New iron(ii) α-iminopyridine complexes and their catalytic activity in the oxidation of activated methylene groups and secondary alcohols to ketones

Article abstract of DOI:10.1039/c1dt10387c

A set of iron(ii) complexes of the general formula [Fe(OTf) ₂L₂] was synthesized in 32 to 78% isolated yields, where L represents a bidentate α-iminopyridine ligand. Four of the iron complexes were characterized structurally, revealing a rich coordination chemistry, because the coordination geometry of the iron complexes strongly depends on the substitution pattern exhibited by the ligands L. The catalytic activity of the new complexes was demonstrated in the oxidation of cyclohexane, activated methylene groups and secondary alcohols to the corresponding ketones utilizing H₂O₂ and t-BuOOH as the oxidants. The oxidation of activated methylene groups and secondary alcohols to the corresponding ketones with t-BuOOH gave isolated yields between 22 and 91% (4 h, room temperature, 3% catalyst load). The influence of the structure of the ligand on the activity of the corresponding metal complex is also reported. Furthermore, UV-vis experiments were performed which provided evidence for the formation of an [Fe-O-O-t-Bu] intermediate. The Royal Society of Chemistry 2011.

Full text of DOI:10.1039/c1dt10387c

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Dalton
Transactions
Cite this: Dalton Trans., 2011, 40, 7617
www.rsc.org/dalton
PAPER
New iron(II) a-iminopyridine complexes and their catalytic activity in the
oxidation of activated methylene groups and secondary alcohols to ketones†
Pushkar Shejwalkar,a Nigam P. Ratha,b and Eike B. Bauer*a
Received 8th March 2011, Accepted 27th April 2011
DOI: 10.1039/c1dt10387c
A set of iron(II) complexes of the general formula [Fe(OTf)2L2] was synthesized in 32 to 78% isolated
yields, where L represents a bidentate a-iminopyridine ligand. Four of the iron complexes were
characterized structurally, revealing a rich coordination chemistry, because the coordination geometry
of the iron complexes strongly depends on the substitution pattern exhibited by the ligands L. The
catalytic activity of the new complexes was demonstrated in the oxidation of cyclohexane, activated
methylene groups and secondary alcohols to the corresponding ketones utilizing H2O2 and t-BuOOH
as the oxidants. The oxidation of activated methylene groups and secondary alcohols to the
corresponding ketones with t-BuOOH gave isolated yields between 22 and 91% (4 h, room temperature,
3% catalyst load). The influence of the structure of the ligand on the activity of the corresponding metal
complex is also reported. Furthermore, UV-vis experiments were performed which provided evidence
for the formation of an [Fe-O-O-t-Bu] intermediate.
Consequently, a number of iron complexes, which are archi-
tecturally reminiscent of the active sites in non-heme enzymes
found in nature have been reported to catalyze oxidation reactions
Introduction
Iron catalysis has emerged as a lively area of research in recent
years.1–7 Iron has a number of advantages over other transition
of alkanes, alkenes, arenes and alcohols using H2O2, t-BuOOH
metals typically employed in catalysis; it is relatively non-toxic,
or PhIO as the oxidants. Iron complexes synthesized from
tris(picolyl)amine (tpa, 1a),23,31 BPMEM (2a, Fig. 1)31 and their
abundant, cheap and environmentally friendly. Consequently,
iron-based catalyst systems have been applied in numerous
derivatives32–35 or from ligand 336 catalyze the oxidation of alkanes
organic transformations, such as carbon–carbon bond form-
or the epoxidation of alkenes. Complex 4 was shown to selectively
ing reactions,8–14 carbon–heteroatom bond forming reactions,15,16
hetero–heteroatom bond forming reactions,17,18 reductions19 and
polymerizations.20
Also, iron-catalyzed oxidation reactions have been intensively
investigated in recent years.21 Iron-containing enzymes found in
nature inspired these research efforts.22–24 The cytochrome P450
enzyme family contains an iron porphyrin heme cofactor, which
oxidizes a range of organic substances in living organisms.25
Non-heme enzymes such as the Rieske dioxygenase or methane
monooxygenase catalyze similar oxidation reactions.26 Research
efforts in the past decades have been directed towards mimicking
these motifs utilizing artificial iron complexes27 resembling the
architectures known from nature.28–30
aUniversity of Missouri - St. Louis, Department of Chemistry and Biochem-
istry, One University Boulevard, St. Louis, MO, 63121, USA
bCenter for Nanoscience, University of Missouri - St. Louis, St. Louis, MO,
63121, USA. E-mail: bauere@umsl.edu; Fax: +1-314-516-5342; Tel: +1-
314-516-5311
† Electronic supplementary information (ESI) available: Experimental
1
details, characterization data and 1H and 13C { H} NMR spectra for the
1
catalysis products; H NMR and UV-vis spectra of all metal complexes.
CCDC reference numbers 804996–804999. For ESI and crystallographic
data in CIF or other electronic format see DOI: 10.1039/c1dt10387c
Fig. 1 Previously employed ligands and complexes for iron catalyzed
oxidation reactions.
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 7617–7631 | 7617
©

View Article Online
oxidize methine groups to tertiary alcohols.37,38 Iron salts in
combination with peroxides and appropriate additives also show
catalytic activity in oxidation reactions of alkanes,39 alkenes40,41 or
arenes42 and in epoxidation reactions.43
states and allow access to low-valent iron complexes with catalytic
activity.57
To the best of our knowledge, the application of iron a-
iminopyridine complexes in catalytic oxidation reactions has not
been investigated. Imines are easy to synthesize, tunable and
less prone to oxidation than amine-based ligands. They are
unsymmetrical and can show flexibility upon coordination to
iron centers. The coordination of two bidentate ligands leaves
two coordination sites on the iron to be utilized for catalysis.
Herein, we describe the synthesis of a set of tuned a-iminopyridine
ligands and their respective iron complexes, four of which were
characterized structurally. We investigated the impact of the
ligand structure on the geometry and physical properties of the
complexes. Finally, we employed the new complexes as catalysts
in oxidation reactions of cyclohexane, activated CH2 groups and
secondary alcohols, compared their activities and identified a
potential catalytic intermediate.
Albeit a number of efficient catalyst systems featuring N-
coordinating ligands have been reported, a major challenge is
the development of synthetically applicable processes. For some
catalytic systems described in the literature, a large excess of
the substrate over the oxidant is employed in order to avoid
overoxidation reactions.31 Oxidative decomposition of the ligands
is a potential concern,44 as CH2NH units and other parts of the
ligands can be oxidized, leading to catalyst deterioration.45
The majority of iron based catalytic systems employ tetraden-
tate N-coordinating ligands. We speculated that bidentate ligands
could be employed in iron catalyzed oxidation reactions as well,
and we wished to employ a ligand system, that was flexible, readily
available in high yields, and tunable. Bidentate a-iminopyridine
ligands seemed to hold great promise (7 in Scheme 1), and have
been described as a “redox noninnocent ligand system”.46 Iron
complexes of bidentate a-iminopyridines are known;46–50 they
exhibit a range of electronic and steric properties51 but only a
few have been structurally characterized,46,49,50,52 some of which
had three bidentate ligands53 or only one bidentate ligand on the
iron.54 The employment of noninnocent or redox-active ligands is
of interest,55,56 as these ligands might stabilize certain oxidation
Results and discussion
Synthesis of a-iminopyridine ligands and iron complexes
First a set of sterically and electronically tuned a-iminopyridine
ligands 7 was synthesized (Scheme 1). They are easily accessible in
high yields by condensation of 2-pyridine-carboxaldehyde 5 and
anilines 6 at room temperature in methanol utilizing molecular
˚
sieves (4 A) to scavenge the water formed during reaction
(Scheme 1). The ligands 7a,58 7d,59 7f60 and 7h61 have previously
been reported; the ligands 7b,c,e, and g are new and the spec-
troscopic properties of the new imines were similar to those of
the known ones. The set of ligands features a range of steric and
electronic properties. Ligands 7c to f are electron rich, whereas
ligand 7b should be electron-poor. The ligands 7c to d and f
are sterically demanding, as they are substituted in both ortho
positions on the aryl ring bonded to the coordinating imine
nitrogen atom.
Next, the bidentate ligands 7 were coordinated to an iron center.
We sought coordination of two ligands 7 to the iron, leaving two
coordination sites for catalysis. As chloro ligands are known to
bridge iron centers,54 we selected the known Fe(OTf)2 with the
-
weakly coordinating triflate (OTf = CF3SO3 ) ligand as the iron
source,62 and which has been successfully employed as starting
material in a variety of iron complex syntheses.32
Accordingly, when the ligands 7 were stirred at room tempera-
ture in CH2Cl2 with Fe(OTf)2 in a 2 : 1 molar ratio, coordination
took place as seen by an immediate color change to red-orange
(Scheme 2). A small amount of CH3CN was added to the reaction
mixture to help dissolve the Fe(OTf)2. The new iron complexes
were isolated by concentrating the crude reaction mixtures and
layering them with Et2O. After a few days at -18 ◦C, precipitates
formed which were isolated and dried under high vacuum to give
the new iron complexes [Fe(OTf)2(7)2] (8) in 32 to 78% isolated
yields as orange to violet powders.
The new complexes were analyzed by MS, IR, X-ray, UV-vis, 1H
and 19F NMR, magnetic measurements and elemental analyses.
The general formulation [Fe(OTf)2(7)2] was confirmed by MS
and elemental analysis (for deviating behavior of the ligands 7d,e
see below). The FAB mass spectra showed a molecular ion peak
for [Fe(OTf)(7)2]+, for which an exact mass was obtained for all
Scheme 1 a-iminopyridine ligands 7 and their synthesis.
7618 | Dalton Trans., 2011, 40, 7617–7631
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Fig. 2 Different coordination modes for the bidentate ligands 7 (N*
denotes the pyridyl nitrogen).
Fig. 2, the bidentate ligands can coordinate to an octahedral iron
center in different ways.
Three different cis (A/A¢, B/B¢ and E/E¢, pairs of enantiomers)
and two trans coordination modes (C and D) are possible. The
complexes 8c and 8f take the trans configuration D and in
complex 8b, the two imine ligands are coordinated in the cis
mode A/A¢ in the solid state, as shown by X-ray (vide infra).
Symmetry considerations suggest that the coordination modes
A/A¢, C, D and E/E¢ give for both ligands one set of signals in the
NMR spectra and B/B¢ two set of signals. Due to the variety of
coordination modes and potential dynamic processes in solution,44
an unambiguous structural assignment for the complexes purely
based on NMR data is not possible (vide infra). However, it is
possible to determine the number of isomers in solution by NMR.
Magnetic moments and 1H NMR spectroscopy
Iron(II) complexes can adopt high spin and low spin configu-
rations, and spin crossover can take place.66,67 We determined
the effective magnetic moments (meff) of the new complexes in
the solid state using a magnetic susceptibility balance (Table 1),
and in addition for some complexes by the Evan’s method.68
It is worth noticing that the complexes 8a and e with ligands
having no substituents in position ortho to the coordinating imine
nitrogen have low effective magnetic moments of 1.19 and 1.84
BM, respectively. All other complexes featuring substituents in the
position ortho to the coordinating imine nitrogen atom and exhibit
meff values between 3.91 and 5.56 BM. Complex 8b converted
in solution from a high spin complex at room temperature to
a low spin complex at 243 K, as seen by a change of the
magnetic moment to 1.20 BM (Evan’s method). In theory, low-spin
diamagnetic Fe(II) complexes have an effective magnetic moment
meff of zero and high-spin paramagnetic Fe(II) complex with S =
2 of 4.9 BM, according to the spin-only formula. However, due
to “admixed” ground states69 or spin crossover phenomena and
orbital contributions to the spin state,70 the measured values can
deviate from the ideal ones. It has been previously reported that
methyl groups in an ortho position in coordinated pyridyl rings of
tpa-type ligands (1b vs. 1a in Fig. 1) can cause a change in the spin
state from low spin to high spin71 and the magnetic behavior of
our complexes appears to follow a similar trend.
