Non-Heme FeII Diastereomeric Complexes Bearing a Hexadentate Ligand: Unexpected Consequences for the Spin State and Catalytic Oxidation Properties

Article abstract of DOI:10.1002/chem.201902423

The reactivity and selectivity of non-heme Fe^II complexes as oxidation catalysts can be substantially modified by alteration of the ligand backbone or introduction of various substituents. In comparison with the hexadentate ligand N,N,N′,N′-tetrakis(pyridin-2-ylmethyl)ethane-1,2-diamine (TPEN), N,N′-bis[1-(pyridin-2-yl)ethyl]-N,N′-bis(pyridin-2-ylmethyl)ethane-1,2-diamine (_2MeL₆²) has a methyl group on two of the four picolyl positions. Fe^II complexation by _2MeL₆² yields two diastereomeric complexes with very similar structures, which only differ in the axial/equatorial positions occupied by the methylated pyridyl groups. In solution, these two isomers exhibit different magnetic behaviors. Whereas one isomer exhibits temperature-dependent spin-state conversion between the S=0 and S=2 states, the other is more reluctant towards this spin-state equilibrium and is essentially diamagnetic at room temperature. Their catalytic properties for the oxidation of anisole by H₂O₂ are very different and correlate with their magnetic properties, which reflect their lability/inertness. These different properties most likely depend on the different steric constraints of the methylated pyridyl groups in the two complexes.

Full text of DOI:10.1002/chem.201902423

A Journal of
Accepted Article
Title: Non-heme Fe(II) diastereomeric complexes bearing a
hexadentate ligand : unexpected consequences on the spin state
and oxidation catalytic properties
Authors: Katell Sénéchal-David, Charlotte Buron, Nathalie Ségaud,
Jean-Noël Rebilly, Amandine Dos Santos, Jonathan Farjon,
Régis Guillot, Christian Herrero, Tanya Inceoglu, and Frederic
Banse
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201902423
Link to VoR: http://dx.doi.org/10.1002/chem.201902423
Supported by

10.1002/chem.201902423
Chemistry - A European Journal
FULL PAPER
Non-heme Fe(II) diastereomeric complexes bearing a hexadentate
ligand : unexpected consequences on the spin state and
oxidation catalytic properties.
Katell Sénéchal-David,*[a] Charlotte Buron,[a] Nathalie Ségaud,[a](b) Jean-Noël Rebilly,[a] Amandine Dos
Santos,[a] Jonathan Farjon,[c] Régis Guillot,[a] Christian Herrero,[a] Tanya Inceoglu,[a] and Frédéric
Banse*[a]
Abstract: The reactivity and selectivity of non-heme FeII complexes
as oxidation catalysts can be substantially modified upon alteration
of the ligand backbone or introduction of various substituents. In
comparison to the hexadentate ligand TPEN (N,N,N',N'-
bonds.[9-12] Catalytic activities with improved regio- or
enantioselectivity and efficiency have also been observed with
more rigid ligand backbones by replacing the ethylenediamine
linker of the ligand by enantiomerically pure bipyrrolidine,[13-16]
bipiperidine[17] or cyclohexane diamines.[18] Additionally, the
presence of bulky substituents has been shown to strongly
influence the site of aliphatic C-H oxidation in different
substrates.[19,20] Thus, it appears that any modification on the
ligand skeleton can result in important tuning of the catalyst
performance. Along this line, we envisioned that methylation of
two of the picolyl groups in the TPEN ligand (N,N,N',N'-
tetrakis(pyridin-2-ylmethyl)ethane-1,2-diamine, Scheme 1) could
influence the catalytic properties of [FeII(TPEN)]2+. The six
coordinate [FeII(TPEN)]2+ displays a spin equilibrium in solution
between low spin (LS, S=0) and high spin (HS, S=2) states
which is in line with its lability. Thus, it readily reacts with
oxidants such as H2O2,[21] oxygen atom donors[22] and
dioxygen[23] yielding reaction intermediates capable of promoting
2
tetrakis(pyridin-2-ylmethyl)ethane-1,2-diamine), 2MeL6 (N,N'-bis(1-
(pyridin-2-yl)ethyl)-N,N'-bis(pyridin-2-ylmethyl)ethane-1,2-diamine)
contains one methyl group on two of the four picolyl positions. FeII
2
complexation by 2MeL6 yields two diastereoisomeric complexes with
very similar structures, only differing in the axial/equatorial positions
occupied by the methylated pyridyl groups. In solution, these two
isomers exhibit different magnetic behavior. Whereas one isomer
displays a temperature dependent spin state conversion between
the S=0 and S=2 states, the other one is more reluctant for this spin
state equilibrium and is essentially diamagnetic at room temperature.
Their catalytic properties for anisole oxidation by H2O2 are very
different and correlate with their magnetic properties which reflect
their lability / inertness. These different properties most likely rely on
the different steric constraints in which the methylated pyridyl groups
are engaged in the two complexes.
the
oxidation
of
olefins,
alkanes[22]
and
aromatic
hydrocarbons.[10,11] Additionally, incorporation of alkyl or
aromatic groups at the 2-pyridinylmethyl positions create chiral
centers whose location has been demonstrated to be beneficial
in asymmetric epoxidation reactions.[24,25]
Introduction
Many mononuclear heme and non heme iron enzymes carry out
the oxidation of inert chemical bonds under mild conditions.[1-4]
With the aim to develop cheap catalytic systems, numerous non
heme iron complexes have been prepared in order to reproduce
the function of these enzymes.[5-7] Following the peroxide shunt
strategy, some FeII complexes supported by polyazadentate
ligands have proven capable of promoting the activation of
hydrogen peroxide leading to the oxidation of olefinic or aliphatic
substrates[8] as well as the hydroxylation of aromatic C-H
N
N
N
N
Me
N
Me
N
N
N
N
N
N
N
2
TPEN
2MeL6
Scheme 1. Ligands used in this study.
