Iron(II) α-aminopyridine complexes and their catalytic activity in oxidation reactions: A comparative study of activity and ligand decomposition

Article abstract of DOI:10.1002/cplu.201200244

New well-defined Fe^II complexes bearing bi- and tridentate α-aminopyridine ligands were synthesized, and their catalytic activity in the oxidation of hydrocarbons and alcohols utilizing peroxide oxidants was investigated. The tridendate bis-(picolyl)amine ligand 6 and its benzylated analogue 7 were converted into complexes [Fe^II(6)₂] OTf₂ (96 %, X-ray; OTf= CF₃SO₃ ^-) and [Fe^II(7)₂]OTf₂ (90 %). The bidentate aminopyridine ligand 8 was converted into [Fe^II(8) ₂(OTf)₂] (93 %, X-ray). The new complexes are catalytically active in the oxidation of secondary alcohols and benzylic methylene groups to the corresponding ketones, of toluene to benzaldehyde, and of cyclohexene to cyclohexene oxide (3 mol% catalyst, tBuOOH (4 equiv), RT, 2-6 h, 28 to 85% yield of isolated product). The catalytic oxidation of cyclohexane with ROOH (R=H, tBu) to an alcohol/ketone mixture with low ratio revealed that these oxidations follow largely a radical mechanism, except when [Fe ^II(6)₂]OTf₂ was employed and H ₂O₂ was added slowly. Together with known bi- and tetradendate iron complexes, a comparative study showed slight reactivity differences for the newly prepared complexes, with the highest observed for [Fe^II(6)₂]OTf₂ and [Fe^II(7) ₂]OTf₂. The reaction of the new complexes with peroxides was followed over time by UV/Visible spectroscopy; this revealed a fast reaction between the two reactants within minutes. Ligand-decomposition pathways were investigated, and revealed that the NCH₂ units of the complexes are rapidly oxidized to the corresponding amides NC=O. The iron complex [Fe ^II(6)₂]OTf₂ showed no decrease in catalytic activity and a moderate decrease in selectivity when first subjected to oxidative conditions similar to those employed in catalysis. Thus, oxidative ligand deterioration had a marginal effect on the catalytic activity of the iron complex [Fe^II(6)₂]OTf₂.

Full text of DOI:10.1002/cplu.201200244

CHEMPLUSCHEM
FULL PAPERS
DOI: 10.1002/cplu.201200244
Iron(II) a-Aminopyridine Complexes and Their Catalytic
Activity in Oxidation Reactions: A Comparative Study of
Activity and Ligand Decomposition
Matthew Lenze,[a] Erin T. Martin,[a] Nigam P. Rath,[a, b] and Eike B. Bauer*[a]
New well-defined FeII complexes bearing bi- and tridentate a-
aminopyridine ligands were synthesized, and their catalytic ac-
tivity in the oxidation of hydrocarbons and alcohols utilizing
peroxide oxidants was investigated. The tridendate bis-
(picolyl)amine ligand 6 and its benzylated analogue 7 were
converted into complexes [FeII(6)2]OTf2 (96%, X-ray; OTf=
CF3SO3À) and [FeII(7)2]OTf2 (90%). The bidentate aminopyridine
ligand 8 was converted into [FeII(8)2(OTf)2] (93%, X-ray). The
new complexes are catalytically active in the oxidation of sec-
ondary alcohols and benzylic methylene groups to the corre-
sponding ketones, of toluene to benzaldehyde, and of cyclo-
hexene to cyclohexene oxide (3 mol% catalyst, tBuOOH
(4 equiv), RT, 2–6 h, 28 to 85% yield of isolated product). The
catalytic oxidation of cyclohexane with ROOH (R=H, tBu) to an
alcohol/ketone mixture with low ratio revealed that these oxi-
dations follow largely a radical mechanism, except when
[FeII(6)2]OTf2 was employed and H2O2 was added slowly. To-
gether with known bi- and tetradendate iron complexes,
a comparative study showed slight reactivity differences for
the newly prepared complexes, with the highest observed for
[FeII(6)2]OTf2 and [FeII(7)2]OTf2. The reaction of the new com-
plexes with peroxides was followed over time by UV/Visible
spectroscopy; this revealed a fast reaction between the two re-
actants within minutes. Ligand-decomposition pathways were
investigated, and revealed that the NCH2 units of the com-
plexes are rapidly oxidized to the corresponding amides NC=
O. The iron complex [FeII(6)2]OTf2 showed no decrease in cata-
lytic activity and a moderate decrease in selectivity when first
subjected to oxidative conditions similar to those employed in
catalysis. Thus, oxidative ligand deterioration had a marginal
effect on the catalytic activity of the iron complex [FeII(6)2]OTf2.
Introduction
Iron catalysis has emerged as a lively area of research in recent
years.[1] Iron is less toxic than other transition metals typically
utilized in catalysis and, owing to its abundance, it is readily
available at a reasonable cost. Consequently, iron salts or com-
plexes have been increasingly applied as catalysts in a variety
of organic transformations, such as in CÀC,[2] CÀN,[3] and CÀP[4]
bond-forming reactions, heteroatom–heteroatom bond-form-
ing reactions,[5] reductions,[6] and polymerizations[7] as well as
other functional-group interconversions.[8]
Iron-based systems have also been investigated as catalysts
in oxidation reactions of organic substrates. Fenton chemis-
try—that is, the oxidative power of iron salts in combination
with peroxides—has been known for over 100 years.[9] Gif and
GoAgg chemistry was popularized at the beginning of the
1980s by Barton and utilizes similarly simple conditions: the
combination of iron or iron salts and acetic acid in pyridine
that oxidatively functionalizes alkanes utilizing dioxygen or
peroxides as the terminal oxidants.[10] The mechanism for both
systems is still the objective of current research,[11] and the
nonradical mechanism originally suggested by Barton[12] has
been subsequently challenged.[13] Owing to the possible in-
volvement of radicals in Fenton and Gif chemistry, their applic-
ability in the synthesis of complex organic molecules is limited.
Consequently, these systems have not been employed exten-
sively in the synthesis of fine chemicals, and Fenton chemistry
is utilized for wastewater treatment, in which the oxidative de-
composition of organic compounds in solution is the final
goal.[14]
[a] M. Lenze, E. T. Martin, Dr. N. P. Rath, Prof. Dr. E. B. Bauer
University of Missouri-St. Louis
Department of Chemistry and Biochemistry
One University Boulevard, St. Louis, MO 63121 (USA)
Fax: (+1)(314)-516-5342
E-mail: bauere@umsl.edu
[b] Dr. N. P. Rath
Center for Nanoscience, University of Missouri-St. Louis
St. Louis, MO 63121 (USA)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cplu.201200244. It contains details of the typi-
cal experiment used to determine the alcohol/ketone ratios in Table 4;
the experimental procedure to determine the activities in Figure 3; the
experimental procedure to record the time-resolved UV/Visible spectra
of the metal complexes after addition of the oxidant in Figures 4, 5, and
6; the representative experimental procedure to determine ligand-de-
composition pathways; experimental details, characterization data, and
Hence, recent research efforts have been directed towards
iron catalysts that can be utilized in synthetically useful oxida-
tion reactions under milder conditions in a more predictable
manner. These research efforts are inspired by heme and non-
heme enzymes such as the cytochrome P450 family[15] or the
1
1H and 13C {1H} NMR spectra for the catalysis products; H NMR and UV/
Visible spectra of all metal complexes; GC–MS traces of the ligand-de-
composition reactions in Scheme 4.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2013, 78, 101 – 116 101

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
methane monooxygenase[16] that can be found in nature.
These enzymes accomplish oxidation reactions under physio-
logical conditions, and feature well-defined, iron-containing co-
factors bearing N-coordinating ligands in their coordination
sphere. Research efforts have been targeted towards artificial
iron complexes mimicking the active sites of the enzymes
found in nature.[15–17] These research efforts afforded a number
of catalyst systems that show high efficiencies and selectivities
in oxidation reactions.[18] Some of these complexes are formed
in situ[19] and have been successfully employed in the oxidation
of alkanes,[20] benzylic methylene groups,[21] and alcohols;[22] in
oxidative cleavage reactions;[23] and in the epoxidation of al-
kenes.[24] It appears that well-defined, preformed nonheme iron
complexes perform notably well in oxidation reactions, for ex-
ample, White’s iron complex 5, which selectively oxidizes terti-
ture can impair the catalytic activity of an iron complex.[33] We
recently reported a series of iron complexes [Fe(4)2(OTf)2]
(OTf=CF3SO3À) bearing a-imino pyridine ligands 4, which we
successfully employed as catalysts in the oxidation of activated
methylene groups and secondary alcohols with tBuOOH to
obtain the corresponding ketones (4 h, RT, 3% catalyst loading,
22–91% yield of isolated product).[33a] Sterically and electroni-
cally tuned ligands 4 were employed in the synthesis of the
corresponding a-imino pyridine iron complexes; however, such
tuning efforts had only limited impact on the catalytic activity,
which was comparable for all complexes of that specific library.
We speculated that ligand architectures could impair the cata-
lytic activity of complexes, and decided to compare the effi-
ciency of different types of ligands in iron-based oxidation sys-
tems.
Herein, we describe the synthesis and characterization of
new iron mono- and bis(2-picolyl)amine complexes and their
catalytic activity. We compare the activities of these new com-
plexes with those of known complexes previously employed in
the title reactions. We furthermore report ligand-decomposi-
tion pathways for the iron complexes and how such ligand de-
terioration impairs the activity of the catalyst system. Finally,
mechanistic studies were performed to identify intermediates
and to assess to what extent radicals may be involved in the
reactions.
Results
Synthesis of new iron complexes
For comparative studies, we first selected a set of nitrogen-
based ligands that differ significantly in their architecture and
denticity and converted them to their respective iron com-
plexes. The tpa complex [FeII(1a)(OTf)2] is known and was syn-
thesized according to literature procedures,[34] and will subse-
quently be referred to as [FeII(tpa)(OTf)2]. Likewise, complex
[FeII(4a)2(OTf)2] has been synthesized and characterized previ-
ously in our laboratory.[33a]
To investigate different denticities, we utilized the commer-
ary CH units to the corresponding alcohols.[25] Iron complexes
of the tris(2-picolyl)amine (tpa) ligands 1 and BPMEN (2a) were
utilized by Que for alkane oxidations with high alcohol/ketone
ratios.[26] The ligands 2b–d,[27] 3,[28] and bispidine[29] have also
been employed in the synthesis of well-defined, catalytically
active iron complexes, and, as a consequence, the application
of these systems in the synthesis of complex organic mole-
cules has become increasingly popular.[30]
cial, tridentate bis(2-pyridinylmethyl)amine (bpa) ligand 6
(Scheme 1). This ligand has been converted previously to iron
complexes of the formula [FeIII(6)Cl3][35] and [FeII(6)2]X2 (X=ClÀ,
BrÀ),[36] and their utilization as catalysts in oxidation reactions is
about to emerge.[37] It is known that LnFeIII complexes can form
relatively stable oxo-bridged iron dimers of the general formu-
la [LnFeÀOÀFeLn].[38] Complex [FeII(6)2]Cl2 dimerizes in solution
and is oxidized in the air to an oxo-bridged FeIII dimer.[36b]
Some oxo-bridged iron complexes show catalytic activity in
alkane oxidations;[39] however, they tend to be less efficient
than the enzymes found in nature,[40] and the stability reported
for some of these systems might, in the present context, slow
down or inhibit catalytic cycles. Simple halide counterions can
firmly coordinate to the iron center and impede substrate co-
ordination to the metal; for this reason, complex [FeIII(tpa)Cl2]-
(ClO4) did not catalyze the oxidation of cyclohexane utilizing
H2O2.[26] Preliminary data from our own laboratory also showed
that complex [FeII(6)2]Cl2 exhibited low catalytic activity in oxi-
Still, existing oxidation systems suffer from some drawbacks,
such as insufficient turnover numbers (TONs) or selectivi-
ties.[20b,31] Potential issues are the involvement of radicals in the
course of the oxidation reactions and oxidative ligand-decom-
position pathways, which potentially can lead to catalyst deac-
tivation.[32] The search for improved iron-based catalyst systems
for oxidation reactions utilizing peroxide oxidants is, thus, on-
going.
As part of our long-standing research interest in iron-based
catalysis, we are especially interested in how the ligand struc-
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2013, 78, 101 – 116 102