Scheme 2 Synthesis of iron imine-pyridyl complexes. The four complexes
on the bottom were characterized by X-ray.
complexes. In addition, a diagnostic fragmentation pattern due to
loss of OTf and/or one ligand 7 was observed. The coordination
of the C N imine functionality to the iron center was observed
by a shift of the imine stretch in the IR from 1626 to 1644 cm-1
in the free ligands to lower wavenumbers between 1590 and 1598
cm-1 upon imine coordination, which are in the same range than
for other iron imine complexes reported in the literature (see Table
1, which compiles key physical data of the new complexes).63
Non-coordinated OTf gives IR stretches at 1268, 1223, 1143,
and 1030 cm-1; the solid state IR spectra of the complexes also
showed these signals; some of the signals are desymmetrized,
as previously reported for iron coordinated OTf.64,65 New bands
associated with metal to ligand charge transfer (MLCT) processes
were observed in the UV-vis spectra, which also indicated the
formation of the new complexes (Table 1, vide infra). As shown in
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 7617–7631 | 7619
©

View Article Online
Table 1 Selected physical data of the new iron complexes
IR, vC N/cm-1
UV-vis, MLCT
Magnetic momentsa
Free ligand
Complex
lmax (CH2Cl2)/nm
e/M-1cm-1
meff/BM
8a
8b
8c
8d
8e
8f
1627
1630
1641
1642
1584
1641
1626
1644
1594
1590
1595
1597
1595
1595
1595
1598
572, 529(sh)
532b
408
1.831 ¥ 103
7.872 ¥ 101
7.074 ¥ 102
1.128 ¥ 103
4.729 ¥ 103
1.208 ¥ 103
2.527 ¥ 103
1.024 ¥ 103
1.19
4.56 (1.20)d
5.14
408
5.28
1.84
577, 522(sh)c
417
5.56 (4.24)e
3.91 (4.85)e
5.39
8g
8h
414
472
a In the solid state at 298 K, determined with a magnetic susceptibility balance. b There was no charge transfer band observed for complex 8b, and the
band reported is assigned to a d–d transition. c In CH3CN. d Determined by Evan’s method (CD3CN, 243 K). e Determined by Evan’s method (CD3CN,
293 K).
Table 2 Key NMR data and structural assignment
19F NMR
Chemical shift range
(Temperature)
Major 1H resonances and
number of sets/ppm
Structural assignment
(see Fig. 2)a
Imine resonances/ppm
d (ppm)
n1/2 Hzb
8a
8b
8c
8d
8e
0 to 12 ppm (298 K)
0 to 12 ppm (243 K)
4 sets between 5.3 and 8.5
2 sets between 6.5 and 8.5
9.31, 9.15, 8.83, 8.81
10.1 (broad) 11.1 (broad)
not observed
mixture of isomers
cis, Ac
-71.2
-71.7
-66.7
-71.6
323
574
595
911
275
-6 to +60 ppm (298 K) 1 set; 56.3, 54.8, -6.1
trans, Cd
-8 to +55 ppm (298 K) 1 set; 55.4, 15.9, 8.2 to 6.9
not observed
trigonal-bipyramidald
0 to 9 ppm (298 K)
between 7.3 and 8.5 and at 5.95, 9.06, 8.87 and 9.41, 9.24,
5.88, 4.92
1 set; 128, 56.1, 54.1, 16.5
1 set; 8.6 and 6.7, 2.90, 1.17.
1 set; 56.2, 52.5, -4.5
mixture of [Fe(OTf)2(7e)2] and -76.5
9.14, 8.90
[Fe(7e)3](OTf)2
8f
8g
8h
0 to 125 ppm (298 K)
0 to 9 ppm (253 K)
0 to 60 ppm (298 K)
not observed
not observed
not observed
trans, Cd
-66.3
-72.0
-70.7
704
425
224
one isomer, A, C, D or E
one isomer A, C, D or E
a Tentative structural assignment based on NMR data. b Width at half height. c Confirmed by X-ray, the NMR data are consistent with a second isomer
in solution, which must be either E, C or D (Fig. 2). d Confirmed by X-ray, the NMR data are consistent with the solid state structures.
Due to the paramagnetic properties of some of the new
iron complexes, their 1H NMR resonances exhibit significant
paramagnetic shifts. Key H NMR data are compiled in Table
and 174 ppm. Complex 8b gave at room temperature a 1H NMR
spectrum with large paramagnetic shifts, which changed at 253 K
into a diamagnetic complex with chemical shifts between 0 and 12
ppm (see also Table 1). The spectrum of complex 8b at 253 K is not
in accordance with the X-ray structure of the complex (vide infra).
In the solid state, 8b takes the A configuration (Fig. 2), but a double
set of signals for the ligands as well as two imine resonances were
observed which suggest a second isomer in solution, either C, D or
E (Fig. 2). The complex 8g also gave a spectrum with large param-
agnetic shifts at room temperature, which changed at 253 K to a 1H
NMR spectrum with shifts between 0 and 9 ppm and showed one
set of signals for the ligands, suggesting one isomer in solution.
The other complexes gave spectra with large paramagnetic
shifts to lower field (up to 128 ppm), and some upfield shifts
(up to -6 ppm) were also observed. This chemical shift pattern
has been observed before for paramagnetic iron complexes with
pyridyl-based ligands.54,72 In the NMR spectra of paramagnetic
iron complexes of this type, sometimes significant line broadening
has been reported previously, and often not all resonances can
be observed.35 The N CH imine proton especially can exhibit
large downfield shifts and line broadening up to invisibility of the
signal,54,67 and this resonance is thus frequently not observed in
our NMR spectra. The paramagnetic metal complexes 8c,d,f and
h behave similar to related paramagnetic complexes previously
1
2 and all spectra are given in the supplementary information.†
The 1H NMR spectra of the complexes 8a and 8e with low ef-
fective magnetic moments in the solid state show resonances in the
diamagnetic region between 0 and 12 ppm as expected. The “par-
ent” compound 8a with no substituents on the phenyl ring showed
four N CH resonances between 9.31 and 8.81 ppm of compara-
ble intensity. The aromatic region of the isolated material showed
a complex, yet well resolved peak pattern (Table 2, for spectrum
see supplementary information†). Peaks at 6.75, 6.63 and 6.09 and
5.32 ppm could be assigned to pyridyl protons based on coupling
constants and by comparison with other iron pyridyl complexes.
The remaining aromatic protons gave four sets of signals for the
ligands, indicating a mixture of isomers, presumably A, B and C
(Fig. 2). The 1H NMR spectrum of complex 8a also revealed, that
only two, but not three ligands coordinated to the iron center.
When a 3.5 : 1 molar ratio of the parent ligand 7a was employed
1
in the synthesis of the diamagnetic complex 8a, the H NMR
spectrum was identical to that one obtained when a 2 : 1 molar
ratio of ligand to metal was employed in synthesis (for comparison
see Figure S4 in the supplementary information†). Complex 8a was
1
the only complex to give a reasonable 13C{ H} NMR spectrum,
1
which is well resolved in the aromatic region, and which also
reported in the literature. Representative H NMR spectra are
exhibited four singlets for the imine C N carbons between 171
shown in Fig. 3.
7620 | Dalton Trans., 2011, 40, 7617–7631
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
supplementary information†). The trimethoxy ligand 7e exhibits
increased electron-donating properties compared to the other
imine ligands 7, facilitating metal coordination. As a consequence,
with ligand 7e a solution equilibrium between [Fe(OTf)2(7e)2] and
[Fe(7e)3](OTf)2 forms, even though a 2 : 1 molar ratio between
ligand and [Fe(OTf)2] was employed. For the other ligands 7,
no evidence for the formation of complexes of the formulation
[Fe(7)3](OTf)2 was observed.
The fluorine atoms of the OTf ligands provide additional
information through 19F NMR spectroscopy. The OTf can give a
range of chemical shifts in the 19F NMR spectra; non-coordinated
OTf ligands have been reported to give solvent-dependent reso-
nances around -70 ppm which shift significantly downfield upon
coordination to Fe(II).73 In the complexes 8, the 19F NMR spectra
resonances were observed between -66.3 and -76.5 ppm (Table
2), which is consistent with free, non-coordinated OTf. All NMR
spectra for the new complexes were recorded in CD3CN (none
of the complexes was well soluble in CDCl3 or CD2Cl2), and
as previously reported, the CD3CN solvent displaced the weakly
coordinating OTf ligand in solution.64 The widths at half height
(n1/2) for the 19F NMR resonances ranged from 224 to 911 Hz for
all complexes. These values are indicative for dynamic processes
in solution;35 the OTf ligand might rapidly exchange with the
CD3CN solvent, leading to significant line broadening of the
resonances. No resonances consistent with iron-coordinated OTf
were observed in all 19F NMR spectra.
Fig. 3 1H NMR spectra of the iron complexes 8c (top), 8f (middle) and
8h (bottom). The schematic representations show only the ligand of the
corresponding iron complexes.
By integration, comparison of the spectra and line shapes and
consideration of literature values, it is possible to analyze the
spectra of the paramagnetic complexes. The three complexes 8c,f
and h show two distinct resonances between 52 and 56 ppm, which
can be assigned to two pyridyl protons; the pyridyl ring is present
in all three complexes and similar shifts have been previously
reported. Complex 8h lacks a second aromatic ring system, and
the t-Bu group gives one broad signal at -4.4 ppm which would
be in accordance with one isomer (Fig. 3, bottom). The other
pyridyl protons and the imine proton were not observed in the
spectrum (up to 200 ppm). The spectra of the complexes 8c and
f are in accordance with their X-ray structure. One set of signals
is observed for the ligands, as expected for the trans coordination
mode C (Fig. 2). For the diethyl complex 8c, in addition to the two
signals at 56.3 and 54.8 ppm, a third pyridyl proton was observed
at -6.12 ppm (Fig. 3, top). The ethyl groups appear at 14.8, 15.6
and 18.7 ppm and the remaining aromatic protons between 7 and
10 ppm, all as broad signals. Complex 8d showed a comparable
spectrum; the CH resonances appear around 4.8 and 2.0 ppm
as broad signals. The pyridyl ring protons appeared at 55.2 ppm
(broad, unsymmetrical signal) and two resonances were observed
around -7 ppm, which would be consistent with the two ligands
7d being equivalent, which would be consistent with the X-ray
structure of 8d. For the trimethyl complex 8f, the three methyl
groups and the two aromatic signals on the aryl ring system appear
as broad signals at 13.5, 16.6 and 17.1 ppm (Fig. 3, middle). In
addition to the two signals at 54.1 and 56.1 ppm, a third, very
broad signal at 128 ppm was observed (not shown in Fig. 3, but
the full spectrum is given in the supplementary information†); we
tentatively assigned this resonance also to the pyridyl ring, albeit
it also could be the imine proton.
