n-butane
4
20(2) A + K 30(3), A + K 39(3),
A/K 18/12 A/K 24/15
a
Reaction conditions: ∼1 mM FeII complex, 20 °C. Each value is an
average of at least three runs. A represents the yield of alcohol. K
represents the yield of ketone.
I
DOI: 10.1021/ic502786y
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Forum Article
FeIVO unit.64,85,86 These observations emphasize that more
remains to be learned about the factors that control the HAT
reactivity, despite a consensus in the computational community
that an S = 2 FeIVO species is more reactive at HAT than its
S = 1 counterpart.49−55,78 In the second part, we report a
further investigation of the HAT reactivity of 28, the putative S
= 2 FeIVO oxidant generated from the reaction of 26a/26b
with dioxygen, and show that it is powerful enough to oxidize
the strong C−H bonds of cyclohexane and even n-butane. The
demonstration that 28 can hydroxylate n-butane represents the
first example of a high-valent iron species derived from
dioxygen that is capable of such a transformation. It not only
provides a steppingstone toward our understanding of how
metalloenzymes work but also highlights the potential such
synthetic systems may hold.
Figure 8. Left: Stacked NMR spectra showing the CH3CH2C(H)-
OHCH3 (3.7 ppm) peak in a reaction solution of 26a under various
conditions. Bottom/blue: Oxidation reaction with n-butane as the
substrate. Middle/red: Reaction mixture upon the addition of 25 μL of
DMSO-d6 and 2 μL of AcOD-d4 (0.18 mM/Fe). Top/green: Reaction
mixture from the middle spiked with 2-butanol (0.49 mM/Fe). Right:
NMR spectra showing the changes in the intensity of the
CH3C(H2)COCH3 (1.8 ppm) peak upon the addition of 2-butanone
to a reaction solution of 26a. Bottom/blue: Oxidation reaction with n-
butane as the substrate (0.12 mM/Fe). Middle/red: Solution in blue
spiked with 2-butanone (0.36 mM/Fe). Top/green: Solution in red
spiked with additional 2-butanone (0.70 mM/Fe).
EXPERIMENTAL SECTION
■
Materials and Methods. All reagents and solvents were
purchased from commercial sources and used without further
purification unless otherwise stated. The preparation and handling of
air-sensitive materials were carried out under an inert atmosphere
using a glovebox. The complexes [Fe(TpPh2)(O2CC(O)Ph)] (26a)
and [Fe(TpPh2)(O2CC(O)CH3)] (26b) were prepared as previously
reported.87
Physical Methods. UV−vis spectra were recorded on a Agilent
Cary 60 UV−vis spectrometer. Gas chromatography (GC) product
analyses were performed on a PerkinElmer Autosystem XL gas
chromatograph (AT-1701, 30 m; DB-1, 30 m) equipped with a flame
ionization detector. GC−mass spectrometry (MS) analyses were
performed on a Agilent 7890A gas chromatograph (HP-5MS, 30 m)
equipped with an Agilent 5975C mass detector in electron impact
mode. NMR analyses were recorded on a Bruker Avance 500
spectrometer.
Product Analysis. Organic products were analyzed after reaction
with oxygen via GC: After the reaction, the mixture was passed
through a short silica column and eluted with tetrahydrofuran. A
known concentration of naphthalene was introduced into the reaction
mixture as an internal standard. Yields of the oxidized products were
determined by a comparison to authentic standards.
should be noted that the signal from the C2 proton of 2-
butanol in the reaction mixture was broad and lacked
resolution, which was similar to the spectrum observed for 2-
butanol alone in benzene (Figure S5 in the SI). As shown in
Figure 8, left, the addition of small amounts of dimethyl
sulfoxide (DMSO) and acetic acid (AcOH) into the reaction
mixture substantially sharpened the C2 proton peak and shifted
it upfield. We attribute the broadness of the 2-butanol C2−H
signal to a hydrogen-bonded cluster of the alcohol in the
nonpolar benzene solvent, which was disrupted upon the
addition of DMSO and AcOH. Integration of the signals
relative to an added tetrachloroethane standard revealed that 2-
butanol and 2-butanone were formed in respective yields of 18
and 12% for the reaction with 26a and 24 and 15% for the
reaction with 26b. In both cases, the addition of aliquots of
either 2-butanol or 2-butanone increased the intensities of the
3.7 and 1.8 ppm peaks, as expected (Figures 8 and S6 in the
SI). No evidence was found in the NMR spectra for any
product arising from C1 oxidation.