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
two equivalents of ligand 7 and one equivalent of [FeII(OTf)2]
afforded, after workup by crystallization, complex [FeII(7)2]OTf2
in 90% yield as a red-tan powder.
It has been suggested in the literature that two tridentate li-
gands in the coordination sphere of iron complexes might in-
hibit catalytic cycles, as they occupy all six coordination sites
at the metal center.[43] For direct comparison with imine li-
gands 4, we synthesized bidentate a-aminopyridyl ligand 8 by
reductive amination of 2-pyridine carboxaldehyde with 2-me-
thoxyaniline by following a literature procedure for the related
ligand 6.[35a] Ligand 8 was converted in a procedure similar to
those described above to the iron complex [FeII(8)2(OTf)2],
which was obtained as a purple solid in 93% yield of isolated
product. We also tried to convert other a-aminopyridyl ligands
to their respective iron complexes, but solubility issues ren-
dered workup efforts and catalytic applications impossible.
Characterization of the new iron complexes
The new complexes [FeII(6)2]OTf2, [FeII(7)2]OTf2, and [FeII(8)2-
1
(OTf)2] were analyzed by MS, IR, X-ray diffraction, UV/Visible, H
and 19F NMR spectroscopy, magnetic measurements, and ele-
mental analyses. The NMR spectroscopy, UV/Visible, magnetic
moment, and IR data for the complexes are compiled in
Table 1.
Scheme 1. Synthesis of iron complexes.
Complex [FeII(6)2]OTf2 has been synthesized previously with
bromide[36a] and iodide counterions.[36b] However, none of
these previously reported complexes have been analyzed by
NMR spectroscopy and were reported to adopt a low-spin con-
dation reactions and its chemistry was, thus, not pursued fur-
ther.
Therefore, next we targeted iron complex [FeII(6)2]OTf2,
which features OTf (=CF3SO3À), a readily displaceable counter-
ion, and two tridentate bpa ligands 6 in its coordination
sphere, potentially inhibiting bimolecular decomposition path-
ways[41] and the formation of stable oxo-bridged dimers.
[FeII(6)2]OTf2 was obtained by combining [FeII(OTf)2] and two
equivalents of 6 in CH3CN. Crystallization with diethyl ether af-
forded iron complex [FeII(6)2]OTf2 as a red solid in 96% yield
(Scheme 1).
figuration.[36] Complex [FeII(6)2]OTf2 exhibited
a magnetic
moment (meff =1.83 BM, determined with a magnetic suscepti-
1
bility balance) and features well-resolved H and 13C NMR spec-
tra in the diamagnetic regions. However, some line broadening
1
was observed in the H NMR spectrum, and it was not in com-
plete agreement with the structure shown in Scheme 1 (as de-
termined by X-ray diffraction, see below). In addition, the mag-
netic moment is not exactly zero as expected for a low-spin
FeII complex. Thus, we recorded low-temperature 1H NMR spec-
tra (provided in the Supporting Information), and at À508C,
a new set of sharp signals appeared in the spectrum, which is
in agreement with the structure of [FeII(6)2]OTf2 as shown in
Scheme 1. It appears that the complex exhibits dynamic be-
havior at room temperature. The trans isomers were deter-
mined by X-ray diffraction for [FeII(6)2]Br2 and [FeII(6)2]Cl2.[36]
The dynamic behavior of [FeII(6)2]OTf2 at room temperature
might explain its intermediate effective magnetic moment of
1.83 BM.
The acidity of the NÀH unit in 6 might increase significantly
upon coordination to a metal center. To investigate the influ-
ence of the NÀH unit on the catalytic activity, we replaced the
hydrogen in ligand 6 by a benzyl group, affording the
known[42] bis(2-pyridylmethyl)benzyl amine (bpba) ligand 7. It
has been employed previously in the synthesis of oxo-bridged
diiron complexes of the general formula [FeIII (O)(OOR)(7)]-
2
(ClO4)2.[40] However, to the best of our knowledge, the related
complex [FeII(7)2]OTf2 has not yet been reported. Combining
Table 1. Selected physical data of the new iron complexes.
IR, vNH [cmÀ1
free ligand/complex
]
UV/Visible[a]
lmax [CH2Cl2] [nm]
Magnetic moments[b]
meff [BM]
19F NMR
e [mÀ1 cmÀ1
]
d [ppm]
n = [Hz]
1
2
[FeII(6)2]2+
[FeII(7)2]2+
[FeII(8)2]2+
3316/3245
–/–
3391/3248
435
ꢀ550
ꢀ550
7450 (MLCT)
!100 (d–d)
!100 (d–d)
1.83
4.40
4.63
À78
À79
À64
38
36
5100
[a] In CH3CN, with concentrations ranging from 0.3ꢁ10À3 to 1.4ꢁ10À3 m. [b] In the solid state at 298 K, determined with a magnetic susceptibility balance.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2013, 78, 101 – 116 103

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
Complex [FeII(7)2]OTf2 also bears two tridentate ligands 7 in
its coordination sphere, and two OTf anions serve as counter-
ions, as confirmed by elemental analysis, 19F NMR spectroscopy,
and MS. As opposed to [FeII(6)2]OTf2, complex [FeII(7)2]OTf2 is
paramagnetic (meff =4.40 BM). As a consequence, its 1H NMR
spectrum shows large paramagnetic shifts between d=À20
and 120 ppm, which have previously been reported in the lit-
erature for related high-spin FeII systems, and often not all res-
onances can be observed.[27a,33a]
The aminopyridyl complex [FeII(8)2(OTf)2] bears two biden-
tate ligands 8 and two OTf ligands in the coordination sphere.
This complex is also paramagnetic (meff =4.63 BM), and features
large paramagnetic shifts between d=0 and 120 ppm in the
1H NMR spectrum. The general formulation [FeII(8)2(OTf)2] is
also confirmed by elemental analysis and MS.
CH3CN). The molar absorptivity e of the MLCT band is
7450mÀ1 cmÀ1 in CH3CN and is typical for MLCT bands (solid
trace in Figure 1); both the wavelengths and the high values
for the molar absorptivity are typical for FeII complexes with N-
coordinating donor ligands and have been reported previously
for structurally related species.[27a,33a,45] The UV/Visible spectra
barely changed when recorded in non-coordinating CH2Cl2
(the spectra are given in the Supporting Information). The simi-
larities of the UV/Visible spectra in coordinating CH3CN and
non-coordinating CH2Cl2 suggest that the structure of the com-
plexes in solution are identical, that is, the CH3CN solvent itself
does not displace the ligands or one of its coordinating N-
donor atoms; similar conclusions were drawn from UV/Visible
data for related iron complexes with nitrogen-donor atoms.[41]
The other two complexes [FeII(7)2]OTf2 and [FeII(8)2(OTf)2] did
not exhibit MLCT bands; they show very weak and broad ab-
sorbances around 550 nm, which could, owing to their low in-
tensity (e !100mÀ1 cmÀ1), be d–d transitions (dotted and
dashed traces in Figure 1).[46]
To assess the position of the OTf counterion in solution,
19F NMR spectra of the three complexes were recorded in
CD3CN. Non-coordinated OTf counterions give solvent-depen-
dent resonances around d=À70 ppm that shift significantly
downfield upon coordination to FeII.[44] In complexes
[FeII(6)2]OTf2 and [FeII(7)2]OTf2, the 19F NMR spectra resonances
were observed between d=À78 and À79 ppm (Table 1),
which is consistent with free, non-coordinated OTf. The widths
at half height (n1/2) for the 19F NMR spectroscopic resonances
for the two complexes are 38 and 36 Hz for the two com-
plexes, respectively. The small widths at half-height are indica-
tive of the absence of dynamic processes in solution involving
the OTf counterion,[27a] as all six coordination sites are occupied
by the two tridentate ligands.
In the amino pyridine complex [FeII(8)2(OTf)2], two coordina-
tion sites are occupied by the weakly coordinating OTf anion,
and its 19F NMR spectrum exhibited a very broad resonance
around d=À64 ppm (n1/2 ꢀ5100 Hz). The OTf ligand appears
to rapidly exchange with the CD3CN solvent, which leads to
significant line broadening of the resonance. The large n1/2
value is indicative of dynamic processes in solution involving
the OTf counterion.[27a]
The IR spectra of the new complexes exhibited strong bands
for the OTf counteranion. Non-coordinated OTf typically exhib-
its IR stretches at 1268, 1223, 1143, and 1030 cmÀ1; the solid-
state IR spectra of the three complexes also showed four com-
parable signals; some of the signals are desymmetrized for
[FeII(8)2(OTf)2] and exhibited shoulders, as previously reported
for OTf coordinated to metal centers.[33c,47] The complexes
[FeII(6)2]OTf2 and [FeII(8)2(OTf)2] feature NÀH units in their li-
gands. Absorptions were observed at 3245 and 3248 cmÀ1, re-
spectively, which are about 100 cmÀ1 lower than the corre-
sponding absorbances of the free ligands (Table 1). This shift in
the absorbances indicates that the NÀH unit is coordinated to
the iron center.
To unequivocally establish the structures of the new iron
complexes, the X-ray structures of complexes [FeII(6)2]OTf2 and
[FeII(8)2(OTf)2] were determined (Figure 2). Crystallographic pa-
rameters are given in Table 2 and key bond lengths and angles
are compiled in Table 3.
For complex [FeII(6)2]OTf2, the bond angles for the N atoms
in cis positions about the iron center range from 82.54(9) to
The UV/Visible spectra of the three complexes were record-
ed in CH3CN and CH2Cl2, and only [FeII(6)2]OTf2 features
a metal-to-ligand charge transfer (MLCT) band at l=435 nm
(Table 1 and Figure 1, which shows the traces recorded in
Figure 2. X-ray structures of [FeII(6)2]2+···OTf (left) and [FeII(8)2(OTf)(CH3CN)]+
(right). Hydrogen atoms, solvent molecules, and one of the OTf counter ions
have been omitted for clarity. Plots giving thermal ellipsoids are given in the
Supporting Information.
Figure 1. UV/Visible spectra of [FeII(6)2]OTf2 (solid line), [FeII(7)2]OTf2, and
[FeII(8)2(OTf)2].
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2013, 78, 101 – 116 104

