Catalytic Alkyl Hydroperoxide and Acyl Hydroperoxide Disproportionation by a Nonheme Iron Complex

Article abstract of DOI:10.1021/acscatal.8b02882

Alkyl hydroperoxides are commonly used as terminal oxidants because they are generally acknowledged to be stable toward disproportionation compared with H₂O₂. We show that alkylperoxide disproportionation is effectively catalyzed by the [Fe(tpena)]²⁺ (tpena = N,N,N′-tris(2-pyridylmethyl)ethylendiamine-N′-acetate). A peroxidase-type mechanism, in other words, involvement of iron(IV)oxo species, is consistent with the rates and product distribution. Accordingly, O₂, tert-butanol, and cumyl alcohol are concurrently produced for substrates tert-butyl hydroperoxide and cumene hydroperoxide, respectively, in the presence of [Fe(tpena)]²⁺ with O₂ yields of 88% and 44%, respectively. Rate constants for initial O₂ production ([Fe] 0.005 mol %) were measured to 3.66(6) and 0.29(3) mM s^-1, respectively. Participating in the mechanism are spectroscopically detectable (UV-vis, EPR, resonance Raman) transient alkyl- and acyl-peroxide adducts, [Fe^IIIOOR(tpenaH)]²⁺ [R = C(CH₃)₃, C(CH₃)₂Ph, C(O)PhCl; T_1/2 = 30 s (5 °C), 20 s (5 °C), 1 s (-30 °C)] with their common decay product [Fe^IVO(tpenaH)]²⁺. Concurrently organic radicals proposed to be ROO^? were detected by EPR spectroscopy. A lower yield of O₂ at 23% with an initial rate of 0.10(3) mM s^-1 for the disproportionation of m-chloroperoxybenzoic acid is readily explained by catalyst inhibition by coordination of the product m-chlorobenzoic acid. Oxidative decomposition of the alkyl groups by a unimolecular β-scission pathway, favored for cumene hydroperoxide, competes with ROOH disproportionation. Despite the fact that the catalytic disproportionation is effective, external C-H substrates - when they are present in excess of ROOH - can be targeted and catalytically and selectively oxidized by ROOH using [Fe(tpena)]²⁺ as the catalyst.

Full text of DOI:10.1021/acscatal.8b02882

Research Article
Cite This: ACS Catal. 2018, 8, 9980−9991
pubs.acs.org/acscatalysis
Catalytic Alkyl Hydroperoxide and Acyl Hydroperoxide
Disproportionation by a Nonheme Iron Complex
Christina Wegeberg,†,‡ Wesley R. Browne, and Christine J. McKenzie*
‡
,†
†
Department of Physics, Chemistry and Pharmacy, University of Southern Denmark, Campusvej 55, 5230 Odense M, Denmark
Molecular Inorganic Chemistry, Stratingh Institute for Chemistry, University of Groningen, Nijenborgh 4, 9747 AG Groningen,
‡
The Netherlands
*
S Supporting Information
ABSTRACT: Alkyl hydroperoxides are commonly used as
terminal oxidants because they are generally acknowledged to
be stable toward disproportionation compared with H O . We
2
2
show that alkylperoxide disproportionation is effectively
catalyzed by the [Fe(tpena)]2 (tpena = N,N,N′-tris(2-
pyridylmethyl)ethylendiamine-N′-acetate). A peroxidase-type
mechanism, in other words, involvement of iron(IV)oxo
species, is consistent with the rates and product distribution.
+
Accordingly, O , tert-butanol, and cumyl alcohol are con-
2
currently produced for substrates tert-butyl hydroperoxide and
cumene hydroperoxide, respectively, in the presence of
2
+
[
Fe(tpena)] with O yields of 88% and 44%, respectively.
2
−
1
Rate constants for initial O production ([Fe] 0.005 mol %) were measured to 3.66(6) and 0.29(3) mM s , respectively.
2
Participating in the mechanism are spectroscopically detectable (UV−vis, EPR, resonance Raman) transient alkyl- and acyl-
III
2+
peroxide adducts, [Fe OOR(tpenaH)] [R = C(CH ) , C(CH ) Ph, C(O)PhCl; T = 30 s (5 °C), 20 s (5 °C), 1 s (−30
3
3
3 2
1/2
IV
2+
•
°
C)] with their common decay product [Fe O(tpenaH)] . Concurrently organic radicals proposed to be ROO were detected
by EPR spectroscopy. A lower yield of O at 23% with an initial rate of 0.10(3) mM s for the disproportionation of m-
−
1
2
chloroperoxybenzoic acid is readily explained by catalyst inhibition by coordination of the product m-chlorobenzoic acid.
Oxidative decomposition of the alkyl groups by a unimolecular β-scission pathway, favored for cumene hydroperoxide,
competes with ROOH disproportionation. Despite the fact that the catalytic disproportionation is effective, external C−H
substrateswhen they are present in excess of ROOHcan be targeted and catalytically and selectively oxidized by ROOH
using [Fe(tpena)]2 as the catalyst.
+
KEYWORDS: nonheme iron, alkylperoxide disproportionation, iron(IV)oxo, O evolution, catalysis
2
INTRODUCTION
aminopyridyl ligands activate a range of oxidants including
7 7,8 9,10 10
■
O , H O ,
PhIO,
and methyl-morpholine-N-oxide,
2
2
2
The most common motif for the active sites of nonheme iron
O activating enzymes is a single endogenous Asp or Glu
with reactivity patterns that are distinct from those found for
counterpart iron complexes based on neutral N4-, N5-, and
N6-donor-only ligands. In addition, for these systems, water
can be primed to act as the O atom donor for the production
2
donor accompanied by 1−3 His and exchangeable water
1
,2
molecules. It can be reasonably expected that the anionic
carboxylato coligand will tune physical properties, not least
redox potentials, with consequences for the activation of
coordinated terminal oxidants. Biology provides examples: the
shift from reversible O2 binding in hemerythrin to O2
activation and catalysis of oxidations by the related enzymes
11
of oxidizing iron(IV)oxo complexes by electrochemical or
12
Ce(IV) activation in aqueous solutions. Water is ultimately
also the terminal source for the O atom of an iron(IV)oxo
complex of a monocarboxylato chelating ligand that can
13
2
oxidize alcoholic substrates in gas phase reactions. These
results point to the carboxylate donor in the first coordination
sphere of iron catalysts and enzymes having an important role
in tuning oxidant activation.
methane monooxygenase and ribonucleotide reductase. The
same effects are observed by tuning of the axial endogeneous
amino acid donor for heme systems. Despite these biological
3
precedents, reports of biomimetic terminal oxidant activation
using iron complexes of multidentate ligands containing a
carboxylato donor remain relatively few by comparison to the
considerable volume of work over the last three decades for
iron complexes based on neutral aminopyridyl N-donor-only
The iron(III) complex of N,N,N′-tris(2-pyridylmethyl)-
ethylendiamine-N′-acetate (tpena, Scheme 1) is remarkable
Received: July 23, 2018
Revised: September 10, 2018
4
−6
ligands.
Our efforts to fill this gap have revealed that iron
complexes of glycyl-substituted tetra-, penta-, and hexadentate
©
XXXX American Chemical Society
9980
DOI: 10.1021/acscatal.8b02882
ACS Catal. 2018, 8, 9980−9991