Overall, the NMR data confirms the general formulation
[FeOTf2(7)2] in solution and matches the solid state structures
of the complexes 8c,d and f, which were analyzed by X-ray. Com-
plexes 8a and 8b form a mixture of structural isomers in solution,
presumably because in these complexes the ortho positions of the
aryl rings bear either no or one small fluoro substituent, facilitating
isomerization processes. Due to the variety of coordination modes
(Fig. 2), an unambiguous structural assignment based on NMR
data was not possible for the other complexes, but they appear
to form only one isomer in solution. However, line broadening in
1
some of the H NMR and all of the 19F NMR spectra suggests
dynamic behavior in solution and it cannot be excluded that for
some of the complexes different structural isomers exhibit a fast
exchange equilibrium.
UV-vis spectroscopy
The iron complexes 8 were isolated as intensely colored solids.
The UV-vis spectra were recorded and are provided in the
Supplementary information (three representative traces are shown
in Fig. 4). The lmax values are compiled in Table 1.
The complexes gave MLCT bands between 408 and 577
nm, which also have been reported for structurally related iron
complexes.35,63,74,75 The bands showed molar absorption coeffi-
cients e between 707 and 4729 M-1cm-1, which are typical for
charge transfer bands.76 In addition, the complexes 8c and f
showed broad bands around 520 nm, which we assigned to d–
d transitions due to their low molar absorption coefficients e
below 50 M-1cm-1.77 The o-fluoro complex 5b did not show a
charge transfer band, but only a weak d–d band at 532 nm (e = 78
M-1cm-1). The low spin complexes 8a and 8e exhibited a completely
different absorption pattern than the high spin complexes (Fig.
4 and S1†); no band around 420 nm was observed, but a very
The 1H NMR also revealed an additional complex in the
coordination chemistry of the trimethoxy ligand 7e. The dia-
magnetic complex 8e gave a 1H NMR spectrum with a well
resolved aromatic region, and two imine resonances and two sets
of signals for the ligands were observed. However, a second set of
signals with reduced intensities was observed in the spectrum.
If during the synthesis of complex 8e a 3.5 : 1 molar ratio of
ligand 7e to the iron precursor [Fe(OTf)2] was used, the intensities
of these signals significantly increased while they simultaneous
decreased for the other set of signals (see Figure S5 in the
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 7617–7631 | 7621
©

View Article Online
same region has been previously reported for structurally related
Fe(II) complexes with tetradentate imine pyridyl ligands, and was
also ascribed to a MLCT transition.63 The shoulders in the UV-vis
spectrum of 8a and 8e might be due to second isomers in solution,
which also were observed by 1H NMR spectroscopy.
X-ray analysis
To unequivocally establish the structures of the new iron com-
plexes, the X-ray structure of complexes 8b,c,d and f was deter-
mined (Fig. 5). Crystallographic parameters are given in Table 3
and key bond length and angles are compiled in Table 4.
For the complexes 8b,c and f, the bond angles for N and O
atoms in cis position about the iron center range from 107.0(2) to
˚
73.6(5) A and for those in trans position range from 163.27(19) to
˚
177.3(7) A. Their structures are, thus, best described as distorted
octahedral. The N(1)–Fe–N(2) and N(3)–Fe–N(4) bond angles
within the ligands range from 73.6(5) to 77.5(4)◦, whereas they
range from 101.9(2) to 100.4(2)◦ between ligands. Thus, the
average “bite angle” for the imine ligands 7b,c and f is 76◦, which
might be responsible for the distortion of the complexes. The
complex 8d takes a trigonal-bipyramidal coordination mode and
will be discussed separately below.
Fig. 4 UV-vis spectra of selected complexes 8.
intense absorption at 572 and 577 nm (which exhibit in addition
shoulders at 529 and 522 nm), whose intensities (e = 1831 and 4729
M-1cm-1) again suggests that it is a MLCT. An absorption in the
Table 3 Crystallographic parameters
Complex
8b
8c
8d
8f
Empirical formula
C26H18F8FeN4O6S2·
(CH3CN)
795.47
C34H36F6FeN4O6S2·
(CH2Cl2)(H2O)0.25
920.07
C38H44F6FeN4O6S2 ·
(CH2Cl2)
971.67
C32H32F6FeN4O6S2
Formula weight
Temperature
Wavelength
802.59
100(2) K
0.71073 A
213(2) K
100(2) K
0.71073 A
296(2) K
0.71073 A
˚
˚
˚
˚
0.71073 A
Crystal system
Space group
Unit cell dimensions
Monoclinic
Cc
Triclinic
P1
Monoclinic
P21/c
Monoclinic
C2
¯
˚
˚
˚
˚
˚
˚
˚
˚
˚
a = 13.7886(13) A
a = 10.7666(3) A
a = 15.8415(13) A
a = 13.7219(17) A
˚
b = 24.535(2) A
b = 12.5610(4) A
b = 14.8314(10) A
b = 14.5742(17) A
˚
˚
c = 10.0567(9) A
c = 15.2293(5) A
c = 20.2777(16) A
c = 18.219(2) A
a = 90◦
a = 90.052(2)◦
b = 98.549(2)◦
a = 90◦
a = 90◦
b = 98.992(3)◦
b = 110.005(4)◦
b = 94.145(5)◦
g = 90◦
g = 91.483(2)◦
g = 90◦
g = 90◦
3
3
3
3
˚
˚
˚
˚
Volume
3360.3(5) A
2036.01(11) A
4476.8(6) A
3633.9(7) A
Z
4
2
4
4
Density (calculated)
Absorption coefficient
F(000)
1.572 Mg m-3
0.666 mm-1
1.501 Mg m-3
1.442 Mg m-3
0.623 mm-1
2008
1.467 Mg m-3
0.608 mm-1
1648
0.680 mm-1
945
1608
Crystal size mm3
Theta range for data collection
Index ranges
0.28 ¥ 0.09 ¥ 0.05
1.71 to 25.17◦
-16 £ h £ 16, -29 £ k £
29,-11 £ l £ 11
35159
0.31 ¥ 0.13 ¥ 0.10
1.91 to 26.51◦
-13 £ h £ 13, -15 £ k £ 15,
-19 £ l £ 19
0.29 ¥ 0.17 ¥ 0.12
0.30 ¥ 0.20 ¥ 0.12
1.74 to 25.00◦.
2.04 to 26.47◦
-18 £ h £ 17, 0 £ k £ 17, 0 -17 £ h £ 17, 0 £ k £ 18, 0
£ l £ 24
144199
£ l £ 22
53003
Reflections collected
47261
Independent reflections
Completeness to theta = 25.00◦
Absorption correction
5765 [R(int) = 0.1103]
8321 [R(int) = 0.0322]
7947 [R(int) = 0.167]
3906 [R(int) = 0.094]
99.0%
99.2%
99.4%
99.9%
Semi-empirical from
equivalents
Semi-empirical from
equivalents
Semi-empirical from
equivalents
Semi-empirical from
equivalents
Max. and min. transmission
Refinement method
0.9687 and 0.8329
0.9332 and 0.8157
0.9307 and 0.8410
0.9328 and 0.8400
Full-matrix least-squares
on F2
Full-matrix least-squares Full-matrix least-squares on Full-matrix least-squares
on F2
F2
on F2
Data/restraints/parameters
Goodness-of-fit on F2
Final R indices [I > 2s(I)]
5765/4/440
1.048
R1 = 0.0606,
wR2 = 0.1453
R1 = 0.0948,
wR2 = 0.1651
0.04(3)
8321/125/553
1.053
R1 = 0.0464,
wR2 = 0.1191
R1 = 0.0644,
wR2 = 0.1329
7947/88/564
1.066
R1 = 0.1061,
wR2 = 0.2305
R1 = 0.1922,
wR2 = 0.2894
3906/71/420
1.020
R1 = 0.0711,
wR2 = 0.1717
R1 = 0.1038,
wR2 = 0.1962
-0.03(7)
R indices (all data)
Absolute structure parameter
Largest diff. peak and hole/e.A
-3
˚
0.976 and -1.180
0.748 and -0.834
1.845 and -0.947
0.562 and -0.321
7622 | Dalton Trans., 2011, 40, 7617–7631
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
◦
˚
Table 4 Key bond lenght (A) and angles ( )
8b
8c
8d
8f
Fe(1)–O(1)
Fe(1)–O(4)
Fe(1)–N(1)
Fe(1)–N(2)
Fe(1)–N(3)
Fe(1)–N(4)
N(2)–C(6)a
N(4)–C(X)a
C(5)–C(6)b
C(X)–C(Y)b
2.112(5)
2.118(5)
2.157(6)
2.167(6)
2.133(6)
2.197(6)
1.286(9)
1.289(9), X = 18
1.467(11)
1.457 (10)
X = 17, Y = 18
94.8(2)
2.083(2)
2.076(2)
2.212(2)
2.233(2)
2.225(2)
2.231(2)
1.267(3)
1.262(4), X = 22
1.459(4)
1.459(4)
X = 21, Y = 22
87.01(8)
89.64(8)
89.84(8)
88.16(8)
106.84(8)
90.81(8)
92.69(8)
176.26(8)
76.17(8)
90.61(8)
91.56(8)
75.59(8)
177.55(8)
101.41(8)
175.44(8)
1.969(6)
—
2.077(6)
2.087(6)
2.181(8)
2.224(11)
2.242(8)
2.233(15)
1.28(2)
1.22(3) X = 21
1.444(19)
1.44(3)
X = 20, Y = 21
89.9(5)
87.8(5)
90.5(5)
91.5(5)
107.0(2)
92.1(5)
90.1(5)
175.1(4)
77.5(4)
89.9(6)
88.2(6)
73.6(5)
2.174(8)
2.130(8)
2.163(7)
2.125(8)
1.266(12)
1.295(13) X = 24
1.459(13)
1.444(13)
X = 23, Y = 24
97.8(3)
—
99.5(3)
—
162.7(3)
115.8(3)
—
94.7(3)
77.3(3)
115.5(3)
—
77.1(3)
95.8(3)
128.7(3)
—
O(1)–Fe(1)–N(3)
O(4)–Fe(1)–N(3)
O(1)–Fe(1)–N(1)
O(4)–Fe(1)–N(1)
N(3)–Fe(1)–N(1)
O(1)–Fe(1)–N(2)
O(4)–Fe(1)–N(2)
N(3)–Fe(1)–N(2)
N(1)–Fe(1)–N(2)
O(1)–Fe(1)–N(4)
O(4)–Fe(1)–N(4)
N(3)–Fe(1)–N(4)
N(1)–Fe(1)–N(4)
N(2)–Fe(1)–N(4)
O(4)–Fe(1)–O(1)
86.6(2)
88.0(2)
101.6(2)
171.4(2)
161.8(2)
84.5(2)
101.9(2)
76.6(2)
90.5(2)
163.27(19)
76.7(2)
95.1(2)
100.4(2)
89.2(2)
179.3(5)
101.9(3)
177.3(7)
a Bond length of the C N imine groups. b Bond length of the C(imine)–C(pyridyl) carbon atoms.
The o-fluoro complex 8b features in the solid state a cis
coordination mode (A in Fig. 2). The C N imine nitrogen atoms
are located trans to the two OTf ligands, and the two pyridyl
rings are also arranged trans to each other. The o-fluoro aryl rings
are twisted with respect to the plane spanned by the imine group
and the phenyl ring connected to it; the fluoro substituents point
towards each other and away from the OTf ligands, presumably to
avoid contact with the neighboring oxygen atoms.