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
98.26(9)8 and those in trans positions range from 174.15(7) to
178.06(8)8. The structure is, thus, best described as distorted
octahedral. The N-Fe-N bond angles within the ligands range
from 82.54(9) to 85.78(9)8, whereas they range from 91.04(9) to
101.9(2)8 between ligands. Thus, the average “bite angle” for
ligand 6 is around 848, which might be responsible for the dis-
tortion of the complex. The FeÀN bond lengths are between
1.984(2) and 2.011(2) ꢂ, which are typical for low-spin FeII com-
plexes. The two OTf counterions are not coordinated to the
iron center. There is a hydrogen bond between the SÀO units
of the OTf and the NÀH group of the ligand (Figure 2, left),
ranging from 2.0 to 2.3 ꢂ, which has been reported previously
for related complexes with ClÀ or BrÀ counterions.[36a] The
bond lengths and angles for [FeII(6)2]OTf2 closely match the
corresponding values of complexes [FeII(6)2]X2 (X=ClÀ, BrÀ),
which have been structurally characterized previously.[36] As
outlined above, there is an interesting difference in the coordi-
nation mode of ligand 6 in these two complexes with halide
counterions and in complex [FeII(6)2]OTf2. In complexes
[FeII(6)2]X2 (X=ClÀ, BrÀ), the two NÀH groups occupy positions
trans to each other.[36] However, the NÀH units are located cis
to each other in complex [FeII(6)2]OTf2, and trans to a pyridyl
ring in the solid state. At room temperature, the complex
shows dynamic behavior in solution, and it appears that the
isomer with the NÀH groups cis to each other crystallizes out
of solution (the crystals for X-ray diffraction were grown at
À188C). Complexes [FeII(6)2]Cl2 and [FeII(6)2]Br2 were not inves-
tigated by NMR spectroscopy in solution.[36] It is possible that
those complexes also show dynamic behavior in solution at
room temperature, but owing to the different counterions, the
other isomers precipitate out of solution.
Table 2. Crystallographic parameters.
[FeII(6)2]OTf2
[FeII(8)2OTf2]
empirical formula
C56H58F12Fe2N14O12S4
C29H31F3FeN5O5S·
(CH3CN)(CF3SO3)
864.62
100(2)
0.71073
Mr
T [K]
1587.10
100(2)
l [ꢂ]
0.71073
monoclinic
Pn
16.9936(10)
9.3511(6)
20.5805(12)
90
crystal system
space group
a [ꢂ]
b [ꢂ]
c [ꢂ]
triclinic
¯
P1
10.5308(6)
13.4426(8)
13.9190(9)
75.702(3)
77.494(3)
79.812(3)
1848.13(19)
2
1.554
0.609
888
0.22ꢁ0.10ꢁ0.09
1.95 to 26.60
a [8]
b [8]
102.006(3)
90
3198.9(3)
2
1.648
0.692
g [8]
V [ꢂ3]
Z
1calcd [MgmÀ3
]
m [mmÀ1
F(000)
]
1624
crystal size [mm3]
q range for data
collection [8]
0.28ꢁ0.27ꢁ0.20
2.02 to 32.18
index ranges
À25ꢁhꢁ25
À13ꢁkꢁ13
À30ꢁlꢁ30
139854
21639 [R(int)=0.0636] 7510 [R(int)=0.0521]
100.0
À13ꢁhꢁ12
À16ꢁkꢁ16
À17ꢁlꢁ17
42411
reflections collected
independent reflections
completeness to
98.6
q=25.008 [%]
absorption correction
semi-empirical from
equivalents
semi-empirical from
equivalents
0.9461 and 0.8791
full-matrix
least-squares on F2
7510/51/516
max. and min. transmission 0.8717 and 0.8277
refinement method
full-matrix
least-squares on F2
21639/15/937
data/restraints/
parameters
GOF on F2
For complex [FeII(8)2(OTf)2], the bond angles for N atoms in
cis positions about the iron center range from 76.37(7) to
94.71(7)8 and for those in the trans positions around 169.48.
The structure is, thus, also best described as distorted octahe-
dral. The two NÀH units are in cis positions to each other, as
observed for complex [FeII(6)2]OTf2. The N-Fe-N bond angles
within the ligands are 76.37(7) and 75.59(7)8, respectively, and
94.71(7) and 98.32(7)8 between ligands. Thus, the average “bite
angle” for the amino pyridyl ligand 8 is around 758, also lead-
ing to the distortion of the octahedral coordination geometry
for the complex. The FeÀN bond lengths are between
2.1463(19) and 2.257(2) ꢂ, which are about 0.2 ꢂ longer than
those for [FeII(6)2]OTf2. Complex [FeII(8)2(OTf)2] has a high-spin
configuration as determined by magnetic measurements
(Table 1), and it has previously been reported that FeIIÀN bond
lengths for high-spin complexes are around 2.2 ꢂ and about
0.2 ꢂ longer than those of corresponding low-spin complexe-
s.[26,33a] The X-ray structure of [FeII(8)2(OTf)2] reveals that one of
the OTf ions is coordinated to the metal center whereas the
other one serves as the counterion. The remaining coordina-
tion site is occupied by a CH3CN solvent molecule. In the origi-
nally isolated complex, there was no evidence for the presence
of CH3CN, as assessed by elemental analysis and IR data. We
assume that its presence in the coordination sphere observed
in the X-ray structure is a result of the crystallization procedure
1.021
1.015
final R indices
[I>2s(I)]
R indices
(all data)
absolute structure
parameter
R1 =0.0468,
wR2 =0.1148
R1 =0.0579,
wR2 =0.1222
0.669(9)
R1 =0.0391,
wR2 =0.0822
R1 =0.0629,
wR2 =0.0921
largest diff. peak and
1.381 and À0.761
0.696 and À0.483
hole [eꢂÀ3
]
Table 3. Key bond lengths [ꢂ] and angles [8].
[FeII(6)2]OTf2
[FeII(8)2]OTf2
Fe2ÀN7
1.964(2)
2.006(2)
1.997(2)
1.986(2)
2.011(2)
1.990(2)
1.487(4)
1.480(4)
1.521(5)
1.479(4)
82.51(10)
97.02(9)
94.33(10)
178.06(8)
175.45(10)
Fe1ÀN1
2.1463(19)
2.257(2)
2.1577(19)
2.249(2)
2.160(2)
2.0879(17)
1.474(3)
1.458(3)
1.497(3)
1.499(3)
76.37(7)
94.71(7)
94.08(8)
169.42(7)
169.47(7)
92.56(7)
Fe2ÀN8
Fe1ÀN2
Fe2ÀN9
Fe1ÀN3
Fe2ÀN10
Fe2ÀN11
Fe2ÀN12
N11ÀC43
N8ÀC30
Fe1ÀN4
Fe1ÀN5
Fe1ÀO1
N2ÀC6
N4ÀC19
C5ÀC6
C29ÀC30
C43ÀC44
N8-Fe2-N9
N9-Fe2-N10
N7-Fe2-N11
N7-Fe2-N10
N8-Fe2-N12
C18ÀC19
N1-Fe1-N2
N2-Fe1-N3
N1-Fe1-N5
N1-Fe1-N3
N2-Fe1-N5
N2-Fe1-O1
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2013, 78, 101 – 116 105

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
employed to obtain X-ray-quality crystals, which involved
CH3CN as solvent.
ketone ratio was much higher (6.6) for [FeII(6)2]OTf2. The alco-
hol/ketone ratio was still >1 for [FeII(7)2]OTf2 and [FeII(8)2(OTf)2]
under these conditions. A high concentration of radicals
formed from the peroxides promote free-radical chain mecha-
nisms, which result in low alcohol/ketone ratios.[48] When H2O2
is added slowly as a dilute solution, the concentration of free
radicals is minimized, and iron-centered [Fen=O] species
formed by a reaction of the peroxide with the iron complex
might be the actual oxidant[48] through insertion of the oxygen
into a CÀH bond of the cyclohexane to give cyclohexanol,
which results in a high alcohol/ketone ratio (see below). It ap-
pears that complex [FeII(6)2]OTf2 promotes a mechanistic path-
way through an iron–oxo species, as seen from the high alco-
hol/ketone ratio of 6.6, which is on the same order as reported
for iron complexes with tetradentate ligands such as [FeII(tpa)-
(OTf)2].[26]
The synthetic efforts described in this section afforded three
well-defined iron complexes [FeII(6)2]OTf2, [FeII(7)2]OTf2, and
[FeII(8)2(OTf)2] which, with the previously synthesized com-
plexes [FeII(4a)2(OTf)2] and [FeII(tpa)(OTf)2] were subsequently
employed in catalytic studies.
Catalysis and catalytic efficiency
The oxidation of cyclohexane is used frequently as a bench-
mark reaction to investigate catalytic selectivity and mechanis-
tic pathways.[11a] In these test reactions, a large excess amount
of the substrate over the peroxide oxidant is employed typical-
ly to avoid overoxidations.[28] The alcohol/ketone ratio of the
oxidation products gives information about the course of the
reaction, which is high for metal-centered mechanistic path-
ways and low for free-radical oxidation reactions (see
below).[28] Accordingly, we employed various cyclohexane/H2O2
or cyclohexane/tBuOOH ratios and the relevant results are
compiled in Table 4.
On the other hand, when tBuOOH was employed as the oxi-
dant, the alcohol/ketone ratio was always low (0.5–0.66) and
showed barely a change with the tBuOOH/cyclohexane ratio or
on how the oxidant was added to the reaction mixtures. It has
been reported that tBuO· radicals readily diffuse and can
induce radical chain reactions involving dioxygen, which result
in a low alcohol/ketone ratio (see below).[13d]
Table 4. Cyclohexane oxidation.[a]
The alcohol/ketone ratios did not significantly change when
the experiments were performed under inert conditions. When
the iron complexes were combined with the peroxide oxidants
without substrate, we typically observed some heat evolution
and occasionally bubbling, which might be evidence for cata-
lase activity, that is, peroxide decomposition to water and
oxygen. The systems could create their own oxygen atmos-
phere under inert conditions, which would explain the out-
come of the observed low alcohol/ketone ratios.
Reaction time
[min]
Cyclohexane oxidation with fast addition of H2O2
Peroxide/substrate
ratio
Alcohol/ketone
ratio
[b]
[FeII(6)2]2+
120
240
120
120
0.1
0.1
0.1
0.1
0.25
0.25
0.25
0.33
[FeII(7)2]2+
[FeII(8)2(OTf)2]
[c]
Cyclohexane oxidation with slow addition of H2O2
Next, we employed complex [FeII(6)2]OTf2 in oxidation reac-
tions of activated methylene groups, secondary alcohols, and
cyclohexene utilizing tBuOOH as the oxidant (Table 5, which
also includes the previously reported yields obtained with [FeII-
(4a)2(OTf2)]).[33a]
At a tBuOOH/substrate ratio of 4:1 and a catalyst load of
3 mol%, after 4 h at room temperature the corresponding
ketone products were obtained in 34 to 85% yield of isolated
product. Toluene was oxidized to benzaldehyde in 57% yield
and cyclohexene to the corresponding epoxide in 28% yield
utilizing similar conditions. Primary alcohols were not oxidized
under the conditions in Table 5. Preliminary GC studies showed
that the conversions for the other two complexes [FeII(7)2]OTf2
and [FeII(8)2(OTf)2] synthesized for this study were comparable.
As [FeII(6)2]OTf2 is the most easily accessible complex, it was
employed for further studies.
The previously synthesized imine complex [FeII(4a)2(OTf)2]
gave yields of isolated product between 22 and 91% for the
reactions illustrated in Table 5. The yields for the oxidation of
“doubly activated” substrates such as diphenylmethane or an-
thracene were comparable for the two complexes [FeII(6)2]OTf2
and [FeII(4a)2(OTf)2]. With imine complex [FeII(4a)2(OTf)2] yields
of isolated product were lower for the oxidation of secondary
alcohols to their corresponding ketones. As opposed to amine
complex [FeII(6)2]OTf2, imine complex [FeII(4a)2(OTf)2] did not
[FeII(6)2]2+
[FeII(7)2]2+
[FeII(8)2(OTf)2]
“19”
25
25
25
25
0.01
0.01
0.01
0.01
6.6
1.1
1.5
3.0
Cyclohexane oxidation with tBuOOH (slow addition of
a dilute solution)[c]
[FeII(6)2]2+
25
25
25
0.1
0.1
0.1
0.5
0.66
0.66
[FeII(7)2]2+
[FeII(8)2(OTf)2]
[a] Peroxide/catalyst ratio=10. Details are given in the Experimental Sec-
tion. [b] Fast addition of H2O2 to the reaction mixture. [c] Slow addition
over 25 min of a dilute 90 mm H2O2 or tBuOOH solution to the reaction
mixture.
For the oxidation reactions with H2O2, the alcohol/ketone
ratios obtained strongly depend on the cyclohexane/H2O2 ratio
and the manner in which the peroxide oxidant was added to
the reaction mixture. When the peroxide solution was added
rapidly to a solution containing the cyclohexane substrate and
the catalyst, a low alcohol/ketone ratio ranging from 0.25 to
0.33 was detected by GC, irrespective of the cyclohexane/oxi-
dant ratio.
However, when H2O2 was added as a dilute solution of
90 mm over the course of 25 min by syringe pump to the reac-
tion mixture at a 0.01 H2O2/cyclohexane ratio, the alcohol/
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2013, 78, 101 – 116 106