ACS Catalysis
Research Article
[Fe(tpena)]2+ catalyzes highly effective alcohol oxidation by
Scheme 1. Selective Switching between Hydrogen Atom
Abstraction (HAT) or Oxygen Atom Transfer (OAT)
Pathways in Substrate Oxidation Using the
H O and, in the absence of alcohol or another substrate, H O
2
2
2
2
8
disproportionation. By contrast the N5/N6 ligand-supported
a
2+
Monocarboxylato Ligand, tpena
[Fe(Rtpen)] systems catalyze neither H O disproportiona-
2
2
tion nor alcohol oxidation by H O . In fact, methanol can be
2
2
III
2+
used as a solvent for observing transient [Fe OOH(Rtpen)]
where half-lives of minutes to hours (rt, room temperature) are
found for these complexes. In understanding the effects of the
introduction of a carboxylate group into the first coordination
sphere of the iron atom for the activation of terminal oxidants
located cis to this donor, we now turn to the activation of alkyl
2
+
hydroperoxides and peracids by [Fe(tpena)] .
Understanding the reactivity of alkyl peroxides is important
because they are used as initiators for curing (polymerization)
18
resins and two-component paints. This process involves
mixing alkyl peroxide solutions with the vinylester/styrene
resin containing any of a wide range of metal catalysts. Alkyl
peroxides are used specifically for these applications since they
are believed not to undergo disproportionation with the
consequent limitations on shelf life, as is the case for the
19
cheaper oxidant H O . Solutions containing cumene hydro-
2
2
peroxide are sold under the trademark Trigonox K-90 or
Trigonox 239a. Alkyl peroxides are also used extensively as
terminal oxidants in industrial- and laboratory-scale organic
a
When tpena acts as a pentadentate ligand, a dangling pyridyl group
20−28
syntheses.
Disproportionation of the alkyl peroxides has
functions as a second coordination sphere base. Undemanding steric
requirements of tpena mean that an additional endogenous
monodentate ligand (e.g., the oxidant PhIO is shown) can coordinate
to a high spin (S = 5/2) seven-coordinated iron atom.
iron(III) resting state is S = 5/2 and S = 1/2 for mer-[Fe(tpena)]
been sporadically suggested as a possible reason for reduced
29−32
yields in catalytic selective substrate oxidations;
however,
9
,12
The
the impact of unproductive reactions has never been evaluated
2
+
in detail nor the release of product O directly measured.
2
2
+
10
and fac-[Fe(tpena)] (not shown), respectively.
Here, we describe the spectroscopic detection of catalytically
2
+
competent iron(III)alkylperoxide adducts of [Fe(tpena)] ,
IV
2+
III
their decay product, [Fe O(tpenaH)] , along with concom-
itantly produced organic radicals that are present in working
solutions. ROOH dismutation is unexpected in contrast to
in its tunable reactivity for substrate oxidation by Fe (tpena)-
oxidant adducts or their iron(IV)-oxo derivatives: Highly
efficient catalytic reactions can be directed either exclusively
toward hydrogen atom abstraction (HAT) or oxygen atom
transfer (OAT) mechanisms depending on the choice of
terminal oxidant and reaction conditions. Both reaction types
are pertinent for rationalizing the impressive scope of reactivity
found for the nonheme enzymes, and the Fe-tpena system is
unique in its ability to effectively model both of these oxidative
pathways. At first sight this is surprising given that tpena is a
potentially coordinatively saturating hexadentate ligand.
However, structural flexibility on the part of the ligand and
geometrical and spin-state flexibility on the part of the iron
play crucial roles. The tpena can act as a hexadentate N5O
ligand in six- or seven-coordinated iron complexes ([Fe-
H O dismutation which is commonly catalyzed by metal salts
2
2
33−37
and complexes.
The quantification of the products of the
disproportionation of ROOH, and an analysis of competing
selective C−H oxidation reactions involving the R group or an
external C−H substrate, allows us to propose a common
2
+
mechanism for [Fe(tpena)] -catalyzed H O and ROOH
2
2
dismutation and external substrate C−H oxidations. Signifi-
cantly, the reactivity patterns imply a radical character for
IV
2+
[
Fe O(tpenaH)] , reinforcing our recent observation when
this species is produced electrochemically in the absence of
11
terminal chemical oxidants. In this case the reactivity of this
carboxylate-coordinated iron(IV)oxo toward C−H substrates
with increasing C−H bond dissociation energies parallels that
2
+
10
2+
9
2+,12
(
tpena)] , [FeOIPh(tpena)] , and [FeOH(tpena)]
)
•
of the hydroxo (HO ) radical, albeit with slower rates, no
and as a pentadentate N4O ligand in six-coordinated iron
2
+ ,
8
doubt because of size. The results presented here broaden
significantly the scope of reactivity observed for nonheme iron
models.
complexes ([FeCl(tpenaH)]
and [Fe (μ-O)-
2
2+,
9
(
tpenaH)2] ). In this latter group one specific pyridyl arm
is uncoordinated and protonated in all crystal struc-
8
−10,12,14
tures.
This allows for the creation of another
EXPERIMENTAL SECTION
important biomimetic motifa second coordination sphere
■
base.
Materials and Preparation. N,N,N′-tris(2-
pyridylmethyl)ethylendiamine-N′-acetic acid (tpenaH),
[Fe O(tpenaH) ](ClO ) (H O) , and [FeCl(Metpen)]PF
Iron(III)-hydroperoxide and iron(III)-peroxide adducts of
the iron complexes of the closely related 2-alkylpyridine-
substituted ethylenediamine-backboned neutral N5/N6 donor
ligands (Rtpen = N′-alkyl-N,N,N′-tris(2-pyridylmethyl)-
2
2
4
4
2
2
6
(Metpen = N-methyl-N,N′,N′-tris(2-pyridylmethyl)ethane-l,2-
9
,38,39
diamine) were prepared as previously described.
i
ethylenediamine, R = Me, Et, Bz, Pr, Pr, Ph, 2-methylpyridine)
Solutions of t-BuOOH (70% in H O), t-BuOOH in decane
2
as well as those for the aforementioned carboxylate-containing
(5.5 M), and cumylOOH (88% in cumene) were used as well
as m-CPBA (77% in m-CBAH). Iodometric titrations and
NMR spectroscopy were used to confirm the concentration
2
+
[
Fe(tpena)]
have been spectroscopically character-
7
,8,15−17
ized.
A significant difference in the chemistry of these
1
8
comparative N5/N6 and N5O systems is evident. For example,
and purity of the peroxide solutions. O−H O was supplied
2
9
981
DOI: 10.1021/acscatal.8b02882
ACS Catal. 2018, 8, 9980−9991

ACS Catalysis
Research Article
2
+
by Rotem Industries Ltd., and all other chemicals were
purchased from Sigma-Aldrich.
(tpena)] , and the alkyl peroxide was injected directly to
the solutions in the sample chamber as the resulting gas
evolution was simultaneously measured. The data were
recorded and processed using Quadstar 422 (Pfeiffer Vacuum,
Asslar, Germany). Volumetric measurements were performed
using a two-neck round-bottom flask with a stopcock-equipped
gas delivery tube connected to a gas-measuring buret (±0.1
mL). The peroxides were injected through a septum, and the
Generation of [Fe(OOR)(tpenaH)]2+ (R = C(CH ) ,
3
3
C(CH ) Ph, C(O)PhCl). [Fe O(tpenaH) ](ClO ) (H O)
3
2
2
2
4
4
2
2
was dissolved in acetonitrile, and the solution was allowed to
equilibrate for 15 min to maximize the concentration of the
solution-state monomeric species [Fe(tpena)]2+ that is derived
4
+
by dehydration of the hemihydrate [Fe O(tpenaH) ] .
2
2
Solutions of alkyl or acyl hydroperoxide were subsequently
evolved O was volumetrically measured as a function of time.
2
2
+
added to generate [Fe(OOR)(tpenaH)] . Spectroscopic
characterization with UV−vis, rRaman, EPR, and Mossbauer
spectroscopy were performed on 2−5 mM [Fe] solutions with
0 equiv of alkyl or acyl hydroperoxide. Catalytic experiments
The rate constants reported represent the mean value of
minimal double determinations which fall within ±5%. Analyte
solutions of peroxide (500 μL), 1.64 M KI(aq) (5 mL), 20%
̈
5
H
2
SO
iodometric titrations (±0.1 mL). An ammonium molybdate
solution [0.5 mL, prepared from (NH Mo 24 (9 g),
NH NO (24 g), and H O (5
4
(aq) (5 mL), and H O (25 mL) were prepared for
2
2+
were performed on 1 mM [Fe(tpena)] (d -MeCN) with 750
3
equiv of alkyl- or acyl-hydroperoxide and 750 equiv of benzyl
alcohol or toluene at rt. Volumetric measurements are
performed at rt on either 0.5 mM [Fe] with 1000 equiv of
oxidant or 25 μM [Fe] with 20 000 equiv of oxidant (time-
dependent detection).
4
)
6
O
7
(aq) (28−30%, 5 mL), NH
3
4
3
2
mL)] and a starch solution (1 g/100 mL) were added as
catalyst and indicator, respectively. The mixture was titrated
with 0.3 M Na
times.
S O (aq), and the analysis was repeated three
2 2 3
Instrumentation. UV−vis spectra were recorded in 1 cm
quartz cuvettes either on an Agilent 8453 spectrophotometer
with an UNISOKU CoolSpeK UV USP-203 temperature
controller or with an Analytikjena Specord S600 with a
Quantum Northwest TC 125 temperature controller. Raman
spectra were recorded in 1 cm quartz cuvettes with
temperature control using a Flash300 instrument (Quantum
Northwest) either at 532 nm (300 mW at source, Cobolt
Lasers) or at 785 nm using a PerkinElmer RamanFlex fiber
optic coupled Raman spectrometer (90 mW at sample). Data
were recorded and processed using Solis (Andor Technology)
with spectral calibration performed using the Raman spectrum
of MeCN/toluene (50:50 v/v). EPR spectra (X-band) were
recorded on a Bruker EMX Plus CW spectrometer
RESULTS AND DISCUSSION
■
Iron(III)-alkyl Peroxides, [Fe(OOR)(tpenaH)]2 R =
+
C(CH ) and CH(CH ) Ph. The solid-state precursor for the
3
3
3 2
solution-state chemistry described below is the dark brown
9
[
(tpenaH)Fe(μ-O)Fe(tpenaH)](ClO ) . On dissolution this
4 4
oxo-bridged complex equilibrates with monomeric species: a
III
2+ 12
hydrate, [Fe (OH)(tpenaH)] ,
[
and a dehydrate,
III
2+ 10
Fe (tpena)] . Cleavage of the oxo-bridge is accompanied
by a color change from yellow−brown (λ 258 nm) to red−
max
orange (λ 360 nm) in acetonitrile and takes ca. 15 min to
max
10
equilibrate at rt. The addition (50 equiv) of tert-butyl
hydroperoxide (t-BuOOH, 70% aq solution) or cumene
hydroperoxide (cumylOOH, 88% solution) to these equili-
brated solutions triggers an immediate color change to purple
(λmax 558 nm) or red (λmax 531 nm), respectively (Figure 1a,
Table 1). Gas evolution is concomitant. This is predominantly
(
modulation amplitude, 10 G; attenuation, 10 dB) on frozen
solutions at 110 K. eview4wr and esimX were used for
4
0
1
13
simulation. H NMR (400.12 MHz) and C NMR (100.61
MHz) spectra were recorded on a Bruker Avance III 400
spectrometer at ambient temperature. Chemical shifts are
denoted relative to the residual solvent peak (d -MeCN, δ =
O , vide infra. The chromophores are ascribed to the transient
2
2
+
3
H
species [Fe(OO-t-Bu)(tpenaH)] and [Fe(OOcumyl)-
2
+
1
.94 ppm and δ = 1.32 ppm), and the catalytic experiments
(tpenaH)] , respectively. With reference to the crystal
8
C
2
+
were performed directly in MeCN-d on 1 mM [Fe] solutions.
structures of [Fe(Cl)(tpenaH)] , [(tpenaH)Fe(μ-O)Fe-
9
3
4
+
2+ 12
Substrate conversion is based on integrals of the aromatic
protons. Headspace FTIR spectra were recorded in sealed 1
cm quartz cuvettes on a JASCO FT-NIR/MIR-4600
(tpenaH)] , [V(O)(tpenaH)] ,
and [Cr(O )-
2
2+
14
(tpenaH)] , the incoming ROOH is proposed to form a
2
+
“charge-separated” adduct with [Fe(tpena)] ; i.e., the
alkylperoxide is deprotonated and bound to iron(III), and
the proton protonates an uncoordinated pyridine group. While
it seems that an initial dehydration of the starting hemihydrate
is necessary to observe these transients, counterintuitively, the
formation of the iron-alkylperoxide adducts is dependent on
the presence of a small amount of water. [Fe(OO-t-
spectrometer, and the CO signal was quantified as previously
2
8
described. Mo
̈
ssbauer spectra were recorded at 80K on a
conventional spectrometer with alternating constant acceler-
ation of the γ-source. Isomer shifts are denoted relative to α-
iron at 298 K, and the sample temperature was maintained
constant in an Oxford Instruments Variox cryostat. The γ-
source (57Co/Rh, 1.8 GBq) was kept at room temperature.
2
+
Bu)(tpenaH)] is not detected if the t-BuOOH is supplied
in decane (5.5 M). If, however, 4 equiv of water is introduced
to the equilibrated acetonitrile solutions containing mono-
40
The mf.SL package was used to fold the spectra hereby
merging the two linear halves of the raw data and to eliminate
the parabolic background as well as fitting the data. ESI-MS
spectra were recorded in high-resolution positive mode with a
Bruker microTOF-QII mass spectrometer. MIMS spectra were
recorded using a Prisma quadrupole mass spectrometer
2
+
2+
meric [Fe(tpena)] /[Fe(OH)(tpenaH)] immediately prior
to the addition of the decane solution of t-BuOOH, the
formation of the iron alkyl peroxide species is reinstated. These
observations suggest that it is the mononuclear hydrate,
2
+
2+
(
(
Pfeiffer Vacuum, Asslar, Germany). A flat sheet membrane
250 μm) of polydimethylsiloxane (Sil-Tec sheeting, Technical
[Fe(OH)(tpenaH)] , rather than [Fe(tpena)] that is the
immediate precursor for reaction with ROOH, and that the
reaction involves substitution of a hydroxo ligand by the
incoming alkylperoxide (Scheme 2) rather than an alkylper-
oxido displacement of a chelating pyridine group as would be
Products, Decatur, GA) separated the vacuum chamber (1 ×
−
6
1
0
mbar) from the solution in the sample chamber (total
volume 2.5 mL), which was equipped with magnetic stirring.
The reaction chamber was filled with solutions of [Fe-
2
+
the case if [Fe(tpena)] was the immediate precursor. We
9
982
DOI: 10.1021/acscatal.8b02882
ACS Catal. 2018, 8, 9980−9991