In this paper, we report the synthesis, characterization, study of
the magnetic behavior in solution and evaluation of the catalytic
abilities of two diastereomeric Fe(II) complexes bearing the
[a]
Dr K. Sénéchal-David, Dr C. Buron, Dr N. Ségaud, Dr J.-N.
Rebilly(s), Mrs A. Dos santos, Dr R. Guillot, Dr. C. Herrero, Mrs T.
Inceoglu, Pr. F. Banse
2
hexadentate ligand 2MeL6 (N,N'-bis(1-(pyridin-2-yl)ethyl)-N,N'-
Institut de Chimie Moléculaire et des Matériaux d'Orsay
Univ. Paris-Sud, Univ. Paris-Saclay, CNRS
91405 Orsay cedex, France
E-mail: katell.senechal-david@u-psud.fr, frederic.banse@u-psud.fr
Dr N. Ségaud
Departement für Chemie und Biochemie
Universität Bern
Freiestrasse 3, 3012 Bern, Switzerland
Dr J. Farjon
bis(pyridin-2-ylmethyl)ethane-1,2-diamine, Scheme 1), in which
chiral centers are introduced by methylation of two of the four
2
pyridinylmethyl positions. These two [FeII(2MeL6 )]2+ complexes
(1a and 1b) exhibit very similar structures, only differing in the
axial/equatorial positions occupied by the methylated pyridyl
groups. Unexpectedly, they exhibit distinct catalytic properties
towards anisole oxidation by hydrogen peroxide. This can be
rationalized by comparing the divergent temperature evolution of
their spin state equilibrium in solution, which reflects their
relative lability. This characteristic itself can be correlated to the
different steric constraints within their coordination sphere. Such
drastic consequences arising from very subtle structural
[b]
[c]
CEISAM
Université de Nantes, CNRS
2, chemin de la Houssinière, 44322 Nantes cedex 3, France
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/chem.201902423
Chemistry - A European Journal
FULL PAPER
changes within twin catalysts have been very rarely
documented.[26]
At higher temperature (293 and 400 K), elongation of the Fe-N
bond distances is observed for both complexes indicative of a
low spin (S=0) to high spin state (S=2) equilibrium for both
complexes. Nonetheless, the variation in the bond distances is
more important for 1b•(PF6)2 suggesting that the spin state
transition occurs at lower temperature for this latter complex
(Supporting Information, Table S3). Magnetic measurements
performed on 1•(PF6)2 shows that this spin state conversion is
not complete at 400 K (Supporting Information, Figure S4).
Results
2
The ligand 2MeL6 was obtained by a conventional reductive
amination route between 3 and pyridine carboxaldehyde in the
presence of NaBH(OAc)3, as described in Scheme 2. The
diamine 3 was obtained by reduction of the isolated diimine 2.
The detailed experimental procedures are given in Supporting
Information.
1a
O
N
H
H
1,2-DCE
CH(OCH3)3
MeOH
NaBH4
N
N
N
N
N
N
N
N
2
+
H2N
NH2
2
3
Λ-SeqSeq
Δ-ReqReq
N
N
H
H
N
N
N
N
O
1,2-DCE
NaBH(OAc)3
N
N
N
+
2
N
1b
N
3
2
2MeL6
2
Scheme 2. Synthesis of ligand 2MeL6
.
2
Δ-ReqSax
Λ-SeqRax
Starting from the diastereomeric mixture of 2MeL6 , three
geometric isomers of [FeII(2MeL6 )]2+ 1 can be anticipated,
2
depending on the axial or equatorial position occupied by each
of the pyridinylethyl group. Among these three isomers (1a, 1b,
and 1c, Supplementary Information, Figure S3), only 1a, which
displays the two 2-pyridinylethyl groups in the equatorial plane,
and 1b, with one 2-pyridinylethyl group in the equatorial plane
and the other in axial position, were observed in solution and in
the solid state (see NMR study below and Experimental Section).
The structures of 1a•(PF6)2 and 1b•(PF6)2 were determined by
single crystal X-ray diffraction at 100 K (Figure 1). In any isomer
of 1•(PF6)2, the ligand wraps the metal center in a hexadentate
fashion and all of the Fe-N bond distances are of ca. 2.0 Å,
indicating a low spin (S=0) FeII at 100 K (Supporting Information,
Table S3). Interestingly, isomer 1a•(PF6)2 contains a S,S/R,R
racemic mixture of the ligand in the unit cell. To avoid steric
congestion between the axial pyridine rings and the methyl
groups of the equatorial 2-pyridinylethyl, the S,S ligand is found
in the Λ complex stereoisomer while the R,R ligand is found in
the Δ complex (Supporting Information, Figure S3). Therefore,
the unit cell of 1a•(PF6)2 only contains the Λ-SS / Δ-RR pair of
enantiomers.