ASSOCIATED CONTENT
* Supporting Information
Oxidation product ratios and yields and H NMR spectra of
■
S
1
butane-derived products. This material is available free of
charge via the Internet at http://pubs.acs.org.
Oxidation of n-butane by 28 represents the only system thus
far to mediate the nonradical oxidation of a gaseous alkane by a
high-valent iron oxidant generated from dioxygen. However,
there are two other nonheme iron-based systems that carry out
AUTHOR INFORMATION
Corresponding Author
*E-mail: larryque@umn.edu.
■
t
similar chemistry on gaseous alkanes, but with BuOOH or
Oxone as the source of oxidizing equivalents.89,90 These are
promising steps toward the functionalization of gaseous
alkanes, but more work is needed to ascertain the nature of
the high-valent iron oxidants implicated in these oxidations.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by National Science Foundation
Grant CHE-1361773 to L.Q.
CONCLUSION
■
In the first part of this Forum Article, we reviewed the C−H
bond cleavage abilities of a series of well-characterized
nonheme oxoiron(IV) complexes and showed that the
observed rates of oxidation depend on the nature of the
supporting ligand and differ by over 6 orders of magnitude. The
most reactive are represented by the S = 1 [FeIV(O)-
(Me3NTB)] complex 11 (described by Nam),63 the recently
reported S = 2 [FeIV(O)(TQA)(NCMe)] complex 19,72 and
the FeIII,IV2(TPA*) complexes 24 and 25 that have an S = 2
REFERENCES
■
(1) Crabtree, R. H. Chem. Rev. 2010, 110, 575−575.
(2) Vedernikov, A. N. Acc. Chem. Res. 2012, 45, 803−813.
(3) Que, L., Jr.; Tolman, W. B. Nature 2008, 455, 333−340.
(4) Sono, M.; Roach, M. P.; Coulter, E. D.; Dawson, J. H. Chem. Rev.
1996, 96, 2841−2887.
(5) Costas, M.; Mehn, M. P.; Jensen, M. P.; Que, L., Jr. Chem. Rev.
2004, 104, 939−986.
J
DOI: 10.1021/ic502786y
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Forum Article
(6) Conley, B.; Tenn, W., III; Young, K.; Ganesh, S.; Steve, M.;
Oxgaard, J.; Gonzales, J.; Goddard, W., III; Periana, R. In Activation of
Small Molecules: Organometallic and Bioinorganic Perspectives; Tolman,
W., Ed.; Wiley-VCH: Weinheim, Germany, 2006; pp 235−285.
(7) Sheldon, R.; Arends, I.; Hanefeld, U. Green Chemistry and
Catalysis; Wiley-VCH: Weinheim, Germany, 2007.
(41) Ray, K.; Pfaff, F. F.; Wang, B.; Nam, W. J. Am. Chem. Soc. 2014,
136, 13942−13958.
(42) Nam, W.; Lee, Y.-M.; Fukuzumi, S. Acc. Chem. Res. 2014, 47,
1146−1154.
(43) Luo, Y.-R. Comprehensive Handbook of Chemical Bond Energies;
CRC Press: Boca Raton, FL, 2007.
(8) Meunier, B. Biomimetic Oxidations Catalyzed by Transition Metal
Complexes; Imperial College Press: London, 2000.
(44) Groves, J. T.; McClusky, G. A. J. Am. Chem. Soc. 1976, 98, 859−
861.