The diethyl complex 8c and the trimethyl complex 8f take a trans
geometry (C in Fig. 2). Here, the C N imine nitrogen atoms are
located trans to the pyridyl ring systems, as has been observed
previously in a related complex bearing three a-iminopyridine
ligands.53 The aryl rings bearing the ethyl and the methyl groups,
respectively, are arranged almost perpendicular to the plane that
includes the imine group and the pyridyl rings, presumably to
release steric strain.
seek a position trans to the weak OTf ligands, which has previously
been observed in the solid state structure of a structurally related
thiocyanate complex.52 Such an arrangement is sterically less
favorable for the diethyl complex 8c and the trimethyl complex
8f, because a cis arrangement of the two ligands as in the o-fluoro
complex 8b might create too much steric congestion caused by
the alkyl groups present on both ortho positions of the aryl rings.
As a consequence, the imine Fe–N bond lengths are significantly
shorter in the o-fluoro complex 8b than in the complexes 8c and
f, where both ortho positions are occupied by alkyl groups. It
can be assumed that the bidentate a-iminopyridine ligands 7
described herein might prefer the cis coordination mode, and bulky
substituents on the aryl ring force the system to adopt the less
favored trans geometry, as seen by elongated Fe–N bond lengths
in the trans complexes 8c and f.
The bond lengths for the C N imine bond for all four
˚
The Fe–N bond length for the complexes range from 2.125(8)
complexes range from 1.22(3) to 1.295(13) A and the C(imine)–
˚
˚
˚
to 2.242(8) A. Bond lengths around 2.2 A are typical for high spin
C(pyridyl) bond lengths from 1.444(19) to 1.467(11) A. These
˚
Fe(II) complexes, longer than about 2.0 A observed for low spin
values are very close to those found in free, closely related
bis(imino)pyridine ligands.46,79 Thus, extensive electron delocal-
ization from the iron(II) center to the ligand does not take place.
Such a formal reduction of the ligand by the iron center has been
reported for related iron(0) species, and resulted in significant
elongation of the C N bond and a concomitant reduction of
the C(imine)–C(pyridyl) bond length.46,79
The complex 8d takes an five-coordinate, distorted trigonal-
bipyramidal coordination mode. The R value of the X-Ray struc-
ture of 8d is somewhat high, but it establishes the connectivity and
geometry of the complex. One of the OTf ligands is coordinated
to the iron center, while the other one serves as counter ion.
complexes.31,78 There is an interesting difference in the Fe–N bond
lengths of the cis complex 8b and the trans complexes 8c and f.
They range from 2.133(6) to 2.197(6) A for the cis complex 8b
and from 2.181(8) to 2.242(8) A for the trans complexes 8c and 8f,
respectively. Thus, the average Fe–N bond lengths are shorter for
the cis complex than for the trans complexes. In turn, for the cis
complex 8b the Fe–O bond lengths (2.112(5) and 2.118(5) A) are
longer than for the trans complexes (2.076(2) to 2.087(6) A).
˚
˚
˚
˚
The coordination geometries of these three complexes in the
solid state can be rationalized by the trans influence of the
coordinated ligands. In the o-fluoro complex 8b, the imine groups
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 7617–7631 | 7623
©

View Article Online
Fig. 5 Molecular structures of 8b (top left), 8c (top right), 8d (bottom left) and 8f (bottom right; the F atoms of the two OTf ligands were omitted for
clarity). Ellipsoids are shown at 25 to 50% probability levels. Hydrogen atoms, solvent molecules and the OTf counter ion for 8d have been omitted for
clarity. For key bond lengths and bond angles, see Table 4.
The structure is, thus, best formulated as [Fe(OTf)(7d)2]+OTf.
Wieghardt described a related iron(I) complex [Fe(7d)2(THF)]+-
[B(ArF)4]-, which also exhibits a trigonal-bipyramidal coordi-
nation mode.46 The two pyridyl nitrogen atoms in 8d occupy
approximately the axial positions, as seen by the N(1)–Fe–N(3)
angle of 162.7(3)◦. The three equatorial positions are coordinated
by the imine nitrogen atoms and the OTf, which form angles
between 115.5(3) and 128.7(3)◦. The average “bite angle” (between
N(1)–Fe–N(2) and N(3)–Fe–N(4)) for the imine ligand 7d is 77◦,
close to those for the other imine ligands described above. It
appears that the bulky i-Pr substituents on the phenyl ring in ligand
7d neither can be accommodated by the octahedral cis nor the trans
coordination geometries. As for the complex 8b, the two pyridyl
nitrogen atoms take trans positions to each other, and the Fe–N
Catalysis
The new complexes were subsequently applied as catalysts in
oxidation reactions employing peroxides as the oxidants. First,
standard conditions for the oxidation of cyclohexane utilizing
H2O2 were employed,36,64,74 and the results for the complexes 8a
and 8c are compiled in Table 5.
In test reactions described in the literature,74 typically a large
excess of the substrate over the peroxide oxidant is employed to
avoid overoxidations, and conversions related to the peroxide oxi-
dant is reported rather than the yield of the oxidized product.44 For
both complexes 8a and 8c, no oxidation products of cyclohexane
were detected by GC when 100 : 1 and 20 : 1 cyclohexane/H2O2
ratios were employed. However, at 2 : 1 and 1 : 1 ratios, alcohol
and ketone products were observed in ratios between 0.37 and
˚
bond lengths (2.125(8) to 2.174(8) A) are also close to those of 8b.
7624 | Dalton Trans., 2011, 40, 7617–7631
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Table 5 Methylene group and alcohol oxidationsa
Catalyst
Reaction time/h
Peroxide/Substrate ratio
Alcohol/Ketone ratio
Conversionb,d
b
Cyclohexane oxidation with H2O2
8c
4
24
4
24
4
24
4
0.5
0.5
1
0.37
0.39
0.48
0.51
0.44
0.44
0.64
0.51
—
12%
16%
4%
1
5%
8a
0.5
0.5
1
11%
11%
3%
24
24
1
4%
1c
—
c
Cyclohexane oxidation with t-BuOOHb
8a
4
24
0.5
0.5
0.19
0.21
18%
38%
Diphenylmethane oxidation to benzophenone with t-BuOOHd
8a
0.5
0.5
0.5
4
—
—
—
73%
56%
69%
4c
4e
c
e
4-Phenyl-2-butanol oxidation to the ketone with t-BuOOHd
8a
1 h
1 h
4
—
—
34%
4c
39%c
a Details are given in the experimental. b For cyclohexane, total yield of alcohol and ketone related to the oxidant, as determined by GC, expressed as mol
product over mol oxidant employed. c In the presence 2,4,6-tri-t-butylphenol (one equivalent related to the substrate). d Yields determined by GC, related
to the diphenyl methane and 4-phenyl-2-butanol starting materials. e Reaction performed under inert conditions.
0.64 (Table 5) and 3 to 16% of the H2O2 oxidant was consumed
for product formation.
Further experiments to better understand the course of the iron
catalyzed oxidation reactions
The low alcohol/ketone ratio and the low efficiency with respect
to the H2O2 turnover are typical for autochain reactions through
oxygen-centered radicals.75 In addition, cyclohexane oxidation
with H2O2 did not take place when 2,4,6-tri-tert-butylphenol was
added to the reaction mixture, which has been reported to be an
efficient trap for oxygen-centered radicals.80
Also in the cyclohexane oxidations with H2O2, the yields do not
increase after 4 h reaction time. Furthermore, an increase of the
H2O2 oxidant leads to a decrease in the yield, which might indicate
catalyst decomposition.
In order to obtain further information about the catalytic effi-
ciencies, we followed the oxidation of 4-phenylbutan-2-ol with t-
BuOOH (entry 8, Table 6) at significantly reduced catalyst loading
of 0.3% over 2.5 h for all iron complexes obtained for this study.
The traces are compiled in Fig. 6, which also displays the observed
rate constant kobs for the different catalysts.
The oxidation of cyclohexane with t-BuOOH shows higher
oxidant efficiency (18%), which increases after 24 h to 38%
(Table 5). The oxidation with t-BuOOH appears to be slower
and no significant ligand decomposition over time seems to
occur. Preliminary experiments with t-BuOOH also showed
that activated methylene groups as in diphenylmethane and
secondary alcohols could be oxidized, which did not decrease
significantly when 2,4,6-tri-tert-butylphenol was added to the
reaction mixture, again demonstrating the absence of freely diffus-
ing, oxygen-centered radicals. The oxidation of diphenylmethane
also barely decreased when performed under inert conditions
(Table 5).
Next, we employed the complex 8a in oxidation reactions of
activated methylene groups and secondary alcohols utilizing t-
BuOOH as the oxidant (Table 6).
At a t-BuOOH/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 22 to 91% isolated yields. The
yields for the other imine complexes synthesized for this study
were comparable. Primary alcohols and methyl groups were not
oxidized under the conditions in Table 6.
Fig. 6 Comparison of the activities of the complexes 8.
There are slight but observable differences in the catalytic
activity of the iron complexes synthesized for the study. Roughly,
the complexes 8a,b,c and f show a somewhat higher catalytic
activity (kobs = 0.15 to 0.21 min-1) than the complexes 8d,e and
h (kobs = 0.05 to 0.09 min-1). Based on the data in Fig. 6, no firm
trend can be established. Among the most active catalysts are
those iron complexes, which show a high or low magnetic moment
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 7617–7631 | 7625
©

View Article Online
Table 6 Oxidation reactions employing iron complex 8a as catalyst and
t-BuOOH as oxidant
evidence for the fast formation of the actual oxidant in solution.
Albeit there are differences in activity for the test reaction in Fig.
6 at lower catalyst loads, no differences in yields were observed
under the conditions in Table 6.
Entry
1
Substrate
Producta
Isolated Yield
91%
The course of iron catalyzed oxidation reactions with peroxide
oxidants strongly depends on the iron complex, and the conditions
employed in the catalysis. A simplified mechanistic picture for t-
BuOOH is outlined in Scheme 3.81,82 Reaction of an iron complex
with t-BuOOH first yields an iron peroxo species [Fe–O–O-tBu]
(9), which has been described to be a “sluggish” oxidant and being
only a precursor of the actual oxidants.83 The further course of the
oxidation reaction depends on the “fate” of the species 9, which
has been reported to depend on the spin state of the iron complex.84
Homolytic cleavage of the [Fe–O–O–t-Bu] bond yields the t-Bu–
O· radical (12), which can promote unfavorable oxygen-centered
radical chain oxidation reactions, giving low alcohol/ketone ratios
in cyclohexane oxidations. Heterolytic cleavage of the iron peroxo
species 9 gives a high valent iron oxo species [Fe O] (10), which is
a metal-centered oxidant.85,86 Its oxidation chemistry is much bet-
ter controlled through supporting ligands on the iron center, and
its existence has been documented for non-heme iron complexes
spectroscopically,87 structurally88 and computationally.89
2
3
74%
51%
4
5
47%
22%
6
7
8
47%
23%
38%b
Scheme 3 Reactivities of iron complexes with t-BuOOH.
a Conditions: Substrate (0.6 mmol), t-BuOOH (2.4 mmol) and 8a (3 mol%),
4 h at room temperature in pyridine (1 mL). The products were isolated
by column chromatography. b The product contained ca 6% of 3-hydroxy-
1-phenylbutan-1-one.
In order to obtain additional information concerning potential
intermediates formed during our catalytic experiments, we fol-
lowed the reaction of the iron complex 8c with a 170-fold excess
of t-BuOOH in CH3CN over time by UV-vis, and the traces are
compiled in Fig. 7. As can be seen in Fig. 7 within the first 30 min
of the reaction, the intensity of the MLCT band around 408 nm
for the complex 8c gradually decreased. At the same time, a new
band at 599 nm appeared which gradually increased over time.