Figure 1. Molecular complexes of the two stereoisomers of [FeII(2MeL62)](PF6)2
1•(PF6)2 determined at 100 K (ellipsoids shown at 50%). (Top) The two
enantiomers of 1a with the pyridinylethyl groups occupying two equatorial
positions (RR/SS ligand); (Bottom) The two enantiomers of 1b with the
pyridinylethyl groups occupying one equatorial and one axial position (RS/SR
ligand). The pyridinylethyl groups are highlighted in green. Only the hydrogen
atoms on the stereogenic carbons and the methyl groups are shown.
Despite numerous attempts under different experimental
conditions, the two diastereomeric complexes 1a•(PF6)2 and
1b•(PF6)2 always crystallized simultaneously as dark red brown
crystals in a 1:1 ratio which mirrors the initial composition of the
diastereomeric ligand mixture (see NMR studies below,
Experimental Section and Supporting Information, Figure S5).
Additionally, any attempt to separate the different stereoisomers
2
of the ligand 2MeL6 by semi-preparative chiral HPLC or by
recrystallization as a salt has been unsuccesful yet. Thus,
solution studies were performed using an equimolar mixture of
the two isomers in most cases. In some instances, larger
crystals of both isomers crystallized and could be manually
sorted, yielding samples of 1a•(PF6)2 (pure) and 1b•(PF6)2 (90%
purity) that were used for further characterization (see below).
In contrast, isomer 1b•(PF6)2 (Figure 1, bottom) contains the
meso form (R,S or S,R) of the ligand. For the same steric
reasons as in 1a, the number of isomers observed is restrained
and the Λ complex contains the SequatorialRaxial ligand while the Δ
complex contains the RequatorialSaxial ligand. Therefore, the unit
cell
of
1b•(PF6)2
only
contains
the
pair
of
Λ-Sequatorial
R
axial / Δ-RequatorialSaxial enantiomers.
This article is protected by copyright. All rights reserved.

10.1002/chem.201902423
Chemistry - A European Journal
FULL PAPER
1 104
8000
6000
4000
2000
0
4 10-5
3 10-5
2 10-5
1 10-5
0
of 1 (i.e. containing both 1a and 1b) and [FeII(TPEN)]2+ were
performed between 233 and 333 K and compared to each other
(Supplementary Information, Figure S6). For both complexes,
the overall decrease of the MLCT band at 420 nm upon heating
is consistent with larger amounts of high spin component.
Indeed, the MLCT for high spin FeII complexes are less
intense.[28]
-1 10-5
-2 10-5
-3 10-5
The 1H NMR spectrum of 1 in CD3CN at room temperature
shows both sharp and broad resonances over the 1 – 14 ppm
region (Figure 3, black trace). The 1H NMR spectrum of 1a
isolated in crystalline form and recorded under the same
conditions displays only the set of sharp resonances spanning
the diamagnetic region, between 1 and 10 ppm (Figure 3, red
trace), and therefore the broad resonances correspond to
complex 1b. This indicates that 1a is essentially diamagnetic
and non fluxional (sharp resonances), whereas 1b which
exhibits shifted resonances displays a larger paramagnetic
contribution at room temperature.
300
400
500
λ (nm)
600
700
0
0.5
1
1.5
2
E (V) / SCE
Figure 2. (Left). CV of [FeII(2MeL62)]2+ 1 (blue) and [FeII(TPEN)]2+ (dashed red)
2 mM in MeCN, TBAPF6 0.1 M, RT, 0.1 V.s-1. (Right) Electronic absorption
spectrum of [FeII(2MeL62)]2+ 1 in MeCN (black line) and in MeOH (green line).
Cyclic voltammetry (CV) of an acetonitrile solution of the
microcrystalline powder of 1•(PF6)2 (i.e. containing an equimolar
mixture of 1a and 1b) displays only one reversible FeIII/FeII
signal at 0.80 V vs SCE indicating that the two isomers have the
same redox properties (Figure 2, left). This signature is also
identical to that of [FeII(TPEN)]2+ (dashed red trace in Figure 2).
Additionally, the electronic absorption spectrum of 1 shows a
spectral feature independent on the solvent, with a MLCT at 420
nm and an extinction coefficient of 9700 M-1cm-1 in both MeOH
and MeCN (Figure 2, right). All these observations support the
conclusion that 1a and 1b retain their ligand bound in an
hexadentate mode as in [FeII(TPEN)]2+ upon dissolution in
MeCN or MeOH.
1b
Me
1a
N
N
Me
C2
N
N
N
N
Fe
N
Fe
N
N
N
N
N
Me
Me
Figure 4. 1H NMR spectrum of 1 in CD3CN at 233 K showing the resonances
of aliphatic protons of 1a (red) and 1b (blue). The resonances are labeled
according to Supplementary information, Scheme S2.
To substantiate this assertion, variable temperature 1H NMR of 1
and [FeII(TPEN)]2+ were performed between 233 and 333K.
At 233 K, 1 displays resonances spanning between 1 and 10
ppm (Supplementary Information, Figure S7). This observation
indicates that the two isomers 1a and 1b are diamagnetic at 233
K. However, one set of signals is sharp, whereas the other is
slightly broader and displays much more resonances. Due to its
C2 axis, 1a is expected to display much less resonances than
1b, for which all protons are differentiated (Figure 4 and
Supporting Information, Figures S8 and S9 with labeling scheme
on Scheme S2). Therefore, the sharp set of resonances can be
unambiguously ascribed to 1a while the broader and more
numerous peaks correspond to 1b (complete assignment for 1a,
1b and [FeII(TPEN)]2+ in Supporting Information, Figures S13-
S19 and S20-22, respectively). For example, the two methyl
groups are equivalent in 1a and appear as a doublet at 1.58
ppm, whereas they are non equivalent in 1b and two different
doublets at 1.58 and 1.70 ppm are observed (Figure 4). Along
Figure 3. 1H NMR spectrum in CD3CN at 293 K of 1 (black) and of 1a (red)
isolated from the crystalline material of 1 (red).