(9) Shaik, S.; Cohen, S.; Wang, Y.; Chen, H.; Kumar, D.; Thiel, W.
Chem. Rev. 2010, 110, 949−1017.
(45) Shaik, S.; Kumar, D.; de Visser, S. P.; Altun, A.; Thiel, W. Chem.
Rev. 2005, 105, 2279−2328.
(10) Dawson, J. H. Science 1988, 240, 433−439.
(46) Kaizer, J.; Klinker, E. J.; Oh, N. Y.; Rohde, J.-U.; Song, W. J.;
(11) Ortiz de Montellano, P. R. Chem. Rev. 2009, 110, 932−948.
(12) Denisov, I. G.; Makris, T. M.; Sligar, S. G.; Schlichting, I. Chem.
Rev. 2005, 105, 2253−2277.
Stubna, A.; Kim, J.; Munck, E.; Nam, W.; Que, L., Jr. J. Am. Chem. Soc.
̈
2003, 126, 472−473.
(47) Cho, K.-B.; Wu, X.; Lee, Y.-M.; Kwon, Y. H.; Shaik, S.; Nam, W.
J. Am. Chem. Soc. 2012, 134, 20222−20225.
(13) Rittle, J.; Green, M. T. Science 2010, 330, 933−937.
(14) Wallar, B. J.; Lipscomb, J. D. Chem. Rev. 1996, 96, 2625−2657.
(15) Merkx, M.; Kopp, D. A.; Sazinsky, M. H.; Blazyk, J. L.; Muller,
J.; Lippard, S. J. Angew. Chem., Int. Ed. 2001, 40, 2782−2807.
(16) Tinberg, C. E.; Lippard, S. J. Acc. Chem. Res. 2011, 44, 280−288.
(17) Beauvais, L. G.; Lippard, S. J. J. Am. Chem. Soc. 2005, 127,
7370−7378.
(48) Russell, G. Reactivity, selectivity, and polar effects in hydrogen atom
transfer reactions; Wiley: New York, 1973.
(49) Finn, M.; Friedline, R.; Suleman, N. K.; Wohl, C. J.; Tanko, J.
M. J. Am. Chem. Soc. 2004, 126, 7578−7584.
(50) Decker, A.; Rohde, J. U.; Klinker, E. J.; Wong, S. D.; Que, L., Jr.;
Solomon, E. I. J. Am. Chem. Soc. 2007, 129, 15983−15996.
(51) Shaik, S.; Chen, H.; Janardanan, D. Nat. Chem. 2011, 3, 19−27.
(52) Wong, S. D.; Bell, C. B.; Liu, L. V.; Kwak, Y.; England, J.; Alp, E.
E.; Zhao, J. Y.; Que, L., Jr.; Solomon, E. I. Angew. Chem., Int. Ed. 2011,
50, 3215−3218.
(18) Shu, L. J.; Nesheim, J. C.; Kauffmann, K.; Munck, E.; Lipscomb,
̈
J. D.; Que, L., Jr. Science 1997, 275, 515−518.
(19) Banerjee, R.; Proshlyakov, Y.; Lipscomb, J. D.; Proshlyakov, D.
A. Nature 2015, 518, 431−434.
(20) Nesheim, J. C.; Lipscomb, J. D. Biochemistry 1996, 35, 10240−
10247.
(53) Hirao, H.; Kumar, D.; Que, L., Jr.; Shaik, S. J. Am. Chem. Soc.
2006, 128, 8590−8606.
(21) Hausinger, R. P. Crit. Rev. Biochem. Mol. Biol. 2004, 39, 21−68.
(22) Koehntop, K. D.; Emerson, J. P.; Que, L., Jr. J. Biol. Inorg. Chem.
2005, 10, 87−93.
(23) Kappock, T. J.; Caradonna, J. P. Chem. Rev. 1996, 96, 2659−
2756.
(24) Rocklin, A. M.; Tierney, D. L.; Kofman, V.; Brunhuber, N. M.