This new band most likely resembles the [Fe–O–O–t-Bu] species
9, which has previously been reported to give bands in the UV
between 590 and 650 nm.36,81,85,90 After five hours, the intensity of
the band around 600 nm did not increase anymore, and after 24 h,
the band had disappeared and the UV-vis spectrum of the mixture
was identical to that one of the starting material 8c (see Figure
S3 in the supplementary information†). No bands around 750 nm
were observed which are typical for the high valent [Fe O] species
10 which has been reported for non-heme systems to be unstable
at room temperature.88 An identical experiment was performed
utilizing H2O2 (spectrum see Figure S2 in the supplementary
(e.g. 8a and 8f) and which bear ligands, where none, one or both
ortho positions of the aryl ring contain substituents. In turn, the
bulky i-Pr substituents in 8d seems to result in somewhat decreased
activity, albeit its structure might favor the formation of an [Fe O]
intermediate and therefore exhibit increased activity. However, the
actual geometry of the complex under catalytic conditions cannot
be readily established and we do not have experimental evidence
for an [Fe O] intermediate. Also, complex 8e with a low magnetic
moment and complex 8h with an aliphatic t-Bu group on the imine
nitrogen atom exhibited lower catalytic activity.
Interestingly, Fe(OTf)2, which was used as precursor for the
synthesis of the iron complexes described herein, did not show
catalytic activity for the alcohol oxidation illustrated in Fig. 6.
Furthermore no induction period was observed, which might be
7626 | Dalton Trans., 2011, 40, 7617–7631
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
but are forced into a trans geometry if substituents are present
in positions ortho to the coordinating imine groups. A bulky i-
Pr substituent in the position ortho to the coordinating imine
ligand forces the complex to adopt a five-coordinate geometry
of the formulation [Fe(OTf)L2]+[OTf]-. The bond lengths of the
coordinated imine moieties are close to those of the free ligand,
indicating that no extensive electron delocalization from the
iron(II) center to the ligand is taking place. The new complexes
showed activity in the oxidation of cyclohexane with H2O2 or
t-BuOOH, which falls short of the activities of iron complexes
with tetradendate N-coordinating ligands. However, the oxidation
of activated methylene groups and secondary alcohols to the
corresponding ketones with t-BuOOH gave isolated yields between
22 and 91%. Kinetic measurements showed a moderate but
measurable influence of the ligand structure on the activity of
the corresponding metal complex in the oxidation of a secondary
alcohol. Spectroscopic evidence for formation of a [Fe–O–O–t-
Bu] intermediate was presented. The a-iminopyridine complexes
presented herein represent another tunable platform in the area of
iron catalyzed oxidation reactions but might also find applications
in cases where the stabilization of certain oxidation states or low-
valent iron complexes are required.
Fig. 7 UV-vis spectra of the reaction of the complex 8c with t-BuOOH.
The spectra were recorded in 3 min intervals for 30 min at room
temperature.
information†). Also, a band around 600 nm was observed, albeit
not as intense as in Fig. 7, which could be due to a [Fe–O–O–H]
species, as previously reported for similar systems.76
Barton has reported the GoAggIV (Fe(III)/t-BuOOH/pyridine)
and the GoAggV (Fe(III)/t-BuOOH/picolinic acid/pyridine) sys-
tems, which perform methylene group to ketone oxidations similar
to those described in Tables 5 and 6.91,92 Barton suggested a
non-radical pathway through an [Fe O] intermediate,93 but this
mechanistic picture has been questioned.94,95 The oxidation of
diphenyl methane to benzophenone (entry 2 in Table 6) also
proceeded in the presence of the radical scavenger 2,4,6-tri-tert-
butylphenol. Furthermore, the same reaction was not inhibited,
when performed under strictly inert conditions (Table 5). Both
experiments would indicate that the reaction does not proceed
through freely diffusing, oxygen-centered radicals, but the data do
not allow firm establishment of a mechanism.
Experimental
General
Chemicals were treated as follows: toluene, diethyl ether, dis-
tilled from Na/benzophenone; CH2Cl2, MeOH distilled from
CaH2, and CHCl3, silica (Aldrich) were used as received. 2,6-
Diisopropylaniline 6d (TCI America), aniline 6a (Fisher), and
other anilines 6 (Aldrich), 2-pyridine-carboxaldehyde (5, Aldrich),
R
Florisilꢀ (Fisher), t-BuOOH (70 wt% in H2O), H2O2 (30 wt%
The iron catalyzed oxidation of secondary alcohols to ketones
utilizing peroxide oxidants is much less common and different
mechanisms have been suggested.96–98 As no alcohol products
were observed in the oxidation of the activated methylene groups
to ketones, we assume that the oxidation of methylene groups
and alcohols does not proceed through the same mechanism and
that alcohols might not be intermediates in the oxidations of the
methylene groups. This assumption is further corroborated by
the fact, that Fe(OTf)2 does not catalyze the oxidation of the
secondary alcohol 4-phenyl-butan-2-ol (entry 7, Table 6). We have
previously shown that Fe(OTf)2 catalyzes the oxidation of fluorene
to fluorenone (entry 1 in Table 6), employing t-BuOOH as the
oxidant.65 A radical mechanism as well as mechanisms through
an [Fe O] intermediate have previously been suggested for iron
catalyzed alcohol oxidations.99,100 As for the oxidation of diphenyl
methane, the oxidation of 4-phenyl-2-butanol was not inhibited
in the presence of 2,4,6-tri-tert-butylphenol, suggesting that the
reaction does not proceed through an oxygen-centered radical,
but again, the data do not allow to firmly establish a mechanism.
in H2O), pyridine (all Aldrich) and other materials were used
as received. The ligands 7a,58 7d,59 7f60 and 7h61 were prepared
according to the literature. All reactions were carried out under
nitrogen employing standard Schlenk techniques, and workups
were carried out in the air.
NMR spectra were obtained at room temperature on a Bruker
Avance 300 MHz or a Varian Unity Plus 300 MHz instrument
and referenced to a residual solvent signal; all assignments are
tentative. Low temperature NMR spectra were recorded on a
Bruker ARX 500 MHz instrument. GC-MS spectra were recorded
on a Hewlett Packard GC-MS System Model 5988A. UV-vis
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 FT-IR spectrometer. Elemental analyses were performed by
Atlantic Microlab Inc., Norcross, GA, USA.
Syntheses of the ligands
(E)-2-Fluoro-N-(pyridin-2-ylmethylene)aniline (7b). In
a
Schlenk flask, 2-pyridine-carboxaldehyde (5, 0.500 g, 4.67 mmol)
was dissolved in methanol (4 mL). A solution of 2-fluoroaniline
(6b, 0.519 g, 4.67 mmol) in methanol (6 mL) was added, followed
by 5 drops of acetic acid. The reaction mixture was stirred for one
Conclusion
In conclusion, we have advanced the iron coordination chemistry
of the bidentate a-iminopyridine ligand (L) by synthesizing a
series of iron complexes of the general formula [Fe(OTf)2L2].
The bidentate ligands appear to prefer a cis coordination mode,
˚
hour and then molecular sieves (4 A, 1 g) was added. The reaction
mixture was stirred overnight and then filtered through basic
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 7617–7631 | 7627
©

View Article Online
alumina. The solution was concentrated to yield the product 7b
as a tan oil (0.759 g, 3.79 mmol, 81%), which solidifies over time
under vacuum. Found: C, 71.88; H, 4.58. Calcd for C12H9FN2: C,
71.99; H, 4.53.
of ligand 7a was added to the solution of Fe(OTf)2. The color of
the solution changed almost instantaneously to violet. After 30
min the solution was concentrated to about 4 mL and layered
with diethyl ether (7 mL). Violet crystals formed after 3 to 4 days.
The solvent was decanted, and the residue was dried under high
vacuum for 2 to 3 days to obtain the product as a violet solid
(0.427 g, 0.595 mmol, 78%). Found: C, 43.91; H, 3.23. Calcd for
C26H20F6N4O6S2Fe: C, 43.47; H, 2.81.101
1H-NMR dH (300.13 MHz; CDCl3; Me4Si) 8.72 (1H, ddd, 1JHH
2
3
= 4.9 Hz, JHH = 1.8 Hz, JHH = 1.0 Hz), 8.67 (1H, s, N CH),
1
2
8.26 (1H, dt, JHH = 7.9 Hz, JHH = 1.0 Hz), 7.86–7.80 (1H, m),
7.41–7.37 (1H, m), 7.27–7.13 (4H, m); 13C-NMR dC (75.5 MHz;
CDCl3; Me4Si) 163.3 (C = N), 149.9, 136.9, 127.8, 127.7, 125.6,
124.8, 124.7, 122.2, 121.9, 116.7, 116.4. IR (neat solid) nmax/cm-1
3058w, 1630 s (C N), 1584 s, 1568w, 1492s.
1H-NMR dH (300.13 MHz; CD3CN; Me4Si; 298 K) 9.31 (1H,
s, N CH), 9.15 (1H, s, N CH), 8.83 (1H, s, N CH), 8.81 (1H,
s, N CH), 8.59–8.53 (2H, m), 8.49–8.46 (1H, m), 8.37–8.35 (2H,
m), 8.25 (1H, t, 1JHH = 7.7 Hz), 7.99 (2H, t, 1JHH = 7.1 Hz), 7.92–7.89
(2H, m), 7.76–7.68 (2H, m), 7.66–7.64 (1H, m), 7.57–7.53 (2H, m),
7.37–7.35 (5H, m), 7.26–7.21 (2H, m), 7.19–7.16 (3H, m), 7.09–
7.07 (1H, m), 6.96 (2H, t, 1JHH = 7.6 Hz), 6.75 (2H, d, 1JHH = 7.4
Hz), 6.63 (2H, d, 1JHH = 7.4 Hz), 6.09 (2H, d, 1JHH = 7.9 Hz), 5.32
(2H, d, 1JHH = 7.9 Hz); 13C-NMR dC (75.5 MHz; CD3CN; Me4Si)
174.0 (N C), 173.4 (N C), 172.4 (N C), 170.9 (N C), 157.8,
157.5, 157.3, 157.0, 155.9, 155.7, 155.2, 154.9, 150.6, 149.6, 147.6,
147.2, 146.1, 138.9, 138.4, 138.1, 137.9, 131.0, 130.4, 129.9, 129.4,
129.3, 128.7, 128.3, 128.2, 128.1, 127.9, 121.5, 120.9, 120.8, 120.4.
HRMS calcd for C25H20F3N4O3S56Fe 569.0557, found 569.0555
(corresponds to [Fe(OTf)(7a)2]+). MS (FAB): m/z 569 (100%, M–
OTf), 420 (50, M–2OTf), 387 (40%, M–2OTf–7a). IR (neat solid)
(E)-2,6-Diethyl-N-(pyridin-2-ylmethylene)aniline (7c). 2-Pyri-
dine-carboxaldehyde (5, 0.500 g, 4.67 mmol) and 2,6-diethylaniline
(6c, 0.711 g, 4.67 mmol) were converted to 7c as described above
for 7b, yellow solid (1.12 g, 4.37 mmol, 94%). Found: C, 80.44; H,
7.79. Calcd for C16H18N2: C, 80.63; H, 7.61.
1H-NMR dH (300.13 MHz; CDCl3; Me4Si) 8.75–8.73 (1H, m),
8.36 (1H, s, N CH), 8.29 (1H, d, 1JHH = 8.1 Hz), 7.89–7.82 (1H,
1
m), 7.44–7.7.40 (1H, m), 7.15–7.05 (3H, m), 2.54 (4H, q, JHH
=
7.5 Hz), 1.16 (6H, t, JHH = 7.5 Hz); 13C-NMR dC (75.5 MHz;
CDCl3; Me4Si) 163.3 (C N), 154.7, 149.9, 149.8, 137.0, 133.0,
126.5, 125.5, 124.5, 121.4, 24.9 (CH2), 14.9 (CH3). IR (neat solid)
1
n
max/cm-1 2964 s, 1641 s (C N), 1584 m, 1467s.
n
max/cm-1 1594 m (C N), 1489 m, 1443 m, 1237 s (OTf), 1222 s
Synthesis of (E)-3,4,5-trimethoxy-N-(pyridin-2-ylmethylene)-
aniline (7e). 2-Pyridine-carboxaldehyde (0.286 g, 2.65 mmol) and
3,4,5-trimethoxyaniline (6e, 0.500 g, 2.65 mmol) were converted
to 7e as described above for 7a, yellow oil (0.728 g, 2.67 mmol,
98%).