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
largely follow a pseudo-zero-order rate law and the kobs values
are derived from the slopes of the linear trends. A trace for ma-
terial “19” that was obtained by pretreatment of [FeII(6)2]OTf2
with tBuOOH is included as well (see below).
Table 5. Oxidation reactions employing iron complexes [FeII(6)2]OTf2 and
[FeII(4a)2(OTf)2] as catalyst and tBuOOH as oxidant.
Entry
Substrate (9)
Product
(10)[a]
Yield of
[FeII(6)2]2+
[%][b]
Yield[33a] of
[Fe(4a)2]+
[%][b]
As can be seen from Figure 3, the catalytic activity is about
comparable for complexes [FeII(6)2]OTf2 and [FeII(7)2]OTf2 (kobs
=
0.19 and 0.1 minÀ1). It is slightly lower for [FeII(tpa)(OTf)2], [FeII-
(4a)2(OTf)2], and [FeII(8)2(OTf)2] (kobs =0.045, 0.059, and
0.018 minÀ1, respectively). Complex [FeII(4a)2(OTf)2] also gave
lower yields for some of the oxidation reactions in Table 5,
which is in accord with its lower activity than [FeII(6)2]OTf2.
However, in general, the catalytic activities of the complexes
do not differ greatly.
1
2
85
76
91
74
3
79
51
Experiments to better understand the course of the oxida-
tion reactions
4
5
34
71
47
22
The course of iron-catalyzed oxidation reactions with peroxide
oxidants is still an objective of current research. A simplified
mechanistic picture for ROOH is outlined in Scheme 2 (R=H,
tBu).[11a,48,49] It seems to be generally accepted that the reaction
of an iron complex with ROOH first yields an iron–peroxo spe-
cies [FeIIIÀOÀOÀR] (11), which has been described as not oxi-
dizing organic substrates and is only a precursor of the actual
oxidants.[50]
6
7
8
69
67
58
–
47
23
The further course of the oxidation reaction depends on the
“fate” of species 11; it decays by either homolytic or heterolytic
bond cleavage of the FeÀO or OÀO bond, which has been re-
ported to depend on the spin state of the iron complex[51] and
on the pH value.[52] Homolytic cleavage of the OÀO or FeÀO
9
57
28
–
–
10
C
C
bond yields RÀO or HÀO radicals eventually. These radicals
can abstract hydrogen atoms from hydrocarbons, which leads
to unfavorable radical chain oxidation reactions following
a Haber–Weis mechanism (Scheme 2, right-hand path);[11a] O2
from the atmosphere (or from iron-catalyzed peroxide decom-
position) is incorporated through the course of the reaction.
The peroxo radical 14 disproportionates through a Russel-type
[a] Typical conditions: Substrate (0.6 mmol), tBuOOH (2.4 mmol), and cata-
lyst (3 mol%), 4 h at room temperature in CH3CN/pyridine (1:1, 2 mL). The
products were isolated by using column chromatography. [b] Yield of iso-
lated product.
show catalytic activity at all for the oxidation of toluene or cy-
clohexene.
To obtain additional informa-
tion for catalytic efficiencies, we
followed the oxidation of diphe-
nylmethane
with
tBuOOH
(entry 2 in Table 5) by GC at the
significantly reduced catalyst
loading of 0.3 mol% over 3 h for
all iron complexes [FeII(6)2]OTf2,
[FeII(7)2]OTf2, and [FeII(8)2(OTf)2]
utilized for this study. The
known complexes [FeII(tpa)-
(OTf)2] and [FeII(4a)2(OTf)2] were
also investigated under the
same conditions. The traces are
compiled in Figure 3, which also
displays the observed rate con-
stant kobs for the different cata-
lysts. At this low catalyst loading,
Figure 3. Activity comparison of different iron complexes. The kobs values were determined from the linear trends
it appears that the reactions of the traces, which are not shown.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2013, 78, 101 – 116 107

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
tri-tert-butylphenol (TBPH), which is an efficient scavenger for
oxygen-centered radicals.[59]
It is largely accepted that iron-catalyzed oxidations with
tBuOOH follow a radical mechanism.[13d] Still, the oxidation re-
actions in Table 5 do not proceed without a catalyst. Only an
8% conversion of diphenylmethane to benzophenone was ob-
served when the radical initiator azobisisobutyronitrile (AIBN)
was employed in place of the catalyst at a significantly higher
reaction temperature of 808C [Eq. (1)]. The iron complexes
C
might, thus, provide a steady supply of tBuO radicals in solu-
tion to keep the reaction running.
In the oxidation of benzylic methylene groups, such as fluo-
rene or diphenylmethane (entries 1 to 5 in Table 5), we never
observed alcohol products. The oxidation reactions are not re-
tarded or inhibited by the radical scavenger TBPH. However, it
has been our observation that benzylic alcohols such as 1-phe-
nylethanol are easily oxidized by peroxides under the reaction
conditions, even without a catalyst. Thus, it is not possible to
establish an alcohol/ketone ratio to elucidate a mechanism. It
appears reasonable to assume that the activated methylene
groups in Table 5 are oxidized first to the alcohol. Alcohols are
in general more easily oxidized[15] and could, under the reac-
tion conditions, be further oxidized rapidly to the ketone
(Scheme 3, top).
Scheme 2. Simplified reactivities of iron complexes with ROOH (R=H, tBu).
termination[53] and peroxide 15 disproportionates promoted by
iron,[54] which gives low alcohol/ketone ratios under alkane oxi-
dations. It has been reported that the reaction can still proceed
through an iron-centered oxidant if the concentration of the
radicals is kept low by slow addition of the oxidant.[11a,48]
Heterolytic cleavage of iron–peroxo species 11 gives a high-
valent iron–oxo species [FeV=O] (16), which is a metal-centered
oxidant.[50,55] Its oxidation chemistry is much better controlled
through supporting ligands on the iron center, and its exis-
tence has been documented for biomimetic, nonheme iron
complexes spectroscopically,[56] structurally,[57] and computa-
tionally.[58] The [FeV=O] species can be stabilized by appropriate
ligands and can insert an oxygen atom into a CÀH bond
To corroborate the faster oxidation of alcohols relative to
benzylic methylene groups, we performed a competition ex-
periment between an alcohol and a hydrocarbon (Scheme 3,
bottom). When a 1:1 mixture of diphenylmethane and cyclooc-
tanol was treated with 0.5 equivalents tBuOOH under condi-
tions identical to those in Table 5, 43% of the cyclooctanol but
only 11% of the diphenylmethane was converted into the cor-
responding ketone, thereby establishing that the alcohol
reacts faster than the benzylic CH2 unit.
through
a
“bound–rebound mechanism”[11a] involving the
bracketed intermediate 17 in Scheme 2, in which the radical in-
termediate R’ presumably never leaves the solvent cage and
does not diffuse freely. This pathway gives specifically alcohols
as the oxidation products, and [FeII] is recovered, which can
participate in another oxidation cycle.
As outlined above, the catalytic oxidation of cyclohexane uti-
lizing tBuOOH appears to proceed through a radical mecha-
nism (as seen by the low alcohol/ketone ratios in Table 4). Only
complex [FeII(6)2]OTf2 gives a high alcohol/ketone ratio of 6.6
(Table 4), when H2O2 is employed as the oxidant; this reaction
presumably involves the iron-centered oxidant 16 (Scheme 2).
Notably, complex [FeII(6)2]OTf2 exhibits an NÀH unit; it may
become significantly acidic upon coordination to the iron
center. A nonradical reaction pathway through an iron–oxo
species might be promoted at low pH values.[52] Furthermore,
for complex [FeII(6)2]OTf2, the alcohol/ketone ratio does not
significantly change when performed in the presence of 2,4,6-
Scheme 3. Potential oxidation sequences (top) and the competition experi-
ment.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2013, 78, 101 – 116 108

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
The oxidation of the benzylic alcohol intermediates as well
as the alcohol substrates in Table 5 to the ketones might pro-
ceed through different mechanisms; in the present context,
a radical mechanism[60] or a cyclic transition state are potential
pathways.[48] In a cyclic transition state the alcohol coordinates
first to the iron center as does the peroxide oxidant [Eq. (2)].[48]
Intramolecular hydride abstraction from the alcohol by the per-
oxide might proceed through six-membered transition state
18 [Eq. (2)]; such a transition state is not possible for saturated
hydrocarbons, as they typically do not coordinate to a metal
center.[48]
Alternatively, the oxidation can proceed through a radical
mechanism following the pathway depicted in Scheme 2 (right
path). Benzylic CÀH bonds and CÀH bonds in alcohols are
weaker than saturated CÀH bonds by about 10 kcalmolÀ1,
which explains why the oxidation of alcohols to ketones is
much faster than the oxidation of alkanes to alcohols
(Scheme 3).[60a] The reactions in Table 5 are not affected by
a radical scavenger. However, it has been suggested that iron-
catalyzed oxidation reactions can follow several mechanistic
pathways simultaneously,[29a,61] and a radical scavenger might
block just one of them.
Figure 4. Time-resolved UV/Visible spectra of [Fe(7)2]OTf2 (top) and
[Fe(8)2(OTf)2] (bottom) in CH3CN after addition of H2O2 (dotted line: before
addition).
Irrespective of the exact mechanism, a common first inter-
mediate for the title reactions is most likely the iron–peroxo
species [FeIIIÀOÀOÀR] (11, R=H, tBu; Scheme 2). It can be ob-
served by UV/Visible spectroscopy, and has been reported to
transfer bands around 450 nm for that species.[17b] A peroxo
species might, under the conditions shown in Figure 4, be very
unstable, and the observed bands could also very well be due
to intermediates resulting from oxidative ligand decomposition
give
a
charge-transfer band between l=550 and
630 nm.[28,50,62,63] We hypothesized that UV-spectroscopic moni-
toring of the formation and decay of the peroxo species [FeIIIÀ (see below). Still, the fast spectral changes within minutes after
OÀOÀR] could give further information about the catalytic ac-
tivity of the iron complexes investigated in this study. Accord-
ingly, we recorded UV/Visible spectra over time after treatment
of solutions of the complexes in CH3CN with the peroxide oxi-
dants ROOH in the presence of cyclohexane.
addition of an oxidant demonstrate a high reactivity of the
two complexes towards the peroxide oxidants.
However, the bpa complex [FeII(6)2]OTf2 underwent signifi-
cant spectroscopic changes in solution when H2O2 or tBuOOH
was added to the complex and UV/Visible spectra were record-
ed before and immediately after addition of the peroxide in
time intervals of one minute. As shown in Figure 5 (top), the
intense charge-transfer band for the complex at 434 nm
(dotted line) completely disappeared immediately after H2O2
was added to the complex. Instead, two new bands formed at
490 and 572 nm. The latter band is characteristic for iron–
peroxo species [FeÀOOH] and its molar absorptivity
e (4000mÀ1 cmÀ1) is also on the same order as previously re-
ported for similar complexes.[28,50,62] This band decayed over
a period of 6 min and had completely disappeared after
15 min. When tBuOOH was employed, after one minute a new
band was observed at 579 nm (e around 700mÀ1 cmÀ1, solid
line in Figure 5, bottom). The band showed a lower absorptivi-
The benzylated amine complex [FeII(7)2]OTf2 gave, immedi-
ately after addition of H2O2, a new, weak shoulder around
480 nm (Figure 4, top), which slowly decayed over the course
of 5 min. Complex [FeII(8)2(OTf)2] gave, immediately after addi-
tion of H2O2, a new band at l=424 nm with a molar absorp-
tion around 700mÀ1 cmÀ1, which barely changed over time
(Figure 4, bottom). The new bands originate presumably—
owing to their intensity—from charge transfer. For both com-
plexes, almost identical results were obtained when tBuOOH
was used as the oxidant (for spectra, see the Supporting Infor-
mation). The wavelength of the new bands for both complexes
are somewhat low for an iron–peroxo complex [FeÀOÀOÀR],
although some complexes have been reported to give charge-
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2013, 78, 101 – 116 109

