Facial triad modelling using ferrous pyridinyl prolinate complexes: Synthesis and catalytic applications

Article abstract of DOI:10.1039/c3dt53266f

A series of new chiral pyridinyl prolinate (RPyProR) ligands and their corresponding Fe(ii) triflate and chloride complexes are reported. The ligands possess an NN′O coordination motif, as found in the active site of non-heme iron enzymes with the so-called 2-His-1-carboxylate facial triad. The coordination behaviour of these ligands towards iron turned out to be dependent on the counter ion (chloride or triflate), the crystallization conditions (coordinating or non-coordinating solvents) and the presence of substituents on the ligand. In combination with Fe(ii)(OTf)₂, coordinatively saturated complexes of the type [Fe(L)₂](OTf)₂ are formed, in which the ligands adopt a meridional coordination mode. The use of FeCl ₂ in a non-coordinating solvent leads to 5-coordinated complexes [Fe(L)(Cl)₂] with a meridional N,N′,O ligand. Crystallization of these complexes from a coordinating solvent leads to 6-coordinated [Fe(L)(solv)(Cl)₂] complexes (solv = methanol or acetonitrile), in which the N,N′,O ligand is coordinated in a facial manner. For RPyProR ligands bearing a 6-Me substituent on the pyridine ring, solvent coordination and, accordingly, ligand rearrangement are prevented by steric constraints. The complexes were tested as oxidation catalysts in the epoxidation of alkene substrates in acetonitrile with hydrogen peroxide as the oxidant under oxidant limiting conditions. The complexes were shown to be especially active in the epoxidation of styrene type substrates (styrene and trans-beta-methylstyrene). In the best case, complex [Fe(6-Me-PyProNH₂)Cl₂] (15) allowed for 65% productive consumption of hydrogen peroxide toward epoxide and benzaldehyde products. This journal is the Partner Organisations 2014.

Full text of DOI:10.1039/c3dt53266f

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Facial triad modelling using ferrous pyridinyl
prolinate complexes: synthesis and catalytic
applications†
Cite this: Dalton Trans., 2014, 43,
6769
Marcel A. H. Moelands,a Daniel J. Schamhart,a Emma Folkertsma,a Martin Lutz,b
Anthony L. Spekb and Robertus J. M. Klein Gebbink*a
A series of new chiral pyridinyl prolinate (RPyProR) ligands and their corresponding Fe(II) triflate and chlor-
ide complexes are reported. The ligands possess an NN’O coordination motif, as found in the active site
of non-heme iron enzymes with the so-called 2-His-1-carboxylate facial triad. The coordination behav-
iour of these ligands towards iron turned out to be dependent on the counter ion (chloride or triflate), the
crystallization conditions (coordinating or non-coordinating solvents) and the presence of substituents on
the ligand. In combination with Fe(II)(OTf)2, coordinatively saturated complexes of the type [Fe(L)2](OTf)2
are formed, in which the ligands adopt a meridional coordination mode. The use of FeCl2 in a non-coor-
dinating solvent leads to 5-coordinated complexes [Fe(L)(Cl)2] with a meridional N,N’,O ligand. Crystalliza-
tion of these complexes from a coordinating solvent leads to 6-coordinated [Fe(L)(solv)(Cl)2] complexes
(solv = methanol or acetonitrile), in which the N,N’,O ligand is coordinated in a facial manner. For
RPyProR ligands bearing a 6-Me substituent on the pyridine ring, solvent coordination and, accordingly,
ligand rearrangement are prevented by steric constraints. The complexes were tested as oxidation cata-
lysts in the epoxidation of alkene substrates in acetonitrile with hydrogen peroxide as the oxidant under
oxidant limiting conditions. The complexes were shown to be especially active in the epoxidation of
styrene type substrates (styrene and trans-beta-methylstyrene). In the best case, complex [Fe(6-Me-
PyProNH2)Cl2] (15) allowed for 65% productive consumption of hydrogen peroxide toward epoxide and
benzaldehyde products.
Received 19th November 2013,
Accepted 16th February 2014
DOI: 10.1039/c3dt53266f
www.rsc.org/dalton
Amongst these enzymes, the binuclear enzyme sMMO has
attracted a lot of attention over the years because of its prop-
Introduction
Non-heme iron enzymes are involved in a large number of oxi- erty to selectively oxidize methane to methanol.3 More
dative processes in biology, and their modes of action consti- recently, a class of mono-nuclear non-heme iron enzymes that
tute a source of inspiration for the design of synthetic possess a so-called 2-His-1-carboxylate facial triad active site
oxidation catalysts.1 In particular, the oxidation of non-acti- have gained widespread interest (Fig. 1).4,5 In the active site of
vated C–H and CvC bonds using dioxygen as the formal these enzymes, the iron centre is surrounded in a facial
oxidant is attractive to synthetic chemists. Efforts towards manner by two histidine ligands and one carboxylate ligand
modelling of the active sites of such enzymes have, therefore, from either glutamate or aspartate. Water molecules occupy
not only focused on a further understanding of the enzyme the other three sites of the octahedron around iron in the
activity and on studying (electronic) structure–activity relation- resting state. These three ‘vacant’ sites are important for the
ships, but have also focused on the design of bio-inspired oxi- overall reactivity of the enzymes, because they provide binding
dation catalysts.2
aOrganic Chemistry & Catalysis, Department of Chemistry, Faculty of Science,
Utrecht University, Universiteitsweg 99, 3584 CG Utrecht, The Netherlands.
E-mail: r.j.m.kleingebbink@uu.nl
bBijvoet Center for Biomolecular Research, Crystal and Structural Chemistry,
Department of Chemistry, Faculty of Science, Utrecht University, Padualaan 8,
3584 CH Utrecht, The Netherlands
†Electronic supplementary information (ESI) available: Details of crystallo-
graphic studies. CCDC 972756–972763. For ESI and crystallographic data in CIF
or other electronic format see DOI: 10.1039/c3dt53266f
Fig. 1 The 2-His-1-carboxylate facial triad.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 6769–6785 | 6769

View Article Online
Paper
Dalton Transactions
sites for substrate or co-factor molecules that are required for
the oxidative transformations carried out by the enzymes.
Most striking, besides the structural aspects of the active
sites, is the wide variety in the reaction patterns that are cata-
lysed by the members of the facial triad enzyme family. This
reactivity pattern includes, e.g., oxidative ring cleavage,
cis-dihydroxylation, hydroxylation, desaturation, ring closure,
and ring expansion. For a single structural entity this palette
of reactivity is unprecedented in biology and rivals the reactiv-
ity diversity displayed by heme enzymes. Not surprisingly, the
facial triad enjoys much attention in the bioinorganic field.
Different approaches have been followed to mimic the
structural and reactivity features of the facial triad and to com-
prehend its overall reactivity. Que et al. reported on the prepa-
ration of a model of the α-ketoglutarate-dependent oxygenases,
based on the hydrotris(3,5-diphenylpyrazol-1-yl)borate(TpPh2
)
Fig. 2 Mixed N,O ligands.
ligand that coordinates to iron(II) in a facial manner via three
nitrogen donor atoms.6,7 Several other all-nitrogen ligands
have been used as well to model the facial triad. Examples
include tris(2-pyridylmethyl)amine (tpa),8 N,N-bis-(2-pyridyl-
methyl)-N,N′-dimethyl-1,2-cyclohexanediamine (bpmcn),9 and
1,4,7-triazacyclononane (tacn).10 Attempts to mimic the struc-
tural features of the facial triad coordination environment in a
more accurate manner, i.e. through the facial coordination of
a combination of two nitrogen-based neutral donors and a
mono- or bidentate carboxylate donor, were more recently
described. Burzlaff and co-workers reported the coordination
chemistry of bispyrazolylacetate heteroscorpionate ligands
towards iron.11,12 Que and coworkers reported the use of Ph-
DPAH (di(2-pyridyl)methylbenzamide) which provides a neutral
facial NNO coordination motif.13 Interestingly, the complex
[FeII(Ph-DAPH)2](OTf)2 is able to carry out the cis-dihydroxyla-
tion of alkenes using H2O2 as the oxidant and thereby mimics
the reactivity of the Rieske dioxygenases. Furthermore, Que and
Tolman have reported an alternative strategy to prepare mono-
nuclear iron(II) complexes with an NNO coordination mode, in
which sterically encumbered carboxylato ligands in combi-
nation with bulky diamine ligands are used.14
Our interest in this field is the development of mixed N,O
ligands and their use in catalytic oxidation chemistry. In an
attempt to closely mimic the electronic properties next to the
structural properties of the facial triad, we recently developed
the bis(alkyl-imidazol-2-yl)propionate (BAIP) ligand family
(Fig. 2A). This ligand family combines a carboxylate donor
moiety with two biologically relevant imidazole donor moieties
in a predisposed facial manner and enables the synthesis of
discrete [M(BAIP)2] and [M(BAIP)(X)2L] complexes, where M =
Fe, Cu, Zn, (X)2 are two anionic ligands or is a bidentate
mono- or dianionic ligand, and L is a neutral donor ligand like
H2O or pyridine.15 Substitution of the imidazole moieties in
the BAIP-ligands allows for the variation of their steric and site
isolating, as well as their solubility properties.15d Complexes of
the type [Fe(BAIP)(cat)(H2O)], where (cat) is a mono-anionic
bidentate (substituted) catechol ligand, were found to react
with oxygen to accomplish cleavage of the catechol ring to
afford extradiol cleavage products in excess of intradiol
cleavage products (2 : 1 ratio, respectively) in 60% yield.16
These complexes represent rather accurate structural and func-
tional models for extra-diol cleaving dioxygenase enzymes.
Upon substituting the carboxylate moiety in the BAIP ligands
for a neutral carboxylic ester moiety (Fig. 2B), the N,N,O facial
capping propensity of the ligands is maintained and com-
plexes of the type [Fe(L)2]2+ are obtained.15c Complexes of this
type were found to be able to catalyze the combined cis-
dihydroxylation and epoxidation of internal and external
olefins. Burzlaff and co-workers have reported independently
on the synthesis and metal coordination properties of
members of the BAIP-ligand family,17,18 while Holland et al.
reported on the structure of [Fe(BAIP)2] complexes.19
In an effort to extend the set of ligands that are able to co-
ordinate to iron in a facial N,N,O manner, we have set out to
investigate the coordination chemistry of a set of chiral ligands
that would in principle allow for the development of enantio-
selective bioinspired oxidation catalysts. The investigated set
of ligands is composed of a (substituted) pyridine moiety con-
nected to a (substituted) prolinate moiety (Fig. 3). Ligands of
this type were earlier reported by Chelucci and co-workers, and
have recently been investigated by Paine et al. in combination
with iron.20 Rutledge and co-workers have also investigated
these and related ligands in the iron-mediated oxidation of
alkenes (in particular cyclohexene) with H2O2.21 These pyridi-
nyl prolinate ligands RPyProR are quite easily synthesized
using proline and its derivatives as cheap chiral pool synthons.
Earlier investigations by us have dealt with analogues of these
ligands in which a central pyridine was combined with two
proline-derived moieties (Fig. 2C, Py(ProMe)2). Those ligands
behave as tridentate meridional or planar pentadentate NN′
NOO ligands and render diastereopure seven-coordinated
metal(II) complexes.22
Here, we report on the synthesis and iron(II) coordination
properties of ligands 1–8, the structural analysis of the result-
ing complexes, and report on an initial evaluation of the
ability of the resulting iron complexes to catalyse the epoxi-
dation of a series of (pro-chiral) olefin substrates.
6770 | Dalton Trans., 2014, 43, 6769–6785
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Fig. 3 RPyProR ligands described in this study.
complex or as a bidentate ligand in combination with solvent
or triflate coordination (Fig. 4).
Results
Ligand synthesis
Accordingly, the 2 : 1 ligand : Fe complexes of ligands 3, 5,
The synthesis of the parent ligand PyProMe (1) was earlier and 7 with Fe(OTf)2·2MeCN were prepared on a preparative
reported by Chelucci et al.23 Following this synthesis route, scale. To this end one equivalent of Fe(OTf)2·2MeCN was
ligands 6-Me-PyProMe (2), PyProPr (3), and 6-Me-PyProPr (4) reacted with two equivalents of the ligand in acetonitrile
were synthesized using the respective prolinate esters and under an inert N2-atmosphere for 1 h. Isolation via precipi-
(substituted) 2-chloromethylpyridine reagents (Scheme 1). tation with Et2O and recrystallization from MeCN–Et2O yielded
Chelucci et al. also reported on the reaction of 1 with either complexes 9–11 as off-white (9 and 11) or yellow (10) powders
NH3 or PhMgBr to afford ligands PyProNH2 (5) and PyPro- in 60–90% yield (Scheme 2). These ferrous complexes are air
Ph2OH (7), respectively. Amide ligand 6-Me-PyProNH2 (6) and sensitive and were stored in an inert atmosphere drybox.
prolinol ligand 6-Me-PyProPh2OH (8) were synthesized in an
ESI-MS analysis of these isolated materials showed mono-
analogous manner starting from the 2-methylpyridinyl ligand cations of the composition [Fe(L)2(OTf)]+ as the parent peak.