The parent complex [FeII(TPEN)]2+ is known to experience a
temperature-dependent spin state conversion in solution
between a low spin (S=0) and a high spin state (S=2).[27] This
spin crossover was shown to be connected to a very fast
fluxional trigonal distortion.[28] As determined by X-ray
crystallography and magnetic susceptibility, 1a•(PF6)2 and
especially 1b•(PF6)2 display a spin state conversion in the solid
state. Thus, variable temperature UV-visible absorption analysis
This article is protected by copyright. All rights reserved.

10.1002/chem.201902423
Chemistry - A European Journal
FULL PAPER
the same line, the H6 pyridine signals in 1a appear as two
doublets at 7.52 and 8.92 ppm whereas four resonances are
detected for 1b at 8.95, 8.83, ca. 7.8 and 7.7 ppm (Supporting
Information, Figures S8). Additionally, integration of the
spectrum of 1 at 233 K shows that 1 is constituted of an
equimolar mixture of 1a and 1b (Figure 4).
information, Figures S10-S12). This indicates a comparable
magnetic behavior in solution for 1b and [FeII(TPEN)]2+.
Consequently, at 273 K, the spectral range observed for 1b is
significantly larger than that of 1a (see dashed vertical lines in
Figure 5). For this latter complex, higher temperatures must be
reached to observe
a significant shift of its resonances
Increasing the temperature up to 333 K results in a progressive
shift of all of the resonances indicating the emergence of a spin
crossover phenomenon between low spin (S=0) and high spin
(S=2) FeII for both 1a and 1b (Supplementary Information,
Figures S10-S12). Nonetheless, while the resonances of 1a
remain essentially sharp (fwhm ca. 20-30 Hz) and moderately
(Supporting Information, Figure S11). Accordingly, the deviation
from Curie behaviour, which reflects the spin crossover
phenomenon, is observed at higher temperatures for 1a than for
1b and [FeII(TPEN)]2+ (Supporting Information, Figure S23).
These observations support that the low spin to high spin
conversion is more difficult for 1a and are in line with the
shifted, those related to 1b are significantly broadened (fwhm ca. observations in the solid state. As reported by Weber and
30-100 Hz) and experience larger displacements.
Walker,[29] the chemical shifts of protons are suitable tools for
following a spin transition in solution. Modeling the 1H NMR data
to equation 1 allows to extract the thermodynamic parameters of
the Boltzmann equilibrium between the high spin and low spin
10
CH
CH2
CH2
8
states, where
δ is the measured chemical shift, δls the
CH3
NCH2
NCH2
corresponding shift in the diamagnetic state, and
C the
6
4
appropriate Curie’s law constant.
!
! = !!" +
(1)
!!
!
!
!
!
![!!!"#
!
!
!
]
2
!
0
This method has proven useful to model the spin crossover
properties of complexes in solution.[30,31] The fits are shown as
solid lines in Figure 5 for 1a and 1b and the values extracted
from these fits are given in Table 1.
-2
240
260
280
T (K)
300
320
340
10
8
Hence, at room temperature, there is a larger fraction of the high
spin configuration for 1b and [FeII(TPEN)]2+ than for 1a. Indeed,
1a remains essentially low spin at 293 K whereas 1b and
[FeII(TPEN)]2+ exhibit a predominant high spin configuration
(Table 1).
NCH2
NCH2
CH2
CH or CH2
CH2
NCH2
NCH2
CH2
6
Table 1. Estimates of the thermodynamic parameters of the spin equilibria in
1a•(PF6)2 and 1b•(PF6)2 in MeCN solution and of the high spin fraction (γHS).
4
ΔH (kJ.mol-1)
ΔS (J.K-1.mol-1)
T1/2
γHS at 293 K
2
Complex
1a
40.1
36.8
39.9
111
134
142
361
274
281
4%
0
240
260
280
T (K)
300
320
340
1b
73%
67%
[FeII(TPEN)]2+
Figure 5. Temperature dependence of the chemical shifts in CD3CN for
several aliphatic protons of 1a (red) and 1b (blue) and [Fe(TPEN)](PF6)2
(green). The solid lines are fits to the data (see text for details). The vertical
dashed line compares the spectral range for the different complexes at 273 K.
The resonances are labeled as indicated in Supplementary information,
Scheme S2.
The catalytic efficiency of [FeII(TPEN)]2+ and related complexes
for the hydroxylation of aromatic substrates by H2O2 has been
previously reported.[10,11] Therefore, we have evaluated the
2
catalytic properties of [FeII(2MeL6 )]2+ 1 in the oxidation of anisole
by H2O2 and compared them to those of the reference complex
[FeII(TPEN)]2+ (Table 2). The oxidation of the phenol products is
much easier than the initial oxygenation of anisole leading to
mixtures that can be difficult to analyze.9 Thus, the catalytic
assays were performed using a large excess of substrate to
prevent over-oxidation of the phenol products.[10,11,32] Thereby,
restricting the catalytic reaction of the substrate to its 2-electrons
oxidation allows comparing the efficacy of different catalysts.