(OTf), 1155 s (OTf), 1027 s (OTf).
[Fe(OTf)2(7b)2] (8b). The complex was synthesized as de-
scribed above from 2-fluoroimine 7b (0.112 g, 0.562 mmol) and
Fe(OTf)2 (0.099 g, 0.281 mmol). The product 8b was obtained
as a violet solid (0.120 g, 0.159 mmol, 57%) by crystallization
from CH2Cl2:diethyl ether (4 : 7), which was dried under high
vacuum for 2 to 3 days. Found: C, 41.95; H, 2.87. Calcd for
C26H18F8N4O6S2Fe: C, 41.39; H, 2.40.101
1H-NMR dH (300.13 MHz; CDCl3; Me4Si) 8.70 (1H, m), 8.64
(1H, s, N CH), 8.20 (1H, d, 1JHH = 7.9 Hz), 7.80 (1H, dt, 1JHH
=
7.8 Hz, 2JHH = 1.6 Hz), 7.37 (1H, ddd, 1JHH = 7.6 Hz, 2JHH = 4.8 Hz,
3JHH = 1.1 Hz), 6.60 (2H, s, aromatic), 3.89 (6H, s, 2OCH3), 3.87
(s, 3H, OCH3). 13C-NMR dC (75.5 MHz; CDCl3; Me4Si) 159.9
(C N), 154.7, 153.8, 149.9, 146.8, 137.5, 136.9, 125.3, 121.9, 98.8
(all s), 61.2 (s, OCH3), 56.3 (s, OCH3). IR (neat solid) nmax/cm-1
2937 m, 2836 m, 1584 s (C N), 1499 s, 1463 s, 1227 s, 1123s. MS
(EI): m/z 238 (80%, M), 257 (100, M–CH3), 242 (5, M–2CH3),
226 (25, M–3CH3).
1H-NMR dH (300.13 MHz; CD3CN; Me4Si, 243 K) 11.15 (1H,
1
broad), 10.13 (1H, broad), 9.41 (1H, d, JHF = 86 Hz), 8.84 (1H,
1
d, JHF = 37 Hz), 8.63 (1H, s), 8.49 (1H, m), 8.33–8.23 (1H, m),
1
8.06 (1H, d, JHF = 26 Hz), 7.91 (1H, s), 7.72–7.69 (1H, m), 7.55
(1H, s), 7.44 (1H, s), 7.24 (2H, d, 1JHF = 26 Hz), 7.12–7.09 (m, 2H),
6.94–6.81 (m, 2H). HRMS calcd for C25H18F5N4O3S56Fe 605.0369,
found 605.0374 (corresponds to Fe(OTf)(7b)2]+). MS (FAB): m/z
605 (100%, M–OTf), 456 (35, M–2OTf), 405 (60, M–OTf–L). IR
(neat solid) nmax/cm-1 1614 w, 1605 w, 1590 w (C N), 1495 m,
1256 s (OTf), 1223 m, 1155 s (OTf), 1029 s (OTf).
Synthesis of (E)-2-t-Butyl-N-(pyridin-2-ylmethylene)aniline
(7g). 2-pyridine-carboxaldehyde (5, 1.00 g, 9.34 mmol) and
2-tert-butylaniline (6g, 1.39 g, 9.34 mmol) was converted to 7g as
described above for 7a, yellow oil (2.07 g, 8.07 mmol, 86%).
1H-NMR dH (300.13 MHz; CDCl3; Me4Si) 8.75 (1H, d, 1JHH
=
[Fe(OTf)2(7c)2] (8c). The complex was synthesized as de-
scribed above from 2,6-diethylphenylimine 7c (0.134 g, 0.564
mmol) and Fe(OTf)2 (0.100 g, 0.283 mmol). The product 8c
was obtained as light red solid (0.134 g, 0.161 mmol, 57%) by
crystallization from CH2Cl2:diethyl ether (4 : 7), which was dried
under high vacuum for 2 to 3 days. Found: C, 49.26; H, 4.38. Calcd
for C34H36F6N4O6S2Fe: C, 49.16; H, 4.37.
1H-NMR dH (300.13 MHz; CD3CN; Me4Si, 298 K) 56.35 (2H,
broad), 54.84 (2H, broad), 18.73 (6H, broad, aromatic), 14.6–14.3
(10H, broad, 4CH2 + 2py-H), 1.83 (12H, broad, 4CH3), -6.12
(2H, br s, py-H). HRMS calcd for C33H36F3N4O3S56Fe 681.1810,
found 681.1818 (corresponds to [Fe(OTf)(7c)2]+). MS (FAB): m/z
681 (95%, M–OTf), 532 (10, M–2OTf), 443 (65, M–OTf–L). IR
(neat solid) nmax/cm-1 1635 m, 1595 m (C N), 1453w, 1316 s, 1233
s (OTf), 1205 s (OTf), 1016 s (OTf).
1
4.4 Hz), 8.53 (1H, s, N CH), 8.29 (1H, d, JHH = 7.9 Hz), 7.85
1
1
(1H, t, JHH = 7.6 Hz), 7.46 (1H, d, JHH = 6.2 Hz), 7.41–7.37
(1H, m), 7.32–7.19 (2H, m), 6.98 (1H, d, 1JHH = 7.1 Hz), 1.52 (9H,
s, 3CH3); 13C-NMR dC (75.5 MHz; CDCl3; Me4Si) 158.8, 155.3,
150.6, 149.7, 143.4, 136.9, 127.3, 126.5, 126.3, 125.1, 121.7, 119.3,
35.8, 30.7. MS (EI): m/z 238 (5%, M), 181 (100, M–t-Bu), 160
(50%, M–py). IR (neat solid) nmax/cm-1 3056w, 2954 s, 2907 m,
1626 s (C N), 1585 s, 1567 s, 1483 s, 1466 m, 1436m.
Synthesis of iron complexes
[Fe(OTf)2(7a)2] (8a). To a Schlenk flask containing Fe(OTf)2
(0.271 g, 0.766 mmol), acetonitrile (0.1 mL) and CH2Cl2 (6 mL)
were added with stirring. In another Schlenk flask, the imine 7a
(0.279 g, 1.53 mmol) was dissolved in CH2Cl2 (4 mL). The solution
7628 | Dalton Trans., 2011, 40, 7617–7631
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
[Fe(OTf)(7d)2][OTf] (8d). The complex was synthesized as
described above from 2,6-diisopropylphenylimine 7d (0.301 g,
1.13 mmol) and Fe(OTf)2 (0.200 g, 0.565 mmol). The product
8d was obtained as light red solid (0.274 g, 0.309 mmol, 55%) by
crystallization from CH2Cl2:diethyl ether (4 : 7), which was dried
under high vacuum for 2 to 3 days. Found: C, 51.01; H, 5.01. Calcd
for C38H44F6N4O6S2Fe: C, 51.47; H, 5.00.101
1H-NMR dH (300.13 MHz; CD3CN; Me4Si, 298 K) 55.4
(2H, broad, py-H), 15.9 (2H, broad, py-H), 8.2–6.9 (6H, broad,
aromatic), 4.8 (4H, broad, 2py-H + 2CH of C3H7), 2.00 (2H,
broad, CH), 1.5–1.1 (24 H, broad, 8CH3), -6.52 (2H, br s, 2py-
H), -7.79 (2H, br s, 2py-H). HRMS calcd for C37H44F3N4O3S56Fe
737.2436, found 737.2424 (corresponds to [Fe(OTf)(7d)2]+). MS
(FAB): m/z 737 (100%, M–OTf), 588 (10, M–2OTf), 471 (65, M–
OTf–L). IR (neat solid) nmax/cm-1 1633 m, 1597 m (C N), 1463w,
1448w, 1251 m (OTf), 1239 m, 1225 m, 1201 s (OTf), 1030 s (OTf).
mmol) and Fe(OTf)2 (0.100 g, 0.283 mmol). The product 8g
was obtained as dark violet solid (0.084 g, 0.102 mmol, 32%) by
crystallization from CH2Cl2:diethyl ether (4 : 7), which was dried
under high vacuum for 2 to 3 days.
1H-NMR dH (300.13 MHz; CD3CN; Me4Si; 253 K) 8.52
(2H, broad), 8.24–8.19 (3H, broad), 7.97 (2H, broad), 7.53 (2H,
broad), 7.19–7.13 (3H, m, broad), 6.99–6.97 (2H, broad), 6.73–
6.71 (1H, broad), 2.92 (3H, broad), 1.18–1.11 (18H, broad, 6CH3).
HRMS calcd for C33H36F3N4O3S56Fe 681.1810, found 681.1817
(corresponds to [Fe(OTf)(8g)2]+). MS (FAB): m/z 681 (85%, M–
OTf), 532 (10, M–2OTf), 443 (100, M–OTf–8g). IR (neat solid)
n
max/cm-1 1632w, 1595 m (C N), 1569w, 1485 m, 1445w, 1284 s
(OTf), 1224 s (OTf), 1163 (OTf), 1027 s (OTf).
[Fe(OTf)2(7h)2] (8h). The complex was synthesized as de-
scribed above from tert-butylimine 7h (0.092 g, 0.564 mmol) and
Fe(OTf)2 (0.100 g, 0.283 mmol). The light red product 8h (0.084
g, 0.124 mmol, 44%) was obtained as a solid by crystallization
from CH2Cl2:diethyl ether (4 : 7), which was dried under high
vacuum for 2 to 3 days. Found: C, 36.79; H, 4.10. Calcd for
C22H28F6N4O6S2Fe: C, 38.95; H, 4.16.101
Attempted Synthesis of [Fe(OTf)2(7e)2] (8e). The complex was
synthesized as described above from 3,4,5-trimethoxyphenylimine
7e (0.154 g, 0.566 mmol) and Fe(OTf)2 (0.100 g, 0.283 mmol).
The product was obtained as dark violet solid as a mixture
of [Fe(OTf)2(7e)2] and [Fe (7e)3](OTf)2 (0.096 g, 0.107 mmol,
38% with respect to [Fe(OTf)2(7e)2]) by crystallization from
CH2Cl2:diethyl ether (4 : 7), which was dried under high vacuum
for 2 to 3 days.