ACS Catalysis
Research Article
Table 1. Spectroscopic Properties for
III
2+
[
Fe OOR(tpenaH)]
λmax
[nm]
ν
Fe−O
ν
O−O
−
1
−1
R
[cm
]
[cm
786
782
]
g-values
C(CH3)3
CH(CH ) Ph
C(O)m-ClPh
558
531
565
675
684
a
2.20, 2.12, 1.97
2.19, 2.12, 1.97
2.23, 2.18, 1.95
3
2
a
a
Lifetime too short for measurement at −40 °C.
have previously shown that two diastereoisomers (a mer and
2
+
fac in terms of the three pyridine donors) for [Fe(tpena)]
8
,10
coexist in solution. Hydration will convert them to the same
conformer, and the simplicity of the spectroscopy, vide infra,
supports the formation of only one diastereoisomer for the
alkylperoxide adducts. Our representation of [Fe(X)-
2
+
(
tpenaH)] , X = OH, OO-t-Bu, OOcumyl (Scheme 2),
depicts therefore the diastereoisomer containing the penta-
dentate tpena conformation analogous to that found in all the
n+ 8−10,12,14
known crystal structures of {MX(tpenaH)} .
In
these, the glycyl donor is located cis to the exogenous ligand
which in turn is trans to the amine donor associated with the
8
,9,12
dipyridylamine moiety.
The iron(III)-alkylperoxide species are short-lived at room
temperature (seconds); however, a T of 30 and 20 s for
1
/2
2
+
2+
[
Fe(OO-t-Bu)(tpenaH)] and [Fe(OOcumyl)(tpenaH)] ,
respectively, is measured at 5 °C (Figure S1). At −30 °C
both species can be detected for several minutes using the
same ratios and concentrations of reagents. Solutions
containing ca. 10% water gave the maximum yield in terms
of the lifetime of the alkyl peroxides, which increased from 3 to
6
min at −15 °C. EPR spectra of frozen solutions show
rhombic signals with g = 2.20, 2.12, and 1.97 and g = 2.19,
2
+
2
.12, and 1.97 for [Fe(OO-t-Bu)(tpenaH)] (Figure 1b) and
2
+
[Fe(OOcumyl)(tpenaH)] (Figure S2), respectively, indicat-
ing low-spin iron(III) (S = 1/2) species. Resonance Raman
−
1
spectra show enhanced bands at 675 and 786 cm for
2
+
−1
[
[
Fe(OO-t-Bu)(tpenaH)] and at 684 and 782 cm for
2
+
Fe(OOcumyl)(tpenaH)] (Figure 1c). On the basis of a
comparison with the spectra of previously reported low-spin
mononuclear nonheme iron(III) alkyl peroxide species, these
bands can be associated with the Fe−O and O−O bonds,
41,42
respectively.
The visible absorbance bands are red-shifted
compared to the hydroperoxide-derived species [Fe(OOH)-
2
+
(
[
tpenaH)] (λmax 520 nm). The O−O stretch of the
2+
Fe(OOR)(tpenaH)] complexes was not shifted compared
to that observed for [Fe(OOH)(tpenaH)] (788 cm ),
whereas the Fe−O stretches are found at 70 cm higher
wavenumbers suggesting a comparatively stronger Fe−O bond.
A similar trend was noted for the Raman shift of the
2
+
−1
8
−
1
Figure 1. Spectroscopic data of the [Fe(OOR)(tpenaH)]2+ generated
2
+
by addition of 50 equiv of t-BuOOH or cumylOOH to [Fe(tpena)]
in MeCN. (a) UV−vis absorption spectra (5 °C) of [Fe(OO-t-
2
+
homologous [Fe(OOR)(TPA)(solvent)] (R = H, t-Bu)
41,43
−1
2+
2
+
Bu)(tpenaH)] (blue, [Fe] = 2 mM) and [Fe(OOcumyl)-
pair.
A band at 484 cm for [Fe(OO-t-Bu)(tpenaH)]
(
tpenaH)]2+ (red, [Fe] = 4 mM). Inset: photograph of frozen
can be assigned to a combined O−C−C/C−C−C bending
2
+
43
samples of [Fe(OO-t-Bu)(tpenaH)]
and [Fe(OOcumyl)-
vibration from the t-Bu group.
2+
(
tpenaH)] . (b) EPR spectrum (black) recorded on a frozen
Evidence for Homolytic FeO−OR Bond Cleavage in
2
+
solution of [Fe(OO-t-Bu)(tpena)] in MeCN at 110 K. Microwave
Alkylperoxide Adducts. The decay of the [Fe(OOR)-
2
+
frequency 9.314 212 GHz. Fitted data for [Fe(OO-t-Bu)(tpenaH)]
and the t-BuOO radical are shown in blue and green, respectively.
2+
(
tpenaH)] complexes proceeds by homolytic cleavage of the
•
III
2
+
Fe O−OR bond. This conclusion is supported by the
(c) Resonance Raman spectra of [Fe(OO-t-Bu)(tpenaH)] (blue,
11,12
IV
evolution of the characteristic 730 nm band
for [Fe O-
λexc = 785, −25 °C, 3 mM) and its decay product (purple, λ = 785),
exc
2+
2
+
(tpenaH)] (Figure 2a). Combined time-resolved UV−vis
and [Fe(OOcumyl)(tpenaH)] (red, λ = 532, −25 °C, 2 mM) and
exc
III
and rRaman spectroscopy of the decaying [Fe OOR-
its decay product (orange, λexc = 532). * = solvent band from MeCN,
2
+
#
= bands from either t-BuOOH or cumylOOH. The spectra are
(tpenaH)] species shows that a resonance-enhanced band
−
1
−1
normalized to the solvent band at 919 cm .
at 863 cm (Figure 1c) can be correlated to the 730 nm band
9
983
DOI: 10.1021/acscatal.8b02882
ACS Catal. 2018, 8, 9980−9991

ACS Catalysis
Research Article
2+
Scheme 2. Interconversion between mer-py and fac-py Diastereoisomers of [Fe(tpena)] via Their Common Hydrated
3
3
2+
Derivative and Substitution of H O in [Fe(OH)(tpenaH)] by ROOH
2
Figure 2. Conversion of [Fe(OO-t-Bu)(tpenaH)]2+ to [Fe O(tpenaH)] (50 equiv of t-BuOOH, [Fe] = 2 mM, 5 °C). (a) Time-dependent UV−
IV
2+
vis absorption spectra. Inset: time-trace for absorbance at 558 nm (blue) and 730 nm (green). (b) Time dependence of absorbance at 558 and 730
2
+
IV
2+
nm as portions of 50 equiv of t-BuOOH are added every 100 s to regenerate [Fe(OO-t-Bu)(tpenaH)] and [Fe O(tpenaH)] .
in the UV−vis spectrum, and it is therefore assigned to a ν
hydrates, an iron(III)alkylperoxide, and an iron(IV)oxo species
is a Mossbauer spectrum (Figure S3) of [Fe(OH)(tpenaH)]
̈
and 50 equiv of cumylOOH, frozen 20 s after mixing. This can
FeO
2
+
stretch. This value is higher than Raman shifts previously
reported for nonheme iron(IV)oxo species which typically
−
1
2+
−1
occur in the range 790−850 cm for both S = 1 and S = 2
be fitted to [Fe(OH)(tpenaH)] (δ = 0.14 mm s , ΔE =
4+
Q
4
4
−1
−1
systems, suggesting an FeO bond that is stronger than
typically found for nonheme iron(IV)oxo systems. We
speculate that the bond possesses partial triple bond character
2.08 mm s , 22%), [Fe O(tpenaH) ] (δ = 0.43 mm s ,
2 2
−
1
IV
2+
ΔE = 1.60 mm s , 26%), and [Fe O(tpenaH)] (δ = 0.00
Q
−
1
−1
mm s , ΔE = 0.90 mm s , 24%) which have been
Q
4
5−47
analogous to that proposed for vanadyl complexes.
Significantly, isotropic signals at g = 2.02 (Figure 1b) and g
independently characterized previously by Mossbauer spec-
̈
12
troscopy. This has allowed for assignment of a fourth species
to the low-spin iron(III)alkylperoxo complex [Fe(OOcumyl)-
2
+
=
[
2.00 (Figure S2) for [Fe(OO-t-Bu)(tpenaH)] and
2+
2+
−1
−1
Fe(OOcumyl)(tpenaH)] , respectively, overlap with the
(tpenaH)] (δ = 0.20 mm s , ΔE = 2.00 mm s , 28%).
These parameters are quite similar to those for [Fe(OOH)-
(tpenaH)] (δ = 0.21 mm s , ΔE = 2.08 mm s ). After
their decay, all species, [Fe(OOR)(tpenaH)] , [Fe O-
Q
signals from the low-spin Fe(III)-hydroperoxide adducts in
the EPR spectra. These are due to production of an organic
2
+
−1
−1
8
Q
III
2+
IV
radical. Homolysis of the Fe O−OR bond produces the
IV
2+
2+
immediate sister products [Fe O(tpenaH)] and the alkoxide
(tpenaH)] , and the organic radicals, are regenerated by
•
•
radicals O-t-Bu or OC(CH ) Ph, respectively. These latter
addition of further portions (50 equiv) of alkyl hydroperoxides
(Figure 2b). The repeatable cyclability is evidence that tpena is
not oxidatively degraded, and this stands in contrast to the
3
2
species can be expected to be too short-lived for detection, and
in the presence of excess ROOH they can be expect to abstract
•
•
8,49
2+
a peroxy H atom to give OO-t-Bu and OOC(CH ) Ph,
dismutation reaction of H O catalyzed by [Fe(tpena)]
3
2
2
2
4
respectively, along with the derived alcohol.
The isotropic
where tpena will start decomposing when the substrate H O
2 2
EPR signals are therefore assigned to these generated peroxy
radicals rather than the directly produced alkoxide radicals.
Signals due to the Fe(IV)oxo species and the organic radicals
is not present in a large excess, or alternatively, if an easily
oxidizable sacrificial substrate is not provided. In the present
case the organic groups of the alkyl hydroperoxides offer built-
in sacrificial exogenous substrates to compete with the alkyl
hydroperoxide dismutation and herewith effectively compet-
itively inhibit tpena degradation. The solid-state precursor
8
2
+
are present concurrently with those for [Fe(OOR)(tpenaH)]
Figures 1b and 2) indicating similar stabilities for the two
iron-based transients. Supporting the coexistence of iron(III)
(
9
984
DOI: 10.1021/acscatal.8b02882
ACS Catal. 2018, 8, 9980−9991