Figure 5 (top, red symbols) displays the chemical shift variation
for the aliphatic protons of 1a between 233 and 333 K. Only very
slight changes are observed up to 273
K from which
temperature the chemical shifts start to deviate. For 1b and
[FeII(TPEN)]2+, the severe line broadening and shift readily starts
from 233 K and increases quickly, preventing to track the
resonances beyond 270 K (Figure 5, bottom, blue and green
symbols for 1b and [FeII(TPEN)]2+, respectively, Supplementary
This article is protected by copyright. All rights reserved.

10.1002/chem.201902423
Chemistry - A European Journal
FULL PAPER
At 293 K, 1 (containing an equimolar mixture of 1a and 1b, see
Figure 4) leads to an overall ca. 50% conversion of the oxidant
into the different phenol products (Table 2). Interestingly, the
reaction also displays a good chemoselectivity for aromatic
hydroxylation (leading to methoxyphenols) vs oxidative
demethylation (yielding phenol). Additionally, reactivity occurs on
enriched in 1b (i.e. 1b/1a 9/1) which displays almost the same
activity as 1 (Table 2). This suggests that during the catalytic run
with sample 1 (1b/1a 1/1 mixture), 1b is responsible for the
nearly whole catalytic activity observed.
the electron enriched ortho- and para- positions with
a
Discussion
regioselectivity in favor of the ortho-methoxyphenol well above
the statistic value of 2. As seen in Table 2, these results are
almost identical to those observed with [FeII(TPEN)]2+. Oxidation
of aromatic substrates by Fenton reagents have been shown to
give rise to more diverse products mixtures.[33] Here, the use of
1a and 1b exhibit the same coordination sphere and display
thermodynamic properties (redox potential) close to those of
[FeII(TPEN)]2+ (Figure 2). Along this line, the reaction
intermediate responsible for the oxidation of anisole generated
from either 1a and 1b must have the same reactivity and,
therefore, should lead to a similar catalytic activity. However,
while the catalytic behaviour of 1b is in line with that of
[FeII(TPEN)]2+, 1a is much less active (Table 2). This distinct
behaviour between 1a and 1b can thus only be rationalized by
kinetic arguments. As determined by NMR, 1b displays a
significant HS fraction at room temperature, and is thus labile,
whereas 1a is mainly diamagnetic and much more inert. Hence,
it can be proposed that 1b reacts faster than 1a with H2O2 to
form the reaction intermediates involved in the hydroxylation of
anisole. The 3 steps to yield the reactive species are :[36]
Fe(OTf)2 instead of
a complex under otherwise identical
conditions did not lead to the formation of methoxyphenols.
Such an ability to promote the oxidation of aromatic molecules
with these significant regio- and chemoselectivities have been
demonstrated to be mechanistic signatures of caged [FeIVO OH•]
active
species
generated
from
LS
(N5)FeIII(OOH)
intermediates.[9,34,35] Thus, this conclusion most probably applies
to 1 as well. Note that EPR detection of LS FeIII(OOH) generated
from 1 also supports this hypothesis (Supplementary Information,
Figure S24).
-
(i)
FeII + H2O2 D FeIII(OH) + HO•
Table 2. Oxidation of anisole by H2O2 catalyzed by [FeII(TPEN)]2+, 1, 1a, and
1b. Catalytic activity reported as yields % vs H2O2 and TON (TON).[a]
(ii) FeIII(OH) + H2O2 D FeIII(OOH) + H2O
(iii) FeIII(OOH) D FeIVO + HO•
O
OH
O
O
O
OH
cat + 20 H2O2
+
OH
+
+
ca. 95% LS at RT
1a
MeCN, 20°C
2 h
inert
N
Me
N
3000
OH
H
H2O
N
meta-
ortho-
para-
phenol
FeII
N
H2O2
H2O2
H
N
N
OH
[FeII(TPEN)]2+
37% (7.4)
1[b]
1a[c]
1b[d]
Me
(ii)
slow
(i)
(N5)FeIII-OOH
(iii)
34%
(7.6)
ortho-
35% (7.0)
13% (2.6)
(N5)FeIII-OH
OH
1b
Me
H2O2
fas
N
para-
meta-
phenol
total
5% (1.0)
1% (0.2)
6% (1.2)
49% (9.6)
7% (1.4)
1% (0.2)
8% (1.6)
2% (0.4)
-
3% (0.7)
-
H
t
OH
[(N5)FeIVO OH]
N
N
FeII
H
N
N
OMe
4% (0.8)
19% (3.9)
5% (1.1)
Me
N
OMe
ca. 70% HS at RT
labile
42%
(9.3)
51% (10.2)
regioselectivity[e]
chemoselectivity[f]
7.4
7.2
5
6.5
3.7
11
Scheme 3. Proposed mechanism to account for the different catalytic activity
of 1a and 1b based on their different lability (the catalytic cycle is adapted from
ref. 9).
5.4
7.4
[a] Yields and TON determined by GC after extraction of the reaction mixture
in Et2O. Values are the average of at least three reactions. [b] Microcrystalline
powder of 1•(PF6)2 containing 1a•(PF6)2 and 1b•(PF6)2 (equimolar mixture).