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
Figure 6. Time-resolved UV/Visible spectra of [FeII(4c)2(OTf2)] in CH3CN after
addition of H2O2 (dotted line: before addition).
only very slowly, as opposed to the corresponding bands for
[FeII(6)2]OTf2, which had completely decayed two minutes after
the addition of the oxidant (Figure 5). A band around 600 nm
was not observed for [FeII(4c)2(OTf)2]. The slow decay of the
band at 418 nm could indicate slow ligand decomposition, as
imines are somewhat more difficult to oxidize than amines.
Figure 5. Time-resolved UV/Visible spectra of [Fe(6)2]OTf2 in CH3CN after ad-
dition of H2O2 (top) and tBuOOH (bottom). Dotted lines: before addition.
Ligand-decomposition pathways and their impact on the
catalytic activity
ty than the band obtained with H2O2 and had disappeared
completely after two minutes.
For [FeII(tpa)]2+, a similar reactivity has been reported previ-
ously with tBuOOH[62a] and H2O2.[50] Upon reaction of the oxi-
dant at À368C, a new band at 600 nm (e=2000mÀ1 cmÀ1) was
As demonstrated in the previous sections, different ligand ar-
chitectures can have an impact on the catalytic activity of their
respective metal complexes. It has been argued that oxidative
ligand decomposition during catalytic applications can de-
observed, which was assigned to an iron–peroxo species [FeÀ crease the catalytic activity of metal complexes,[64] although
OOH] and that also decayed over time to give a [FeIV=O] spe-
cies.
the opposite has been reported for pyridin-2-yl-type ligands
and their manganese complexes.[65] We were, thus, interested
in determining what decomposition products are formed as
a consequence of ligand oxidation under catalytic conditions
from the iron complexes investigated in this study and how
the oxidation of the ligands impairs the catalytic activity.
However, the transient species that formed in the experi-
ments in Figure 5 cannot be unambiguously assigned to
a structure. The spectra in Figure 5 were recorded at room
temperature. The new bands around 572 nm are in the range
for other [FeÀOOR] species reported in the literature, but these
species are typically investigated at much lower temperatures,
although studies at room temperature were also reported.[62c]
Furthermore, [Fe(6)2]OTf2 does not immediately have an open
coordination site available. It might be that the complex opens
a coordination site and forms a peroxo species, but the spec-
tral changes observed over time might also be due to ligand
decomposition.
Accordingly, the [FeII(6)2]OTf2 complex was treated with
tBuOOH under catalytic conditions identical to those employed
in the reactions shown in Table 5 but without the presence of
substrate. The oxidative decomposition products of the iron
complex were isolated as a mixture, which was complex and
could not be analyzed by NMR spectroscopy or MS [although
a species bearing two oxidized ligands was identified by MS;
Eq. (3)]. The mixture will be referred to subsequently as “19”.
We were, thus, looking for a method to detach the ligands
from the iron center after oxidation to identify ligand-decom-
position products. Accordingly, [FeII(6)2]OTf2 was treated with
tBuOOH under conditions otherwise identical to the catalysis
experiment shown in Table 5, but without a substrate. The li-
gands were subsequently displaced from the iron center by
adding aqueous NaCN to the reaction mixture. The displaced
ligands (or their oxidation products, respectively) were isolated
A different behavior was observed for the iron–imine com-
plex [FeII(4c)2(OTf)2], for which an experiment similar to those
in Figures 4 and 5 was performed with tBuOOH (Figure 6).[33a]
After addition of the tBuOOH oxidant, the charge-transfer
band of the complex at 418 nm intensified, presumably owing
to the fact that the polarity of the solution increased after the
addition of the aqueous solution of the oxidant. Over the
course of 10 min, the charge-transfer band at 418 nm decayed
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2013, 78, 101 – 116 110

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
Ligands 6 in [FeII(6)2]OTf2 were cleanly oxidized to the corre-
sponding diamide 20, which was the only product detected
after isolation. For [FeII(7)2]OTf2, a complex mixture of decom-
position products was observed, and only the three major
components are displayed in Scheme 4. Diamide 20 was de-
tected, together with some 21 and 22; the benzyl or picolyl
units can be oxidatively cleaved from the ligands after oxida-
tion of the CH2 units, as observed previously in copper com-
plexes bearing structurally related ligands.[66] The imine unit in
[FeII(4a)2(OTf)2] also was oxidized cleanly to the corresponding
amide 23. Complex [FeII(tpa)(OTf)2] behaves slightly differently;
it was only oxidized partially to compounds 6, 24, 25, and 26,
and the material isolated after oxidation also contained the
original tpa ligand 1a.
by extraction and analyzed by GC–MS, NMR spectroscopy, and
IR spectroscopy. Similar reactions were performed with the
other complexes [FeII(tpa)(OTf)2], [FeII(7)2]OTf2, and [FeII(4a)2-
(OTf)2], in which the material obtained was analyzed by GC–
MS. The results are compiled in Scheme 4, and the GC–MS
traces are given in the Supporting Information. In all cases,
about 90% of the expected mass was recovered, and, thus,
little material was lost during workup.
It appeared that the NCH2 unit of the ligands is commonly
oxidized to an amide NC=O for the three complexes, presuma-
bly through an intramolecular oxygen transfer from an Fe=O
species 27 (Scheme 5), as previously suggested for an intramo-
Scheme 5. Potential pathway for an intramolecular NHC2 oxidation.
lecular aryl hydroxylation reaction involving the iron complex
[FeII(L)(CH3CN)2](ClO4)2 (L=modified tpa).[67] Copper[68] and
iron[69] complexes of diamide 20 are known; in all these cases,
the amide ligands are coordinated through the N atoms to the
iron center, as shown by X-ray studies. Related iron–amide
complexes are still capable of catalytically activating H2O2.[70]
Thus, it is likely that the decomposed ligands in Scheme 4
were still coordinated to the iron, after they had been oxidized
to the corresponding amides.
Interestingly, we did not observe any chemical changes of
the aryl or pyridyl ring systems of the ligands (although small
amounts of bipyridine were sometimes detected, which might
have resulted from oxidative dimerization of the pyridine sol-
vent). When iron complex [FeII(6)2]OTf2 was subjected to ligand
displacement without prior exposure to the tBuOOH oxidant,
only the unoxidized ligand 6 was recovered. Thus, the oxida-
tion of the ligand does not take place during the displacement
procedure.
We were finally interested to determine whether or not the
oxidized ligands could impair the catalytic selectivity and activ-
Scheme 4. Experiments to determine ligand-decomposition products under
oxidative conditions.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2013, 78, 101 – 116 111

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
ity, and material “19” was employed for that purpose, which
was obtained by oxidizing [FeII(6)2]OTf2 utilizing tBuOOH as de-
scribed above [Eq. (3)]. Material “19” was subsequently em-
ployed in the determination of the catalytic activity in the oxi-
dation of diphenylmethane in the same way as it was per-
formed for the unoxidized iron complexes in Figure 3. Some-
what surprisingly, the catalytic activity of “19” is comparable to
the unoxidized complex [FeII(6)2]OTf2 (Figure 3, kobs =0.11).
However, the exact composition of the isolated oxidized mate-
rial “19” is not known, thus making a direct comparison of ab-
solute values difficult.
The catalytic activity of “19”, which was obtained by oxida-
tion of [FeII(6)2]OTf2 [Eq. (3)] shows catalytic activity in the test
reaction for Figure 3 that is comparable to the other com-
plexes. The ligand-decomposition experiments (Scheme 4) re-
vealed that in the ligands the NCH2 units are oxidized to the
amides NC=O as reported for other iron complexes,[72] but the
ligands largely remain di-, tri-, or tetradentate. Consequently,
the pyridyl units for the ligands can still coordinate to the iron
center through the nitrogen atoms, as shown for related sys-
tems,[69,70] even after oxidative degradation of the ligand,
which seems to be the major aspect for retaining catalytic ac-
tivity. It has been described before that in manganese-based
catalyst systems the oxidation of pyridyl-based ligands to pyri-
dine-2-carboxylic acid takes place prior to the onset of the cat-
alytic oxidation of organic substrates.[65] In our catalytic sys-
tems, the oxidation of the ligand appears to be very fast and
our reactions do not exhibit an induction period (Figure 3).
Thus, we cannot determine whether or not the oxidation of
the ligands is a necessary prerequisite for developing catalytic
activity, but oxidation of the ligands, in turn, does not shut
down the activity.
Likewise, when oxidized material “19” was employed in the
cyclohexane oxidations in Table 4, the alcohol/ketone ratio de-
creased, relative to the unoxidized complex [FeII(6)2]OTf2, from
6.6 to 3. The selectivity of “19” is somewhat lower than that of
its precursor, but still higher than those of the other complexes
synthesized for this study. Thus, oxidative ligand decomposi-
tion slightly decreases the selectivity of the complex in the oxi-
dation of cyclohexane but has no measurable negative impact
on the catalytic activity for the oxidation of diphenylmethane
under the conditions in Figure 3.
Discussion
Conclusion
The preceding data establish that the iron complexes
[FeII(6)2]OTf2 and [FeII(7)2]OTf2 with tridentate aminopyridine li-
gands are another efficient class of iron catalysts active for the
oxidation of CH2 groups and alcohols to the corresponding ke-
tones with peroxide oxidants. The mild reaction conditions
(4 h reaction time at room temperature) make them an attrac-
tive starting point for further investigation and optimization
effort.
In conclusion, we have synthesized a series of complexes of
the general formula [FeIILn]OTf2 (n=1, 2) bearing bi- and tri-
dentate aminopyridyl ligands L, two of which were structurally
characterized. The new complexes showed catalytic activity in
the oxidation of benzylic CH2 groups and secondary alcohols
to the corresponding ketones (3% catalyst load, 4 equivalents
ROOH, RT, 4 h, 28 to 85% yield of isolated product). The cata-
lytic activity of the new complexes was determined and com-
pared with previously synthesized iron complexes [FeII(tpa)-
(OTf)2] and [FeII(iminopyridyl)2(OTf)2] bearing the tetradentate
tris(picolyl)amine (tpa) and bidentate iminopyridyl ligands in
their coordination sphere.
It has been argued that two tridentate ligands in the coordi-
nation sphere of iron complexes might decrease the catalytic
activity, as all six coordination sites are occupied by the li-
gands.[43] Indeed, the majority of biomimetic, nonheme iron
catalysts bear one ligand with four or five nitrogen or oxygen
donor atoms.[1d,16a,17b] However, complexes [FeII(6)2]OTf2 and
[FeII(7)2]OTf2 show catalytic activity that is comparable (even
slightly higher) than the previously established complex [FeII-
(tpa)(OTf)2] in the oxidation of diphenylmethane in Figure 3.
Complex [FeII(6)2]OTf2 needed to detach one of the pyridyl
rings to initiate catalytic activity if the reaction proceeds
through an iron–peroxo intermediate [FeÀOOR]. However,
complex [FeII(6)2]OTf2 with two tridentate ligands could also
operate through an outer-sphere electron-transfer mecha-
nism,[71] and the species observed in the UV/Visible spectrum is
an intermediate of the ligand decomposition that is taking
place during the course of the reaction.
Complex [FeII(8)2(OTf)2] with two bidentate ligands 8 exhibits
a somewhat diminished catalytic activity, and so does the
imine complex [FeII(4a)2(OTf)2]; there is a slight dependency of
the catalytic activity on the denticity of the ligands coordinat-
ed to the iron center, as the complexes with tris- or tetraden-
tate ligands perform slightly better than the complexes bear-
ing bidentate ligands.
The denticity of the ligands has a slight impact on the cata-
lytic activity, as the iron complexes featuring tri- and tetraden-
tate ligands in their coordination sphere performed slightly
better than their bidentate counterparts. Ligand-decomposi-
tion products were identified that form under the oxidative,
catalytic conditions. It appears that primarily NCH2 units are
oxidized to the corresponding amides NC=O. However, the
overall denticity of the ligands largely stays intact, and no oxi-
dations of the pyridyl units were observed. When [FeIIL2]OTf2
(L=di(picolyl)amine) was treated first with peroxide to obtain
“oxidized” material, it was determined that its catalytic activity
is not diminished appreciably relative to the “unoxidized” start-
ing material.
The findings presented herein can assist in the future design
of catalytically active iron complexes for the title reaction; em-
ployment of ligands with a high denticity appears to have
a beneficial impact on the performance of the complexes, and
oxidative ligand decomposition does not inevitably lead to cat-
alyst deactivation.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2013, 78, 101 – 116 112