ACS Catalysis
Research Article
Figure 3. Detection of evolved O and CO , when ROOH is added to [Fe(OH)(tpenaH)] in MeCN. (a) MIMS spectra of [Fe(OH)(tpenaH)]2+
2
+
2
2
(
black) and with the addition of 50 equiv of cumylOOH after 2 min (red) and 12 min (blue). The broad peak between m/z = 35 and 42 is due to
acetonitrile. The m/z = 43 signal originates from unreacted cumylOOH passing through the membrane. Inset: time dependence of the ion current
for the ions O2 (m/z 32). [Fe] = 0.5 mM. (b) Time-trace of the O evolution (1556 cm ) with headspace Raman spectroscopy (λ = 532 nm)
•
+
−1
2
exc
with the addition of two portions of 100 equiv of t-BuOOH. [Fe] = 2 mM. The arrows indicate addition of t-BuOOH. Build-up of pressure in the
sealed cuvette prevented further additions of the peroxide to the closed system; however, the gas evolution can be reinitiated more than 10 times
using an open vessel. (c) Time-dependent volumetric detection of the gas release with the addition of 20 000 equiv of oxidant to 25 μM
2
+
[
Fe(OH)(tpenaH)] solutions in MeCN: t-BuOOH (blue), cumylOOH (red), and m-CPBA (green). (d) Time-resolved headspace FTIR
−
1
spectroscopy showing the evolution of CO over time after addition of 1000 equiv of cumylOOH. Time-dependence of absorbance at 2360 cm .
Red, 1000 equiv of cumylOOH; blue, 1000 equiv of t-BuOOH. [Fe] = 0.5 mM. The bands at 2253 and 2292 cm are due to acetonitrile.
2
−
1
[
(tpenaH)Fe(μ-O)Fe(tpenaH)](ClO ) can be quantitatively
water is the sole solvent; however, in this case the iron-based
transients cannot be observed. The oxo-bridged complex
dominates the speciation in these solutions, and this
observation supports the conclusion that this cannot directly
activate ROOH and corroborates the deduction that the
catalytically competent resting-state complex is [Fe(OH)-
4
4
recovered after the consumption of thousands of equivalents of
alkyl peroxides.
Catalytic Alkyl Hydroperoxide Disproportionation
and Competing Radical Reactions. Simultaneous with
detection of [Fe (OOR)(tpenaH)] and [Fe O(tpenaH)] ,
III
2+
IV
2+
2
+
colorless gas production is visible as bubbles when the alkyl
(tpenaH)] . The addition of 1000 equiv of t-BuOOH (aq
2
+
2+
peroxides are added in excess to [Fe(OH)(tpenaH)] in
acetonitrile (e.g., 50 equiv of ROOH to 0.5 mM [Fe]) (see
Video S1). Membrane-introduction mass spectrometry
70%) to a [Fe(tpena)] solution (3 mL 0.5 mM) results in the
production of 15.5(2) mL of gas (Table 2). The addition of a
second portion of 1000 equiv results in a release of another
(
MIMS) (m/z 32, Figure 3a) and headspace Raman
15.5(2) mL. This includes a very small amount of CO , vide
2
−
1
spectroscopy (OO band at 1556 cm , Figure 3b)
infra, and when this is accounted for, the yield of O is
2
confirmed that this is predominantly O . That this was the
approximately 88% (assuming that 2 mol of t-BuOOH are
2
result of water oxidation of adventitious water, and that
provided in the aqueous solutions of peroxides, was excluded
because of inappropriately large amounts of O2 released
required to yield 1 mol of O ). The equivalent experiment
2
using cumylOOH gives a lower yield of O at 44% (8.0 mL).
2
Concurrent tert-butanol and cumyl alcohol production as the
major organic products, respectively, was confirmed using
1
8
compared to the water available, and the lack of
O
50
18
incorporation in the evolved O2, when O-labeled water
was added to mixtures. In addition, this experiment
demonstrates that the rate of ROOH disproportionation is
faster than any potential O atom exchange between water and
NMR spectroscopy to follow in situ reactions in d -MeCN.
3
Spectra recorded 1 h after the addition of alkyl hydroperoxide
showed that the t-BuOOH and cumylOOH were fully
consumed, and gas evolution had ceased (Figures S4 and S5).
IV
2+
51
the [Fe O(tpenaH)] : The evolved O derives exclusively
Sugimoto and Sawyer have shown that simple iron
2
from the disproportionation of the alkyl hydroperoxides, and
perchlorates catalyze H O dismutation in acetonitrile under
2
2
the reaction is catalytic. O , in lower yields, evolves also when
anhydrous conditions. In a significant contrast they do not
2
9
985
DOI: 10.1021/acscatal.8b02882
ACS Catal. 2018, 8, 9980−9991

ACS Catalysis
Research Article
Table 2. Volumetric Measurements of O and CO Were
Performed on 0.5 mM [Fe(OH)(tpenaH)] with the
adducts have been characterized for this and related Rtpen (R
2
2
2+
15−17,53−55
= Me, Bz, Et, Pr, i-Pr, Ph) species.
Using the same
+
Addition of Either t-BuOOH, CumylOOH, or m-CPBA
ratios of [FeCl(Metpen)] :t-BuOOH and conditions as those
above for the Fe-tpena system ([Fe] = 0.5 mM, 1000 equiv of
tBuOOH) we found no detectable disproportionation of t-
a
(
1000 equiv)
yield of O
ratio of ROOH and
2
b
BuOOH; i.e., over 30 min, no O had evolved.
adjusted for CO
release
products,
2
2
Vgas [mL]
ROOH:ROH:ketone
Volumetric monitoring of the release of the O as a function
2
t-BuOOH
cumylOOH
m-CPBA
15.5 ± 0.2
8.0 ± 0.4
4.0 ± 0.4
15.1
7.6
3.7
0:0.95:0.05
0:0.70:0.30
0.50:0.50:0
of time revealed that the initial rate constants are 3.66(6) and
−
1
0.29(8) mM s (3 mL of 25 μM [Fe] and 20 000 oxidant/
substrate) for the addition of t-BuOOH and cumylOOH,
respectively; hence, the initial rate is 12 times faster for t-
a
The ratio of organic products was determined by in situ NMR
experiments using [Fe(OH)(tpenaH)] (1 mM in MeCN-d ) and
the addition of 750 equiv of oxidant. After 1 h of reaction time in the
absence of external substrates.
2
+
BuOOH (Figure 3c). The O release ceases after 5 min when
3
2
b
employing t-BuOOH, whereas this takes several hours when
cumylOOH is used under the same concentrations. At
constant complex concentration saturation kinetics were not
observed on the addition of 6500−30 000 equiv of t-BuOOH,
and the relationship between catalyst concentration and the
rate constants is nonlinear: A doubling of the iron
concentration results in a more than doubling of the rate
promote t-BuOOH or meta-chloroperoxybenzoic acid (m-
52
CPBA) disproportionation. Our checks as a control for the
present context verify this earlier work, also when water is
present. In lieu of a known catalyst for t-BuOOH
disproportionation with which to compare the efficiency of
constants (Figures S6−S8). The initial lag time before the O
2
2
+
[
Fe(tpena)] , we tested [FeCl(Metpen)]PF for its ability to
releases is rationalized by the need to establish a steady-state
6
IV
2+
catalyze this reaction. This complex is a highly pertinent
system for comparison since with its ethylene backbone it is
structurally related, and importantly, iron(III) hydroperoxide
concentration of [Fe O(tpenaH)] and the alkoxy radical. As
•
•
mentioned, both t-BuO and cumylO will abstract an H atom
from the excess t-BuOOH and cumylOOH, respectively, to
Scheme 3. Catalytic Alkyl Hydroperoxide Disproportionation Using the [Fe(tpena)]2+ Systema
a
III
2+
Entry in the cycle occurs by the charge separation addition of an alkyl hydroperoxide to form [Fe OOCR (tpenaH)] , see Scheme 2. (a)
3
IV
2+
IV
2+
Production of [Fe O(tpenaH)] and alkoxide radical by FeO−OCR homolysis. (b) Reaction of alkylperoxide radical with [Fe O(tpenaH)] to
form the putative [FeOOOCR (tpenaH)] (c) β-Scission of alkoxy radical. (d) Production of alkylperoxide radical and product alcohol or
carboxylic acid. (e) Reaction of alkyl hydroperoxide with [Fe O(tpenaH)] to give alkylperoxide radical. (f) Ligand exchange of the alcohol
adduct to give the alkyl hydroperoxide adduct of [Fe (tpena)] . (g) O expulsion and formation of the alkoxy adduct of the resting-state catalyst,
3
2
+
3
IV
2+
III
2+
2
III
2+
[
Fe OR(tpenaH)] .
9
986
DOI: 10.1021/acscatal.8b02882
ACS Catal. 2018, 8, 9980−9991

