Bioinspired nonheme iron complexes derived from an extended series of N,N,O-ligated BAIP ligands

Article abstract of DOI:10.1021/ic400096e

A series of mononuclear Fe(II) triflate complexes based on the 3,3-bis(1-alkylimidazole-2-yl)propionate ester (BAIP) ligand scaffold are reported. In these complexes, the tripodal N,N,O-BAIP ester ligand is varied by (i) changing the ester moiety (i.e., n-Pr, tert-Bu esters, n-Pr amide), (ii) changing the methylimidazole moieties to methylbenzimidazole moieties, and (iii) changing the methylimidazole moieties to 1-ethyl-4-isopropylimidazole moieties. The general structure of the resulting complexes comprises two facially capping BAIP ligands around a coordinatively saturated octahedral Fe(II) center, with either a transoid or cisoid orientation of the N,N,O-donor manifold that depends on the combined steric and electronic demand of the ligands. In the case of the sterically most encumbered ligand, a four-coordinate all N-coordinate complex is formed as well, which cocrystallizes with the six-coordinate complex. In combination with the catalytic properties of the new complexes in the epoxidation/cis-dihydroxylation of cyclooctene with H₂O₂, in terms of turnover number and cis-diol formation, these studies provide a number of insights for further ligand design and catalyst development aimed at Fe-mediated cis-dihydroxylation.

Full text of DOI:10.1021/ic400096e

Article
pubs.acs.org/IC
Bioinspired Nonheme Iron Complexes Derived from an Extended
Series of N,N,O-Ligated BAIP Ligands
Marcel A. H. Moelands,† Sjoerd Nijsse,† Emma Folkertsma,† Bas de Bruin,‡ Martin Lutz,‡,§
Anthony L. Spek,‡,§ and Robertus J. M. Klein Gebbink*,†
†Organic Chemistry and Catalysis, Debye Institute for Nanomaterials Science, Faculty of Science, Utrecht University,
Universiteitsweg 99, 3584 CG Utrecht, The Netherlands
‡Van’t Hoff Institute for Molecular Sciences (HIMS), University of Amsterdam, P.O. Box 94720, 1090 GE Amsterdam, The
Netherlands
§Bijvoet Center for Biomolecular Research, Crystal and Structural Chemistry, Faculty of Science, Utrecht University, Padualaan 8,
3584 CH Utrecht, The Netherlands
S
* Supporting Information
ABSTRACT: A series of mononuclear Fe(II) triflate complexes based on the 3,3-
bis(1-alkylimidazole-2-yl)propionate ester (BAIP) ligand scaffold are reported. In
these complexes, the tripodal N,N,O-BAIP ester ligand is varied by (i) changing the
ester moiety (i.e., n-Pr, tert-Bu esters, n-Pr amide), (ii) changing the methylimidazole
moieties to methylbenzimidazole moieties, and (iii) changing the methylimidazole
moieties to 1-ethyl-4-isopropylimidazole moieties. The general structure of the
resulting complexes comprises two facially capping BAIP ligands around a
coordinatively saturated octahedral Fe(II) center, with either a transoid or cisoid
orientation of the N,N,O-donor manifold that depends on the combined steric and
electronic demand of the ligands. In the case of the sterically most encumbered ligand,
a four-coordinate all N-coordinate complex is formed as well, which cocrystallizes with
the six-coordinate complex. In combination with the catalytic properties of the new
complexes in the epoxidation/cis-dihydroxylation of cyclooctene with H2O2, in terms
of turnover number and cis-diol formation, these studies provide a number of insights for further ligand design and catalyst
development aimed at Fe-mediated cis-dihydroxylation.