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
dried under high vacuum to give [Fe(6)2]OTf2 as red crystals
(0.351 g, 0.466 mmol, 96%). H NMR (300 MHz, CD3CN, 258C, TMS):
Experimental Section
1
General
d=9.12 (brs, 4H; pyridylÀH), 7.97 (brs, 2H; pyridylÀH), 7.82 (t,
J(H,H)=7.2 Hz, 8H; pyridylÀH), 7.64 (brs, 2H; pyridylÀH), 5.95 (s,
2H; NH), 5.36–5.25 ppm (m, 8H; 4CH2); 1H NMR (300 MHz,
[D6]acetone, À508C, TMS): d=9.11 (d, 2J(H,H)=5.0 Hz, 2H; NCH
pyridyl), 8.29 (d, 2J(H,H)=5.0 Hz, 2H; NCH, pyridyl), 7.93 (t,
3J(H,H)=7.3 Hz, 2H; pyridyl), 7.79 (t, 3J(H,H)=7.3 Hz, 2H; pyridyl),
7.71 (d, 2J(H,H)=7.4 Hz, 4H; pyridyl), 7.29 (t, 3J(H,H)=6.0 Hz, 2H;
pyridyl), 6.05 (s, 2H; NH), 5.19 (dd, 2J(H,H)=8.4 Hz, 2H; NCHH’),
4.44 (d, J(H,H)=16.5 Hz, 2H; NCHH’), 4.35 (d, J(H,H)=18.5 Hz, 2H;
NCHH’), 4.27 ppm (d, 2J(H,H)=11.5 Hz, 2H; NCHH’); 13C{1H} NMR
(75 MHz, [D6]acetone, 258C, TMS): d=164.4 (pyridyl), 156.0 (pyrid-
yl), 137.8 (pyridyl), 128.5 (pyridyl), 124.3 (pyridyl), 59.0 ppm (CH2);
IR (neat solid): n˜ =3245 (m, NÀH), 1254 (s, OTf), 1162 (sh, OTf),
1028 cmÀ1 (sh, OTf); MS (FAB): m/z (%): 603 (30) [Fe(6)2OTf+], 404
(100) [Fe(6)OTf+], 255 (20) [Fe(6)+]; HRMS: m/z calcd for
C25H26F356FeN6O3S: 603.10883; found: 603.1091 (corresponds to
[Fe(6)2(OTf)]+); elemental analysis calcd (%) for C26H26F6FeN6O6S2: C
41.50, H 3.48, found: C 41.64, H 3.49.
Chemicals were treated as follows: toluene and diethyl ether dis-
tilled from Na/benzophenone; CH2Cl2, MeOH, and CH3CN distilled
from CaH2. CHCl3, silica, pyridine, tri(2-picolyl)amine (tpa, 1a), bis(2-
picolyl)amine (6, all Aldrich), and other materials were used as re-
ceived. Ligand 7[42] and [Fe(OTf)2][73] were prepared according to
the literature. All reactions were carried out under nitrogen and by
employing standard Schlenk techniques, and workups were carried
out in the air.
2
2
NMR spectra were obtained at room temperature on a Bruker
Avance 300 MHz or a Varian Unity Plus 300 MHz instrument at
room temperature and were referenced to a residual solvent
signal; all assignments are tentative. GC–MS spectra were recorded
on a Hewlett Packard GC–MS System Model 5988A. UV/Visible
spectra were recorded on Varian Cary 50 Bio spectrophotometer.
Exact masses were obtained on a JEOL MStation [JMS-700] mass
spectrometer. IR spectra were recorded on a Thermo Nicolet
360 FTIR spectrometer. Elemental analyses were performed by At-
lantic Microlab Inc., Norcross, GA, USA. Magnetic moments,
19F NMR spectra, and UV/Visible data are listed in Table 1.
[Fe(7)2]OTf2: Acetonitrile (2 mL) was added to a Schlenk flask con-
taining benzylbis(2-picolyl)amine (7, 0.082 g, 0.283 mmol). Subse-
quently, [Fe(OTf)2] (0.062 g, 0.141 mmol) was added to the red-
orange solution. The red-orange solution was then heated at 608C
for 1 h. The resulting solution was cooled down to 08C and layered
with diethyl ether (8 mL) and stored at À188C overnight to yield
a red-tan precipitate. The supernatant was decanted and the re-
sulting solid was washed with cold diethyl ether (3ꢁ5 mL) and
dried under high vacuum to give the product [Fe(7)2]OTf2 as a red-
tan solid (0.119 g, 0.128 mmol, 90%). 1H NMR (300 MHz, CD3CN,
258C, TMS): d=111.6 (brs, 2H; pyridylÀH), 75.9 (brs, 2H; pyridylÀ
H), 64.6 (brs, 2H; pyridylÀH), 58.0 (brs, 2H; pyridylÀH), 44.5 (brs,
2H; pyridylÀH), 42.5 (brs, 2H; pyridylÀH), 41.3 (brs, 2H; pyridylÀH),
15.4 (s, 3H), 13.7 (s, 3H), 8.36 (brs, 2H), 7.36 (brs, 4H), 6.20 (brs,
2H), 4.81–4.17 (m, 4H), 3.39 (s, 1H), 1.99 (s, 1H), À10.2 (s, 1H; pyr-
idylÀH), À16.4 (s, 1H; pyridylÀH), À19.0 ppm (s, 2H; pyridylÀH); IR
(neat solid) n˜ =1251 (s, OTf), 1152 (m, OTf), 1028 cmÀ1 (s-sh, OTf);
MS (FAB): m/z (%): 783 (30) [Fe(7)2(OTf)]+, 494 (100) [Fe(7)(OTf)]+;
HRMS: m/z calcd for C39H38F356FeN6O3S: 783.20270; found:
783.2018 (corresponds to [Fe(7)2(OTf)]+); elemental analysis calcd
(%) for C40H38F6FeN6O6S2: C 51.51, H 4.11; found: C 50.68, H 4.07.
Synthesis
Ligand 8: 2-Pyridine carboxaldehyde (1.0 g, 9.34 mmol) was dis-
solved in methanol (10 mL), cooled with an ice bath, while 2-me-
thoxyaniline (1.16 g, 9.43 mmol) in methanol (10 mL) was added
dropwise. The ice bath was then removed, allowing the reaction to
stir for 1 h at room temperature. Then NaBH4 (0.741 g, 19.61 mmol)
was added in 3 portions at 08C. The reaction mixture was then al-
lowed to stir overnight at room temperature. After 12 h, aqueous
HCl solution (5m) was added until a pH of 4 was reached. The re-
action mixture was then allowed to stir at room temperature for
an additional hour. Then aqueous NaOH (2m) solution was added
until a pH of 12 was reached. Then CH2Cl2 (30 mL) was added and
the two layers were transferred to a separating funnel. The organic
phase was collected and the product was extracted from the aque-
ous layer, using CH2Cl2 (2ꢁ30 mL). The combined organic layers
were dried over sodium sulfate and the solvent was removed
under vacuum to yield a yellow oil that was subsequently purified
by means of column chromatography (CH2Cl2/Et3N 90:10 as eluent)
to give ligand 8 as a red oil (1.87 g, 8.73 mmol, 94%). 1H NMR
(300 MHz, CDCl3, 258C, TMS): d=8.55 (d, J(H,H)=2.9 Hz, 1H; pyrid-
ylÀH), 7.53 (t, J(H,H)=6.7 Hz, 1H; pyridylÀH), 7.28 (d, J(H,H)=
7.2 Hz, 1H; pyridylÀH), 7.10 (d, J(H,H)=7.2 Hz, 1H; pyridylÀH),
6.82–6.75 (m, 2H; ArÀH), 6.67–6.62 (m, 1H; ArÀH), 6.51 (d, J(H,H)=
7.6 Hz, 1H; ArÀH), 5.19 (s, 1H; NH), 4.45 (s, 2H; CH2), 3.82 ppm (s,
3H; OCH3); 13C{1H} NMR (300 MHz, CDCl3, 258C, TMS): d=159.5 (s;
aromatic), 149.6 (s; aromatic), 147.4 (s; aromatic), 138.2 (s; aromat-
ic), 137.1 (aromatic), 122.4 (s; aromatic), 121.8 (s; aromatic), 117.2
(s; aromatic), 110.6 (s; aromatic). 109.9 (s; aromatic), 55.8 (s; NCH2),
49.6 ppm (s; OCH3); IR (neat oil): n˜ =3391 cmÀ1 (m, NH); MS (FAB):
m/z (%): 214 (100) [8+].
[Fe(8)2(OTf)2]: Acetonitrile (3 mL) was added to a Schlenk flask con-
taining [Fe(OTf)2] (0.265 g, 0.607 mmol). Ligand
8 (0.260 g,
1.22 mmol) was added to the yellow transparent solution and the
resulting red-purple solution was allowed to stir for 1.5 h at room
temperature. This mixture was then concentrated to about 1.5 mL,
cooled down to 08C, layered with diethyl ether (10 mL), and stored
at À188C overnight to yield a purple crystalline precipitate. The su-
pernatant was decanted and the precipitate was dried under high
vacuum to give [Fe(8)2(OTf)2] as purple crystals (0.443 g,
1
0.566 mmol, 93%). H NMR (300 MHz, CD3CN, 258C, TMS): d=53.9
(brs, 2H; pyridylÀH), 45.3 (brs, 2H; pyridylÀH), 11.9 (brs, 2H; pyrid-
ylÀH), 6.6–4.4 (brs, ca. 18H), À20.8 ppm (s, 2H; pyridylÀH); IR (neat
solid): n˜ =3268 (w, NÀH), 1239 (s, OTf), 1157 (m, OTf), 1024 cmÀ1 (s
sh, OTf); MS (FAB): m/z (%): 633 (30) [Fe(8)2(OTf)+], 419 (100)
[Fe(8)(OTf)+]; HRMS: m/z calcd for C27H28F356FeN4O5S: 633.10809;
found: 633.1081 (corresponds to [Fe(8)2(OTf)]+); elemental analysis
calcd (%) for C28H28F6FeN4O8S2: C 42.98, H 3.61; found: C 42.57, H
3.74.
[Fe(6)2]OTf2: Acetonitrile (3 mL) was added to a Schlenk flask con-
taining [Fe(OTf)2] (0.211 g, 0.483 mmol). Ligand
6 (0.192 g,
0.967 mmol) was added to the resulting yellow transparent solu-
tion and the resulting red solution was allowed to stir for 30 min
at room temperature. This mixture was then concentrated to
about 1 mL, cooled down to 08C, layered with diethyl ether
(10 mL), and stored at À188C overnight to yield a red crystalline
precipitate. The supernatant was decanted and the precipitate was
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2013, 78, 101 – 116 113