ACS Catalysis
Research Article
•
form the EPR detectable secondary radicals t-BuOO and
3 path g). The beauty of this proposal is that all the detected
products are accountable by the pathways in Scheme 3.
Concurrently with the alkyl hydroperoxide decomposition
•
cumylOO . Acetone and acetophenone, respectively, are
detected by NMR spectroscopy as the expected minor organic
products from the disproportionation reactions (Table 2,
Figures S4 and S5). The formation of these ketone products
occurs by β-scission reactions of the alkoxyl radical. In the case
of t-BuOOH, the acetone formation amounts to ca. 5% of the
organic products, whereas for cumylOOH the yield of
acetophenone is greater at approximately 30%. This is readily
and generation of O (m/z = 32), an increase of the peaks at
2
m/z 29, 30, 31, and 44 in MIMS spectra is observed suggesting
formation of one-carbon-containing molecules (Figure 3a).
Decomposition of tpena as the carbon source is excluded,
because of (i) the repeatability over many cycles without loss
of catalytic efficiency, (ii) the fact that [(tpenaH)Fe(μ-
O)Fe(tpenaH)](ClO ) can be quantitatively recovered, and
•
rationalized by the lower stability of cumylO compared with t-
4
4
•
56
BuO . Hence the β-scission pathway, being unimolecular, is
more pronounced when cumylOOH is a substrate, and
cumylOOH disproportionation is less favored than the
disproportionation of t-BuOOH. The addition of further
portions of alkyl hydroperoxides reinitiates O2 evolution
(iii) the ratio of the alkyl derived products of the reactions.
Acetonitrile oxidation as the source is also discounted because
of the relatively high BDEs for H−CH CN and H C−CN
2
3
−
1
60
(406 and 522 kJmol ) as well as the obvious lack of
correlation to concentration. The remaining and most
plausible option is that the methyl radical produced via the
β-scission pathway is the source (Scheme 3, path c). This
methyl radical can react with O2 (which saturates the
solutions) to give carbon monoxide, carbon dioxide, form-
(
Figure 3b) concomitant with the visible regeneration of the
iron(III)-peroxide species (Figure 2b). The catalyst is robust.
As mentioned in the Introduction, the cause of low yields in
redox-active metal (Rh, Fe)-complexes catalyzing oxidations of
organic substrates by alkyl hydroperoxides have sporadically
been ascribedwithout experimental verificationto a
61,62
aldehyde, and methanol.
The exact ratio between these
products will be pressure- and temperature-dependent;
however, the growth of ions at m/z 29, 30, 31, and 44 in
the MIMS spectra for both alkyl hydroperoxides is consistent
2
9−31
putative background dismutation.
Aside from the
oxidation of organic substrates, alkyl hydroperoxides are
routinely used in the preparation of high-valent metal
complexes. In these syntheses it would seem that ample
opportunities for the observation of disproportionation have
been available to the many workers in the area of high-valent
biomimetic first-row transition metal complexes: The dearth of
observations of this specific reaction is surprising. Over two
decades ago Caudle et al. showed that Mn (2-OHsalpn) (2-
OHsalpn = 1,3-bis(salicylideneamino)-2-propanol) reacts
with t-BuOOH to give a Mn (O)Mn complex. These
authors assumed that the concurrently produced t-butoxy
radical (undetected) reacts with excess t-BuOOH to give the t-
butylperoxyl radical (eq 1). Subsequent coupling of two t-
butylperoxyl radicals was proposed as the source of the
with formation of formaldehyde, methanol, and CO (Figure
2
S9). Crucially, the quantity of CO is directly proportional to
2
the amount of ROOH that was added initially. The yield was
quantified by headspace FTIR spectroscopy to be approx-
imately 1 CO per 30 ROOH consumed (Figure 3d). This
2
CO accounts for a minor proportion of the gas evolved during
2
III
2
2
the decomposition, and as mentioned above, the determination
57
of O yields has taken this into account.
2
III
IV
Competitive and Selective C−H Substrate Oxidation
Reactions. If toluene (750 equiv) is added before addition of
2
+
t-BuOOH (750 equiv) to [Fe(OH)(tpenaH)] , NMR
spectroscopy verifies that it is oxidized to benzaldehyde
(
Figure S10: δ (CHO) = 10.00 ppm and δ (CHO) = 193.7
H C
observed O (eq 2). O evolution from alkyl hydroperoxides
2
2
ppm). After 1 h only trace amounts of t-BuOOH remain, with
t-BuOH being the major product with traces of acetone. On
the basis of the comparison of the integrals of the five aromatic
protons in toluene and benzaldehyde, a 10% conversion has
occurred (TON = 38). Benzylalcohol, which is an intermediate
on the reaction pathway from toluene to benzaldehyde, is not
detected, presumably because of its facile onward oxidation.
Benzoic acid is not detected suggesting that benzaldehyde is
not easily oxidized under these conditions. Control reactions
show that t-BuOOH cannot oxidize toluene, benzyl alcohol, or
2
+
was noted also when they are reacted with [Fe(tpa)X2] and
Fe O(tpa) (H O) ] (X = Cl, Br; tpa = tris(2-pyridylmethyl)-
[
2
2
2
2 4
5
8
amine) ). In these cases, peroxy H atom abstraction from
ROOH to produce ROO by the iron(III)-tpa complex with
concurrent reduction of the iron to the +2 state was proposed
to initialize the reaction. Again, O extrusion after the ROO
coupling reaction in eq 2 was proposed. In contrast to the
tpena-based system here, high-valent iron is not involved. O2
was not quantified in either of these studies so we cannot
compare yields or rates. Neither of these reactions were
studied in detail from the viewpoint of actual turnover or
catalysis.
•
•
2
2
+
benzaldehyde in the absence of [Fe(tpena)] .
2
+
The [Fe(OH)(tpenaH)] /ROOH system furnishes highly
selective catalysis of the oxidation of appropriate C−H bonds
under appropriate limiting reagent conditions. A conversion of
•
•
O‐t‐Bu + t‐BuOOH → t‐BuOH + OO‐t‐Bu
(1)
4
7% to benzaldehyde after 1 h is measured when benzyl
•
•
alcohol is added to acetonitrile solutions in equivalent amounts
2
OO‐t‐Bu → t‐BuOOOO‐t‐Bu → O + 2 O‐t‐Bu (2)
2
to the subsequently added t-BuOOH (PhCH OH:t-BuOOH:
2
•
2+
While the coupling of OOR radicals (eq 2) might be one
pathway to O , the reaction mixtures contain an abundance of
competing substrates to potentially quench OOR, not least
the oxy-radical-like [Fe O(tpenaH)] . We propose that the
major O releasing step could predominantly involve the
coupling of [Fe O(tpenaH)] , with the abundantly present
[Fe(OH)(tpenaH)] = 750:750:1, TON = 352). Volumetric
2
detection of the gas release under these conditions shows a
decrease to 8.0 mL compared with the 15.5 mL that could be
recovered if an external substrate is not added. Quantification
of the CO shows that the same amount of CO is produced as
•
IV
2+ 11
2
2
2
IV
2+ 11
that in the absence of benzyl alcohol; thus, the β-scission
pathway is not suppressed in any significant way, and an
•
OOCR . Extrusion of O2 , from
a
putative
3
III
2+
[
Fe OOOCR (tpenaH)] reminiscent of the Russell mech-
anism’s dialkyltetroxide intermediate in eq 2, will collapse to
O and an iron(III)-alkoxide, [Fe OCR (tpenaH)] (Scheme
adjusted yield of 44% O in the presence of equimolar amounts
3
2
59
of t-BuOOH and benzyl alcohol can be determined. This
correlates nearly perfectly with the 47% conversion of the
III
2+
2
3
9
987
DOI: 10.1021/acscatal.8b02882
ACS Catal. 2018, 8, 9980−9991