Unfortunately, these ruthenium systems are not as selective as
the osmium ones, and ruthenium is considered to be an
expensive metal. Alternatively, KMnO4 may be used for alkene
cis-dihydroxylation, although stoichiometric amounts of this
oxidant have to be used.8,9 Catalytic systems based on
manganese have been reported by Feringa10 and Che.11
The answer to finding a good alternative to osmium in cis-
dihydroxylation reactions might be found in nature, where
several classes of metalloenzymes catalyze a wide variety of
oxidation reactions. Rieske oxygenases are one important class
of enzymes from the perspective of cis-dihydroxylations.12
These dioxygenases are found in soil bacteria and catalyze the
first step in the biodegradation of aromatic compounds by
performing a cis-dihydroxylation on the arene substrate.13
Because this reaction is regio- and stereospecific, it was
investigated extensively over the past two decades, resulting in a
range of crystallographic and mechanistic studies. The enzyme
that has been most studied is naphthalene dioxygenase (NDO),
which catalyzes the cis-dihydroxylation of naphthalene to cis-
(1R,2S)-1,2-dihydronaphtalene-1,2-diol. The combined crystal-
INTRODUCTION
■
Oxidation reactions are important in the production of bulk
and fine chemicals. More than 20% of all organic products
made in the chemical industry are obtained via catalytic
oxidation reactions.1 The catalysts used in these reactions have
an additional role when the stereochemistry of the oxidation
product is of interest, for instance, in the production of building
blocks for the pharmaceutical industry. Of particular interest in
this respect are catalysts that are able to bring about the
stereoselective cis-dihydroxylation of alkenes to their corre-
sponding vicinal diols.
Several procedures are known to oxidize selectively alkene
substrates to their corresponding cis-diols, but in most cases
these reactions are associated with several unwanted drawbacks
or problems.2 For example, osmium tetraoxide can be used
catalytically in the presence of a secondary oxidant (e.g.,
hydrogen peroxide) to facilitate alkene cis-dihydroxylation.3
The Sharpless AD-mixes α and β are commercially available
osmium-based reagent mixtures for the asymmetric dihydrox-
ylation of alkenes.4 Although these systems are very reliable, the
obvious disadvantages of the use of osmium are its price and
toxicity. Several examples of alkene cis-dihydroxylation using
ruthenium as an alternative to osmium have been reported.5−7
Received: January 14, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ic400096e | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
lographic data on Rieske dioxygenases show that the catalytic
sites of these dioxygenases are very similar. Each consists of a
mononuclear iron center anchored to the enzyme by two
histidine residues and a mono- or bidentate carboxylate ligand
from either an aspartate or a glutamate residue (Figure 1).
A typical feature of the above systems is that they all contain
an all-nitrogen-ligand donor set, instead of the mixed N,O
donor set found in the Rieske dioxygenases. In addition, these
systems do not follow the facially capping tridentate-designer
mold laid down by the 2-His-1-carboxylate facial triad. In 2005,
Que and co-workers reported on the Ph-DPAH ligand (DPAH
= di(2-pyridyl)methylbenzamide) that has an N,N,O donor
set.16 Corresponding iron complex [Fe(II)(Ph-DPAH)2]-
(OTf)2 has an octahedral geometry in which the two carbonyl
oxygen donors are trans to each other. Although this iron
complex is coordinatively saturated, it does catalyze the
oxidation of olefins. During these reactions, one ligand most
likely dissociates from the iron center to make three sites
available for substrate and hydrogen peroxide binding. The
complex forms the cis-diol as the major product for a range of
olefins, with conversions of 50−80% depending on the amount
of hydrogen peroxide added, and does so with a retention of
the olefin configuration.
Along a similar vein, other N,N,O ligands are currently being
explored to mimic the structural aspects of the 2-His-1-
carboxylate facial triad more accurately. Examples of ligands
that are used in these studies in combination with iron include
the scorpionate-type bis(pyrazol-1-yl) acetates19 and mixed
proline−pyridine ligands.20 Iron complexes of bispyrazolyl
acetate ligands were quite extensively studied by Burzlaff and
coworkers,19c whereas more recently, Jones et al. have
developed some sterically encumbered versions of these
ligands.21
In recent articles, we have reported on the syntheses of
iron(II) complexes derived from a new class of facial N,N,O
ligands.22−24 The tripodal 3,3-bis(1-alkylimidazole-2-yl)-
propionate (BAIP) ligands contain two imidazole groups and
a carboxylate moiety to mimic the histidine and aspartate
residues in the active site of NDO (Figure 2). These anionic
N,N,O ligands form mononuclear iron complexes in the
presence of an additional anionic ligand, for instance,
catecholate. In the absence of such additional ligands,
coordinatively saturated [Fe(BAIP)2] complexes are formed
that were found to be inactive as oxidation catalysts.23
Changing the carboxylate into a carboxylate ester changes the
coordination properties of these ligands. Depending on the
counterion and the solvent that are used, either a trans bis-
ligand complex or bis-solvent bis-ligand adduct is formed
(Figure 2). In the latter adduct, the BMIPnPr ligands (BMIPnPr
= propyl 3,3-bis(1-methylimidazole-2-yl)propionate) act as
N,N donors and two MeOH molecules coordinate to iron in
Figure 1. Structure of the monoiron(II) active site in naphthalene 1,2-
dioxygenase.