6-Me-PyProMe (2) (Scheme 1).
Elemental analysis corroborated the anticipated Fe(L)2(OTf)2
composition of the isolated materials. UV-Vis spectra recorded
in MeCN showed typical ligand π–π* and MLCT bands for octa-
hedral Fe(II) complexes below 300 nm and in the 330–360 nm
region, respectively. The solid state IR-spectrum of complex 9
showed a distinctive vibration band at 1658 cm−1 (Fig. 5), indi-
cating that the carbonyl fragments in the complex are co-
ordinated to the metal ion. For the free ligand (3), the carbonyl
band is found at 1728 cm−1. The symmetric and asymmetric
Fe-triflate complexes
We initially studied the complexation behaviour of ligands 1–8
towards Fe(II) triflate, thereby regarding the triflates as non-
coordinating anions.
A
Job plot for ligand
1
and
Fe(OTf)2·2MeCN in acetonitrile shows the preferential for-
mation of a 2 : 1 ligand : metal complex, suggesting that 1 acts
either as a tridentate ligand to form a hexacoordinated Fe-
Scheme 1 Synthesis of ligands 1–8.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 6769–6785 | 6771

View Article Online
Paper
Dalton Transactions
carbonyl band at 1673 cm−1 also indicates coordination of this
group to iron. In the free ligand (5), this band is also found at
a relatively low energy (1669 cm−1), which is most likely caused
by intermolecular hydrogen bonding between carbonyl and
amino groups. The position of the triflate bands also indicates
that these ions are not coordinated to iron in complex 10;
vibration bands for the triflate groups are found at 1270
(νas SO3), 1227 (νs CF3), 1158 (νas CF3) and 1031 (νs SO3) cm−1
.
For complex 11, the situation is slightly different as the
complex does not contain carbonyl groupings, but instead con-
tains tertiary alcohol groupings. In the IR-spectra of complex
11 a broad peak is observed for the OH group around
3339 cm−1. The peak of the OH group in the free ligand is
found at a higher wavenumber (3395 cm−1). The ROH groups
in this complex most likely coordinate to iron, whereas they
are not deprotonated according to the parent peak observed in
ESI-MS ([Fe(PyProPh2OH)2(OTf)]+, m/z = 893.269). The triflate
groups in this complex are just as for complexes 9 and 10 not
coordinated towards the iron centre; triflate vibrations are
found at 1277 (νas SO3), 1223 (νs CF3), 1157 (νas CF3) and 1027
(νs SO3) cm−1 for complex 11.
Magnetic susceptibility measurements in solution using
Evans’ method25,26 indicated µeff values of 4.86 (9, acetone),
4.46 (10, MeCN) and 4.58 (11, MeCN) µβ for these complexes.
While these values are somewhat below the theoretical value
of 4.92 for an S = 2 system, these do indicate a high spin con-
figuration for these iron(II) triflate complexes.
Fig. 4 Job plot between Fe(OTf)2·2MeCN and ligand 1; the black dots
represent the data points.
Twinned crystals suitable for X-ray diffraction were obtained
for complex [Fe(PyProNH2)2](OTf)2 (10) by slow vapour
diffusion of diethyl ether into an acetonitrile solution of the
complex. Crystal structure determination showed a very distorted
octahedral geometry for the iron complex (Fig. 6); selected bond
lengths and angles are depicted in Table S1.† Two PyProNH2
ligands coordinate in a meridional fashion around the iron
centre, corresponding to the ligand to metal ratio found in the
Scheme 2 Synthesis of the triflate complexes 9–11.
Fig. 5 Solid state IR spectrum of complex 9.
vibrations of the CF3 and SO3 groups in 9 are found at 1255
(νas SO3), 1223 (νs CF3), 1148 (νas CF3) and 1028 (νs SO3) cm−1
,
which points at the presence of non-coordinating triflate
groups.24
Similar observations were made in the solution IR-spectrum
(MeCN) of complex 9. This indicates that no structural
changes related to both carbonyl or triflate coordination take
place and that the overall solid-state structure of 9 is main-
tained in solution. For solid complex 10, the position of the
Fig. 6 Displacement ellipsoid plot (50% probability) of 10; C–H hydro-
gen atoms and non-coordinated triflate anions are omitted for clarity.
6772 | Dalton Trans., 2014, 43, 6769–6785
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Job plot (Fig. 4). From each ligand, the nitrogen atoms of the triflate complexes due to the weak intensity of the Fe-Cl
pyridine and the proline ring coordinate to iron, and the triden- vibrational peaks. The solid-state IR spectrum of [Fe(PyProMe)-
tate nature of the ligand is complemented by coordination of Cl2] (12) does show a clear vibration band of the carbonyl
the carbonyl moiety. The triflate moieties are found as dis- group at 1697 cm−1. The corresponding vibration of the free
ordered, non-coordinating counter ions. The crystal structure ligand is found at 1732 cm−1. This shift in energy clearly indi-
showed furthermore that complex 10 is involved in a hydrogen- cates the coordination of the carbonyl group to the iron centre.
bonded one-dimensional chain (ESI Fig. S1 and Table S2†).
The solution IR spectrum of [Fe(PyProMe)Cl2] in MeCN shows
The meridional coordination mode of the tridentate the carbonyl vibration at 1702 cm−1, which indicates that in
ligands is most likely caused by steric reasons and gives rise to solution the carbonyl group remains coordinated to iron.
a trans orientation of the proline rings; the two Fe–O bonds are
For the 6-methylpyridine complex [Fe(6-Me-PyProMe)Cl2]
oriented cis to each other. The Fe–N distances vary between (13), the same trend in the vibration of the carbonyl group was
2.132(5) and 2.197(5) Å, which are typical bond distances for observed for the free ligand, the prepared complex and the
high spin iron(II) complexes. The deviation from ideal octa- complex in solution (MeCN). Complexes [Fe(PyProNH2)Cl2] (14)
hedral geometry is reflected in the bond angles around iron and [Fe(6-Me-PyProNH2)Cl2] (15) showed a carbonyl vibration at
(angular variance is 257.62 deg2), which is caused by the 1652 and 1656 cm−1, respectively, in their solid state IR spectra.
geometrical restrictions of the ligands.
In the free ligands, these bands are also found at a relatively low
energy, which is most likely caused by the intermolecular
hydrogen bonding between the carbonyl and the amino groups.
For the prolinol complexes [Fe(PyProPh2OH)Cl2] (16) and [Fe(6-
Me-PyProPh2OH)Cl2] (17) a characteristic broad band around
3060 cm−1 for the ROH group was observed, which points to
coordination of this moiety towards the metal centre.
Magnetic susceptibility measurements in solution using
Evans’ method25,26 showed the formation of high spin (S = 2)
iron(II) chloride complexes. Detailed spectroscopic character-
istics for complexes 12–17 are summarized in Table 1.
Fe-chloride complexes
The preparation of 1 : 1 iron to ligand complexes using the
PyProR ligands 1–8 was envisioned through the use of an
iron(II) source that contains coordinating counter ions. Indeed,
complexes 12–15 were prepared via the reaction of one equi-
valent of the corresponding ligand with one equivalent of
ferrous chloride in methanol under an inert N2-atmosphere
for 1 h. After precipitation with Et2O, the complexes were iso-
lated as yellow powders in 90–94% yield (Scheme 3). Com-
plexes 12–15 were stored in an inert atmosphere drybox
because of their air sensitivity.
Ferrous chloride complexes 16 and 17 derived from the
diphenylprolinol ligands 7 and 8 were prepared in a similar
manner using a solvent mixture of methanol and dichloro-
methane (4 : 5, v/v) because of solubility reasons. Both com-
plexes were isolated as yellow powders after precipitation in a
yield of 90 and 85%, respectively (Scheme 3). Complexes 16
and 17 were also stored in an inert atmosphere drybox because
of their air sensitivity.
Structural features of the Fe chloride complexes in the solid
sate (X-ray crystal structures)
Fe(PyProMe)Cl2 (12). Single crystals of 12 were obtained by
either slow vapor diffusion of ether into a solution of 12 in
dichloromethane, or by ether diffusion into a solution of 12 in
methanol. These crystallization conditions lead to different
crystal forms in which 12 adopts different coordination geome-
tries (Fig. 7).
From a non-coordination solvent, complex 12a crystallizes
in a very distorted five-coordinated geometry, which cannot be
ESI-MS analysis of the chloride complexes 12–17 showed a
singly charged parent peak of composition [Fe(L)(Cl)]+ in all
cases and pointed at the formation of complexes with a metal
to ligand ratio of 1 : 1. Elemental analysis corroborated the for-
mation of complexes with an [Fe(L)Cl2] composition.
Table 1 Spectroscopic properties of complexes 12–17
CvO
vibration
CvO
vibration
ESI-MS
For these ferrous chloride complexes, IR spectroscopy is
somewhat less informative than in the case of the ferrous
free ligand complex
UV-Visa
(nm)
(m/z)
Evans
Complex (cm−1
)
(cm−1
1697
1687
)
[M − Cl]+ μeff (μβ)
12
13
1732
1732
291(866),
363(501)
207(9910),
267(4282),
369(338)
385(469)
385(456)
255(7096),
292(5319)
311.025
325.041
4.62
5.07
14
15
16
1668
1625
1652
1657
296.026
310.040
425.091
4.33
5.10
4.88
17
205(21 889), 449.106
261(7282),
5.23
316(4139)
a Extinction coefficients (M−1 cm−1) are given in parentheses after each
feature.
Scheme 3 Synthesis of the chloride complexes 12–17.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 6769–6785 | 6773

View Article Online
Paper
Dalton Transactions
Fig. 7 Molecular structures of complex 12 crystallized from a dichloromethane solution (12a, left) or a methanol solution (12b, right). Displacement
ellipsoid plot (50% probability); C–H hydrogen atoms are omitted for clarity.
described by a Berry pseudorotation pathway. The two largest
angles at Fe1 are 149.78(6) (N1–Fe1–O1) and 126.96(5)°
(Cl1–Fe1–Cl2), while one would expect 180°/120° for an ideal
trigonal bipyramid and 150°/150° for a square pyramid.
Using a coordinating solvent like methanol, the complex
crystallizes in an octahedral geometry with a facial coordi-
nation mode of the PyProMe ligand. In the octahedral geome-
try a methanol solvent molecule is coordinated trans to the
pyridine ring. Selected bond lengths and angles of 12a and
12b are depicted in Table S3.†
The observed Fe–N bond lengths are all in agreement with
high spin (S = 2) iron(II) complexes. The bond length towards
Cl1 in 12b is much longer than the other Fe–Cl bond lengths.
Fig. 8 Molecular structure of complex 13 crystallized from a methanol
solution. Displacement ellipsoid plot (50% probability); all hydrogen
Cl1 is the acceptor of an intermolecular hydrogen bond with
the hydroxyl atom of the coordinated methanol of a neigh-
bouring complex as a hydrogen bond donor (ESI Fig. S2 and
Table S4†).
atoms are omitted for clarity.