[c] Single crystals of 1a•(PF6)2 manually sorted. [d] 1b•(PF6)2/1a•(PF6)2 9/1.
TONs corrected for the 90% purity in 1b•(PF6)2. [e] ortho-/para-. [f] (ortho- +
para- + meta-)/phenol.
With
most
non-heme
FeII
complexes,[36-39]
including
[FeII(TPEN)]2+, the rate determining step (rds) is step (iii), and
the faster step (i), allowing the accumulation of the FeIII(OOH)
intermediate. With 1, the FeIII(OOH) intermediates can be
observed by EPR (Supporting Information, Figure S24) but they
do not accumulate enough to be observed by UV-visible
suggesting step (iii) is most probably not the rds for 1.
Nonetheless, the rate constant for step (i) can be extracted by
UV-visible monitoring of the oxidation of 1 by H2O2. Interestingly,
the decay of its 420 nm MLCT is constituted of two first order
Interestingly, catalytic assays using a batch of 1a complex
isolated from the crystalline material of 1 lead to similar regio-
and chemoselectivities to those observed with 1 but with much
lower conversions suggesting that 1b is much more reactive
than 1a. This is confirmed by catalytic runs using a sample
This article is protected by copyright. All rights reserved.

10.1002/chem.201902423
Chemistry - A European Journal
FULL PAPER
kinetic regimes (Supplementary Information, Figure S25). One of Conclusions
these phases is fast and attributed to 1b, whereas the other is
2
much slower and attributed to the reaction of 1a with H2O2.
Ferrous complexes prepared from the hexadentate 2MeL6 ligand
The mechanism of aromatic hydroxylation
by nonheme
exist as two diastereomeric complexes, 1a and 1b, structurally
characterized at the solid state as their PF6 salts. In the solid
FeII/H2O2 systems has been previously elucidated[9,34,35] and
serves as basis to account for the different reactivity of 1a and
1b (Scheme 3). Both 1a and 1b react with H2O2 following two
parallel inner sphere oxidation to form FeIII(OH) species (step (i))
but at very different rates (Figure S25). Being more reactive by
more than one order of magnitude, 1b enters much more rapidly
in the catalytic cycle to form the FeIII(OOH) intermediate and the
resulting [FeIVO •OH]. Whatever the 1a/1b ratio, 1/1 or 1/9,
present in the mixture, complex 1b is responsible for most of the
H2O2 consumption and oxidation of the substrate. Along this line,
the same conversions must be obtained starting from either of
these samples, in agreement with the experimental results
reported in Table 2.
-
state, these complexes exhibit very similar bond distances and
angles but differ in the axial or equatorial coordination positions
occupied by the 2-pyridinylethyl or 2-pyridinylmethyl groups.
Such apparently subtle changes lead to a different steric
congestion in the two complexes, which is probably a source for
their distinct magnetic behaviour in solution. Indeed, 1a is
essentially diamagnetic (low spin FeII) from low to room
temperature whereas 1b undergoes a conversion between
diamagnetic and paramagnetic states (high spin FeII S=2).
Furthermore, these slight structural changes also affect
reactivity: 1b is a better catalyst in the oxidation of anisole by
H2O2 than 1a. This can be assigned to the increased lability of
1b. Thus, 1b reacts fastly with H2O2 whereas 1a is sluggish. The
relationships between structural properties, spin states and
reactivity of coordination complexes is often evoked but there
are rather few reported experimental data to support the claims.
Our present study provides clear experimental evidences in that
direction.
Future work in our lab is aimed at incorporating two CH3 groups
at the picolyl positions instead of one to create a steric
congestion whose relaxation would favor the stabilization of the
high spin state, which is beneficial for an efficient catalytic
activity.
2.01 Å
Experimental Section
Synthesis of [Fe(2MeL62)](PF6)2 1•(PF6)2. In a glovebox, Fe(OTf)2 (1.70 g,
4.8 mmol) dissolved in a minimum volume of MeOH was added dropwise
to the ligand 2MeL62 (2.16 g, 4.8 mmol) previously dissolved in a minimum
volume of MeOH. The resulting orange-brown solution was allowed to stir
for 30 minutes, then 2.2 equiv. NaPF6 (1.59 g, 10.7 mmol) dissolved in a
minimum of MeOH was added dropwise. A red-brown solid precipitated
and the reaction mixture was allowed to stir for one night. The solid was
collected by filtration, washed with Et2O and dried under vacuum to yield
[Fe(2MeL62)](PF6)2 1•(PF6)2 as an orange-brown powder with a yield of
68 %. Elemental Analysis (%) found C 41,5, H 4.0, N 10.2; expected C
42.1, H 4,0, N 10.5. This indicates that the microcrystalline powder is
solvent free, in agreement with the crystal structures of 1a•(PF6)2 and
1b•(PF6)2. HR-ESI-MS (CH3CN): m/z 254.1036 (calcd. 254.1013 for
[Fe(2MeL62)]2+). UV-visible : (MeOH) λmax (nm) [ε (M-1.cm-1)]: 420 (8600),
250 (>25 000); (CH3CN) λmax (nm) [ε (M-1.cm-1)]: 420 (9000), 250 (14000).