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
Representative catalytic procedure (Table 5)
Acknowledgements
The substrate fluorene (0.100 g, 0.601 mmol) and the catalyst
[FeII(6)2]OTf2 (0.014 g, 0.018 mmol) were dissolved in pyridine
(1.0 mL) and CH3CN (1.0 mL). The oxidant tBuOOH (0.344 mL,
70 wt% in H2O, 2.4 mmol) was added dropwise and the tan solu-
tion was shaken for 5 h at room temperature. The pyridine and
CH3CN solvents were removed under vacuum. The product fluore-
none was isolated by column chromatography as a colorless oil
(0.092 g, 0.510 mmol, 85%) using silica gel and CHCl3 as eluent.
Funding from the National Science Foundation for the purchase
of the NMR spectrometer (CHE-9974801), the purchase of the
ApexII diffractometer (MRI, CHE-0420497), and the purchase of
the mass spectrometer (CHE-9708640) is acknowledged.
Keywords: biomimetic catalysis · homogeneous catalysis ·
ligand effects · N ligands · oxidation
1
Spectroscopic data and copies of the H and 13C{1H} NMR spectra
of all catalysis products in Table 5 are given in the Supporting In-
formation.
[1] a) C.-L. Sun, B.-J. Li, Z.-J. Shi, Chem. Rev. 2011, 111, 1293–1314; b) W. M.
ˇ
Czaplik, M. Mayer, J. Cvengros, A. Jacobi von Wangelin, ChemSusChem
2009, 2, 396–417; c) A. A. O. Sarhan, C. Bolm, Chem. Soc. Rev. 2009, 38,
2730–2744; d) E. B. Bauer, Curr. Org. Chem. 2008, 12, 1341–1369; e) S.
Enthaler, K. Junge, M. Beller, Angew. Chem. 2008, 120, 3363–3367;
Angew. Chem. Int. Ed. 2008, 47, 3317–3321; f) A. Correa, O. Garcia Man-
cheÇo, C. Bolm, Chem. Soc. Rev. 2008, 37, 1108–1117; g) B. D. Sherry, A.
Fꢃrstner, Acc. Chem. Res. 2008, 41, 1500–1511; h) C. Bolm, J. Legros, J.
Le Paih, L. Zani, Chem. Rev. 2004, 104, 6217–6254.
X-ray crystal structure determinations of [Fe(6)2]OTf2 and
[Fe(8)2(OTf)2]
Suitable crystals of appropriate dimensions were mounted on
Mitgen loops in random orientations. Preliminary examination and
data collection were performed using a Bruker Kappa Apex-II
charge-coupled device (CCD) detector system single-crystal X-ray
diffractometer equipped with an Oxford Cryostream LT device.
Data were collected using graphite-monochromated MoKa radiation
(l=0.71073 ꢂ) from a fine-focus sealed-tube X-ray source. Prelimi-
nary unit-cell constants were determined with a set of 36 narrow
frame scans. Typical data sets consist of combinations of w and f
scan frames with a typical scan width of 0.58 and counting time of
15–30 s/frame at a crystal to detector distance of approximately
3.5 to 4.0 cm. The collected frames were integrated using an orien-
tation matrix determined from the narrow frame scans. Apex II and
SAINT software packages[74] were used for data collection and data
integration. Analysis of the integrated data did not show any
decay. Final cell constants were determined by global refinement
of reflections from the complete data set. Data were corrected for
systematic errors using SADABS[74] based on the Laue symmetry
using equivalent reflections. Structure solutions and refinement
were carried out using the SHELXTL-PLUS software package.[75] The
structures were refined with full-matrix least-squares cycles by min-
imizing Sw(F2oÀF2c)2. All non-hydrogen atoms were refined aniso-
tropically to convergence. Typically, H atoms were added at the
calculated positions in the final refinement cycles. Specific details
of refinement for the compounds are listed below.
[2] a) M. Lin, X.-l. Chen, T. Wang, P. Yan, S.-x. Xu, Z.-p. Zhan, Chem. Lett.
2011, 40, 111–113; b) G. C. Midya, S. Paladhi, K. Dhara, J. Dash, Chem.
Commun. 2011, 47, 6698–6700; c) M. Mayer, W. M. Czaplik, A. Jacobi
von Wangelin, Adv. Synth. Catal. 2010, 352, 2147–2152; d) T. Hata, S.
Iwata, S. Seto, H. Urabe, Adv. Synth. Catal. 2012, 354, 1885–1889; e) S.
Gꢃlak, A. Jacobi von Wangelin, Angew. Chem. 2012, 124, 1386–1390;
Angew. Chem. Int. Ed. 2012, 51, 1357–1361.
[3] J. Bonnamour, C. Bolm, Org. Lett. 2011, 13, 2012–2014.
[4] W. Han, P. Mayer, A. R. Ofial, Adv. Synth. Catal. 2010, 352, 1667–1676.
[5] O. G. Mancheno, J. Dallimore, A. Plant, C. Bolm, Org. Lett. 2009, 11,
2429–2432.
[6] a) A. Berkessel, S. Reichau, A. von der Hçh, N. Leconte, J.-M. Neudçrfl,
Organometallics 2011, 30, 3880–3887; b) J. Yang, T. D. Tilley, Angew.
Chem. 2010, 122, 10384–10386; Angew. Chem. Int. Ed. 2010, 49, 10186–
10188; c) M. Flꢃckiger, A. Togni, Eur. J. Org. Chem. 2011, 4353–4360;
d) J. M. S. Cardoso, B. Royo, Chem. Commun. 2012, 48, 4944–4946.
[7] A. M. Tondreau, C. Milsmann, A. D. Patrick, H. M. Hoyt, E. Lobkovsky, K.
Wieghardt, P. J. Chirik, J. Am. Chem. Soc. 2010, 132, 15046–15059.
[8] a) S. Enthaler, Eur. J. Org. Chem. 2011, 2011, 4760–4763; b) C. Wang, X.
Li, F. Wu, B. Wan, Angew. Chem. 2011, 123, 7300–7304; Angew. Chem.
Int. Ed. 2011, 50, 7162–7166; c) Y.-Y. Lin, Y.-J. Wang, C.-H. Lin, J.-H.
Cheng, C.-F. Lee, J. Org. Chem. 2012, 77, 6100–6106; d) D. Bꢄzier, G. T.
Venkanna, L. C. Misal Castro, J. Zheng, T. Roisnel, J.-B. Sortais, C. Darcel,
Adv. Synth. Catal. 2012, 354, 1879–1884.
[9] H. J. H. Fenton, J. Chem. Soc. Trans. 1894, 65, 899–910.
[10] a) D. H. R. Barton, Tetrahedron 1998, 54, 5805–5817; b) D. H. R. Barton,
Tetrahedron 1994, 50, 1011–1032; c) D. H. R. Barton, D. Doller, Pure Appl.
Chem. 1991, 63, 1567–1576; d) D. H. R. Barton, E. Csuhai, D. Doller, N.
Ozbalik, G. Balavoine, Proc. Natl. Acad. Sci. USA 1990, 87, 3401–3404.
[11] a) F. Gozzo, J. Mol. Catal. A 2001, 171, 1–22; b) B. Singh, J. R. Long, F.
Fabrizia de Biani, D. Gatteschi, P. Stavropoulos, J. Am. Chem. Soc. 1997,
119, 7030–7047.
[Fe(6)2]OTf2 was refined as a twin in the space group Pn with BASF
refining to 0.669. Some of the CF3 and SO3 groups of the anion are
disordered. The disorder was modeled with partial occupancy
atoms and refined with restraints. The NH hydrogen atoms were
located and refined with a riding model. The structure can be
solved and refined in P2/n but this results in poor refinement and
geometry.
[12] D. H. R. Barton, Chem. Soc. Rev. 1996, 25, 237–239.
[13] a) P. Stavropoulos, R. C¸elenligil-C¸etin, A. E. Tapper, Acc. Chem. Res. 2001,
34, 745–752; b) D. W. Snelgrove, P. A. McFaul, K. U. Ingold, D. D. M.
Wayner, Tetrahedron Lett. 1996, 37, 823–826; c) M. J. Perkins, Chem. Soc.
Rev. 1996, 25, 229–236; d) F. Minisci, F. Fontana, S. Araneo, F. Recupero,
S. Banfi, S. Quici, J. Am. Chem. Soc. 1995, 117, 226–232.
[14] E. Neyens, J. Baeyens, J. Hazard. Mater. 2003, 98, 33–50.
[15] P. R. Ortiz de Montellano, Chem. Rev. 2010, 110, 932–948.
[16] a) E. P. Talsi, K. P. Bryliakov, Coord. Chem. Rev. 2012, 256, 1418–1434;
b) P. C. A. Bruijnincx, G. van Koten, R. J. M. Klein Gebbink, Chem. Soc. Rev.
2008, 37, 2716–2744.
[17] a) L. Que, Jr., W. B. Tolman, Nature 2008, 455, 333–340; b) M. Costas,
M. P. Mehn, M. P. Jensen, L. Que, Jr., Chem. Rev. 2004, 104, 939–986.
[18] a) A. Beck, B. Weibert, N. Burzlaff, Eur. J. Inorg. Chem. 2001, 521–527;
b) M. Costas, K. Chen, L. Que, Jr., Coord. Chem. Rev. 2000, 200–202,
517–544; c) J. T. Groves, J. Inorg. Biochem. 2006, 100, 434–474; d) P. D.
Oldenburg, L. Que, Jr., Catal. Today 2006, 117, 15–21; e) N. Burzlaff,
¯
[Fe(8)2(OTf)2] was refined in the space group P1. The coordinated
CF3SO3 group is disordered. The disorder was modeled with partial
occupancy atoms and refined with restraints. The NH hydrogen
atoms were located and refined with a riding model.
CCDC-852959 ([FeII(6)2]OTf2) and 852960 ([FeII(8)2(OTf)2]) contains
the supplementary crystallographic data for this paper. These data
can be obtained free of charge from The Cambridge Crystallo-
graphic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2013, 78, 101 – 116 114