The methyl group is probably also the reason for the larger
Fe(6-Me-PyProMe)Cl2 (13). Crystals of 13 suitable for X-ray Cl–Fe–Cl angle in 13 (127.01(2)°) compared to 12a (113.01(2)°).
diffraction were obtained by slow vapour diffusion of ether Selected bond lengths and angles for 13 are shown in
into a solution of 13 in methanol. Despite the presence of the Table S5.†
coordinating solvent methanol, the crystal structure of 13
Fe(PyProNH2)Cl2 (14) and Fe(6-Me-PyProNH2)Cl2 (15). Crys-
shows a five-coordinated iron centre (Fig. 8) and is therefore tals suitable for X-ray analysis were obtained by slow vapor
similar to 12a (from CH2Cl2) and not to 12b (from MeOH). diffusion of ether into an acetonitrile solution of 14 and a
Again, the coordination polyhedron of 13 is very distorted and methanol solution of complex 15. Compound 14 crystallizes in
does not follow a Berry pseudorotation pathway. The largest an octahedral geometry with a facial coordination of the
angles at Fe1 are 153.71(6) (N1–Fe1–O1) and 127.01(2)° PyProNH2 ligand and with the additional coordination of an
(Cl1–Fe1–Cl2).
acetonitrile molecule (Fig. 9). The acetonitrile ligand is co-
The observation that complex 13 does not pick up a metha- ordinated trans to the proline nitrogen, while in complex 12b
nol molecule in its coordination sphere upon crystallization the methanol ligand is coordinated trans to the pyridine nitro-
from methanol most likely arises from the presence of the gen. The crystal structure also showed that complex 14 is able
methyl group at the 6-position of the pyridine ring in 6-Me- to form a hydrogen-bonded two-dimensional network in the
PyProMe. This methyl group provides some steric bulk and crystallographic a,b-plane (ESI Fig. S3 and Table S7†). On the
seems to prevent the rearrangement of the 6-Me-PyProMe other hand compound 15 was obtained as a five-coordinated
ligand from a mer to a fac coordination mode by blocking iron complex with a square-pyramidal geometry (τ = 0.28)27
the concomitant rearrangement of the two chloride ligands. (Fig. 9). The basal plane is formed by the tridentate ligand and
6774 | Dalton Trans., 2014, 43, 6769–6785
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Fig. 9 Molecular structures of complex 14 crystallized from an acetonitrile solution (left) and complex 15 crystallized from a methanol solution
(right). Displacement ellipsoid plot (50% probability); C–H hydrogen atoms are omitted for clarity.
the chlorine atom Cl2. The iron centre Fe1 is 0.73 Å above this compared to oxygen. This difference causes the amide carbo-
plane. The largest angles at Fe1 are 147.29(3) and 130.66(4)°. nyl group to bind more strongly to the iron centre.
The amide group of complex 15 is involved in a one-dimen-
Fe(PyProPh2OH)Cl2 (16) and Fe(6-Me-PyProPh2OH)Cl2
sional hydrogen-bonding network towards the chloride atoms (17). For the last two complexes crystals suitable for X-ray ana-
of neighbouring complexes (ESI Fig. S4 and Table S8†). Prob- lyses were obtained by diffusion of ether into a dichloro-
ably also in this case, just like for complex 13, the presence of methane solution of the corresponding complex. Fig. 10 shows
the methyl group on the pyridine ring prevents the coordi- the two molecular structures. For both complexes 16 and 17,
nation of a solvent molecule, giving rise to a five coordinated two independent residues (res 1 and res 2) are present in the
iron center.
asymmetric unit. Both complexes also turned out to form
The proline ring in both structures has the envelope confor- hydrogen-bonded dimers between the two independent mole-
mation, for complex 14 (cosform 0.966, sinform 0.034)28 and cules (ESI Fig. S5 and Table S10† for complex 16 and Fig. S6
for complex 15 (cosform 0.804, sinform 0.196).27 Selected and Table S11† for complex 17).
bond lengths and angles of complexes 14 and 15 are depicted
in Table S6.†
In 16, both independent molecules have a five-coordinated
iron centre with a geometry between square pyramidal and tri-
The Fe–O bond lengths of these two complexes are signifi- gonal bipyramidal (τ = 0.43/0.37).27 The two independent
cantly shorter than the other reported Fe–O bond lengths. This molecules in 17 have a square pyramidal geometry (τ = 0.28/
difference can be explained by the presence of the amide 0.24)27 with the basal plane formed by the tridentate ligand
group. The carbonyl group in an amide bond has a higher and one chlorine. The iron centre is 0.82/0.81 Å above this
partial negative charge compared to the carbonyl group in an plane. Selected bond lengths and angles for both residues of
ester bond due to the lower electronegativity of nitrogen complexes 16 and 17 are depicted in Table S9.†
Fig. 10 Molecular structures of complexes 16 and 17. Displacement ellipsoid plot (50% probability); C–H hydrogen atoms and the co-crystallized
CH2Cl2 solvent molecule in 16 are omitted for clarity. For both structures, only one of two independent residues is shown.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 6769–6785 | 6775

View Article Online
Paper
Dalton Transactions
Scheme 4 General representation of catalytic conditions.
enantioselectivity was found. In the best case a conversion
towards the epoxide of 35% is reached based on the amount of
oxidant used, with the total productive consumption of hydro-
gen peroxide of 51% (formation of epoxide and benzaldehyde).
Benzaldehyde formation was observed both in the reactions
with styrene and trans-beta-methylstyrene. In fact, in the case
of 9 and 11 more benzaldehyde than epoxide is formed out of
styrene after 1 night (9 equiv. for 9 and 4.5 equiv. for 11).
Complex 10 produces an equal amount of epoxide and benz-
aldehyde after 1 night. In the reactions with trans-beta-methyl-
styrene the epoxide is the predominant product after 1 night.
Overall, the activity of these triflate complexes is rather poor,
except perhaps for complex 10 in combination with styrenes.
The low activity of the complexes is likely caused by the fact
that these complexes are coordinatively saturated; it seems
that at least one dative bond needs to be broken before the
complexes can interact with either the substrate or the
oxidant.
The chloride complexes 12–17 showed an overall higher
reactivity profile compared to the triflate complexes 9–11 in
the oxidation reactions (Table 3). At the same time, several
trends are rather similar amongst the two series of complexes.
In all cases, aliphatic olefins and in particular external olefins
(1-octene) are converted in only a small number of turnovers,
while styrenes give higher and in some cases good turnover
numbers.
In addition, the amide-appended complexes 10 and 15 give
the highest activities in these series, while the most bulky
complexes 11, 16, and 17 give the lowest activities. For trans-
beta-methylstyrene, amide complex 15, [Fe(6-Me-PyProNH2)-
Cl2], gives 50 turnovers in epoxide formation, which corres-
ponds to 50% productive H2O2 consumption. Significant benz-
aldehyde formation was also observed with styrene substrates
in the case of complexes 12–17 (Table 3).
In order to provide an overview of the overall activity of the
chloride complexes, the cumulative formation of epoxides and
aldehydes as a measure for productive H2O2 consumption is
shown in Fig. 12. The graph clearly shows the overall effect of
the 6-Me-pyridine substituent on the catalytic activity. For
12 and 13 only minor changes are observed, whereas for 16
and 17 reactivity clearly goes down. In contrast, for amide com-
plexes 14 and 15 inclusion of a methyl-substituent leads to a
substantial increase in reactivity. Clearly, the highest activity is
obtained for complex 15, with a maximum of 65% productive
H2O2 conversion with methylstyrene. Also for this series of
catalysts, a mediocre enantioselectivity was found in the
epoxide formation out of styrenes. No attempts were made to
further optimize these ee values.
Fig. 11 Quaternion fit of all residues in the crystals of 16 and 17 (16: red
(res 1) and black (res 2); 17: green (res 1) and blue (res 2)).
The bond lengths differ slightly between the two complexes.
The bond length Fex–N1x is slightly longer for complex 17,
which also in this case may be due to the presence of the
methyl group in the sixth position on the pyridine ring. Fur-
thermore, the angle between the two chloride atoms is larger
for the complex with the methyl group at the sixth position of
the pyridine ring. For comparison, a quaternion plot of the
different residues in crystals of 16 and 17 is presented in
Fig. 11, which clearly shows the similarity of the four struc-
tures. The most apparent differences are the orientation of the
two phenyl rings of the diphenylprolinol moiety and the Cl–
Fe–Cl angle. Overall the pyridine methyl group has no large
structural influence in this case.
Oxidation catalysis
Iron(II) complexes 9–17 were tested as catalysts for the epoxi-
dation of alkene substrates. For the catalytic screening aceto-
nitrile was chosen as a solvent and hydrogen peroxide was
used as a sacrificial oxidant. A ratio between catalyst, oxidant
and substrate of 1 : 100 : 500 (oxidant limiting conditions) was
used during these screenings (Scheme 4). In total four
different benchmark substrates were used, cyclooctene,
1-octene, styrene and trans-beta-methylstyrene. trans-Beta-
methylstyrene was used to examine the enantio-inducing
ability of the chiral iron(II) complexes. Product formation was
monitored by GC after 1 and 3 h, and after 1 night.
Table 2 shows the conversion data (TON in epoxide/benz-
aldehyde) in the reactions with the triflate complexes 9–11
derived from PyProMe, PyProNH2 and PyProPh2OH. Among
these complexes, 10 showed the highest activity towards all
four substrates, while complex 9 and in particular the more
bulky complex 11 showed poor activities. As it turned out,
1-octene is not a suitable substrate for these complex systems,
as there is hardly any conversion to the epoxide product. In
the best case, for complex 10, a conversion towards the
epoxide of 3.3% is obtained based on the amount of added
hydrogen peroxide.
The highest conversion to epoxide products was observed
with trans-beta-methylstyrene, although no appreciable
6776 | Dalton Trans., 2014, 43, 6769–6785
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Table 2 Oxidation of alkenes catalysed by complexes 9–11a
TONb
Styrene
trans-β-Methylstyrene
(benzaldehyde)
Cyclooctene
1-Octene
1 h
(benzaldehyde)
Complex
1 h
1.1
4.9
0.6
3 h
2.3
7.9
1.1
1 Night
4.8
3 h
0.6
2.5
0.2
1 Night
0.7
1 h
3 h
1 Night
1 h
3 h
1 Night
9
0.4
1.3
(4.8)
6.6
(9.8)
0.9
2.4
(7.0)
7.9
(11.4)
0.9
2.7
(9.0)
14.1
(14.4)
1.0
5.7
10.0
(9.8)
24.4
(14.0)
6.7
22.9
(13.9)
35.0
(15.9)
8.7
CO2nPr
10
(8.4)
20.6
(13.6)
3.5
14.5
1.3
1.8
3.3
CONH2
11
0.1
0.3
CPh2OH
(2.8)
(3.7)
(4.5)
(5.0)
(7.0)
(7.0)
a Reaction conditions: 0.5 mL of 700 mM H2O2 solution in acetonitrile (0.35 mmol, 100 eq., diluted from 35% aqueous H2O2) was slowly added
over 30 min to a stirred solution of 3 mL acetonitrile containing catalyst (3.5 µmol) and substrate (1.75 mmol) at ambient temperature, stirring
was continued for another 30 min. After 1 hour from the start internal standard was added (10 μL, cyclooctene: 1,2-dibromobenzene, all other
substrates: bromobenzene) and the first samples were taken and analyzed by GC. b Moles of product/moles of catalyst.
Table 3 Oxidation of alkenes catalyzed by complexes 12–17a
TONb
Styrene
trans-β-Methylstyrene
Cyclooctene
1-Octene
1 h
(benzaldehyde)
(benzaldehyde)
Complex
1 h
5.6
1.2
4.3
5.8
2.3
2.0
3 h
8.6
1 Night
12.1
9.9
3 h
1.9
0.5
1.3
3.2
0.9
0.6
1 Night
3.5
1 h
3 h
1 Night
1 h
3 h
1 Night
12
CO2Me
13
1.2
8.5
(7.0)
1.4
(5.9)
6.1
13.1
(10.9)
2.3
(8.5)
8.6
15.1
(11.8)
13.6
(9.6)
9.1
(13.7)
28.2
(13.6)
10.0
(10.7)
5.7
20.8
(9.6)
3.8
29.4
(12.8)
7.0
29.1
(10.6)
30.5
3.5
5.4
0.3
2.9
MeCO2Me
14
CONH2
(10.2)
18.2
(4.0)
10.7
(5.5)
10.4
(6.0)
6.5
(13.7)
25.9
(5.9)
27.3
(11.5)
17.0
(8.7)
9.8
(11.6)
24.7
(13.3)
49.2
7.7
0.9
2.7
(2.6)
6.0
(3.6)
3.9
(4.0)
13.1
(7.8)
6.0
15
13.2
3.8
17.5
7.3
1.1
7.0
MeCONH2
16
CPh2OH
(15.0)
24.9
(11.1)
12.5
0.6
1.7
(2.9)
2.2
(5.5)
2.9
17
2.7
6.4
0.4
1.9
MeCPh2OH
(2.3)
(3.6)
(9.3)
(4.5)
(6.1)
(7.8)
a For reaction conditions see Table 2. b Moles of product/moles of catalyst.
Discussion
In the present investigations we have explored the coordi-
nation chemistry of a series of simple pyridinyl prolinate and
pyridinyl diphenylproline ligands with respect to Fe(II).