Single crystals were obtained by slow diffusion of tert-butyl ethylether in a
CH3CN solution of the complex. 1a•(PF6)2 and 1b•(PF6)2 always
crystallized simultaneously as dark red brown crystals which complicated
their isolation. In some instances, larger crystals of both isomers
crystallized and could be manually sorted, taking advantadge of their
different shape. 1a•(PF6)2 appear as plates whereas 1b•(PF6)2 are
needle-shaped crystals (Supporting Information, Figure S5). As shown by
NMR (see Figure 4 in the main text and NMR section in the Supporting
Information), any crop of solid 1•(PF6)2 contained an equimolar mixture
of isomers 1a•(PF6)2 and 1b•(PF6)2.
Figure 6. Stick diagram and space filling view of 1b showing the close
contacts present between H atoms.
Looking into more details into the molecular structures of 1a and
1b may provide insights into this reactivity difference. In 1b, the
distance between the H atom of the axial pyridinylethyl and the
H
atom of a neighbouring CH2 group of the equatorial
ethylenediamine fragment is extremely short (Figure 6 and
Figure S26). Such very short contacts, way beyond the sum of
Van der Waals radii (2.4 Å)[40] are absent from the structure of
1a (Supplementary Information, Figure S26). They suggest a
significant steric congestion in 1b, unlike 1a.
Relaxation of the steric clash in 1b can be achieved by
lengthening of the Fe-ligand bonds, thus weakening the ligand
field and favoring the low spin to high spin state conversion, in
concordance with our observations.
Conversely, 1a does not display such a steric congestion in the
solid state and, compared to 1b, does not exhibit a large
paramagnetic contribution at room temperature.
The relaxation of steric congestion thus appears as a possible
driving force for the low spin to high spin state conversion and,
hence, for the faster formation of FeIII(OOH) (and the
subsequent anisole oxidation) with 1b than with 1a.
Catalytic assays were carried out with the following samples
composition : catalyst (1mM)/H2O2/anisole 1/20/3000 in MeCN for 2h at
This article is protected by copyright. All rights reserved.

10.1002/chem.201902423
Chemistry - A European Journal
FULL PAPER
[18]
[19]
[20]
M. Costas, A. K. Tipton, K. Chen, D.-H. Jo, L. Que, J. Am. Chem.
Soc. 2001, 123, 6722–6723.
P. E. Gormisky, M. C. White, J. Am. Chem. Soc. 2013, 135,
14052–14055.
D. Font, M. Canta, M. Milan, O. Cussó, X. Ribas, R. J. M. Klein
Gebbink, M. Costas, Angew. Chem. Intl. Ed. 2016, 128, 5870–
5873.
20°C. The products were analyzed and quantified by GC with respect to
commercial standards after extraction of the reaction mixture in Et2O.
The results given are the mean of 3 experiments. Blank experiments with
FeII(OTf)2 instead of the catalyst did not yield any oxidation product.
[21]
[22]
[23]
A. J. Simaan, S. Döpner, F. Banse, S. Bourcier, G. Bouchoux, A.
Boussac, P. Hildebrandt, J.-J. Girerd, Eur. J. Inorg. Chem. 2000,
2000, 1627–1633.
M. Martinho, F. Banse, J.-F. Bartoli, T. A. Mattioli, P. Battioni, O.
Horner, S. Bourcier, J.-J. Girerd, Inorg. Chem. 2005, 44, 9592–
9596.
N. Ségaud, E. Anxolabéhère-Mallart, K. Sénéchal-David, L.
Acosta-Rueda, M. Robert, F. Banse, Chem. Sci. 2015, 6, 639–
647.
M. Wu, Bin Wang, S. Wang, C. Xia, W. Sun, Org. Lett. 2009, 11,
3622–3625.
M. Wu, C.-X. Miao, S. Wang, X. Hu, C. Xia, F. E. Kühn, W. Sun,
Adv. Synth. Catal. 2011, 353, 3014–3022.
M. Grau, A. Kyriacou, F. C. Martinez, I. M. de Wispelaere, A. J. P.
White, G. J. P. Britovsek, Dalton Trans. 2014, 43, 17108–17119.
H. Toftlund, Monatshefte Fur Chemie 2001, 132, 1269–1277.
H. R. Chang, J. K. McCusker, H. Toftlund, S. R. Wilson, A. X.
Trautwein, H. Winkler, D. N. Hendrickson, J. Am. Chem. Soc.
1990, 112, 6814–6827.
Acknowledgements
Eric Rivière and Jean-Pierre Baltaze are acknowledged for their
precious help. This work is supported by a public grant overseen
by the French National Research Agency (ANR) as part of the
“Investissements d’Avenir” program (Labex charmmmat,
reference: ANR-11-LABX-0039-grant). The ECOSTBio Action
CM1305 is also acknowledged for support
[24]
[25]
[26]
Keywords: iron • N ligands • spin states • NMR • catalysis
[27]
[28]
[1]
I. Gamba, Z. Codolà, J. Lloret-Fillol, M. Costas, Coord. Chem.
Rev. 2016, 1–23.
[29]
[30]
[31]
B. Weber, F. A. Walker, Inorg. Chem. 2007, 46, 6794–6803.
S. E. Creutz, J. C. Peters, Inorg. Chem. 2016, 55, 3894–3906.
H.-J. Lin, D. Siretanu, D. A. Dickie, D. Subedi, J. J. Scepaniak, D.
Mitcov, R. Clérac, J. M. Smith, J. Am. Chem. Soc. 2014, 136,
13326–13332.
[2]
[3]
P. R. Ortiz de Montellano, Chem. Rev. 2010, 110, 932–948.
C. Krebs, D. Galonić Fujimori, C. T. Walsh, J. M. Bollinger, Acc.
Chem. Res. 2007, 40, 484–492.