ACS Catalysis
Research Article
benzyl alcohol to benzaldehyde detected by NMR spectrosco-
py. Hence, PhCH OH (BDE for methylene C−H, 331 kJ
2
−
1
60
−160
mol
) and t-BuOOH (BDE t-BuOO-H, 352 kJ mol
compete equally as substrates for [Fe O(tpenaH)] despite
)
IV
2+
the lower BDE for PhCH OH. This implies a polar effect on
2
the HAT process promoting t-BuOOH as an equal substrate. It
is easy to envisage that the dangling pyridinium arm might
participate through a supramolecular H bonding interaction
with the distal O atom of the t-BuOOH. A 1:5 ratio of t-
BuOOH:PhCH OH reduced the gas release further to a 25%
2
of the yield found in the absence of benzyl alcohol. The in situ
oxidants responsible for the oxidation of toluene and benzyl
IV
2+
•
alcohol are proposed to be all of [Fe O(tpenaH)] , RO , and
•
ROO . These will all abstract H atoms from the aliphatic C−H
bonds of these substrates to regenerate [Fe OH(tpenaH)]
III
2+
and ROOH, and produce ROH, along with benzyl radicals
with which the generated O might react directly.
2
O Evolution Is Suppressed in the Reaction of Fe(III)-
2
III
tpena with a Peracid. The reaction of [Fe (OH)-
tpenaH)] with m-CPBA (50 equiv) generates [Fe O-
tpenaH)] more rapidly compared to H O2 and alkyl
2
+
IV
(
(
2
+
2
hydroperoxides as made evident by the appearance of an
absorption band at 730 nm, detected only at low temperatures
IV
2+
(
−30 °C) where [Fe O(tpenaH)] shows a half-life of
around 20 s. At −40 °C a barely perceptible shoulder at 565
nm is revealed (<2 s, Figure 4a). This is likely due to
Fe(OOC(O)m-PhCl)(tpenaH)] . The electron-withdrawing
2+
[
properties of the chloro-substituted aromatic acylated system
produces a more labile O−O bond than that observed for t-
BuOOH and cumylOOH. The same tendency in O−O bond
stability has been observed for iron(III) porphyrin complexes
63
of m-CPBA and t-BuOOH. The EPR spectra (Figure 4b)
III
recorded on flash-frozen samples of mixtures of [Fe (OH)-
2
+
(
tpenaH)] and 50 equiv of m-CPBA show three distinct
eff
signals: a high-spin iron(III) (g = 8.2, 5.2, 4.2); a rhombic
low-spin iron(III) with g = 2.23, 2.18, and 1.95 (red); and an
isotropic signal with g = 2.00 (blue). The latter two signals can,
analogously with the alkyl peroxide reactions, be associated
with the transient peroxyacid adduct [Fe(OOC(O)m-PhCl)-
2
+
•
(
tpenaH)] and its daughter m-ClPhC(O)O or derived m-
2
+
•
Figure 4. (a) UV−vis spectra of [Fe(tpena)] (orange) with the
addition of m-CPBA (50 equiv, [Fe] = 2 mM) at −40 °C to generate
ClPhC(O)OO radical. The high-spin signal observed in the
EPR spectrum has no counterpart in the reactions with t-
BuOOH or cumylOOH, and we propose that this is due to
2
+
IV
2+
[
(
[
Fe(OOC(O)m-PhCl)(tpenaH)] (red) and [Fe O(tpenaH)]
green). (b) EPR spectra recorded on a frozen solution of
Fe(OH)(tpenaH)] with the addition of 50 equiv of m-CPBA in
III
2+
[
Fe (OC(O)PhCl)(tpenaH)] , vide infra.
While some O is detected, the reaction of [Fe(OH)-
2+
2
MeCN at 110 K. Microwave frequency: 9.309 872 GHz. Fitted data
2
+
2
+
•
(
tpenaH)] with 1000 equiv of m-CPBA does not give rise to
for [Fe(OOC(O)m-PhCl)(tpenaH)] and the m-ClPhC(O)O or
derived m-ClPhC(O)OO• radical are shown in red and blue,
respectively. (c) ESI-MS of a solution of [Fe(tpena)] and m-CBA
(1:50, MeCN). Assignments as follows. m/z: calcd for
the release of large amounts of O as it does in the reactions
2
2
+
with t-BuOOH and cumylOOH (Table 2). A yield of 23% of
the theoretical O production (based on the mechanism in
2
III
+
C H ClFeN O , [Fe (tpena)(m-CBA)] , 601.117; found,
Scheme 3) is volumetrically detected after 30 min. The
29 28
5
4
III
6
01.114. m/z: calcd for C H ClFeN O , [Fe (tpena)(m-CBA)-
28 28 5 2
products of the disproportionation of m-CPBA are O and m-
2
+
CO ] , 557.128; found, 557.124. m/z: calcd for C H FeN ,
2
20 22
5
chlorobenzoic acid (m-CBAH). Thus, X in Scheme 3 will be
m-chlorobenzoyl. Corroborating, NMR spectra of mixtures
II
+
[
Fe (tpena)−CH COO] , 388.122; found, 388.119. m/z: calcd for
2
I
I
+
C H ClFeN , {[Fe (Cl)(tpena)−CH COO](MeCN)} , 465.125;
2
2
26
6
2
containing Fe-tpena and m-CPBA (1:750) in d -MeCN show
3
found, 465.075.
that m-CBAH is produced concurrently with consumption of
the m-CPBA. After 1 h of reaction time, half of the m-CPBA
has disappeared (Figure S11), but 25% of the m-CPBA
remains unreacted after 3 days. It is not unexpected that the
product m-CBAH might inhibit the catalytic reaction to a far
greater degree compared with the product alcohols from the
alkylhydroperoxide disproportionation reactions, i.e.; the
2
+
counterpart [Fe(O-t-Bu)(tpenaH)] and [Fe(Ocumyl)-
2
+
(tpenaH)] complexes. This was verified by recording ESI-
mass spectra of [Fe(tpena)] solutions in the presence of 50
equiv of t-BuOH or m-CBAH. Both show a base peak at m/z =
446.1256 corresponding to [Fe (tpena)] ; however, the
spectrum of the mixture containing m-CBAH contains also
two intense peaks at m/z = 601.114 (59%; calcd for
2
+
II
+
2
+
species [Fe(OX)(tpenaH)] in Scheme 3, in this case
III
2+
[
Fe (OC(O)PhCl)(tpenaH)] , is more stable than the
9
988
DOI: 10.1021/acscatal.8b02882
ACS Catal. 2018, 8, 9980−9991

ACS Catalysis
Research Article
C H ClFeN O , 601.117) and m/z = 557.124 (39%; calcd
imental design, the selective oxidation of an external substrate
can be favored.
2
9
28
5
4
for C H ClFeN O , 557.128) which correspond to the m-
2
8
28
5
2
III
+
III
CBA adduct ions, [Fe (m-CBA)(tpena)] and [Fe (m-
CBA)(tpena) − CO ] , respectively (Figure 4c). These results
show that the product m-CBAH inhibits step f in the catalytic
cycle by formation of a relatively stable [Fe (OC(O)PhCl)-
tpenaH)] (Scheme 3). Commercial m-CPBA is significantly
Ultimately, our results demonstrate that disproportionation
of alkyl hydroperoxides and peroxyperacids can potentially be a
significant background reaction, when these reagents are used
as terminal oxidants in metal-catalyzed reactions. It is clear,
however, given the number of applications of these reagents,
including commercial, that significant unproductive decom-
position of alkyl hydroperoxide is in fact relatively rare. Finally,
as an effective catalyst for alkyl hydroperoxide disproportiona-
+
2
III
2
+
(
contaminated by m-CBAH, so this inhibitor is actually present
from the first addition of its derived peroxide. Consistently
with the ESI-MS, UV−vis and EPR spectra of acetonitrile
solutions of [Fe(tpena)]2 are not affected by the addition of t-
BuOH whereas the addition of m-CBAH results in the
+
2+
tion, [Fe(tpena)] might find useful application in quenching
unwanted hydrogen alkyl peroxides in cases of excesses or in
chemical spills.
2
+
disappearance of the characteristic band of [Fe(tpena)] at
λmax = 360 nm. The EPR spectrum of this latter solution
reproduces the high-spin signal (g = 8.5, 5.2, 4.2, 3.0) of the
working solutions where m-CPBA is the substrate (Figure 4b;
eff
ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acscatal.8b02882.
■
*
S
Figure S12). It is therefore unambiguous that the EPR signal
III
can be assigned to high-spin (S = 5/2) [Fe (tpenaH)(m-
2
+
CBA)] .
EPR, UV−vis, Mossbauer, and NMR spectra of reaction
̈
mixtures; MIMS data; and supporting plots for the
volumetric experiments (PDF)
CONCLUSION
■
O evolution observed on t-BuOOH disproportionation
2
The Fe-tpena system, uniquely compared to all other known
nonheme models, contains a chelating single carboxylato
donor cis to the peroxide binding site. This is biomimetic with
respect to the propensity of the occurrence of mono Asp/Glu
(AVI)
AUTHOR INFORMATION
Corresponding Author
*E-mail: mckenzie@sdu.dk. Phone: 45 6550 2518.
ORCID
Christina Wegeberg: 0000-0002-6034-453X
Wesley R. Browne: 0000-0001-5063-6961
Christine J. McKenzie: 0000-0001-5587-0626
■
coordination cis to O activating enzymatic nonheme sites.
2
Transient low-spin iron(III)-alkylperoxides and iron(III)-
III
2+
acylperoxides [Fe OOR(tpenaH)] (R = C(CH ) , C-
3
3
(
CH ) Ph, C(O)PhCl) form on the replacement of the
3 2
III
2+
water equivalent in [Fe (OH)(tpenaH)] by ROOH. The
complex is bifunctional: The ingoing and outgoing coligands
are charge separated into a XO (X = H, RO) ligand, and a
−
Notes
proton on the noncoordinated pyridinium arm. This system is
the first nonheme iron model incorporating a biomimetic
second coordination sphere base for enabling this outcome.
The authors declare no competing financial interest.
IV
2+
ACKNOWLEDGMENTS
The ν
for the derived [Fe O(tpenaH)] is hypsochromi-
■
FeO
−
1
cally shifted, by around 30 cm , compared to the iron(IV)oxo
This work was supported by the Danish Council for
Independent ResearchNatural Sciences (Grant 4181-00329
to C.McK.). C.W. thanks the COST Action CM1305
44
complexes of N-donor-only aminopyridyl ligands. The way in
which these structural and electronic features are important for
the indisputable inherent radical character of the unique
(
ECOSTBio) for financial support (STSM 37381 and the
ECOSTBio Winter school 2016). We thank Dr. Eckhard Bill
for help with Mossbauer spectroscopy.
IV
2+11
[
Fe O(tpenaH)]
is yet to be determined; however, it is
̈
pertinent to note that carboxylato donors are redox non-
innocent and this will influence the activation of the bonds to
and in cis ligands. For example, the lability of the FeO−OR
bond in the alkyl and acyl peroxide adducts of Fe-tpena is
greater than that of analogous iron complexes supported by the
more mundane N-donor-only ligands. These factors are
intrinsic for the unified peroxidase-like mechanism we propose
REFERENCES
■
(1) Kovaleva, E. G.; Lipscomb, J. D. Versatility of Biological Non-
Heme Fe(II) Centers in Oxygen Activation Reactions. Nat. Chem.
Biol. 2008, 4, 186−193.
(
9
(
2) Stenkamp, R. E. Dioxygen and Hemerythrin. Chem. Rev. 1994,
4, 715−726.
(Scheme 3). The speed of the reaction and the extensive
3) Dawson, J. H. Probing Structure-Function Relations in Heme-
characterization of the iron-based and organic species in our
working solutions lead us to propose that the evolved O2
Containing Oxygenases and Peroxidases. Science 1988, 240, 433−439.
(4) Costas, M.; Mehn, M. P.; Jensen, M. P.; Que, L. Dioxygen
Activation at Mononuclear Nonheme Iron Active Sites: Enzymes,
Models, and Intermediates. Chem. Rev. 2004, 104, 939−986.
(5) Nam, W. Synthetic Mononuclear Nonheme Iron-Oxygen
Intermediates. Acc. Chem. Res. 2015, 48, 2415−2423.
6) Que, L.; Tolman, W. B. Biologically Inspired Oxidation
Catalysis. Nature 2008, 455, 333−340.
7) Nielsen, A.; Larsen, F. B.; Bond, A. D.; McKenzie, C. J.
Regiospecific Ligand Oxygenation in Iron Complexes of a
Carboxylate-Containing Ligand Mediated by a Proposed Fe(V)-Oxo
Species. Angew. Chem., Int. Ed. 2006, 45, 1602−1606.
(8) Wegeberg, C.; Lauritsen, F. R.; Frandsen, C.; Mørup, S.; Browne,
W. R.; McKenzie, C. J. Directing a Non-Heme Iron(III)-Hydro-
predominantly derives from the O extrusion from a putative
2
III
Fe OOOR species obtained from coupling of the radical
•
iron(IV)oxo complex with ROO .
With its resting-state +3 oxidation state and readily
accessible +4 state the iron-tpena system is highly applicable
to one-electron reactions in cycles, i.e., those involving HAT
reactions as represented by the top section of Scheme 1 in the
presence of peroxides as terminal oxidants. All the reactions
described here support this conclusion. Interestingly, despite
(
(
•
•
•
the presence of several radical species (ROO , RO , R ,
[
IV
2+
Fe O(tpenaH)] ) in these reactions, with judicial exper-
9
989
DOI: 10.1021/acscatal.8b02882
ACS Catal. 2018, 8, 9980−9991