1H-NMR dH (300.13 MHz; CD3CN; Me4Si, 296 K). Peak as-
signment was performed by comparison with the NMR of isolated
material, where a three to one molar ratio of 7e to [Fe(OTf)2] was
employed in synthesis, see supplementary information, Figures S4
and S5†). [Fe(7e)3](OTf)2: 9.41 (s), 9.24 (s), 9.14 (s), 8.90 (s), 8.59
(t, JHH = 8.3 Hz), 8.48 (d, JHH = 7.8 Hz), 8.44 (d, JHH = 6.1 Hz),
8.34 (d, JHH = 1.3 Hz), 8.30 (d, JHH = 1.3 Hz), 8.25 (d, JHH = 1.1
Hz), 8.07 (m), 7.95 (d, JHH = 1.1 Hz), 5.94 (s), 5.89 (s), 5.79 (s), 5.45
(s), 3.66 (m), 3.58 (s). [Fe(OTf)2(7e)2]: 9.06 (s), 8.87 (s), 8.01 (m),
7.56 (d, JHH = 5.5 Hz), 7.46 (m), 7.06 (d, JHH = 5.5 Hz), 6.67 (s),
6.49 (s), 5.45 (s), 4.13 (s), 3.93 (d), 3.74 (s), 3.39 (s). HRMS calcd
for C31H32F3N4O9S56Fe 749.1192, found 749.1179 (corresponds to
[Fe(OTf)(7e)2]+). MS (FAB): m/z 749 (85%, M–OTf), 600 (10, M–
2OTf). IR (neat solid) nmax/cm-1 1659 m, 1595 s (C N), 1503 m,
1460 m, 1250 m (OTf), 1231 m (OTf), 1129 s (OTf), 1030 s (OTf).
1H-NMR dH (300.13 MHz; CD3CN; Me4Si, 298 K) 56.2 (2H,
broad, 2 py-H), 52.5 (2H, broad, 2py-H), -4.4 (18H, broad, tBu).
HRMS calcd for C21H28F3N4O3S56Fe 529.1184, found 529.1174
(corresponds to [Fe(OTf)(7h)2]+). MS (FAB): m/z 529 (100%, M–
OTf), 380 (15, M–2OTf), 367 (90, M–2OTf–7h). IR (neat solid)
n
max/cm-1 1638w, 1598 m (C N), 1383w, 1306 s (OTf), 1221 s
(OTf), 1167 s (OTf), 1029 s (OTf).
Standard experimental procedure for oxidation of cyclohexane,
diphenylmethane and 4-phenyl-2-butanol with H2O2 and
t-BuOOH (Table 5)
In a screw capped vial, complex 8a (4.3 mg, 0.006 mmol),
cyclohexane (0.507 g, 6.02 mmol), and bromobenzene (50 ml as
internal standard) were dissolved in pyridine (1 ml), and H2O2
(6.8 ml, 0.0602 mmol, 30wt% in H2O) was added. After the time
indicated in Table 5, an aliquot (0.1 ml) was taken out of the
solution, filtered through a short pad of silica gel (2 cm), which was
rinsed with CH2Cl2 (1 mL), and subsequently a GC was recorded.
Standard experimental procedure for oxidations in Table 6
[Fe(OTf)2(7f)2] (8f). The complex was synthesized as described
above from Fe(OTf)2 (0.100 g, 0.282 mmol) and trimethylim-
ine 7f (0.127 g, 0.564 mmol). The product was obtained as
orange crystals (0.109 g, 0.136 mmol, 48%) by crystallization
from CH2Cl2:diethyl ether (4 : 7), which was dried under high
vacuum for 2–3 days. Found: C, 47.67; H, 3.97. Calcd for
C32H32F6N4O6S256Fe: C, 47.89; H, 4.02.
1H-NMR dH (300.13 MHz; CDCl3; Me4Si; 298 K) 128 (2H,
broad), 56.1 (2H, s, py-H), 54.1 (2H, s, py-H), 17.1 (7H, broad),
16.5 (3H, broad), 14.5-14.4 (10 H, broad). (6 protons were not
observed in the spectrum, possibly 2 imine, 2 pyridyl and 2
aromatic protons). HRMS calcd for C31H32F3N4O3SFe 653.1497,
found 653.1519 (corresponds to [Fe(OTf)(7f)]+). MS (FAB): m/z
653 (50%, M–OTf), 504 (35, M–2OTf), 429 (85, M–2OTf–7f). IR
(neat solid) nmax/cm-1 1636 m, 1595 m (C N), 1479w, 1322 m,
1309 m, 1235 m (OTf), 1205 s (OTf), 1169 s (OTf), 1020 s (OTf).
Benzophenone. The substrate diphenylmethane (0.100 g, 0.595
mmol) and the catalyst 8a (0.013 g, 0.018 mmol) were dissolved
in pyridine (1.0 mL). The oxidant t-BuOOH (0.62 mL, 70 wt% in
H2O, 2.4 mmol) was slowly added within 20 to 30 min, and the
tan solution was shaken for 4 h at room temperature. The pyridine
solvent was removed under vacuum. The product benzophenone
was isolated by column chromatography as a colorless liquid
(0.080 g, 0.439 mmol, 74%, TOF 6.1), using silica gel and
CH2Cl2 as eluent. Experimental details and characterization data
1
including H and 13C NMR spectra of the catalysis products are
given in the supplementary information.†
Procedure for the kinetic experiment in Fig. 6. Complex 8a
(0.0014 g, 0.0020 mmol), 4-Phenyl-2-butanol (0.100 g, 0.67 mmol)
and bromobenzene (50 mL, internal standard) were dissolved in
pyridine (1 ml) and t-BuOOH (0.344 g, 70 wt% in H2O, 2.68 mmol)
was added in one portion at room temperature. In time intervals of
20 min., a sample (0.1 mL) was taken, filtered through a short pad
[Fe(7g)2(OTf)2] (8g). The complex was synthesized as de-
scribed above from 2-tert-butylphenylimine 7g (0.154 g, 0.566
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 7617–7631 | 7629
©

View Article Online
of silica gel (which was rinsed with 2 mL CH2Cl2), and analyzed
by GC. The other complexes were analyzed under strictly identical
conditions.
Acknowledgements
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.
Experimental for the UV-vis spectra in Fig. 7. A 1.83 mmol
l-1 solution of complex 8c was prepared, and a UV was recorded.
To this solution (0.5 mL, 0.91 ¥ 10-3 mmol 8c), t-BuOOH (20
mL, 70 wt% in H2O, 0.155 mmol) was added and UV-vis spectra
were recorded over time in 3 min intervals at room temperature.
An identical experiment was performed using H2O2 (10 mL, 30
wt%, aqueous solution), see Figure S2 of the Supplementary
information.
References
1 W. M. Czaplik, M. Mayer, J. Cvengrosˇ and A. Jacobi von Wangelin,
ChemSusChem, 2009, 2, 396.
2 E. B. Bauer, Curr. Org. Chem., 2008, 12, 1341.
3 S. Enthaler, K. Junge and M. Beller, Angew. Chem., Int. Ed., 2008, 47,
3317.
4 A. Correra, O. Garcia Manchen˜o and C. Bolm, Chem. Soc. Rev., 2008,
37, 1108.
X-ray Crystallography for 8b, 8c, 8f and 8d
Crystals were grown from CH2Cl2 solutions, which were layered
with diethyl ether and stored at -18 ◦C for several days. 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 Mo-Ka
5 B. D. Sherry and A. Fu¨rstner, Acc. Chem. Res., 2008, 41, 1500.
6 C. Bolm, J. Legros, J. Le Paih and L. Zani, Chem. Rev., 2004, 104,
6217.
7 A. A. O. Sarhan and C. Bolm, Chem. Soc. Rev., 2009, 38, 2730.
8 W. M. Czaplik, M. Mayer and A. Jacobi von Wangelin, Angew. Chem.,
Int. Ed., 2009, 48, 607.
9 W. M. Czaplik, M. Mayer, S. Grupe and A. Jacobi von Wangelin, Pure
Appl. Chem., 2010, 82, 1545.
10 R. B. Bedford, M. A. Hall, G. R. Hodges, M. Huwe and M. C.
Wilkinson, Chem. Commun., 2009, 6430.
11 S.-K. Xiang, L.-H. Zhang and N. Jiao, Chem. Commun., 2009, 6487.
12 S. Ito, T. Itoh and M. Nakamura, Angew. Chem., Int. Ed., 2011, 50,
454.
13 M. Lenze, N. P. Rath and E. B. Bauer, Tetrahedron Lett., 2010, 2855.
14 J. Jankowska, J. Paradowska, B. Rakiel and J. Mlynarski, J. Org.
Chem., 2007, 72, 2228.
15 W. Han, P. Mayer and A. R. Ofial, Adv. Synth. Catal., 2010, 352, 1667.
16 S. Prateeptongkum, I. Jovel, R. Jackstell, N. Vogl, C. Weckbecker and
Matthias Beller, Chem. Commun., 2009, 1990.
17 O. G. Mancheno, J. Dallimore, A. Plant and C. Bolm, Org. Lett., 2009,
11, 2429.
18 J. Yang and T. D. Tilley, Angew. Chem., Int. Ed., 2010, 49, 10186.
19 P. O. Lagaditis, A. J. Lough and R. H. Morris, Inorg. Chem., 2010, 49,
10057.
20 A. M. Tondreau, C. Milsmann, A. D. Patrick, H. M. Hoyt, E.
Lobkovsky, K. Wieghardt and P. J. Chirik, J. Am. Chem. Soc., 2010,
132, 15046.
21 A. C. Mayer and C. Bolm, in Iron Catalysis in Organic Chemistry, ed.
B. Plietker, Wiley-VCH, Weinheim, 1st edn, 2008, ch. 3, pp. 73–123.
22 L. Que, Jr. and W. B. Tolman, Nature, 2008, 455, 333.
23 P. D. Oldenburg and L. Que, Jr., Catal. Today, 2006, 117, 15.
24 C. He and Y. Mishina, Curr. Opin. Chem. Biol., 2004, 8, 201.
25 P. R. Ortiz de Montellano, Chem. Rev., 2010, 110, 932.
26 P. C. A. Bruijnincx, G. van Koten and R. J. M. Klein Gebbink, Chem.
Soc. Rev., 2008, 37, 2716.
27 A. Beck, B. Weibert and N. Burzlaff, Eur. J. Inorg. Chem., 2001, 521.
28 M. Costas, K. Chen and L. Que, Jr., Coord. Chem. Rev., 2000, 200–202,
517.
29 J. T. Groves, J. Inorg. Biochem., 2006, 100, 434.
30 N. Burzlaff, Angew. Chem., Int. Ed., 2009, 48, 5580.
31 K. Chen, M. Costas and L. Que, Jr., J. Chem. Soc., Dalton Trans.,
2002, 672.
˚
radiation (l = 0.71073 A) from a fine focus sealed tube X-ray
source. Preliminary unit cell constants were determined with a set
of 36 narrow frame scans. Typical data sets consist of combinations
of v and f scan frames with typical scan width of 0.5◦ and
counting time of 15–30 s/frame at a crystal to detector distance
of ~3.5 to 4.0 cm. The collected frames were integrated using an
orientation matrix determined from the narrow frame scans. Apex
II and SAINT software packages were used for data collection
and data integration.102 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 based on the Laue
symmetry using equivalent reflections.102 Structure solutions and
refinement were carried out using the SHELXTL- PLUS software
package.103 The structures were refined with full matrix least-
squares refinement by minimizing Rw(Fo -Fc2)2. All non-hydrogen
2
atoms were refined anisotropically 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.
Complex 8b was refined in the space group Cc. A molecule of
CH3CN was located in the lattice. However, the C atom of the
solvent could not be refined with anisotropic thermal parameters.