W.; Hoffman, B. M.; Christoffersen, R. E.; Reich, N. O.; Lipscomb, J.
D.; Que, L., Jr. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 7905−7909.
(25) Mirica, L. M.; Klinman, J. P. Proc. Natl. Acad. Sci. U.S.A. 2008,
105, 1814−1819.
(26) Baldwin, J. E.; Bradley, M. Chem. Rev. 1990, 90, 1079−1088.
(27) Roach, P. L.; Clifton, I. J.; Hensgens, C. M. H.; Shibata, N.;
Schofield, C. J.; Hajdu, J.; Baldwin, J. E. Nature 1997, 387, 827−830.
(28) Schenk, W. A. Angew. Chem., Int. Ed. 2000, 39, 3409−3411.
(29) Krebs, C.; Fujimori, D. G.; Walsh, C. T.; Bollinger, J. M. Acc.
Chem. Res. 2007, 40, 484−492.
(30) Riggs-Gelasco, P. J.; Price, J. C.; Guyer, R. B.; Brehm, J. H.; Barr,
E. W.; Bollinger, J. M.; Krebs, C. J. Am. Chem. Soc. 2004, 126, 8108−
8109.
(54) Louwerse, M. J.; Baerends, E. J. Phys. Chem. Chem. Phys. 2007, 9,
156−166.
(55) Srnec, M.; Wong, S. D.; Solomon, E. I. Dalton Trans. 2014, 43,
17567−17577.
(56) Hirao, H.; Que, L., Jr.; Nam, W.; Shaik, S. Chem.Eur. J. 2008,
14, 1740−1756.
(57) Mandal, D.; Ramanan, R.; Usharani, D.; Janardanan, D.; Wang,
B.; Shaik, S. J. Am. Chem. Soc. 2015, 137, 722−733.
́
(58) Wilson, S. A.; Chen, J.; Hong, S.; Lee, Y. M.; Clemancey, M.;
Garcia-Serres, R.; Nomura, T.; Ogura, T.; Latour, J. M.; Hedman, B.;
Hodgson, K. O.; Nam, W.; Solomon, E. I. J. Am. Chem. Soc. 2012, 134,
11791−11806.
(59) Rohde, J. U.; In, J. H.; Lim, M. H.; Brennessel, W. W.;
Bukowski, M. R.; Stubna, A.; Munck, E.; Nam, W.; Que, L. Science
̈
2003, 299, 1037−1039.
(60) Sastri, C. V.; Lee, J.; Oh, K.; Lee, Y. J.; Jackson, T. A.; Ray, K.;
Hirao, H.; Shin, W.; Halfen, J. A.; Kim, J.; Que, L., Jr.; Shaik, S.; Nam,
W. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 19181−19186.
(61) Hong, S.; So, H.; Yoon, H.; Cho, K. B.; Lee, Y. M.; Fukuzumi,
S.; Nam, W. Dalton Trans. 2013, 42, 7842−7845.
(31) Price, J. C.; Barr, E. W.; Glass, T. E.; Krebs, C.; Bollinger, J. M. J.
Am. Chem. Soc. 2003, 125, 13008−13009.
(62) Wang, D.; Ray, K.; Collins, M. J.; Farquhar, E. R.; Frisch, J. R.;
(32) Proshlyakov, D. A.; Henshaw, T. F.; Monterosso, G. R.; Ryle, M.
J.; Hausinger, R. P. J. Am. Chem. Soc. 2004, 126, 1022−1023.
(33) Hoffart, L. M.; Barr, E. W.; Guyer, R. B.; Bollinger, J. M.; Krebs,
C. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 14738−14743.
(34) Galonic, D. P.; Barr, E. W.; Walsh, C. T.; Bollinger, J. M.; Krebs,
C. Nat. Chem. Biol. 2007, 3, 113−116.
́
Gomez, L.; Jackson, T. A.; Kerscher, M.; Waleska, A.; Comba, P.;
Costas, M.; Que, L., Jr. Chem. Sci. 2013, 4, 282−291.