Depending on the coordination mode of the carboxylate, this
active-site architecture is referred to as either the 2-His-1-
carboxylate facial triad (monodentate carboxylate)14 or the 2-
His-1-carboxylate structural motif (bidentate carboxylate bind-
ing).12 Because iron adopts an octahedral geometry in these
enzymes, there are respectively three or two vacant sites left to
bind the substrate, dioxygen, and/or cofactor ligands. If none of
these are bound, then the sites are either occupied by weakly
bonded solvent molecules or are vacant.
Interest in the catalytic properties of the Rieske dioxygenase
has led to several endeavors that are trying to grasp its structure
and reactivity in synthetic model systems. The structure of the
2-His-1-carboxylate facial triad structural motif has, in addition,
become a designer platform for the development of synthetic
Fe-based dihydroxylation catalysts. The group of Que and
coworkers was the first to synthesize a successful catalyst based
on the active site of NDO.15 Compound [Fe(II)(6-Me3-
tpa)(CH3CN)2](ClO4)2 (6-Me3-tpa = tris[(6-methylpyridin-2-
yl)methyl]amine) is able to convert a number of aliphatic
alkenes to their corresponding cis-diol products (in combina-
tion with some epoxide byproducts) with the use of hydrogen
peroxide as the oxidant. As a critical structural feature, the Fe(6-
Me3-tpa) complex contains two labile cis-coordination sites that
are occupied by CH3CN molecules. Que and co-workers later
reported on the synthesis of the chiral Fe(II) complex [Fe(6-
Me-BPBP)(OTf)2] (BPBP = bis(pyridinylmethyl)-
bipyrrolidine) that is able to form cis-diols from olefins in up
to 97% ee.17 These systems are currently the most selective
iron-based catalysts for the dihydroxylation of olefins. Other Fe-
based systems able to carry out cis-dihydroxylations include
systems based on bispidine and on (bi)cyclic tetraaza ligands.18
Figure 2. Ligands 3,3-bis(1-methylimidazole-2-yl)propionate (BMIP) and propyl 3,3-bis(1-methylimidazole-2-yl)propionate (BMIPnPr) and some
of their iron(II) complexes.
B
dx.doi.org/10.1021/ic400096e | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 3. Overview of BAIP-type ligands.
Scheme 1. Synthesis of Ligands 3−6
a
a
(i) (a) n-BuLi, THF, −78 °C, 1 h; (b) propyl bromoacetate, −78 °C to rt, overnight. (ii) (a) n-BuLi, THF, −78 °C, 1 h; (b) tert-butyl
bromoacetate, −78 °C to rt, overnight.
a
Scheme 2. Synthesis of Amide Ligands 9 and 12
a
(i) (a) One equiv KOH, THF, rt; (b) one equiv HCl. (ii) NHS, DCC, pyridine, THF, 35−40 °C . (iii) Propylamine, CH2Cl2, rt.
a cis fashion. When the trans bis-ligand complex [Fe-
(BMIPnPr)2](OTf)2 is used in the oxidation of alkenes with
H2O2, in addition to the formation of the epoxide product, the
formation of the cis-diol product is also observed.23 The
formation of the cis-diol products indicates the flexibility of the
ligand system because for the cis-diol product to be formed
two-labile cis sites are deemed necessary on the metal center.
The [Fe(BMIPnPr)2](OTf)2 complex most likely rearranges in
solution to a structure that is similar to that observed for the
MeOH adduct.
divided into three different groups (Figure 3). In the first
group, the two 1-methylimidazole donors of the BMIP
structure were combined with different ester or amide donors.