ACS Catalysis
Research Article
peroxide Species on a Trifurcated Reactivity Pathway. Chem. - Eur. J.
(27) Ouyang, X.-H.; Song, R.-J.; Li, J.-H. Iron-Catalyzed Oxidative
1,2-Carboacylation of Activated Alkenes with Alcohols: A Tandem
Route to 3-(2-Oxoethyl)indolin-2-ones. Eur. J. Org. Chem. 2014,
2014, 3395−3401.
2
(
018, 24, 5134−5145.
9) Lennartson, A.; McKenzie, C. J. An Iron(III) Iodosylbenzene
V
Complex: A Masked Non-Heme Fe O. Angew. Chem., Int. Ed. 2012,
1, 6767−6770.
10) de Sousa, D. P.; Wegeberg, C.; Vad, M. S.; Mørup, S.;
5
(
(28) Saka, E. T.; Bıyıklıoglu, Z. Co(II) and Fe(II) phthalocyanines:
̆
Synthesis, Characterization and Catalytic Activity on Cyclohexene
Oxidation with Different Oxygen Source. J. Organomet. Chem. 2013,
745−746, 50−56.
Frandsen, C.; Donald, W. A.; McKenzie, C. J. Halogen-Bonding-
Assisted Iodosylbenzene Activation by a Homogenous Iron Catalyst.
Chem. - Eur. J. 2016, 22, 3810−3820.
(29) Coelho, J. A. S.; Trindade, A. F.; Wanke, R.; Rocha, B. G. M.;
Veiros, L. F.; Gois, P. M. P.; Pombeiro, A. J. L.; Afonso, C. A. M. N-
Heterocyclic Carbene Dirhodium(II) Complexes as Catalysts for
Allylic and Benzylic Oxidations. Eur. J. Org. Chem. 2013, 2013, 1471−
(11) De Sousa, D. P.; Miller, C. J.; Chang, Y.; Waite, T. D.;
McKenzie, C. J. Electrochemically Generated cis-Carboxylato-
Coordinated Iron(IV) Oxo Acid-Base Congeners as Promiscuous
Oxidants of Water Pollutants. Inorg. Chem. 2017, 56, 14936−14947.
1
(
478.
30) Napoly, F.; Kieffer, R.; Jean-Gerard, L.; Goux-Henry, C.; Draye,
́
(12) Vad, M. S.; Lennartson, A.; Nielsen, A.; Harmer, J.; McGrady, J.
M.; Andrioletti, B. Fe(TAML)Li/ tert-butyl hydroperoxide as a new
E.; Frandsen, C.; Mørup, S.; McKenzie, C. J. An Aqueous Non-Heme
Fe(IV)oxo Complex with a Basic Group in the Second Coordination
Sphere. Chem. Commun. 2012, 48, 10880−10882.
combination for benzylic C−H oxidation. Tetrahedron Lett. 2015, 56,
2
(
517−2520.
31) Hodges, G. R.; Lindsay Smith, J. R.; Oakes, J. The Oxidation of
(13) Donald, W. A.; McKenzie, C. J.; O’Hair, R. A. J. C-H Bond
Azo Dyes by Peroxy Acids and tert-Butyl Hydroperoxide in Aqueous
Solution Catalysed by Iron(III) 5,10,15,20-tetra(2,6-dichloro-3-
sulfonatophenyl)porphyrin: Product Studies and Mechanism. J.
Chem. Soc., Perkin Trans. 2 1999, 1943−1952.
Activation of Methanol and Ethanol by a High-Spin Fe(IV)O
Biomimetic Complex. Angew. Chem., Int. Ed. 2011, 50, 8379−8383.
(14) De Sousa, D. P.; Bigelow, J. O.; Sundberg, J.; Que, L.;
McKenzie, C. J. Caught! Crystal trapping of a Side-on Peroxo bound
(
32) Kang, C.; Redman, C.; Cepak, V.; Sawyer, D. T. Iron(II)/
Hydroperoxide (Fenton Reagent)-induced Activation of Dioxygen for
A) the Direct Ketonization of Methylenic Carbon and (B) the
Dioxygenation of cis-Stilbene. Bioorg. Med. Chem. 1993, 1, 125−140.
33) Palopoli, C.; Duhayon, C.; Tuchagues, J.-P.; Signorella, S.
to Cr(IV). Chem. Commun. 2015, 51, 2802−2805.
(15) Hazell, A.; McKenzie, C. J.; Nielsen, L. P.; Schindler, S.;
(
Weitzer, M. Mononuclear Non-Heme Iron(III) Peroxide Complexes:
Syntheses, Characterisation, Mass Spectrometric and Kinetic Studies.
J. Chem. Soc., Dalton Trans. 2002, 310.
(
Synthesis, Characterization, and Reactivity Studies of a Water-Soluble
Bis(alkoxo)(carboxylato)-Bridged diMn(III) Complex Modeling the
Active Site in Catalase. Dalton Trans. 2014, 43, 17145−17155.
(34) Palopoli, C.; Bruzzo, N.; Hureau, C.; Ladeira, S.; Murgida, D.;
Signorella, S. Synthesis, Characterization, and Catalase Activity of a
Water-Soluble diMn(III) Complex of a Sulphonato-Substituted Schiff
Base Ligand: An Efficient Catalyst for H O Disproportionation.
(16) Bernal, I.; Jensen, I. M.; Jensen, K. B.; McKenzie, C. J.;
Toftlund, H.; Tuchagues, J.-P. Iron(II) Complexes of Polydentate
Aminopyridyl Ligands and an Exchangeable Sixth Ligand; Reactions
1
1
with Peroxides. Crystal Structure of [FeL (H O)][PF ] ·H O [L =
2
6
2
2
N,N′-bis-(6-methyl-2-pyridylmethyl)-N,N′-bis(2-pyridylmethyl)-
ethane-1,2-diamine]. J. Chem. Soc., Dalton Trans. 1995, 3667−3675.
2
2
(17) Jensen, K. B.; McKenzie, C. J.; Nielsen, L. P.; Zacho Pedersen,
Inorg. Chem. 2011, 50, 8973−8983.
J.; Svendsen, H. M. Deprotonation of Low-Spin Mononuclear
Iron(III)−Hydroperoxide Complexes give Transient Blue Species
Assigned to High-Spin Iron(III)−Peroxide Complexes. Chem.
Commun. 1999, 1313−1314.
(
35) Sengupta, K.; Chatterjee, S.; Dey, A. Catalytic H O
2
2
Disproportionation and Electrocatalytic O Reduction by a Functional
2
Mimic of Heme Catalase: Direct Observation of Compound 0 and
Compound I in situ. ACS Catal. 2016, 6, 1382−1388.
(
18) Mageli, O. L.; Kolczynski, J. R. Organic Peroxides. Ind. Eng.
Chem. 1966, 58, 25−32.
19) Salem, I. A.; El-Maazawi, M.; Zaki, A. B. Kinetics and
(36) Chen, W.-H.; Wei, H.-H.; Lee, G.-H.; Wang, Y. Dimeric
Iron(III) Complexes with Salicylaldimine Ligands: Structures,
Magnetic Properties, and Activity of H O Disproportionation.
(
2
2
Mechanisms of Decomposition Reaction of Hydrogen Peroxide in
Polyhedron 2001, 20, 515−521.
Presence of Metal Complexes. Int. J. Chem. Kinet. 2000, 32, 643−666.
(
37) Gelasco, A.; Askenas, A.; Pecoraro, V. L. Catalytic
(20) Nakanishi, M.; Bolm, C. Iron-Catalyzed Benzylic Oxidation
Disproportionation of Hydrogen Peroxide by the Tetranuclear
with Aqueous tert-Butyl Hydroperoxide. Adv. Synth. Catal. 2007, 349,
61−864.
21) Umbreit, M. A.; Sharpless, K. B. Allylic Oxidation of Olefins by
ll
Manganese Complex [Mn (2-OHpicpn)] . Inorg. Chem. 1996, 35,
4
8
(
1
(
419−1420.
38) Vad, M. S.; Nielsen, A.; Lennartson, A.; Bond, A. D.; McGrady,
Catalytic and Stoichiometric Selenium Dioxide with tert-Butyl
Hydroperoxide. J. Am. Chem. Soc. 1977, 99, 5526−5528.
J. E.; McKenzie, C. J. Switching on Oxygen Activation by Cobalt
Complexes of Pentadentate Ligands. Dalton Trans. 2011, 40, 10698−
(22) Traylor, T. G.; Tsuchiya, S.; Byun, Y. S.; Kim, C. High-Yield
1
(
0707.
Epoxidations with Hydrogen Peroxide and tert-Butyl Hydroperoxide
Catalyzed by Iron(III) Porphyrins: Heterolytic Cleavage of Hydro-
peroxides. J. Am. Chem. Soc. 1993, 115, 2775−2781.
39) Wegeberg, C.; McKee, V.; McKenzie, C. J. A Coordinatively
Flexible Hexadentate Ligand gives Structural Isomeric Complexes
M (L)X (M = Cu, Zn; X = Br, Cl). Acta Crystallogr., Sect. C: Struct.
2
3
(23) Lu, X.; Liu, Y.; Sun, B.; Cindric, B.; Deng, L. Catalytic
Chem. 2016, 72, 68−74.
40) Bill, E. Max-Planck-Institute for Chemical Energy Conversion
in Mulheim, 2016; available from the author by mail to eckhard.bill@
cec.mpg.de.
41) Lehnert, N.; Ho, R. Y.; Que, L.; Solomon, E. I. Spectroscopic
Enantioselective Peroxidation of α,β-Unsaturated Ketones. J. Am.
Chem. Soc. 2008, 130, 8134−8135.
(
̈
(24) Lu, Y.; Zheng, C.; Yang, Y.; Zhao, G.; Zou, G. Highly
Enantioselective Epoxidation of α,β-Unsaturated Ketones Catalyzed
by Primary-Secondary Diamines. Adv. Synth. Catal. 2011, 353, 3129−
(
Properties and Electronic Structure of Low-Spin Fe(III)-Alkylperoxo
Complexes: Homolytic Cleavage of the O-O Bond. J. Am. Chem. Soc.
2001, 123, 8271−8290.
3
(
133.
25) Terent’ev, A. O.; Sharipov, M. Y.; Krylov, I. B.; Gaidarenko, D.
V.; Nikishin, G. I. Manganese Triacetate as an Efficient Catalyst for
(42) Menage, S.; Wilkinson, E. C.; Que, L.; Fontecave, M.
́
Bisperoxidation of Styrenes. Org. Biomol. Chem. 2015, 13, 1439−
Formation of an Alkylperoxoiron(III) Complex during Oxidations
Catalyzed by μ-Oxodiiron(III) Complexes. Angew. Chem., Int. Ed.
Engl. 1995, 34, 203−205.
1
(
445.
26) Wang, Q.; Geng, H.; Chai, W.; Zeng, X.; Xu, M.; Zhu, C.; Fu,
R.; Yuan, R. Copper-Catalyzed Formation of C-O Bonds by Oxidative
Coupling of Benzylic Alcohols with Ethers. Eur. J. Org. Chem. 2014,
(43) Ho, R. Y. N.; Roelfes, G.; Feringa, B. L.; Que, L. Raman
Evidence for a Weakened O−O Bond in Mononuclear Low-Spin
Iron(III)−Hydroperoxides. J. Am. Chem. Soc. 1999, 121, 264−265.
2014, 6850−6853.
9
990
DOI: 10.1021/acscatal.8b02882
ACS Catal. 2018, 8, 9980−9991