Data for the structure 8c was collected at 213 K as the crystal
breaks apart at 100 K. The structure was refined in the space
¯
group P1. The anion and the solvent CH2Cl2 in the structures
32 G. J. P. Britovsek, J. England, S. K. Spitzmesser, A. J. P. White and D.
J. Williams, Dalton Trans., 2005, 945.
exhibit disorder which were modeled using various constraints and
restrains. A partial occupancy water molecule was also located in
the lattice and was modeled with 25% occupancy.
33 G. J. P. Britovsek, J. England and A. J. P. White, Inorg. Chem., 2005,
44, 8125.
34 G. J. P. Britovsek, J. England and A. J. P. White, Dalton Trans., 2006,
1399.
The structures 8d and 8f were found to be two component
twins and were refined with twin model (HKLF 5/BASF). The
compounds crystallize in the monoclinic space groups P21/c and
C2 respectively. In both cases the anions and the solvent CH2Cl2
in the case of 8d were disordered. The disorder was modeled with
partial occupancy atoms. Even with multiple data sets relatively
poor quality data could only be obtained for these two compounds
due to twinning and disorder resulting in higher R values.
35 J. England, R. Gondhia, L. Bigorra-Lopez, A. R. Petersen, A. J. P.
White and G. J. P. Britovsek, Dalton Trans., 2009, 5319.
36 S. Gosiewska, H. P. Permentier, A. P. Bruins, G. van Koten and R. J.
M. Klein Gebbink, Dalton Trans., 2007, 3365.
37 M. S. Chen and M. C. White, Science, 2007, 318, 783.
38 M. S. Chen and M. C. White, Science, 2010, 327, 566.
39 C. Pavan, J. Legros and C. Bolm, Adv. Synth. Catal., 2005, 347, 703.
40 K. Schro¨der, S. Enthaler, B. Join, K. Junge and M. Beller, Adv. Synth.
Catal., 2010, 352, 1771.
7630 | Dalton Trans., 2011, 40, 7617–7631
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
41 P. C. A. Bruijnincx, I. L. C. Buurmans, S. Gosiewska, M. A. H.
Moelands, M. Lutz, A. L. Spek, G. van Koten and R. J. M. Klein
Gebbink, Chem.–Eur. J., 2008, 14, 1228.
42 K. Mo¨ller, G. Wienho¨fer, K. Schro¨der, B. Join, K. Junge and M. Beller,
Chem.–Eur. J., 2010, 16, 10300.
43 K. Schro¨der, B. Join, A. J. Amali, K. Junge, X. Ribas, M. Costas and
M. Beller, Angew. Chem., Int. Ed., 2011, 50, 1425.
44 J. England, C. R. Davies, M. Banaru, A. J. P. White and G. J. P.
Britovsek, Adv. Synth. Catal., 2008, 350, 883.
45 T. J. Collins, Acc. Chem. Res., 1994, 27, 279.
46 C. C. Lu, E. Bill, T. Weyhermu¨ller, E. Bothe and Karl Wieghardt, J.
Am. Chem. Soc., 2008, 130, 3181.
47 Y. Oso and T. Ishida, Chem. Lett., 2009, 38, 604.
48 J. Y. Wu, B. N. Stanzl and T. Ritter, J. Am. Chem. Soc., 2010, 132,
13214.
74 F. Li, M. Wang, A. Gao, H. Chen and L. Sun, Dalton Trans., 2006,
2427.
75 A. N. Biswas, A. Pariyar, S. Bose, P. Das and P. Bandyopadhyay, Catal.
Commun., 2010, 11, 1008.
76 V. Balland, F. Banse, E. Anxolabe´he`re-Mallart, M. Nierlich and J.-J.
Girerd, Eur. J. Inorg. Chem., 2003, 2529.
77 T. Buchen and P. Gu¨tlich, Inorg. Chim. Acta, 1995, 231, 221.
78 M. M. Bhadbhade and D. Srinivas, Polyhedron, 1998, 17, 2699.
79 S. C. Bart, K. Chłopek, E. Bill, M. W. Bouwkamp, E. Lobkovsky, F.
Neese, K. Wieghardt and P. J. Chirik, J. Am. Chem. Soc., 2006, 128,
13901.
80 D. H. R. Barton, V. N. Le Gloahee, H. Patin and F. Launay, New J.
Chem., 1998, 559.
81 M. P. Jensen, A. M. i Payeras, A. T. Fiedler, M. Costas, J. Kaizer, A.
Stubna, E. Mu¨nck and L. Que, Jr., Inorg. Chem., 2007, 46, 2398.
82 C. Nguyen, R. J. Guajardo and P. K. Mascharak, Inorg. Chem., 1996,
35, 6273.
49 K. Nienkemper, V. V. Kotov, G. Kehr, G. Erker and R. Fro¨hlich, Eur.
J. Inorg. Chem., 2006, 366.
50 C. Vedder, F. Schaper, H.-H. Brintzinger, M. Kettunen, S. Babik and
G. Fink, Eur. J. Inorg. Chem., 2005, 1071.
51 Malcolm A. Halcrow, Polyhedron, 2007, 26, 3523.
52 J.-F. Le´tard, M. Kollmansberger, C. Carbonera, M. Marchivie and P.
Guionneau, C. R. Chim., 2008, 11, 1155.
53 M.-D. S¸erb, B. Calmuschi-Cula, F. Dumitru, U. Englert and C. Guran,
Acta Crystallogr., Sect. E: Struct. Rep. Online, 2008, E64, m212–m213.
54 V. C. Gibson, R. K. O’Reilly, D. F. Wass, A. J. P. White and D. J.
Williams, Dalton Trans., 2003, 2824.
55 P. J. Chirik and K. Wieghardt, Science, 2010, 327, 794.
56 T. D. Manuel and J.-U. Rohde, J. Am. Chem. Soc., 2009, 131, 15583.
57 O. Garc´ıa Manchen˜o, Angew. Chem. Int. Ed., 2011, 50, 2216.
58 M. Schulz, M. Klopfleisch, H. Go¨rls, M. Kahnes and M. Wester-
hausen, Inorg. Chim. Acta, 2009, 362, 4706.
83 M. S. Seo, T. Kamachi, T. Kouno, K. Murata, M. J. Park, K.
Yoshizawa and W. Nam, Angew. Chem., Int. Ed., 2007, 46, 2291.
84 M. R. Bukowski, H. L. Halfen, T. A. Van Den Berg, J. A. Halfen and
L. Que, Jr., Angew. Chem., Int. Ed., 2005, 44, 584.
85 M. S. Seo, T. Kamachi, T. Kouno, K. Murata, M. J. Park, K.
Yoshizawa and W. Nam, Angew. Chem., Int. Ed., 2007, 46, 2291.
86 C. Krebs, D. G. Fujimori, C. T. Walsh and J. M. Bollinger, Jr., Acc.
Chem. Res., 2007, 40, 484.
87 M. P. Jensen, M. Costas, R. Y. N. Ho, J. Kaizer, A. M. i Payeras, E.
Mu¨nck, L. Que, Jr., J.-U. Rohde and A. Stubna, J. Am. Chem. Soc.,
2005, 127, 10512.
88 J.-U. Rohde, J. H. In, M. H. Lim, W. W. Brennessel, M. R. Bukowski,
A. Stubna, E Mu¨nck, W. Nam and L. Que, Jr., Science, 2003, 299,
1037.
59 Z. Benko¨, S. Burck, D. Gudat, M. Nieger, L. Nyula´szi and N Shorea,
Dalton Trans., 2008, 4937.
60 C. Burstein, C. W. Lehmann and F. Glorius, Tetrahedron, 2005, 61,
6207.
89 A. Decker, J.-U. Rohde, E. J. Klinker, S. D. Wong, L. Que, Jr. and E.
I. Solomon, J. Am. Chem. Soc., 2007, 129, 15983.
90 F. Namuswe, G. D. Kasper, A. A. Narducci Sarjeant, T. Hayashi, C.
M. Krest, M. T. Green, P. Monne-Loccoz and D. P. Goldberg, J. Am.
Chem. Soc., 2008, 130, 14189.
91 D. H. R. Barton and W. Chavasiri, Tetrahedron, 1994, 50, 19.
92 D. H. R. Barton and T.-L. Wang, Tetrahedron, 1994, 50, 1011.
93 U. Schuchardt, M. J. D. M. Jannini, D. T. Richens, M. C. Guerreiro
and E. V. Spinace´, Tetrahedron, 2001, 57, 2685.
94 D. W. Snelgrove, P. A. McFraul, K. U. Ingold and D. D. M. Wayner,
Tetrahedron Lett., 1996, 37, 823.
95 P. Stavropoulos, R. C¸ elenligil-C¸ etin and A. E. Tapper, Acc. Chem.
Res., 2001, 34, 745.
96 A. Al-Hunaiti, T. Niemi, A. Sibaouih, P. Pihko, M. Leskela¨ and T.
Repo, Chem. Commun., 2010, 46, 9250.
97 S. A. Moyer and T. W. Funk, Tetrahedron Lett., 2010, 51, 5430.
98 S. E. Mart´ın and A. Garrone, Tetrahedron Lett., 2003, 44, 549.
99 E. Balogh-Hergovich and G. Speier, J. Mol. Catal. A: Chem., 2005,
230, 79.
100 J. H. Han, S.-K. Yoo, J. S. Seo, S. J. Hong, S. K. Kimb and C. Kim,
Dalton Trans., 2005, 402.
61 K. Kloc, E. Kubicz, J. Młochowski and L. Syper, Synthesis, 1987, 12,
1084.
62 K. S. Hagen, Inorg. Chem., 2000, 39, 5867. The synthesis of
[Fe(CF3SO3)2]·2CH3CN is described but we used material that was
dried for several days under vacuum and we had no evidence for
leftover coordinating CH3CN.
63 V. R. Souza, H. R. Rechenberg, J. A. Bonacin and H. E. Toma,
Spectrochim. Acta, Part A, 2008, 71, 1296.
64 S. Gosiewska, J. J. L. M. Cornelissen, M. Lutz, A. L. Spek, G. van
Koten and R. J. M. Klein Gebbink, Inorg. Chem., 2006, 45, 4214.
65 M. Lenze and E. B. Bauer, J. Mol. Catal. A: Chem., 2009, 309, 117.
66 A. Hauser, J. Jeftic´, H. Romstedt, R. Hinek and H. Spiering, Coord.
Chem. Rev., 1999, 190–192, 471.
67 B. Weber and F. A. Walker, Inorg. Chem., 2007, 46, 6794.
68 M. L. Naklicki, C. A. White, L. L. Plante, C. E. B. Evans and R. J.
Crutchley, Inorg. Chem., 1998, 37, 1880.
69 C. A. Reed and F. Guiset, J. Am. Chem. Soc., 1996, 118, 3281.
70 K. P. Bryliakov, E. A. Duban and E. P. Talsi, Eur. J. Inorg. Chem.,
2005, 72.
71 Y. Zang, J. Kim, Y. Dong, E. C. Wilkinson, E. H. Appelman and L.
Que, Jr., J. Am. Chem. Soc., 1997, 119, 4197.
101 Some of the elemental analysis were slightly high on carbon, which we
attribute to residual solvent. The NMR spectra typically contain trace
amounts of solvents, which could not be removed even after extended
periods of drying under high vacuum.
72 Y. Zang and Lawrence Que, Jr., Inorg. Chem., 1995, 34, 1030.
73 D. W. Blakesley, S. C. Payne and K. S. Hagen, Inorg. Chem., 2000, 39,
1979.
102 Bruker Analytical X-ray, Madison, WI, 2008.
103 G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr., 2008,
A64, 112.
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 7617–7631 | 7631
©