Whereas the parent ligand of the series reported here (S)-methyl-
1-(pyridin-2-ylmethyl)pyrrolidine-2-carboxylate (PyProMe) has
been studied by Paine et al. in relation with iron-mediated
catechol oxidation,20 the current extended series of pyridinyl
proline ligands (RPyProR) allows for
a more general
description of the coordination properties of this ligand
family. The specific interest in these ligands combines the
ease of synthesis of these chiral ligands and their anticipated
mimicry of the 2-His-1-carboxylate facial triad coordination
mode as found in certain mono-nuclear non-heme iron
enzymes. Our synthetic efforts have resulted in a full structural
characterization of most of the reported Fe(II) complexes and
Fig. 12 Productive consumption of an oxidant after 1 night for com-
plexes 12–17.
show the coordinative flexibility of the N,N,O pyridinyl proline
fragment acting either as a meridional or as a facial ligand.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 6769–6785 | 6777

View Article Online
Paper
Dalton Transactions
Closed-shell, octahedral structures of the type [Fe(L)2]2+ complexes in order to arrive at a more detailed structural
were formed when iron triflate was used as the iron source. description of such complexes.
This is a rather common observation for tridentate ligands
The coordination flexibility of the PyPro ligands is shown
and has also been observed for related tridentate ligands of by the complexes that form when their 5-coordinated com-
the N,N,O type. In our earlier studies on BAIP and BAIPR type plexes are crystallized from coordinating solvents like metha-
ligands, the formation of similar octahedral closed-shell struc- nol or acetonitrile. In such solvents, these complexes
tures was observed in cases where non-coordinating counter accommodate one solvent molecule in the coordination sphere
ions were used, albeit that in these cases the ligands coordi- of iron and, accordingly, form octahedral complexes of the
nate in a facial rather than in the meridional manner to iron type [Fe(L)(solv)Cl2]. In these cases the coordination geometry
like observed for complex 10.15c,29
of the RPyProR ligand becomes facial. The coordination of
In particular when anionic N,N,O ligands are used, the for- solvent molecules is only observed in complexes that contain a
mation of octahedral bis-ligand complexes is difficult to avoid. PyPro ligand without an unsubstituted pyridine moiety. Intro-
Attempts to form discrete [Fe(II)(L)X] or [Fe(III)(L)X2] complexes duction of a methyl group at the 6-position of this moiety pre-
derived from the family of BAIP-ligands have proven difficult vents the coordination of solvent molecules at the cis-position
so far. In most reported cases the thermodynamic [Fe(L)2]0/1+ with respect to the pyridine N-donor, most likely due to steric
complexes did form. Burzlaff and co-workers also reported on constraints. Solvent coordination at the trans-position is also
their difficulties in synthesizing mono-ligand complexes prevented because the size of the methyl substituent prevents
derived from the bis(pyrazolyl)proprionate ligand family. Ulti- the reorientation of the chloride ligands, i.e. positioning of the
mately, they succeeded in synthesizing [Fe(L)Cl3]− in which Fe–Cl vector within the plane defined by the pyridine ring is
iron is in the oxidation state 3+.12 The major differences not possible. Where the pyridine 6-position in nitrogen
between the RPyProR ligands and the BAIP as well as the bis- ligands is often used to steer substrate approach to the metal
(pyrazolyl)proprionate ligand families are the predisposed center38 and the electronics at the metal (spin state),8b
facial coordination modes of the latter ligands and their inclusion of a substituent at this position in the RPyProR
anionic nature in comparison to the neutral PyPro ligands.
ligand framework prevents a meridional to facial ligand
Using FeCl2 as the source of iron in combination with a coordination change and locks the RPyProR framework in a
PyPro ligand leads to the formation of discrete [Fe(L)Cl2] com- meridional conformation.
plexes that have a penta-coordinated iron centre. For com-
All prepared complexes 9–17 were tested in the catalytic oxi-
plexes 15–17 the geometry would at best be described as dation of different alkenes using hydrogen peroxide as the
square pyramidal (τ values vary from 0.24 to 0.45). Following oxidant. During the catalytic reactions oxidant limiting con-
the square pyramidal description the PyPro ligands coordinate ditions were used to allow putative short-live oxygenated metal
to iron as a N,N,O meridional ligand in these cases, position- species to be trapped rapidly by the excess of substrate. It
ing the two chloride ligands in ‘equatorial’ positions. turned out that the triflate complexes perform far less com-
Examples of discrete square pyramidal structures in non-heme pared to the chloride-based complexes. This can be explained
iron chemistry include complexes derived from pyridine(bis- by the fact that the triflate complexes are closed-shell com-
imine) ligands.30,31 Other examples of meridionally co- plexes with a fully occupied octahedral coordination sphere
ordinated N,N,O ligands are reported with cobalt, zinc and around iron. These complexes can, therefore, only react via
copper.32–34
inner-sphere pathways with external reagents such as alkene
Detailed analysis of the 5-coordinated structures reported substrates or oxidants after one or more existing coordination
here (7 in total) shows that none of these structures obeys bonds are broken. The chloride complexes turned out to be
exactly the Berry pseudo-rotation trajectory (ESI Table S12†). much more reactive. Several of the crystal structures of these
The deviation is particularly large for complexes 12 and 13, complexes nicely show how solvent molecules can be accom-
which is why τ values would not properly describe their geome- modated in their coordination sphere to obtain six-coordinate
try. This again shows that the coordination geometry of these complexes. The formation of oxygenated complexes of the type
complexes is ill-defined and can, accordingly, actually not be [Fe(L)(Cl)2(ox)] is therefore not restricted due to the lack of
labelled as square pyramidal. An analysis of the CSD-data- coordination sites.
base35 revealed 14 other 5-coordinated FeCl2 complexes
The observation of the lower activity of the triflate com-
bearing an N,N,O ligand.36 Of these complexes also none plexes is also somewhat surprising. We have earlier reported
obeys exactly the Berry trajectory. A graphical representation of that related coordinatively saturated [Fe(BAIP)2]2+ complexes
the 14 structures found in the CSD-database and our 7 struc- are active as epoxidation/dihydroxylation catalysts and have
tures reported here is depicted in the ESI (Graph S1).† Vahren- postulated that open coordination sites in these systems may
kamp and co-workers have earlier noticed this feature in form by means of ester moiety dissociation.15c In addition, the
5-coordinated Zn(II) complexes and have proposed an adapted [Fe(Ph-DPAH)2]2+ system that was studied by Que et al. has
description of the Berry trajectory in these systems.37 The similar coordination features and at the same time is very
current set of Fe(II) complexes is limited in number and does active in the dihydroxylating of olefins (TONs of 8 are
not allow for a similar analysis at this stage. Current efforts in obtained, ratio catalyst : oxidant : substrate 1 : 10 : 1000).13 In
our lab involve the expansion of the set of 5-coordinated FeCl2 this case the authors have assumed that full dissociation of
6778 | Dalton Trans., 2014, 43, 6769–6785
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
one of the facial N,N,O ligands takes place prior to any cataly- the case of the chloride complexes 12–17, the reactivity
tic turn-over. Given the lower turn-over numbers of the PyPro- remains about equal between 12 and 13; it increases consider-
derived iron triflate complexes ligand dissociation does not ably between 14 and 15, but decreases between 16 and 17.
seem to take place readily for these complexes and accordingly Although there is no clear trend, it seems that complexes
TONs are lower.
without the 6-methyl substituent on pyridine have a higher
The overall structure of the RPyProR ligands has a distinct initial activity compared to the complexes with the methyl
effect on the catalytic activity of their complexes. Complexes group (Table 3). On the other hand, complexes bearing the
derived from bulky diphenylprolinol ligands gave the lowest 6-methyl group do stay active over the whole course of the reac-
activities, both in the case of the triflate as well as the chloride tion. This reactivity difference can be explained by assuming a
complexes and for all substrates. A combination of alcohol longer lifetime of species involved in the catalytic cycle for the
instead of ester or amide coordination and enhanced steric complexes with the methyl group due to steric congestion. The
bulk could lead to a less electron-rich metal centre (electronic fact that complexes derived from the methyl-substituted
effect) and a less favourable olefin substrate attack on the oxy- ligands do not accommodate a solvent molecule as the sixth
genated active species (steric effect). In all cases, complexes ligand could also implicate a higher barrier for the formation
derived from amide-appended RPyProR ligands gave the better of six-coordinate oxygenated intermediates for these com-
catalysts, i.e. higher productive consumption of H2O2 was plexes. This would either suggest that these systems operate
found. The higher activity of these complexes may be via a different pathway or that different kinetics are at play
explained by the stronger Fe–O bonds that are formed by the within a similar pathway.
amide ligands, which could lead to enhanced catalyst life-
Few other Fe-based catalysts based on a mixed N,N,O
times. On the other hand, the short Fe–O bond lengths to the ligand manifold, in particular those based on proline, are
amide moieties indicate an enhanced electron donation, known to perform the epoxidation of alkenes using H2O2.
which will lead to more electron-rich Fe-centres and, in turn, Examples include the BAIP and Ph-DAPH systems (vide supra).
to lower oxidation potentials. Finally, the amino groups of the Paine et al. reported on a study using the parent ligand
primary amide may have a stabilizing effect on putative high- (PyProMe) in relationship with iron-catalyzed catechol oxi-
valent iron–oxygen intermediates by means of acting as a dation instead of the epoxidation of alkenes.20 Beller and co-
hydrogen bond donor. Similar effects were recently found for workers reported in 2007 on the use of ligands derived from
amide BAIPR Fe(II) complexes.29
the L-proline backbone.39 These ligands were used in the
Concerning the reactivity of the RPyProR systems towards epoxidation of aromatic alkenes (trans-stilbene) and gave for
different olefin substrates, it is clear that these show a higher the first time ee’s above 20%. Unfortunately no reactions were
reactivity towards the more electron-deficient styrenes than to done on non-aromatic alkenes. We have previously reported on
the more electron-rich aliphatic olefins. This observation the use of iron complexes derived from bis-proline substituted
seems to point to a nucleophilic nature of the active species in pyridine ligands for the oxidation of alkenes with hydrogen
the oxidations. This seems to make sense given the neutral peroxide as an oxidant.22 These complexes showed a low
nature of the more reactive Fe(II)Cl2 complexes and of their activity towards the epoxidation of various alkenes.
putative high-valent intermediates, in particular in the pres-
ence of coordinating chlorides. The comparison of the relative
reaction rates of a series of para-substituted styrene substrates
did not show an appreciable difference in the reaction rate.
Conclusions
While this observation does not prove the nucleophilic nature In conclusion, we have developed a series of readily available
of the active species, it may indicate that other reaction steps chiral tripodal N,N,O ligands derived from proline. The struc-
besides substrate attack are involved in determining the tural characterization of several Fe(II)complexes has shown
overall reaction rate. In the oxidation experiments with the that these ligands can adopt a meridional or a facial coordi-
pro-chiral trans-beta-methylstyrene substrate no appreciable nation mode in 6-coordinated complexes. On the other hand,
enantioselectivity was observed. This can probably be rational- the 5-coordinated complexes show ill-defined geometries that
ized by the involvement of a radical-based reaction pathway, do not obey a Berry-type description. Depending on the size
which tends to result in racemic product distributions. This and nature of the substituents at both the proline and at the
assumption is corroborated by the large extent of benz- pyridine ring, the coordination mode of these ligands may
aldehyde formation in the oxidation of styrenes by the PyPro change upon binding of a solvent molecule. This allows for
systems. On the other hand, the use of trans-beta-methylstyr- the formation of complexes of the type [Fe(L)(solv)Cl2], where
ene as the substrate did result in the highest TONs for these L is a facial and neutral N,N,O ligand. The further develop-
systems. With amide complex 15, a useful conversion of the ment of such complexes is of interest in modelling the features
oxidant towards the epoxide of almost 50% was reached, and of the 2-His-1-carboxylate facial triad found in mono-nuclear
an overall productive conversion of the oxidant of almost 65% non-heme iron enzymes.
was found.
The use of iron complexes derived from the PyPro ligands
Finally, the role of the substitution of the pyridine moiety results in reasonable turnover numbers for both epoxide and
in the PyPro ligands on their reactivity may be pointed out. In benzaldehyde products, albeit no substantial enantioselectivity
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 6769–6785 | 6779

View Article Online
Paper
Dalton Transactions
has been observed thus far. These results may be used as added dropwise for 30 min with stirring to a colourless solu-
guidelines in the further development of the bio-inspired non- tion of pyridin-2-ylmethanol (2.20 g, 20.2 mmol) in anhydrous
heme iron oxidation catalyst based on cheap chiral building CH2Cl2 (50 mL) at 0 °C. After 60 min stirring at 0 °C, the white
blocks.
turbid solution was allowed to reach ambient temperature.
After 60 min, the solution was quenched with n-propanol
(3.5 mL), stirred for 15 min, evaporated, and dried in vacuo to
leave a white powder, which was used in the subsequent reac-
tion without further purification (3.31 g, quant.).