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
Angew. Chem. 2009, 121, 5688–5690; Angew. Chem. Int. Ed. 2009, 48,
5580–5582.
[40] K. Visvaganesan, E. Suresh, M. Palaniandavar, Dalton Trans. 2009, 3814–
3823.
[19] B. Join, K. Mçller, C. Ziebart, K. Schrçder, D. Gçrdes, K. Thurow, A. Span-
nenberg, K. Junge, M. Beller, Adv. Synth. Catal. 2011, 353, 3023–3030.
[20] a) A. Company, L. Gꢅmez, X. Fontrodona, X. Ribas, M. Costas, Chem. Eur.
J. 2008, 14, 5727–5731; b) J. Tang, P. Gamez, J. Reedijk, Dalton Trans.
2007, 4644–4646; c) F. Li, M. Wang, C. Ma, A. Gao, H. Chen, L. Sun,
Dalton Trans. 2006, 2427–2434.
[21] a) M. Nakanishi, C. Bolm, Adv. Synth. Catal. 2007, 349, 861–864; b) C.
Pavan, J. Legros, C. Bolm, Adv. Synth. Catal. 2005, 347, 703–705.
[22] a) R. R. Fernandes, J. Lasri, M. F. C. Guedes da Silva, J. A. L. da Silva, J. J. R.
Frausto da Silva, A. J. L. Pombeiro, J. Mol. Catal. A 2011, 351, 100–111;
b) S. A. Moyer, T. W. Funk, Tetrahedron Lett. 2010, 51, 5430–5433; c) E.
Balogh-Hergovich, G. Speier, J. Mol. Catal. A 2005, 230, 79–83; d) H.
Hosseini-Monfared, C. Nꢆther, H. Winkler, C. Janiak, Inorg. Chim. Acta
2012, 391, 75–82.
[41] A. Thibon, J.-F. Bartoli, S. Bourcier, F. Banse, Dalton Trans. 2009, 9587–
9594.
[42] H. R. Lucas, L. Li, A. A. Narducci Sarjeant, M. A. Vance, E. I. Solomon, K. D.
Karlin, J. Am. Chem. Soc. 2009, 131, 3230–3245.
[43] A. C. Mayer, C. Bolm in Iron Catalysis in Organic Chemistry, 1st ed. (Ed.: B.
Plietker), Wiley-VCH, Weinheim, 2008, Chapter 3, pp. 73–123.
[44] D. W. Blakesley, S. C. Payne, K. S. Hagen, Inorg. Chem. 2000, 39, 1979–
1989.
[45] V. Balland, F. Banse, E. Anxolabꢄhꢇre-Mallart, M. Nierlich, J.-J. Girerd, Eur.
J. Inorg. Chem. 2003, 2529–2535.
[46] T. Buchen, P. Gꢃtlich, Inorg. Chim. Acta 1995, 231, 221–223.
[47] S. Gosiewska, J. J. L. M. Cornelissen, M. Lutz, A. L. Spek, G. van Koten,
R. J. M. Klein Gebbink, Inorg. Chem. 2006, 45, 4214–4227.
[48] J. Kim, R. G. Harrison, C. Kim, L. Que, Jr., J. Am. Chem. Soc. 1996, 118,
4373–4379.
[23] a) P. Comba, H. Wadepohl, S. Wunderlich, Eur. J. Inorg. Chem. 2011,
5242–5249; b) T. M. Shaikha, F.-E. Hong, Adv. Synth. Catal. 2011, 353,
1491–1496.
[49] a) M. P. Jensen, A. Mairata i Payeras, A. T. Fiedler, M. Costas, J. Kaizer, A.
Stubna, E. Mꢃnck, L. Que, Jr., Inorg. Chem. 2007, 46, 2398–2408; b) C.
Nguyen, R. J. Guajardo, P. K. Mascharak, Inorg. Chem. 1996, 35, 6273–
6281.
[24] a) M. Wu, C.-X. Miao, S. Wang, X. Hu, C. Xia, F. E. Kꢃhn, W. Sun, Adv.
Synth. Catal. 2011, 353, 3014–3022; b) K. Schrçder, B. Join, A. Jose Ama-
li, K. Junge, X. Ribas, M. Costas, M. Beller, Angew. Chem. 2011, 123,
1461–1465; Angew. Chem. Int. Ed. 2011, 50, 1425–1429; c) F. G. Gelal-
cha, G. Anilkumar, M. K. Tse, A. Brꢃckner, M. Beller, Chem. Eur. J. 2008,
14, 7687–7698; d) P. C. A. Bruijnincx, I. L. C. Buurmans, S. Gosiewska,
M. A. H. Moelands, M. Lutz, A. L. Spek, G. van Koten, R. J. M. Klein Geb-
bink, Chem. Eur. J. 2008, 14, 1228–1237; e) F. Oddon, E. Girgenti, C.
Lebrun, C. Marchi-Delapierre, J. Pꢄcaut, S. Mꢄnage, Eur. J. Inorg. Chem.
2012, 85–96.
[50] M. S. Seo, T. Kamachi, T. Kouno, K. Murata, M. J. Park, K. Yoshizawa, W.
Nam, Angew. Chem. 2007, 119, 2341–2344; Angew. Chem. Int. Ed. 2007,
46, 2291–2294.
[51] M. R. Bukowski, H. L. Halfen, T. A. van den Berg, J. A. Halfen, L. Que, Jr.,
Angew. Chem. 2005, 117, 590–593; Angew. Chem. Int. Ed. 2005, 44,
584–587.
[52] F. Shi, M. K. Tse, Z. Li, M. Beller, Chem. Eur. J. 2008, 14, 8793–8797.
[53] G. A. Russell, J. Am. Chem. Soc. 1957, 79, 3871–3877.
[54] A. Brꢄhꢄret, C. Lambeaux, S. Mꢄnage, M. Fontecave, F. Dallemer, E.
Fache, J.-L. Pierre, P. Chautemps, M.-T. Averbusch-Pouchot, C. R. Acad.
Sci. Chemistry 2001, 4, 27–34.
[25] a) M. S. Chen, M. C. White, Science 2007, 318, 783–787; b) M. S. Chen,
M. C. White, Science 2010, 327, 566–571.
[26] K. Chen, M. Costas, L. Que, Jr., J. Chem. Soc. Dalton Trans. 2002, 672–
679.
[55] C. Krebs, D. G. Fujimori, C. T. Walsh, J. M. Bollinger, Jr., Acc. Chem. Res.
2007, 40, 484–492.
[56] M. P. Jensen, M. Costas, R. Y. N. Ho, J. Kaizer, A. Mairata i Payeras, E.
Mꢃnck, L. Que, Jr., J.-U. Rohde, A. Stubna, J. Am. Chem. Soc. 2005, 127,
10512–10525.
[27] a) J. England, R. Gondhia, L. Bigorra-Lopez, A. R. Petersen, A. J. P. White,
G. J. P. Britovsek, Dalton Trans. 2009, 5319–5334; b) J. England, C. R.
Davies, M. Banaru, A. J. P. White, G. J. P. Britovsek, Adv. Synth. Catal.
2008, 350, 883–897; c) G. J. P. Britovsek, J. England, A. J. P. White, Dalton
Trans. 2006, 1399–1408; d) G. J. P. Britovsek, J. England, A. J. P. White,
Inorg. Chem. 2005, 44, 8125–8134; e) G. J. P. Britovsek, J. England, S. K.
Spitzmesser, A. J. P. White, D. J. Williams, Dalton Trans. 2005, 945–955.
[28] S. Gosiewska, H. P. Permentier, A. P. Bruins, G. van Koten, R. J. M. Klein
Gebbink, Dalton Trans. 2007, 3365–3368.
[57] J.-U. Rohde, J. H. In, M. H. Lim, W. W. Brennessel, M. R. Bukowski, A.
Stubna, E. Mꢃnck, W. Nam, L. Que, Jr., Science 2003, 299, 1037–1039.
[58] A. Decker, J.-U. Rohde, E. J. Klinker, S. D. Wong, L. Que, Jr., E. I. Solomon,
J. Am. Chem. Soc. 2007, 129, 15983–15996.
[29] a) P. Comba, M. Maurer, P. Vadivelu, Inorg. Chem. 2009, 48, 10389–
10396; b) M. R. Bukowski, P. Comba, A. Lienke, C. Limberg, C. Lopez de
Laorden, R. Mas-Ballestꢄ, M. Merz, L. Que, Jr., Angew. Chem. 2006, 118,
3524–3528; Angew. Chem. Int. Ed. 2006, 45, 3446–3449.
[30] M. Christmann, Angew. Chem. 2008, 120, 2780–2783; Angew. Chem. Int.
Ed. 2008, 47, 2740–2742.
[31] A. N. Biswas, A. Pariyar, S. Bose, P. Das, P. Bandyopadhyay, Catal.
Commun. 2010, 11, 1008–1011.
[32] a) T. J. Collins, Acc. Chem. Res. 1994, 27, 279–285; b) A. Marques, M.
Marin, M.-F. Ruasse, J. Org. Chem. 2001, 66, 7588–7595.
[33] a) P. Shejwalkar, N. P. Rath, E. B. Bauer, Dalton Trans. 2011, 40, 7617–
7631; b) P. Shejwalkar, N. P. Rath, E. B. Bauer, Molecules 2010, 15, 2631–
2650; c) M. Lenze, S. L. Sedinkin, N. P. Rath, E. B. Bauer, Tetrahedron Lett.
2010, 51, 2855–2858; d) M. Lenze, E. B. Bauer, J. Mol. Catal. A 2009,
309, 117–123.
[34] A. Diebold, K. S. Hagen, Inorg. Chem. 1998, 37, 215–223.
[35] a) N. M. F. Carvalho, A. Horn, Jr., A. J. Bortoluzzi, V. Drago, O. A. C. An-
tunes, Inorg. Chim. Acta 2006, 359, 90–98; b) D. Mandon, A. Nopper, T.
Litrol, S. Goetz, Inorg. Chem. 2001, 40, 4803–4806.
[36] a) A. Malassa, H. Gçrls, A. Buchholz, W. Plass, M. Westerhausen, Z. Anorg.
Allg. Chem. 2006, 632, 2355–2362; b) C. J. Davies, G. A. Solan, J. Faw-
cett, Polyhedron 2004, 23, 3105–3114.
[37] a) N. M. F. Carvalho, O. A. C. Antunes, A. Horn, Jr., Dalton Trans. 2007,
1023–1027; b) K. Visvaganesan, R. Mayilmurugan, E. Suresh, M. Pala-
niandavar, Inorg. Chem. 2007, 46, 10294–10306.
[59] D. H. R. Barton, V. N. Le Gloahec, H. Patin, F. Launay, New J. Chem. 1998,
22, 559–563.
[60] a) X. Baucherel, L. Gonsalvi, I. W. C. E. Arends, S. Ellwood, R. A. Sheldon,
Adv. Synth. Catal. 2004, 346, 286–296; b) F. Minisci, F. Recupero, A. Cec-
chetto, C. Gambarotti, C. Punta, R. Faletti, R. Paganelli, G. Franco Pedulli,
Eur. J. Org. Chem. 2004, 109–119.
[61] G. Roelfes, M. Lubben, R. Hage, L. Que, Jr., Chem. Eur. J. 2000, 6, 2152–
2159.
[62] a) J. Kaizer, M. Costas, L. Que, Jr., Angew. Chem. 2003, 115, 3799–3801;
Angew. Chem. Int. Ed. 2003, 42, 3671–3673; b) A. Mairata i Payeras,
R. Y. N. Ho, M. Fujita, L. Que, Jr., Chem. Eur. J. 2004, 10, 4944–4953; c) Y.
He, J. D. Gordon, C. R. Goldsmith, Inorg. Chem. 2011, 50, 12651.
[63] F. Namuswe, G. D. Kasper, A. A. Narducci Sarjeant, T. Hayashi, C. M. Krest,
M. T. Green, P. Moꢈnne-Loccoz, D. P. Goldberg, J. Am. Chem. Soc. 2008,
130, 14189–14200.
[64] D. M. Pearson, N. R. Conley, R. M. Waymouth, Organometallics 2011, 30,
1445–1453.
[65] D. Pijper, P. Saisaha, J. W. de Boer, R. Hoen, C. Smit, A. Meetsma, R.
Hage, R. P. van Summeren, P. L. Alsters, B. L. Feringa, W. R. Browne,
Dalton Trans. 2010, 39, 10375–10381.
[66] a) M. C. Brçhmer, S. Mundinger, S. Brꢆse, W. Bannwarth, Angew. Chem.
2011, 123, 6299–6301; Angew. Chem. Int. Ed. 2011, 50, 6175–6177;
b) S. S. Massoud, F. R. Louka, M. Mikuriya, H. Ishida, F. A. Mautner, Inorg.
Chem. Commun. 2009, 12, 420–425.
[67] S. J. Lange, H. Miyake, L. Que, Jr., J. Am. Chem. Soc. 1999, 121, 6330–
6331.
[38] J. Cui, M. S. Mashuta, R. M. Buchanan, C. A. Grapperhaus, Inorg. Chem.
2010, 49, 10427–10435.
[68] H. Chowdhury, S. H. Rahaman, R. Ghosh, S. Kumar Sarkar, H.-K. Fun, B.
Kumar Ghosh, J. Mol. Struct. 2007, 826, 170–176.
[39] E. Balogh-Hergovich, G. Speier, M. Rꢄglier, M. Giorgi, E. Kuzmann, Attila
Vꢄrtes, Eur. J. Inorg. Chem. 2003, 1735–1740.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2013, 78, 101 – 116 115

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
[69] a) D. Wu, Acta Crystallogr. Sect. E 2009, 65, m1340; b) T. Kajiwara, R.
Sensui, T. Noguchi, A. Kamiyama, T. Ito, Inorg. Chim. Acta 2002, 337,
299–307.
[73] K. S. Hagen, Inorg. Chem. 2000, 39, 5867–5869.
[74] Bruker Analytical X-Ray, Madison, WI, 2008.
[75] G. M. Sheldrick, Acta Crystallogr. Sect. A 2008, 64, 112–122.
[70] C. Panda, M. Ghosh, T. Panda, R. Banerjee, S. S. Gupta, Chem. Commun.
2011, 47, 8016–8018.
´
[71] K. Barbusinski, Ecol. Chem. Eng. S 2009, 16, 347–358.
Received: September 17, 2012
[72] J. M. Rowland, M. M. Olmstead, P. K. Mascharak, Inorg. Chem. 2002, 41,
2754–2760.
Published online on December 19, 2012
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2013, 78, 101 – 116 116