(63) Seo, M. S.; Kim, N. H.; Cho, K. B.; So, J. E.; Park, S. K.;
́
Clemancey, M.; Garcia-Serres, R.; Latour, J. M.; Shaik, S.; Nam, W.
Chem. Sci. 2011, 2, 1039−1045.
(64) Xue, G.; Pokutsa, A.; Que, L., Jr. J. Am. Chem. Soc. 2011, 133,
16657−16667.
(35) Fujimori, D. G.; Barr, E. W.; Matthews, M. L.; Koch, G. M.;
Yonce, J. R.; Walsh, C. T.; Bollinger, J. M.; Krebs, C.; Riggs-Gelasco, P.
J. J. Am. Chem. Soc. 2007, 129, 13408−13409.
(36) Matthews, M. L.; Krest, C. M.; Barr, E. W.; Vaillancourt, F. H.;
Walsh, C. T.; Green, M. T.; Krebs, C.; Bollinger, J. M. Biochemistry
2009, 48, 4331−4343.
(37) Panay, A. J.; Lee, M.; Krebs, C.; Bollinger, J. M.; Fitzpatrick, P.
F. Biochemistry 2011, 50, 1928−1933.
(38) Eser, B. E.; Barr, E. W.; Frantorn, P. A.; Saleh, L.; Bollinger, J.
M.; Krebs, C.; Fitzpatrick, P. F. J. Am. Chem. Soc. 2007, 129, 11334−
11335.
(39) Que, L., Jr. Acc. Chem. Res. 2007, 40, 493−500.
(40) McDonald, A. R.; Que, L., Jr. Coord. Chem. Rev. 2013, 257,
414−428.
(65) Lim, M. H.; Rohde, J. U.; Stubna, A.; Bukowski, M. R.; Costas,
M.; Ho, R. Y. N.; Munck, E.; Nam, W.; Que, L. Proc. Natl. Acad. Sci.
̈
U.S.A. 2003, 100, 3665−3670.
́
(66) Company, A.; Prat, I.; Frisch, J. R.; Mas-Balleste, R.; Guell, M.;
̈
Juhas
Chem.Eur. J. 2011, 17, 1622−1634.
(67) Planas, O.; Clemancey, M.; Latour, J.-M.; Company, A.; Costas,
́
z, G.; Ribas, X.; Munck, E.; Luis, J. M.; Que, L., Jr.; Costas, M.
̈
́
M. Chem. Commun. 2014, 50, 10887−10890.
(68) Hong, S.; Lee, Y. M.; Cho, K. B.; Sundaravel, K.; Cho, J.; Kirn,
M. J.; Shin, W.; Nam, W. J. Am. Chem. Soc. 2011, 133, 11876−11879.
(69) Comba, P.; Wunderlich, S. Chem.Eur. J. 2010, 16, 7293−
7299.
K
DOI: 10.1021/ic502786y
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Forum Article
(70) England, J.; Martinho, M.; Farquhar, E. R.; Frisch, J. R.;
Bominaar, E. L.; Munck, E.; Que, L., Jr. Angew. Chem., Int. Ed. 2009,
̈
48, 3622−3626.
(71) England, J.; Guo, Y. S.; Van Heuvelen, K. M.; Cranswick, M. A.;
Rohde, G. T.; Bominaar, E. L.; Munck, E.; Que, L., Jr. J. Am. Chem. Soc.
̈
2011, 133, 11880−11883.
(72) Biswas, A. N.; Puri, M.; Meier, K. K.; Oloo, W. N.; Rohde, G. T.;
Bominaar, E. L.; Munck, E.; Que, L., Jr. J. Am. Chem. Soc. 2015, 137,
̈
2428−2431.
(73) England, J.; Bigelow, J. O.; Van Heuvelen, K. M.; Farquhar, E.
R.; Martinho, M.; Meier, K. K.; Frisch, J. R.; Munck, E.; Que, L., Jr.