Experimental section
General
1H NMR (400 MHz, DMSO-d6, 25 °C): δ = 4.98 (s, 2H,
3
3
Air-sensitive organic reactions and reactions with metal salts ArCH2), 7.75 (dd, 1H, JHH = 7.6 Hz, JHH = 5.6 Hz, PyH(5)),
3
3
were carried out under an inert, N2 atmosphere using standard 7.92 (d, 1H, JHH = 8.0 Hz, PyH(3)), 8.29 (t, 1H, JHH = 7.8 Hz,
Schlenk techniques. Solvents were dried and distilled before PyH(4)), 8.76 (d, 1H, JHH = 5.4 Hz, PyH(6)) ppm. 13C{1H} NMR
3
use. Chemicals were either commercially obtained and used as (100 MHz, DMSO-d6, 25 °C): δ = 42.9, 125.5, 125.8, 143.2,
received or reproduced from the literature. 1H and 13C NMR 145.2, 153.1 ppm.
spectra were recorded on a Varian 400 spectrometer at
2-(Chloromethyl)-6-methylpyridinium
chloride. SOCl2
400 MHz and 100 MHz, respectively, operating at 25 °C. Infra- (1.2 mL, 17 mmol) in anhydrous CH2Cl2 (28 mL) was reacted
red spectra were recorded with a Perkin-Elmer Spectrum One with (6-methylpyridin-2-yl)methanol (1.41 g, 11.4 mmol) in
FT-IR instrument. Solution IR measurements were recorded anhydrous CH2Cl2 (28 mL) in the same manner as described
with a Mettler Toledo ReactIR™ 1000 spectrometer with a for 2-(chloromethyl)pyridinium chloride. After the solution
SiComp™ probe under a N2 atmosphere. ESI-MS was measured reached ambient temperature it was stirred for an additional
on a Waters LCT Premier XE. UV-Vis spectra were recorded on 60 min at 60 °C after which the reaction was quenched with
a Varian Cary 50. Solution magnetic moments were deter- nPrOH (2.0 mL), stirred for 15 min, evaporated, and dried
mined by the Evans’ NMR method in acetone-d6–cyclohexane in vacuo to leave a white powder, which was used in the sub-
(95/5 v/v) or in acetonitrile-d3–cyclohexane (95/5 v/v) at sequent reaction without further purification (2.04 g, quant.).
25 °C.25,26 GC analyses were performed on a Perkin-Elmer
1H NMR (400 MHz, DMSO-d6, 25 °C): δ = 2.69 (s, 3H,
Clarus 500 GC (30 m, Econo-Cap EC-5) with a FID detector. ArCH3), 5.01 (s, 2H, ArCH2), 7.69 (d, 1H, 3JHH = 7.6 Hz, PyH(5)),
3
3
Elemental microanalyses were carried out by the Mikroanaly- 7.78 (d, 1H, JHH = 8.0 Hz, PyH(3)), 8.26 (t, 1H, JHH = 8.0 Hz,
tisches Laboratorium KOLBE, Mülheim an der Ruhr, Germany. PyH(4)) ppm. 13C{1H} NMR (100 MHz, DMSO-d6, 25 °C): δ =
Fe(OTf)2·2MeCN,40
(S)-2-(methoxycarbonyl)pyrrolidinium 41.9, 62.5, 123.5, 126.2, 143.9, 151.8, 155.5 ppm.
chloride,41 (S)-methyl-1-(pyridin-2-ylmethyl)pyrrolidine-2-car-
(S)-Methyl-1-((6-methylpyridin-2-yl)methyl)pyrrolidine-2-car-
boxylate (PyProMe) (1),23 1-(pyridin-2-ylmethyl)pyrrolidine-2- boxylate (6-Me-PyProMe) (2). A yellow suspension of 2-(chloro-
carboxamide (PyProNH2) (5),23 diphenyl(1-(pyridin-2-ylmethyl)- methyl)-6-methylpyridinium chloride (9.62 g, 54.0 mmol),
pyrrolidin-2-yl)methanol (PyProPh2OH) (7)23 were prepared (S)-2-(methoxycarbonyl)pyrrolidinium
chloride
(12.2
g,
according to published procedures.
73.5 mmol), NaI (3.08 g, 20.5 mmol) and Na2CO3 (19.0 g,
(S)-2-(n-Propoxycarbonyl)pyrrolidinium
chloride. SOCl2 179 mmol) in anhydrous DMF (140 mL) was stirred at 50 °C.
(2.6 mL, 36 mmol) was added dropwise for 15 min with stir- After 16 h the reaction mixture was taken up in H2O (280 mL)
ring to anhydrous n-propanol (12 mL) at 0 °C. S-Proline and extracted with CH2Cl2 (5 × 100 mL). The organic phases
(3.24 g, 28.1 mmol) was added to this solution at 0 °C. The were combined, washed with H2O (120 mL), dried (Na2SO4),
white suspension was stirred and heated to 60 °C. After 24 h filtered and evaporated. The remaining orange oil was sub-
the light brown solution was evaporated and dried in vacuo jected to column chromatography (SiO2 (900 mL), Et2O–Et3N
(<0.1 mbar) with stirring at 75 °C for 4 h to leave a light brown, 100 : 1 → 100 : 2 (v/v), Rf = 0.3–0.4). The remaining impurity
viscous oil, which was used in the subsequent reaction (1H NMR (CDCl3): 6.82–6.88 (m) and 7.17–7.21 (m)) was
without further purification (5.32 g, 98%).
removed by heating the oil in vacuo (<0.5 mbar) at 90 °C for
2 h to leave a yellow oil (8.57 g, 70%).
3
1H NMR (300 MHz, DMSO-d6, 25 °C): δ = 0.90 (t, 3H, JHH
=
3
7.2 Hz, CH3), 1.64 (sxt, 2H, JHH = 7.2 Hz, CH2CH2CH3),
1H NMR (400 MHz, CDCl3, 25 °C): δ = 1.77–2.01 (m, 3H,
1.83–2.03 (m, 3H, CH2 ring, γ and β to CO), 2.19–2.33 (m, 1H, CH2 ring, γ and β to CO), 2.10–2.21 (m, 1H, CH2 ring, β to CO),
CH2 ring, β to CO), 3.13–3.28 (m, 2H, NCH2 ring), 4.06–4.19 2.53 (s, 3H, ArCH3), 2.53–2.59 (m, 1H, NCHH ring), 3.09–3.14
(m, 2H, CH2CH2CH3), 4.33–4.38 (m, 1H, CH α to CO), 9.15 (s, (m, 1H, NCHH ring), 3.42–3.46 (m, 1H, CH α to CO), 3.65 (s,
2
b, 1H, NHH), 10.33 (s, b, 1H, NHH) ppm. 13C {1H} NMR 3H, CH3), 3.80 (d, 1H, JHH = 13.6 Hz, ArCHHN), 4.01 (d, 1H,
3
(100 MHz, CDCl3, 25 °C): δ = 10.4, 22.0, 23.8, 29.0, 46.1, 59.4, 2JHH = 13.6 Hz, ArCHHN), 7.01 (d, 1H, JHH = 7.6 Hz, PyH(5)),
3
3
68.6, 169.1 ppm. IR (film): ν = 2965.4, 2879.0, 1690.4, 1739.8, 7.28 (d, 1H, JHH = 7.6 Hz, PyH(3)), 7.54 (t, 1H, JHH = 7.6 Hz,
1560.4, 1458.4, 1396.4, 1354.3, 1222.8, 1092.6, 1057.4, 998.9, PyH(4)) ppm. 13C{1H} NMR (100 MHz, CDCl3, 25 °C): δ = 23.4,
968.5, 927.9, 903.7, 758.8, 676.1 cm−1. [α]D21 −43.3 deg cm3 g−1 24.5, 29.5, 51.9, 53.7, 60.3, 65.5, 120.5, 121.8, 137.0, 157.6,
dm−1 (c 0.952, CHCl3).
158.0, 174.5 ppm. IR (film): ν = 2951.6, 1732.2, 1591.4, 1577.9,
2-(Chloromethyl)pyridinium chloride. A cooled solution of 1455.8, 1435.4, 1358.2, 1275.0, 1196.0, 1169.4, 1087.1, 1038.4,
SOCl2 (2.1 mL, 29 mmol) in anhydrous CH2Cl2 (50 mL) was 995.4, 933.2, 890.9, 783.4, 758.1 cm−1. UV-Vis (MeCN) [λmax
,
6780 | Dalton Trans., 2014, 43, 6769–6785
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
2
nm (ε, M−1 cm−1)]: 207 (7587), 265 (4096). [α]2D1 −72.1 deg cm3 ArCHHN), 3.97–4.07 (m, 2H, CH2CH2CH3), 4.04 (d, 1H, JHH
=
3
g−1 dm−1 (c 1.1, CHCl3). Anal. for C13H18N2O2 (234.29): calc. 14.0 Hz, ArCHHN), 7.00 (d, 1H, JHH = 7.6 Hz, PyH(5)), 7.29 (d,
3
3
C 66.64, H 7.74, N 11.96; found C 66.54, H 7.83, N 11.90.
1H, JHH = 7.6 Hz, PyH(3)), 7.52 (t, 1H, JHH = 7.6 Hz, PyH(4))
n-Propyl-1-(pyridin-2-ylmethyl)pyrrolidine-2-carboxylate ppm. 13C{1H} NMR (100 MHz, CDCl3, 25 °C): δ = 10.5, 22.1,
(PyProPr) (3). An orange suspension of 2-(chloromethyl)pyridi- 23.4, 24.5, 29.5, 53.6, 60.3, 65.5, 66.2, 120.3, 121.7, 136.9,
nium chloride (3.30 g, 20.1 mmol), (S)-2-(n-propoxycarbonyl)- 157.5, 158.3, 174.2 ppm. IR (film): ν = 2967.7, 1728.6, 1591.5,
pyrrolidinium chloride (5.32 g, 26.8 mmol), NaI (1.19 g, 1578.1, 1456.4, 1357.7, 1267.0, 1173.4, 1086.8, 1059.2, 1040.2,
7.94 mmol) and Na2CO3 (7.29 g, 68.8 mmol) in anhydrous 991.7, 935.4, 900.8, 782.9, 757.5 cm−1. UV-Vis (MeCN) [λmax
,
DMF (54 mL) was stirred at 50 °C. After 25 h the resulting pink nm (ε, M−1 cm−1)]: 207 (8018), 265 (4347). [α]D21 −65.8 deg cm3
suspension was taken up in H2O (110 mL) and extracted with g−1 dm−1 (c 0.7, CHCl3). Anal. for C15H22N2O2 (262.35): calc.
CH2Cl2 (5 × 30 mL). The organic phases were combined, C 68.67, H 8.45, N 10.68; found C 68.57, H 8.65, N 10.63.
washed with H2O (50 mL), dried (Na2SO4), filtered and evapor-
ated. The remaining red oil was subjected to column chrom- (6-Me-PyProNH2) (6). A colorless solution of
1-((6-Methylpyridin-2-yl)methyl)pyrrolidine-2-carboxamide
(3.48 g,
2
atography (SiO2 (250 mL), Et2O–hexanes–Et3N 75 : 25 : 2 (v/v), 14.9 mmol) in anhydrous MeOH (60 mL) was saturated with
Rf = 0.3). The remaining impurity (1H NMR (CDCl3): 6.83–6.87 NH3 at 0 °C for 30 min. The solution was allowed to reach
(m) and 7.18–7.22 (m)) was removed by heating the oil in vacuo ambient temperature and after 6 days was evaporated to
(<0.1 mbar) at 110 °C for 1 h to leave a dark yellow oil (3.99 g, dryness. The remaining yellow oil was dissolved in a small
80%).
amount of eluent and subjected to column chromatography
(SiO2 (400 mL), Et2O–MeOH–Et3N 30 : 1 : 1 → 30 : 3 : 1 (v/v), Rf =
3
1H NMR (400 MHz, CDCl3, 25 °C): δ = 0.91 (t, 3H, JHH
=
7.2 Hz, CH3), 1.63 (sxt, 2H, JHH = 7.2 Hz, CH2CH2CH3), 0.3–0.4). The remaining impurity (1H NMR (CDCl3): 6.83–6.86
1.78–2.03 (m, 3H, CH2 ring, γ and β to CO), 2.12–2.23 (m, 1H, (m) and 7.17–7.21 (m)) was removed by heating in vacuo
CH2 ring, β to CO), 2.53–2.60 (m, 1H, NCHH ring), 3.08–3.13 (<0.5 mbar) at 100 °C for 2 h to leave a yellow, highly viscous
(m, 1H, NCHH ring), 3.43–3.46 (m, 1H, CH α to CO), 3.80 (d, oil, which solidified upon standing for several weeks (2.65 g,
3
2
1H, JHH = 13.6 Hz, ArCHHN), 3.99–4.08 (m, 2H, CH2CH2CH3), 81%).