ACS Catalysis
Research Article
(
44) McDonald, A. R.; Que, L. High-Valent Nonheme Iron-Oxo
Complexes: Synthesis, Structure, and Spectroscopy. Coord. Chem. Rev.
013, 257, 414−428.
45) Kraft, B. J.; Coalter, N. L.; Nath, M.; Clark, A. E.; Siedle, A. R.;
(61) Martin, R. B.; Noyes, W. A. The Reactions of Methyl Radicals
1
with Oxygen . J. Am. Chem. Soc. 1953, 75, 4183−4185.
2
(
(62) Barnard, J. A.; Cohen, A. Reaction of Methyl Radicals with
Oxygen. Reaction of Methyl Radicals with Oxygen. Trans. Faraday
Soc. 1968, 64, 396.
Huffman, J. C.; Zaleski, J. M. Photothermally Induced Bergman
Cyclization of Metalloenediynes via Near-Infrared Ligand-to-Metal
Charge-Transfer Excitation. Inorg. Chem. 2003, 42, 1663−1672.
(63) Nam, W.; Han, H. J.; Oh, S.-Y.; Lee, Y. J.; Choi, M.-H.; Han, S.-
Y.; Kim, C.; Woo, S. K.; Shin, W. New Insights into the Mechanisms
of O−O Bond Cleavage of Hydrogen Peroxide and tert-Alkyl
Hydroperoxides by Iron(III) Porphyrin Complexes. J. Am. Chem.
Soc. 2000, 122, 8677−8684.
(
46) Cooney, J. J. A.; Carducci, M. D.; McElhaney, A. E.; Selby, H.
D.; Enemark, J. H. New Oxovanadium Bis(1,2-dithiolate) Com-
pounds That Mimic the Hydrogen-Bonding Interactions at the Active
Sites of Mononuclear Molybdenum Enzymes. Inorg. Chem. 2002, 41,
(64) Russell, G. A. Deuterium-isotope Effects in the Autoxidation of
Aralkyl Hydrocarbons. Mechanism of the Interaction of Peroxy
7
(
086−7093.
1
Radicals . J. Am. Chem. Soc. 1957, 79, 3871−3877.
47) Pacigova, S.; Gyepes, R.; Tatiersky, J.; Sivak, M. Interpretation
́
́
2
(65) Howard, J. A.; Ingold, K. U. Self-Reaction of sec-Butylperoxy
Radicals. Confirmation of the Russell Mechanism. J. Am. Chem. Soc.
of the multiple vanadium-oxygen bonds in the central VO(η -O )+
2
group. Synthesis, structure, supramolecular interactions and DFT
studies for complexes with 2,2′-bipyridine, 1,10-phenanthroline,
pyrazinato(1-) and pyrazinamide ligands. Dalton Trans. 2008, 121−
1968, 90, 1056−1058.
1
(
30.
48) Ingold, K. U.; Morton, J. R. Electron Spin Resonance Spectra of
Organic Oxy Radicals. J. Am. Chem. Soc. 1964, 86, 3400−3402.
49) Symons, M. C. R. Electron Spin Resonance Spectra of Organic
Oxy Radicals. The Radical (CH ) CO . J. Am. Chem. Soc. 1969, 91,
(
3
3
3
5
(
924−5924.
50) Poulsen, A. K.; Rompel, A.; McKenzie, C. J. Water Oxidation
Catalyzed by a Dinuclear Mn Complex: A Functional Model for the
Oxygen-Evolving Center of Photosystem II. Angew. Chem., Int. Ed.
2
(
005, 44, 6916−6920.
51) Sugimoto, H.; Sawyer, D. T. Iron(II)-Induced Activation of
2
+
1
Hydrogen Peroxide to Ferryl Ion (FeO ) and Singlet Oxygen ( O )
2
in Acetonitrile: Monoxygenations, Dehydrogenations, and Dioxyge-
nations of Organic Substrates. J. Am. Chem. Soc. 1984, 106, 4283−
4
(
285.
52) Sugimoto, H.; Sawyer, D. T. Iron(II)-Induced Activation of
Hydroperoxides for the Dehydrogenation and Monooxygenation of
Organic Substrates in Acetonitrile. J. Am. Chem. Soc. 1985, 107,
5
(
712−5716.
53) De Sousa, D. P.; McKenzie, C. J. Molecular Iron-based Oxidants
and their Stochiometric Reactions in Topics in Organometallic Chemistry:
Iron Catalysis II; Springer, 2015; Vol. 50, pp 311−356.
(
54) Horner, O.; Jeandey, C.; Oddou, J.-L.; Bonville, P.; McKenzie,
2
+
C.; Latour, J.-M. Hydrogenperoxo-[(bztpen)Fe(OOH)] and Its
Deprotonation Product Peroxo-[(bztpen)Fe(O )] , Studied by EPR
+
2
and Mo
Structures of Peroxo Model Complexes. Eur. J. Inorg. Chem. 2002,
002, 3278−3283.
55) Nebe, T.; Beitat, A.; Wu
̈
ssbauer Spectroscopy - Implications for the Electronic
2
(
rtele, C.; Ducker-Benfer, C.; van Eldik,
̈ ̈
R.; McKenzie, C. J.; Schindler, S. Reinvestigation of the Formation of
a Mononuclear Fe(III) Hydroperoxido Complex using High Pressure
Kinetics. Dalton Trans. 2010, 39, 7768−7773.
(56) Avila, D. V.; Brown, C. E.; Ingold, K. U.; Lusztyk, J. Solvent
Effects on the Competitive β-scission and Hydrogen Atom
Abstraction Reactions of the Cumyloxyl Radical. Resolution of a
Long-Standing Problem. J. Am. Chem. Soc. 1993, 115, 466−470.
(57) Caudle, M. T.; Riggs-Gelasco, P.; Gelasco, A. K.; Penner-Hahn,
J. E.; Pecoraro, V. L. Mechanism for the Homolytic Cleavage of Alkyl
Hydroperoxides by the Manganese(III) Dimer MnIII2(2-OHsalpn)2.
Inorg. Chem. 1996, 35, 3577−3584.
(58) Kim, J.; Harrison, R. G.; Kim, C.; Que, L. Fe(TPA)-Catalyzed
Alkane Hydroxylation. Metal-Based Oxidation vs Radical Chain
Autoxidation. J. Am. Chem. Soc. 1996, 118, 4373−4379.
(
59) The reader should be aware that the coupling reaction of
64,65
alkylperoxyl radicals, known as the Russell mechanism,
was
formulated on the basis of experiments involving solutions of
alkylhydroperoxides (containing an α-H atom) and their reaction
with radical initiators. These solutions do not contain significant
amounts of other radicals, like the high-valent iron oxo species.
(60) Luo, Y.-R. Comprehensive handbook of chemical bond energies;
CRC Press: Boca Raton, 2007.
9
991
DOI: 10.1021/acscatal.8b02882
ACS Catal. 2018, 8, 9980−9991