In the second group, the 1-methylimidazole donors were
substituted for 1-methylbenzimidazole donors to induce more
steric bulk around the nitrogen donors and to make the ligand
more soluble in apolar solvents. In the third group, the 1-
methylimidazole donors of the ligand were replaced with 1-
ethyl-4-isopropylimidazole donors. The coordination behavior
of these N,N,O ligands toward iron and the catalytic behavior
of the resulting complexes in the oxidation of cyclooctene were
investigated.
Here, we report on an extension of this work by the synthesis
of several analogues of the BMIPnPr ligand in the search for
more selective cis-dihydroxylation catalysts based on iron. A
number of different ligands were investigated, which can be
C
dx.doi.org/10.1021/ic400096e | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Scheme 3. Synthesis of Iron Triflate Complexes 14−20 (Generalized Structure Shown)
prepared complexes were stored as dry powders under a
nitrogen atmosphere. Benzimidazole amide complex 19 proved
quite sensitive toward oxidation, even under these storage
conditions. The color of the dry powder of 19 turned red
overnight, indicating the oxidation of Fe(II) to Fe(III) in this
complex.
High-resolution mass spectroscopy analysis (ESI-MS) proved
very conclusive in the composition of the isolated materials. In
all cases, monocations of the composition [Fe(L2)(OTf)]+
were recorded as the parent peak. This composition is in
agreement with the anticipated composition of the complexes
as [Fe(L2)](OTf)2 on the basis of the first-generation complex
[Fe(BMIPnPr)2](OTf)2.23
IR spectra were recorded for the different complexes, both in
the solid state and in solution. The IR spectrum of
[Fe(BMIPnPr)2](OTf)2 (14) in the solid state showed a clear
vibrational band of the carbonyl group at 1691 cm−1, whereas
the frequency of the carbonyl vibration in the free ligand was
found at 1727 cm−1. This frequency shift indicates the
coordination of the carbonyl group to the iron center. The
symmetric and asymmetric vibrations of the CF3 and SO3
groups of the triflate anions in [Fe(BMIPnPr)2](OTf)2 were
found at 1259 (νas SO3), 1216 (νs CF3), 1152 (νas CF3), and
1030 (νs SO3) cm−1, which indicates that these are non-
coordinating ions.32 Similar observations were made in the
solid state for complexes 15, 18, and 19, pointing to solid-state
structures that are similar that of 14 (Table 1).
RESULTS AND DISCUSSION
Ligand Synthesis. Ligands BMIPnPr (3) and BMBIPnPr
■
(4) were synthesized according to a synthesis route previously
reported by our group25 that was also employed for the
synthesis of ligands BMIPtBu (5) and BMBIPtBu (6) (Scheme
1). The starting bis(1-methylimidazole-2-yl)methane (1)26,27
and bis(1-methylbenzimidazole-2-yl)methane (2)28 com-
pounds can be easily synthesized on a multigram scale.
Lithiation of these bis(imidazole)methanes at the methylene
position using n-BuLi at −78 °C followed by the addition of
either n-propylbromoacetate or tert-butylbromoacetate yielded
propyl 3,3-bis(1-methylimidazole-2-yl)propionate, BMIPnPr
(3), propyl 3,3-bis(1-methylbenzimidazole-2-yl)propionate,
BMBIPnPr (4), tert-butyl 3,3-bis(1-methylimidazole-2-yl)-
propionate, BMIPtBu (5), and tert-butyl 3,3-bis(1-methylbenzi-
midazole-2-yl)propionate, BMBIPtBu (6), respectively, in good
yields (83−94%).