2
3
4.10 (d, 1H, JHH = 13.6 Hz, ArCHHN), 7.13 (dd, 1H, JHH
7.6 Hz, JHH = 4.8 Hz, PyH(5)), 7.46 (d, 1H, JHH = 7.6 Hz, CH2 ring, γ to CO), 1.94–2.02 (m, 1H, CH2 ring, β to CO),
=
1H NMR (400 MHz, CDCl3, 25 °C): δ = 1.72–1.85 (m, 2H,
3
3
3
3
PyH(3)), 7.64 (t, 1H, JHH = 7.6 Hz, PyH(4)), 8.50 (d, 1H, JHH
=
2.21–2.31 (m, 1H, CH2 ring, β to CO), 2.48–2.55 (m, 1H, NCHH
4.8 Hz, PyH(6)) ppm. 13C{1H} NMR (100 MHz, CDCl3, 25 °C): ring), 2.53 (s, 3H, ArCH3), 3.07–3.12 (m, 1H, NCHH ring), 3.37
δ = 10.5, 22.1, 23.3, 29.4, 53.5, 60.2, 65.5, 66.3, 122.2, 123.5, (s (br), 1H, CH α to CO), 3.72 (d, 1H, 2JHH = 13.6 Hz, ArCHHN),
2
136.6, 149.0, 158.9, 174.1 ppm. IR (film): ν = 2967.0, 1728.2, 3.99 (d, 1H, JHH = 13.6 Hz, ArCHHN), 5.58 (s (br), 1H, NHH),
3
3
1588.8, 1569.9, 1473.9, 1433.3, 1360.5, 1269.9, 1174.1, 1086.8, 7.03 (d, 1H, JHH = 7.6 Hz, PyH(5)), 7.06 (d, 1H, JHH = 7.6 Hz
1059.3, 1046.3, 993.5, 934.6, 897.1, 757.7 cm−1. UV-Vis (MeCN) PyH(3)), 7.53 (t, 1H, JHH = 7.6 Hz, PyH(4)), 8.01 (s (br), 1H,
3
[λmax, nm (ε, M−1 cm−1)]: 202 (7617), 261 (3024). [α]D21 −65.4 NHH) ppm. 13C{1H} NMR (100 MHz, CDCl3, 25 °C): δ = 24.4,
deg cm3 g−1 dm−1 (c 0.8, CHCl3). Anal. for C14H20N2O2 24.6, 30.7, 54.3, 61.1, 67.4, 119.8, 122.1, 137.1, 157.7, 158.4,
(248.32): calc. C 67.71, H 8.12, N 11.28; found C 67.64, H 8.17, 177.9 ppm. IR (solid): ν = 3388.2, 3191.1, 2972.0, 2920.6,
N 11.25.
2872.6, 2825.5, 1624.5, 1594.4, 1576.8, 1460.0, 1402.2, 1374.8,
n-Propyl-1-((6-methylpyridin-2-yl)methyl)pyrrolidine-2-carboxy- 1331.9, 1310.0, 1275.1, 1235.0, 1196.8, 1156.3, 1111.2, 983.3,
late (6-Me-PyProPr) (4). A yellow suspension of 2-(chloro- 899.8, 785.6 cm−1. UV-Vis (MeCN) [λmax, nm (ε, M−1 cm−1)]:
methyl)-6-methylpyridinium chloride (2.02 g, 11.3 mmol), 199 (8820), 265 (4012). [α]2D1 −42.3 deg cm3 g−1 dm−1 (c 0.6,
(S)-2-(n-propoxycarbonyl)pyrrolidinium chloride (2.99 g, CHCl3). Anal. for C12H17N3O (219.28): calc. C 65.73, H 7.81,
15.4 mmol), NaI (0.68 g, 4.5 mmol) and Na2CO3 (4.13 g, N 19.16; found C 65.64, H 7.76, N 19.05.
39.0 mmol) in anhydrous DMF (31 mL) was stirred at 50 °C.
(1-((6-Methylpyridin-2-yl)methyl)pyrrolidin-2-yl)diphenyl-
After 25 h the faintly yellow suspension was taken up in H2O methanol (6-Me-PyProPh2OH) (8). Cut and oven-dried Mg
(60 mL) and extracted with CH2Cl2 (5 × 25 mL). The organic turnings were stirred and heated under N2. After 30 min they
phases were combined, washed with H2O (20 mL), dried were wetted with anhydrous THF and treated with 8 drops of
(Na2SO4), filtered and evaporated. The remaining orange oil 1,2-dibromoethane. The mixture was heated for 5 min, stirred
was subjected to column chromatography (SiO2 (200 mL), for 30 min, diluted with THF (15 mL), and subsequently
Et2O–hexanes–Et3N 75 : 25 : 2 (v/v), Rf = 0.4). The remaining treated dropwise for 40 min with a solution of anhydrous
impurity was removed by heating the oil in vacuo (<0.1 mbar) bromobenzene in THF (30 mL) with stirring. After stirring for
at 110 °C for 60 min to leave a yellow oil (2.65 g, 89%).
20 h the mixture was concentrated in vacuo (to 24 mL). The
grey suspension was cooled to 0 °C and subsequently treated
3
1H NMR (400 MHz, CDCl3, 25 °C): δ = 0.91 (t, 3H, JHH
=
3
7.2 Hz, CH2CH2CH3), 1.62 (sxt, 2H, JHH
= 7.2 Hz, dropwise for 60 s via a cannula with a solution of 2 (2.14 g,
CH2CH2CH3), 1.75–2.01 (m, 3H, CH2 ring, γ and β to CO), 9.14 mmol) in THF (10 mL) with vigorous stirring. The green
2.10–2.20 (m, 1H, CH2 ring, β to CO), 2.51–2.57 (m, 1H, NCHH solution was allowed to reach ambient temperature. After
ring), 2.52 (s, 3H, ArCH3), 3.07–3.12 (m, 1H, NCHH ring), 60 min of stirring the solution was acidified with 2 M HCl
2
3.39–3.43 (m, 1H, CH α to CO), 3.76 (d, 1H, JHH = 14.0 Hz, (40 mL), and after 30 min the aqueous phase was basified to
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 6769–6785 | 6781

View Article Online
Paper
Dalton Transactions
pH 8 with 2 M NaOH. After separation of the organic phase, RT the solvent was evaporated in vacuo to leave a bright yellow
the aqueous phase was extracted with CH2Cl2 (2 × 80 mL, 1 × foam. The product was precipitated by the addition of Et2O
40 mL). The combined organic phases were washed with H2O (40 mL) to a solution of the product in MeCN (5 mL) with stir-
(60 mL), dried (Na2SO4), filtered, and evaporated. The remain- ring for 60 min. The colourless supernatant was removed via a
ing yellow solid was dissolved in CH2Cl2 (10 mL), purified by cannula. The remaining yellow powder was recrystallized from
column chromatography (SiO2 (250 mL), hexanes–EtOAc 4 : 1 MeCN (5 mL) by the addition of Et2O (25 mL) and dried
→ 0 : 1 (v/v), Rf (hexanes–EtOAc 2 : 1) = 0.4), and dried in vacuo in vacuo (752 mg, 90%). Slow vapour diffusion of Et2O into a
to leave a white powder (2.80 g, 85%).
concentrated solution of 10 in MeCN yielded yellow, needle-
1H NMR (400 MHz, CDCl3, 25 °C): δ = 1.66 (s (br), 2H, CH2 shaped crystals suitable for X-ray diffraction.
ring, γ to CO), 1.79 (s (br), 1H, CH2 ring, β to CO), 1.92–2.02
IR (solid): ν = 3353.1, 1673.6, 1592.4, 1481.8, 1439.7, 1249.9,
(m, 1H, CH2 ring, β to CO H), 2.44–2.65 (m, 1H, NCHH ring), 1224.5, 1149.2, 1106.5, 1055.0, 1026.8, 991.7, 926.6, 843.5,
2.49 (s, 3H, ArCH3), 3.03 (s (br), 1H, NCHH ring), 3.40 (s (br), 781.6, 757.2 cm−1. IR (MeCN): ν = 2250 (νCN), 1668 (νCO),
2H, CH α to CO and ArCHHN), 4.12 (s (br), 1H, ArCHHN), 5.16 1270 (νasSO3), 1227 (νsCF3), 1158 (νasCF3), 1031 (νsSO3) cm−1
.
(s (br), 1H, OH), 6.93 (d, 1H, 3JHH = 7.6 Hz PyH(5)), 6.97 (d, 1H, UV-Vis (MeCN) [λmax, nm (ε, M−1 cm−1)]: 365 (785). [α]2D1 +208
3
3JHH = 7.6 Hz, PyH(3)), 7.08 (t, 1H, JHH = 7.2 Hz, p-PhH), 7.16 deg cm3 g−1 dm−1 (c 0.3, MeCN). Anal. for C24H30F6FeN6O8S2
3
3
(t, 1H, JHH = 7.6 Hz, p-PhH), 7.22 (t, 2H, JHH = 7.6 Hz, (764.50): calc. C 37.71, H 3.96, N 10.99; found C 37.85, H 4.06,
3
3
m-PhH), 7.29 (t, 2H, JHH = 7.6 Hz, m-PhH), 7.48 (t, 1H, JHH
7.6 Hz, PyH(4)), 7.58 (d, 2H, 3JHH = 7.6 Hz, o-PhH), 7.68 (d, 2H, Solution magnetic moment (Evans’ method): µeff = 4.46µB.
3JHH = 7.6 Hz, o-PhH) ppm. 13C{1H} NMR (100 MHz, CDCl3,
[Fe(PyProPh2OH)2](OTf)2 (11). A solution of Fe(OTf)2·
=
N 11.05. ESI-MS: m/z = 615.125 ([M − OTf]+, calc. 615.130).
25 °C): δ = 24.5, 29.8, 55.9, 62.3, 71.3, 78.2, 119.5, 121.5, 125.9, 2MeCN (636 mg, 1.46 mmol) in anhydrous, degassed MeCN
126.4, 126.6, 128.2, 128.3, 136.8, 146.5, 157.4 ppm. IR (solid): (20 mL) was added via a cannula with stirring to a solution of
ν = 3267.6, 2966.6, 2947.0, 2821.8, 1592.1, 1577.5, 1487.2, 7 (1.01 g, 2.93 mmol) in CH2Cl2 (22 mL). Immediately upon
1449.1, 1393.0, 1300.0, 1250.0, 1206.3, 1155.8, 1135.0, 1106.4, addition a colour change to bright yellow was observed. After
1083.3, 1062.1, 1032.5, 962.3, 944.3, 924.9, 895.6, 870.1, 832.1, 60 min stirring at RT the solvent was evaporated in vacuo to
795.0, 767.4, 753.3, 711.1, 703.3, 658.7 cm−1. UV-Vis (MeCN) leave a yellow solid. The product was recrystallized twice by the
[λmax, nm (ε, M−1 cm−1)]: 203 (55 625), 265 (7885). [α]D21 +54.3 addition of Et2O (50 mL, 65 mL) to a solution of the product in
deg cm3 g−1 dm−1 (c 0.6, CHCl3). Anal. for C24H26N2O (358.48): MeCN (10 mL, 12 mL). The yellow supernatants were removed
calc. C 80.41, H 7.31, N 7.81; found C 80.46, H 7.37, N 7.78.
via a cannula. The remaining off-white powder was dried
[Fe(PyProPr)2](OTf)2 (9). A solution of Fe(OTf)2·2MeCN in vacuo (939 mg, 62%). IR (solid): ν = 3338.5 (b), 2972.4, 1610.1,
(725 mg, 1.66 mmol) in anhydrous, degassed MeCN (23 mL) 1574.4, 1491.5, 1449.3, 1430.2, 1277.0, 1243.6, 1222.6, 1156.8,
was added via a cannula with stirring to a solution of 3 (0.83 g, 1110.6, 1074.9, 1055.2, 1027.0, 936.1, 889.4, 844.3, 768.9,
3.3 mmol) in MeCN (23 mL). Immediately upon addition a 747.8, 707.3, 696.3, 663.9 cm−1. UV-Vis (MeCN) [λmax, nm (ε,
colour change to yellow was observed. After 60 min stirring at M−1 cm−1)]: 204 (65 048), 259 (12 570), 331 (3900). [α]D21 –21 deg
RT the solvent was evaporated in vacuo to leave an orange cm3 g−1 dm−1 (c 0.5, MeCN). Anal. for C48H48F6FeN4O8S2
foam. The product was purified twice by the addition of Et2O (1042.88): calc. C 55.28, H 4.64, N 5.37; found C 55.20, H 4.99,
(2 × 30 mL) to a solution of the product in MeCN (2 × 7 mL) N 5.46. ESI-MS: m/z = 893.269 ([M − OTf]+, calc. 893.265). Solu-
with stirring (for 16 h, for 1 h). The colourless supernatants tion magnetic moment (Evans’ method): µeff = 4.58µB.
were removed via a cannula. The remaining red-brown oil was
[Fe(PyProMe)Cl2] (12). A solution of FeCl2 (81 mg,
dried in vacuo to leave an orange foam. Stirring in vacuo 0.64 mmol) in anhydrous, degassed MeOH (4 mL) was added
resulted in a white powder with an orange-pink hue (1.11 g, via a cannula with stirring to a solution of 1 (140 mg,
78%).