̈
Chem. Sci. 2014, 5, 1204−1215.
(74) Klinker, E. J.; Kaizer, J.; Brennessel, W. W.; Woodrum, N. L.;
Cramer, C. J.; Que, L., Jr. Angew. Chem., Int. Ed. 2005, 44, 3690−3694.
(75) Klinker, E. J.; Shaik, S.; Hirao, H.; Que, L., Jr. Angew. Chem., Int.
Ed. 2009, 48, 1291−1295.
(76) Zhou, Y. M.; Shan, X. P.; Mas-Balleste,
Stubna, A.; Chakrabarti, M.; Slominski, L.; Halfen, J. A.; Munck, E.;
́
R.; Bukowski, M. R.;
̈
Que, L., Jr. Angew. Chem., Int. Ed. 2008, 47, 1896−1899.
(77) Ghosh, M.; Singh, K. K.; Panda, C.; Weitz, A.; Hendrich, M. P.;
Collins, T. J.; Dhar, B. B.; Sen Gupta, S. J. Am. Chem. Soc. 2014, 136,
9524−9527.
(78) Kundu, S.; Thompson, J. V.; Shen, L. Q.; Mills, M. R.;
Bominaar, E. L.; Ryabov, A. D.; Collins, T. J. Chem.Eur. J. 2015, 21,
1803−1810.
(79) Janardanan, D.; Wang, Y.; Schyman, P.; Que, L., Jr.; Shaik, S.
Angew. Chem., Int. Ed. 2010, 49, 3342−3345.
(80) Pestovsky, O.; Stoian, S.; Bominaar, E. L.; Shan, X. P.; Munck,
̈
E.; Que, L., Jr.; Bakac, A. Angew. Chem., Int. Ed. 2005, 44, 6871−6874.
(81) Pestovsky, O.; Bakac, A. J. Am. Chem. Soc. 2004, 126, 13757−
13764.
(82) England, J.; Guo, Y. S.; Farquhar, E. R.; Young, V. G.; Munck,
̈
E.; Que, L., Jr. J. Am. Chem. Soc. 2010, 132, 8635−8644.
(83) Lacy, D. C.; Gupta, R.; Stone, K. L.; Greaves, J.; Ziller, J. W.;
Hendrich, M. P.; Borovik, A. S. J. Am. Chem. Soc. 2010, 132, 12188−
12190.
(84) Bigi, J. P.; Harman, W. H.; Lassalle-Kaiser, B.; Robles, D. M.;
Stich, T. A.; Yano, J.; Britt, R. D.; Chang, C. J. J. Am. Chem. Soc. 2012,
134, 1536−1542.
(85) Xue, G.; De Hont, R.; Munck, E.; Que, L., Jr. Nat. Chem. 2010,
̈
2, 400−405.
(86) Xue, G.; Geng, C.; Ye, S.; Fiedler, A. T.; Neese, F.; Que, L., Jr.
Inorg. Chem. 2013, 52, 3976−3984.
(87) Mehn, M. P.; Fujisawa, K.; Hegg, E. L.; Que, L., Jr. J. Am. Chem.
Soc. 2003, 125, 7828−7842.
(88) Mukherjee, A.; Martinho, M.; Bominaar, E. L.; Munck, E.; Que,
̈
L., Jr. Angew. Chem., Int. Ed. 2009, 48, 1780−1783.
(89) Tse, C. W.; Chow, T. W. S.; Guo, Z.; Lee, H. K.; Huang, J. S.;
Che, C. M. Angew. Chem., Int. Ed. 2014, 53, 798−803.
(90) Vincent, J. B.; Huffman, J. C.; Christou, G.; Li, Q.; Nanny, M.
A.; Hendrickson, D. N.; Fong, R. H.; Fish, R. H. J. Am. Chem. Soc.
1988, 110, 6898−6900.
L
DOI: 10.1021/ic502786y
Inorg. Chem. XXXX, XXX, XXX−XXX