S. M. Barry, G. L. Challis, ACS Catal. 2013, 3, 2362–2370.
W. Nam, Y.-M. Lee, S. Fukuzumi, Acc. Chem. Res. 2014, 47,
1146–1154.
P. C. A. Bruijnincx, G. van Koten, R. J. M. Klein Gebbink, Chem.
Soc. Rev. 2008, 37, 2716.
X. Engelmann, I. Monte-Pérez, K. Ray, Angew. Chem. Int. Ed.
2016, 55, 7632–7649.
[4]
[5]
[32]
[33]
[34]
[35]
[36]
D. Mathieu, J.-F. Bartoli, P. Battioni, D. Mansuy, Tetrahedron
2004, 60, 3855–3862.
[6]
C. Walling, R. A. Johnson, J. Am. Chem. Soc. 1975, 97, 363–
367.
A. S. Faponle, M. G. Quesne, C. V. Sastri, F. Banse, S. P. de
Visser, Chem. Eur. J. 2015, 21, 1221–1236.
A. S. Faponle, F. Banse, S. P. de Visser, J Biol Inorg Chem 2016,
21, 453–462.
J. Chen, A. Draksharapu, D. Angelone, D. Unjaroen, S. K.
Padamati, R. Hage, M. Swart, C. Duboc, W. R. Browne, ACS
Catal. 2018, 8, 9665–9674.
G. Roelfes, M. Lubben, K. Chen, R. Y. N. Ho, A. Meetsma, S.
Genseberger, R. M. Hermant, R. Hage, S. K. Mandal, V. G.
Young, et al., Inorg. Chem. 1999, 38, 1929–1936.
A. J. Simaan, F. Banse, P. Mialane, A. Boussac, S. Un, T.
Kargar-Grisel, G. Bouchoux, J. J. Girerd, Eur. J. Inorg. Chem.
1999, 993–996.
[7]
[8]
G. Olivo, O. Cussó, M. Borrell, M. Costas, J Biol Inorg Chem
2017, 22, 425–452.
[9]
A. Thibon, V. Jollet, C. Ribal, K. Sénéchal-David, L. Billon, A. B.
Sorokin, F. Banse, Chem. Eur. J. 2012, 18, 2715–2724.
A. Thibon, J.-F. Bartoli, R. Guillot, J. Sainton, M. Martinho, D.
Mansuy, F. Banse, J. Mol. Catal. A 2008, 287, 115–120.
V. Balland, D. Mathieu, N. Pons-Y-Moll, J.-F. Bartoli, F. Banse, P.
Battioni, J.-J. Girerd, D. Mansuy, J. Mol. Catal. A 2004, 215, 81–
87.
O. V. Makhlynets, E. V. Rybak-Akimova, Chem. Eur. J. 2010, 16,
13995–14006.
M. S. Chen, M. C. White, Science 2007, 318, 783–787.
P. Spannring, V. Yazerski, P. C. A. Bruijnincx, B. M.
Weckhuysen, R. J. M. Klein Gebbink, Chem. Eur. J. 2013, 19,
15012–15018.
O. Cussó, I. Garcia-Bosch, X. Ribas, J. Lloret-Fillol, M. Costas, J.
Am. Chem. Soc. 2013, 135, 14871–14878.
M. Mitra, O. Cussó, S. S. Bhat, M. Sun, M. Cianfanelli, M. Costas,
E. Nordlander, Dalton Trans. 2019, 48, 6123–6131.
E. A. Mikhalyova, O. V. Makhlynets, T. D. Palluccio, A. S. Filatov,
E. V. Rybak-Akimova, Chem. Commun. 2012, 48, 687–689.
[10]
[11]
[37]
[38]
[39]
[40]
[12]
[13]
[14]
V. Balland, F. Banse, E. Anxolabéhère-Mallart, M. Ghiladi, T. A.
Mattioli, C. Philouze, G. Blondin, J.-J. Girerd, Inorg. Chem. 2003,
42, 2470–2477.
A. Bondi, J. Phys. Chem. 1964, 68, 441–451.
[15]
[16]
[17]
This article is protected by copyright. All rights reserved.

10.1002/chem.201902423
Chemistry - A European Journal
FULL PAPER
Entry for the Table of Contents (Please choose one layout)
Layout 1:
FULL PAPER
Two new diastereoisomeric FeII
complexes exhibit different magnetic
behavior in solution. Whereas one
isomer displays a temperature
dependent spin state conversion
between the S=0 and S=2 states,
Katell Sénéchal-David,* Charlotte
Buron, Nathalie Ségaud, Jean-Noël
Rebilly, Amandine Dos Santos,
Jonathan Farjon, Régis Guillot,
Christian Herrero, Tanya Inceoglu,
and Frédéric Banse*
the other one is essentially
RS meso ligand
70% Fe(II) S=2 at RT
RR/SS ligand
4% Fe(II) S=2 at RT
Page No. – Page No.
diamagnetic at room temperature.
Their catalytic properties for anisole
oxidation by H2O2 are very different
and correlate with their magnetic
properties which reflect their lability /
inertness.
Non-heme Fe(II) diastereomeric
complexes bearing a hexadentate
ligand : unexpected consequences
on the spin state and oxidation
catalytic properties.
high catalytic activity
poor catalytic activity
This article is protected by copyright. All rights reserved.