0.64 mmol) in MeOH (4 mL). Immediately upon addition a
IR (solid): ν = 2973.1, 1658.7, 1608.8, 1447.4, 1415.6, 1365.6, colour change to yellow was observed. After 60 min stirring at
1254.8, 1222.9, 1148.4, 1104.7, 1081.9, 1052.9, 1028.0, 930.0, RT the solvent was evaporated in vacuo to leave a yellow solid/
845.9, 767.9 cm−1. IR (MeCN): ν = 2250 (νCN), 1664 (νCO), oil. The product was precipitated by the addition of Et2O to a
1274 (νasSO3), 1224 (νsCF3), 1158 (νasCF3), 1034 (νsSO3) cm−1
.
solution of the product in MeOH. The colorless supernatant
UV-Vis (MeCN) [λmax, nm (ε, M−1 cm−1)]: 205 (13 387), 260 was removed via a cannula. The remaining yellow powder was
(6357), 354 (593). [α]2D1 0 deg cm3 g−1 dm−1 (c 0.6, MeCN). Anal. dried in a stream of N2 and subsequently in vacuo (206 mg,
for C30H40F6FeN4O10S2 (850.62): calc. C 42.36, H 4.74, N 6.59; 93%). Slow vapour diffusion of Et2O into a concentrated solu-
found C 42.18, H 4.75, N 6.62. ESI-MS: m/z = 701.191 ([M − tion of 12 in MeOH yielded yellow crystalline needles suitable
OTf]+, calc. 701.192). Solution magnetic moment (Evans’ for X-ray diffraction. Slow solvent diffusion of Et2O into a con-
method): µeff = 4.86µB.
centrated solution of 12 in CH2Cl2 at RT yielded yellow, blocky
[Fe(PyProNH2)2](OTf)2 (10). A solution of Fe(OTf)2·2MeCN crystals suitable for X-ray diffraction.
(476 mg, 1.09 mmol) in anhydrous, degassed MeCN (15 mL)
IR (solid): ν = 3285.4 (b), 2966.7, 2907.4, 1696.6, 1602.3,
was added via a cannula with stirring to a solution of 5 (0.46 g, 1568.7, 1476.8, 1442.0, 1388.4, 1366.0, 1346.2, 1296.9, 1271.8,
2.2 mmol) in MeCN (15 mL). Immediately upon addition a 1254.5, 1171.2, 1099.4, 1077.4, 1050.4, 1025.8, 996.4, 958.6,
colour change to yellow was observed. After 60 min stirring at 931.9, 912.1, 872.8, 859.0, 825.6, 783.1, 771.3, 665.2 cm−1. IR
6782 | Dalton Trans., 2014, 43, 6769–6785
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
(MeCN): ν = 2247 (νCN), 1668 (νCO) cm−1. UV-Vis (MeCN)
IR (solid): ν = 3060.0, 2988.8, 1605.0, 1483.4, 1444.4, 1369.4,
[λmax, nm (ε, M−1 cm−1)]: 291 (866), 363 (501). [α]2D1 +24.9 deg 1343.9, 1325.3, 1275.4, 1176.8, 1152.4, 1066.2, 1049.7, 1011.8,
cm3 g−1 dm−1 (c 0.42, MeCN). Anal. for C13H20Cl2FeN2O3 960.8, 940.3, 905.4, 855.7, 778.4, 752.1, 696.5, 665.0 cm−1
.
(379.06): calc. C 41.19, H 5.32, N 7.39; found C 41.11, H 5.71, UV-Vis (MeCN) [λmax, nm (ε, M−1 cm−1)]: 255 (7096), 292
N 7.26. ESI-MS: m/z = 311.025 ([M − Cl]+, calc. 311.025). Solu- (5319). [α]2D1 –32 deg cm3 g−1 dm−1 (c 0.1, MeCN). Anal. for
tion magnetic moment (Evans’ method): µeff = 4.62µB.
C23H24Cl2FeN2O (471.20): calc. C 58.63, H 5.13, N 5.95; found
[Fe(6-Me-PyProMe)Cl2] (13). Reacting FeCl2 (100 mg, C 58.58, H 5.20, N 5.86. ESI-MS: m/z = 435.091 ([M − Cl]+, calc.
0.79 mmol) with 2 (185 mg, 0.79 mmol) in the manner 435.093). Solution magnetic moment (Evans’ method): µeff
described for 12 yielded 13 as a yellow powder (258 mg, 90%). 5.44µB.
=
Slow vapour diffusion of Et2O into a concentrated solution of
13 in MeOH yielded yellow, blocky crystals suitable for X-ray 0.59 mmol) with 8 (212 mg, 0.59 mmol) in CH2Cl2 (5 mL) in
diffraction. the manner described for 12 yielded 17 as a yellow powder
[Fe(6-Me-PyProPh2OH)Cl2] (17). Reacting FeCl2 (75 mg,
IR (solid): ν = 2982.4, 2925.3, 1687.4, 1604.5, 1571.2, 1442.6, (245 mg, 85%). Slow solvent diffusion of Et2O into a concen-
1391.1, 1313.1, 1266.0, 1223.0, 1182.3, 1162.0, 1082.4, 1049.8, trated solution of 17 in CH2Cl2 yielded yellow, blocky crystals
1033.1, 1012.0, 984.0, 941.0, 921.5, 872.8, 839.9, 818.2, 786.5, suitable for X-ray diffraction.
733.3 cm−1. IR (MeCN): ν = 2250 (νCN), 1702 (νCO) cm−1
.
IR (solid): ν = 3059.0, 2974.3, 1606.2, 1574.6, 1491.4, 1463.1,
UV-Vis (MeCN) [λmax, nm (ε, M−1 cm−1)]: 207 (9910), 267 1446.7, 1382.0, 1345.3, 1321.9, 1274.4, 1160.9, 1084.7, 1067.3,
(4282), 369 (338). [α]2D1 +29.6 deg cm3 g−1 dm−1 (c 0.448, 1037.2, 1005.9, 907.7, 853.9, 782.9, 747.4, 705.5, 695.0,
MeCN). Anal. for C13H18Cl2FeN2O2 (361.05): calc. C 43.25, H 657.4 cm−1. UV-Vis (MeCN) [λmax, nm (ε, M−1 cm−1)]: 205
5.03, N 7.76; found C 43.18, H 5.14, N 7.83. ESI-MS: m/z = (21 889), 261 (7282), 316 (4139). [α]2D1 +19 deg cm3 g−1 dm−1
325.041 ([M − Cl]+, calc. 325.041). Solution magnetic moment (c 0.1, MeCN). Anal. for C24H26Cl2FeN2O (485.23): calc.
(Evans’ method): µeff = 5.07µB.
C 59.41, H 5.40, N 5.77; found C 59.33, H 5.36, N 5.66. ESI-MS:
[Fe(PyProNH2)Cl2] (14). Reacting FeCl2 (104 mg, 0.82 mmol) m/z = 449.106 ([M − Cl]+, calc. 449.108). Solution magnetic
with 5 (168 mg, 0.82 mmol) in the manner described for moment (Evans’ method): µeff = 5.23µB.
12 yielded 14 as a yellow powder (245 mg, 90%). Slow
Protocol for catalytic olefin oxidation
vapour diffusion of Et2O into a concentrated solution of 14 in
MeCN yielded yellow, blocky crystals suitable for X-ray To a solution of Fe-complex (3.5 μmol) in anhydrous, degassed
diffraction.
acetonitrile was added substrate (1.75 mmol) to obtain a total
IR (solid): ν = 3315.1, 3162.0, 2972.0, 1652.3, 1602.1, 1476.1, volume of 3.0 mL. Subsequently, 0.5 mL of oxidant solution
1444.5, 1383.9, 1299.6, 1155.1, 1102.8, 1052.1, 1019.1, 963.2, (700 mM solution in acetonitrile diluted from 35% aqueous
927.8, 766.5 cm−1. IR (MeCN): ν = 2247 (νCN), 1668 (νCO) H2O2) was added portionwise for 30 min. The reaction mixture
cm−1. UV-Vis (MeCN) [λmax, nm (ε, M−1 cm−1)]: 385 (469). was stirred at room temperature and after 1 h (from start of
[α]2D1 +60 deg cm3 g−1 dm−1 (c 0.3, MeOH). ESI-MS: m/z = oxidant addition) internal standard (10 μL, cyclooctene: 1,2-
296.026 ([M − Cl]+, calc. 296.025). Solution magnetic moment dibromobenzene, all other substrates: bromobenzene) was
(Evans’ method): µeff = 4.33µB.
added and the first sample was taken. To an aliquot of the
[Fe(6-Me-PyProNH2)Cl2] (15). Reacting FeCl2 (102 mg, reaction mixture was added diethyl ether. The sample was ana-
0.80 mmol) with 6 (176 mg, 0.80 mmol) in the manner lysed by GC. The products were identified and quantified by
described for 12 yielded 15 as a yellow powder (260 mg, 94%). comparison with authentic compounds.
Slow vapour diffusion of Et2O into a concentrated solution of
X-ray crystal structure determinations
15 in MeOH yielded yellow, needle-shaped crystals suitable for
X-ray diffraction.
Reflections were measured on a Nonius KappaCCD diffract-
IR (solid): ν = 3322.0, 3250.9, 3170.8, 2964.0, 2900.9, 1656.9, ometer with a rotating anode and a graphite monochromator
1604.7, 1591.4, 1467.2, 1455.7, 1384.6, 1357.3, 1304.8, 1218.3, (λ = 0.71073 Å). Intensity data were integrated with the software
1168.5, 1123.8, 1101.0, 1015.0, 895.1, 797.6, 781.8 cm−1
.
Eval1442 (compound 10), Eval1543 (compound 15) or
UV-Vis (MeCN) [λmax, nm (ε, M−1 cm−1)]: 385 (456). [α]2D1 HKL200044 (compounds 12a, 12b, 13, 14, 16, 17). Absorption
+71 deg cm3 g−1 dm−1 (c 0.4, MeOH). Anal. for correction and scaling was performed based on multiple
C12H17Cl2FeN3O (346.03): calc. C 41.65, H 4.95, N 12.14; found measured reflections with SADABS45 (10–15) or SORTAV46 (16,
C 41.48, H 4.91, N 11.96. ESI-MS: m/z = 310.040 ([M − Cl]+, 17). The structures were solved by direct methods using the
calc. 310.041). Solution magnetic moment (Evans’ method): programs SHELXS-9747 (10, 12b, 13–17) or SIR-9748 (12a).
µeff = 5.10µB.
Least-squares refinement was performed using SHELXL-9747
[Fe(PyProPh2OH)Cl2] (16). Reacting FeCl2 (72 mg, against F2 of all reflections. Non-hydrogen atoms were refined
0.57 mmol) with 7 (197 mg, 0.57 mmol) in CH2Cl2 (5 mL) in freely with anisotropic displacement parameters. Hydrogen
the manner described for 12 yielded 16 as a yellow powder atoms were introduced in calculated positions (10, 12a, 13) or
(242 mg, 90%). Slow solvent diffusion of Et2O into a concen- located in difference Fourier maps (12b, 14–17). Hydrogen
trated solution of 16 in CH2Cl2 yielded yellow, blocky crystals atoms were refined with a riding model, the N–H or O–H
suitable for X-ray diffraction.
hydrogen atoms of 12b, 14, and 15 were refined freely with
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 6769–6785 | 6783

View Article Online
Paper
Dalton Transactions
isotropic displacement parameters. The crystal structure of 10 10 (a) K. Wieghardt, K. Pohl and W. Gebert, Angew. Chem., Int.
was refined as a twin with (−1,0,0/0,−1,0/1,0,1) as twin matrix
(pseudo-orthorhombic C-centred lattice). The twin fraction
refined to 0.4877(13). The triflate anions in 10 were refined
Ed. Engl., 1983, 22, 727; (b) G. B. Shul’pin, G. V. Nizova,
Y. N. Kozlov, L. Gonzales Cuervo and G. Süss-Fink, Adv.
Synth. Catal., 2004, 346, 317.
with disorder models using restraints for distances and angles 11 A. Beck, B. Weibert and N. Burzlaff, Eur. J. Inorg. Chem.,
and for the approximation of isotropic behaviour. Geometry 2001, 521.
calculations and checking for higher symmetry were per- 12 A. Beck, A. Barth, E. Hübner and N. Burzlaff, Inorg. Chem.,
formed using the PLATON program.49 Further details of
2003, 42, 7182.
the crystal structure determinations are given in the ESI 13 P. D. Oldenburg, A. A. Shteinman and L. Que Jr., J. Am.
Tables S8–S12.† Chem. Soc., 2005, 127, 15672.
CCDC 972756–972763 contain the supplementary crystallo- 14 S. J. Friesse, B. E. Kucera, L. Que Jr. and W. B. Tolman,
graphic data for this paper.
Inorg. Chem., 2006, 45, 8003.
15 (a) P. C. A. Bruijnincx, M. Lutz, A. L. Spek, E. E. van
Faassen, B. M. Weckhuysen, G. van Koten and
R. J. M. Klein Gebbink, Eur. J. Inorg. Chem., 2005, 779;
(b) P. C. A. Bruijnincx, M. Lutz, J. P. den Breejen,
A. L. Spek, G. van koten and R. J. M. Klein Gebbink, J. Biol.
Inorg. Chem., 2007, 12, 1181; (c) 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; (d) M. A. H.
Moelands, S. Nijsse, E. Folkertsma, B. de Bruin, M. Lutz,
A. L. Spek and R. J. M. Klein Gebbink, Inorg. Chem., 2013,
52, 7394–7410.
Acknowledgements
The work described here was financially supported by the
National Research School Combination-Catalysis (NRSC-C).
Notes and references
1 B. Meunier, Biomimetic Oxidations Catalyzed by Transition
Metal Complexes, Imperial College Press, London, 1999.
2 M. Costas, M. P. Mehn, M. P. Jensen and L. Que Jr., Chem. 16 P. C. A. Bruijnincx, M. Lutz, A. L. Spek, W. R. Hagen,
Rev., 2004, 104, 939.
B. M. Weckhuysen, G. van Koten and R. J. M. Klein
3 (a) A. C. Rosenzweig, C. A. Frederick, S. J. Lippard and
Gebbink, J. Am. Chem. Soc., 2007, 129, 2275.
P. Nordlund, Nature, 1993, 366, 537; (b) A. C. Rosenzwieg, 17 L. Peters, E. Hubner and N. Burzlaff, J. Organomet. Chem.,
P. Nordlund, P. M. Takahara, C. A. Frederick and 2005, 690, 2009.
S. J. Lippard, Chem. Biol., 1995, 2, 409; (c) N. Elango, 18 N. Burzlaff, Angew. Chem., Int. Ed., 2009, 48, 5580.
R. Radhakrishnam, W. A. Froland, B. J. Waller, 19 S. S. Rocks, W. W. Brennessel, T. E. Machonkin and
C. A. Earhart, J. D. Lipscomb and D. H. Ohlendorf, Protein
Sci., 1997, 6, 556.
4 P. C. A. Bruijnincx, G. van Koten and R. J. M. Klein
Gebbink, Chem. Soc. Rev., 2008, 37, 2716.
5 (a) E. G. Kovaleva and J. D. Lipscomb, Nat. Chem. Biol.,
2008, 4, 186; (b) K. D. Koehntop, J. P. Emerson and L. Que
Jr., J. Biol. Inorg. Chem., 2005, 10, 87.
6 E. L. Hegg, R. Y. K. Ho and L. Que Jr., J. Am. Chem. Soc.,
1999, 121, 1972.
P. L. Holland, Inorg. Chim. Acta, 2009, 362, 1387.
20 S. Paria, P. Halder and T. K. Paine, Inorg. Chem., 2010, 49,
4518.
21 (a) V. J. Dungan, B. M.-L. Poon, E. S. Barrett and
P. J. Rutledge, Tetrahedron Lett., 2013, 54, 1236;
(b) V. J. Dungan, S. M. Wong, S. M. Barry and P. J. Rutledge,
Tetrahedron, 2012, 68, 3231. No structural data are
presented in these studies.
22 (a) S. Gosiewska, J. L. M. Cornelissen, M. Lutz, A. L. Spek,
G. van Koten and R. J. M. Klein Gebbink, Inorg. Chem.,
2006, 45, 4214; (b) S. Gosiewska, M. Lutz, A. L. Spek and
R. J. M. Klein Gebbink, Inorg. Chim. Acta, 2007, 360, 405.
7 M. P. Mehn, K. Fujisawa, E. L. Hegg and L. Que Jr., J. Am.
Chem. Soc., 2003, 125, 7828.
8 (a) Y. Dong, H. Fujii, M. P. Hendrich, R. A. Leising, G. Pan,
C. R. Randall, E. C. Wilkinson, Y. Zang, L. Que Jr., 23 G. Chelucci, M. Falorni and G. Giacomelli, Tetrahedron:
B. G. Fox, K. Kauffmann and E. Münck, J. Am. Chem. Soc., Asymmetry, 1990, 1, 843.
1995, 117, 2778; (b) Y. Zang, J. Kim, Y. Dong, 24 P. Å. Bergström and R. French, J. Phys. Chem., 1995, 99,
E. C. Wilkinson, E. H. Appelman and L. Que Jr., J. Am. 12603.
Chem. Soc., 1997, 119, 4197; (c) C. Kim, K. Chen, J. Kim 25 G. J. P. Britovsek, V. C. Gibson, S. K. Spitzmesser,
and L. Que Jr., J. Am. Chem. Soc., 1997, 119, 5964;
(d) K. Chen and L. Que Jr., J. Am. Chem. Soc., 2001, 123,
K. P. Tellmann, A. J. P. White and D. J. Williams, J. Chem.
Soc., Dalton Trans., 2002, 1159.
6327; (e) K. Chen, M. Costas, J. Kim, A. Tipton and L. Que 26 D. F. Evans, J. Chem. Soc., 1959, 2003.
Jr., J. Am. Chem. Soc., 2002, 124, 3026; (f) J. Y. Ryu, J. Kim, 27 A. W. Addison, T. N. Rao, J. Reedijk, J. van Rijn and
M. Costas, K. Chen, W. Nam and L. Que Jr., Chem.
Commun., 2003, 1288.
G. C. Verschoor, J. Chem. Soc., Dalton Trans., 1984,
1349.
9 M. Costas, A. K. Tipon, K. Chen, D.-W. Jo and L. Que Jr., 28 G. G. Evans and J. A. Boeyens, Acta Crystallogr., Sect. B:
J. Am. Chem. Soc., 2001, 123, 6722.
Struct. Sci., 1989, 45, 581.
6784 | Dalton Trans., 2014, 43, 6769–6785
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
29 M. A. H. Moelands, S. Nijsse, E. Folkertsma, B. de Bruin,
M. Lutz, A. L. Spek and R. J. M. Klein Gebbink, Inorg.
Chem., 2013, 52, 7394.
30 B. L. Small and M. Brookhart, Macromolecules, 1999, 32,
2120.
31 F. Pelascini, M. Wesolek, F. Peruch and P. J. Lutz,
Eur. J. Inorg. Chem., 2006, 21, 4309.
32 C. Dowling, V. J. Murphy and G. Parkin, Inorg. Chem., 1996,
35, 2415.
D. J. Williams, J. Am. Chem. Soc., 2003, 125, 8450;
( j) F. A. R. Kaul, G. T. Puchta, G. D. Frey, E. Herdtweck and
W. A. Herrmann, Organometallics, 2007, 26, 988;
(k) N. J. Hardman, X. Fang, B. L. Scott, R. J. Wright,
R. L. Martin and G. J. Kubas, Inorg. Chem., 2005, 44, 8306;
(l) W.-S. Wu and S.-X. Liu, Chem. J. Chin. Univ., 2003, 24,
2137; (m) Y. Yamaguchi, H. Ando, M. Nagaya, H. Hinago,
T. Ito and M. Asami, Chem. Lett., 2011, 40, 983.
37 H. Vahrenkamp, Acc. Chem. Res., 1999, 32, 589.
33 M. Biswas, G. Pilet, M. S. El Fallah, J. Ribas and S. Mitra, 38 A. Company, L. Gómez, X. Fontrodona, X. Ribas and
Inorg. Chim. Acta, 2008, 361, 387. M. Costas, Chem.–Eur. J., 2008, 14, 5727.
34 A. J. Amoroso, M. W. Burrows, S. J. Coles, R. Haigh, 39 F. G. Gelalcha, B. Bitterlich, G. Anilkumar, M. K. Tse and
R. D. Farley, M. B. Hursthouse, M. Jones, K. M. A. Malik
and D. M. Murphy, Dalton Trans., 2008, 506.
M. Beller, Angew. Chem., Int. Ed., 2007, 46, 7293.
40 K. S. Hagen, Inorg. Chem., 2000, 39, 5867.
35 F. H. Allen, Acta Crystallogr., Sect. B: Struct. Sci., 2002, 58, 41 G. Suster, S. A. Stoykova, A. Linden and H. Heimgartner,
380–388. Helv. Chim. Acta, 2000, 83, 2961.
36 (a) G. Argay, A. Kalman and V. M. Leovac, Acta Crystallogr., 42 A. J. M. Duisenberg, L. M. J. Kroon-Batenburg and
Sect. C: Cryst. Struct. Commun., 1999, 55, 1440; A. M. M. Schreurs, J. Appl. Crystallogr., 2003, 36, 220.
(b) M. A. Yampol’skaya, S. G. Shova, Yu. A. Simonov, 43 A. M. M. Schreurs, X. Xian and L. M. J. Kroon-Batenburg,
N. V. Gerbeleu, K. I. Turte and V. E. Zavodnik, Russ. J. Inorg. J. Appl. Crystallogr., 2010, 43, 70.
Chem., 1985, 30, 1221; (c) P. V. Bernhardt, P. Chin, 44 Z. Otwinowski and W. Minor, Methods in Enzymology,
P. C. Sharpe, J.-Y. C. Wang and D. R. Richardson, J. Biol.
Inorg. Chem., 2005, 10, 761; (d) D. Urakami, K. Inoue and
Volume 276, ed. C. W. Carter Jr. and R. M. Sweet, Academic
Press, 1997, pp. 307–326.
S. Hayami, Acta Crystallogr., Sect. E: Struct. Rep. Online, 45 G. M. Sheldrick, SADABS: Area-Detector Absorption Correc-
2009, 65, m800; (e) H. Liu, Y. Kou, L. Feng, D. Li, C.-Y. Gao, tion, v2.10, Universität Göttingen, Germany, 1999.
J. Tian, J. Zhang and S. Yan, Appl. Organomet. Chem., 2010, 46 R. H. Blessing, Acta Crystallogr., Sect. A: Fundam. Crystal-
24, 636; (f) Z. D. Tomic, V. M. Leovac, Z. K. Jacimovic, logr., 1995, 51, 33.
G. Giester and S. D. Zaric, Inorg. Chem. Commun., 2006, 9, 47 G. M. Sheldrick, Acta Crystallogr., Sect. A: Fundam. Crystal-
833; (g) N. Kleigrewe, W. Steffen, T. Blomker, G. Kehr, logr., 2008, 64, 112.
R. Frohlich, B. Wibbeling, G. Erker, J.-C. Wasilke, G. Wu 48 A. Altomare, M. C. Burla, M. Camalli, G. L. Cascarano,
and G. C. Bazan, J. Am. Chem. Soc., 2005, 127, 13955;
(h) V. C. Gibson, C. Redshaw, G. A. Solan, A. J. P. White and
C. Giacovazzo, A. Guagliardi, A. G. G. Moliterni, G. Polidori
and R. Spagna, J. Appl. Crystallogr., 1999, 32, 115.
D. J. Williams, Organometallics, 2007, 26, 5119; 49 A. L. Spek, Acta Crystallogr., Sect. D: Biol. Crystallogr., 2009,
(i) R. K. O’Reilly, V. C. Gibson, A. J. P. White and
65, 148.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 6769–6785 | 6785