The preparation of the amide-appended ligands 3,3-bis(1-
methylimidazol-2-yl)-N-propylpropanamide, BMIPAnPr (9),
and 3,3-bis(1-methylbenzimidazol-2-yl)-N-propylpropanamide,
BMBIPAnPr (12), proceeded via the preparation of an active
ester intermediate (Scheme 2). First, propyl ester ligands 3 and
4 were hydrolyzed using 1 equiv of KOH.25 The resulting
carboxylic acids were turned into active esters by DCC coupling
according to a modified procedure by Suijkerbuijk et al.29 The
yield of 8 (43%) was much lower in comparison to the yield of
11 (85%), which was caused by the low solubility of acid 7 in
THF. In the final step, the active esters were reacted with
propylamine to yield the desired amide ligands 9 and 12 in 25
and 90% isolated yields after recrystallization, respectively. The
preparation of the BEiPrIPnPr ligand (13) with two 1-ethyl-4-
isopropylimidazole donors was previously reported by us.30
Fe Triflate Complexes. For the set of seven ligands, the
corresponding iron bis-ligand complexes were synthesized in
the same way as was reported for the BMIPnPr complex
(Scheme 3).23 For the preparation of the complexes, the
appropriate ligand was mixed in a 2:1 ratio with Fe-
(OTf)2·2MeCN31 in methanol for 1 h. After the solvent was
removed, recrystallization of the resulting brown or (off) white
powders from an acetonitrile/diethyl ether mixture yielded iron
complexes [Fe(BMIPnPr)2](OTf)2 (14),23 [Fe(BMIPtBu)2]-
(OTf)2 (15), [Fe(BMBIPnPr)2](OTf)2 (17), [Fe-
(BMBIPtBu)2](OTf)2 (18), [Fe(BMBIPAnPr)2](OTf)2 (19),
and [Fe(BEiPrIPnPr)2](OTf)2 (20) as crystalline solids in
high yields (79−95%); however, the yield of 20 was much
lower (34%) because of the two consecutive crystallization
steps in its purification. Corresponding iron bis-ligand complex
16, derived from ligand BMIPAnPr, was not synthesized as
isolated material but was instead prepared in situ for catalytic
testing.
Table 1. Solid-State IR Vibrations of Complexes 14, 15, 18,
and 19
triflate
ligand
complex
vibrations
complex
ν(CO)(cm−1
)
ν(CO)(cm−1
)
ν(cm−1
)
[Fe(BMIPnPr)2]
(OTf)2 (14)
[Fe(BMIPtBu)2]
(OTf)2 (15)
[Fe(BMBIPtBu)2]
(OTf)2 (18)
[Fe(BMBIPAnPr)2]
(OTf)2 (19)
1727
1726
1720
1656
1691
1692
1665
1628
1259, 1216,
1152, 1030
1258, 1224,
1144, 1029
1260, 1226,
1150, 1030
1244, 1223,
1151, 1028
In the case of benzimidazole complex [Fe(BMBIPnPr)2]-
(OTf)2 (17), two carbonyl-stretching vibrations were found: a
strong vibration at 1694 cm−1 and a shoulder at 1709 cm−1 that
both point to coordinating carbonyl groups (free ligand at 1734
cm−1). It is likely that both the cis and trans isomers of the
complex are present in the solid state. For [Fe(BEiPrIPnPr)2]-
(OTf)2 (20), several carbonyl-stretching vibrations were
observed. Just as in 17, two bands that correspond to
coordinating carbonyl groups were found (1705, 1691 cm−1),
again implying the presence of both cis and trans isomers. In
addition, a strong vibration was found for a noncoordinating
carbonyl moiety at 1736 cm−1 (free ligand 1734 cm−1). This
These paramagnetic complexes were characterized by
different techniques, including single-crystal X-ray structure
determination, IR spectroscopy, and ESI-MS analysis. All
D
dx.doi.org/10.1021/ic400096e | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 4. Solution IR spectra of complexes 14 and 17 (left) and 15 and 18 (right) in acetonitrile. Vibrational energies are given in wavenumbers
(cm−1).
indicates that at least three different forms of the complex are
present in the solid state. On the basis of these IR data, the
triflate groups do not coordinate to iron in the solid state in
complexes 17 and 20.
moieties point away from the iron center and do not exert a
noticeable effect on the coordination geometry around iron.
Amide ligand BMIPAnPr coordinates in the same way to iron as
does the BMIPnPr ligand (i.e., the amide moiety coordinates to
iron via its carbonyl group, and the propyl amide tails point
away from iron). Bruijnincx et al. have earlier reported on the
structure of complex 14.23 In that case, the crystals were
obtained from a MeOH/Et2O mixture, and a different structure
was obtained [Fe(BMIPnPr)2(MeOH)2](OTf)2 in which the
two ligands coordinate to iron via the nitrogen-donor atoms
only and two molecules of MeOH are coordinated in a cis
fashion to the iron center (Figure 2). A structure of the
corresponding tetraphenylborate complex [Fe(BMIMPnPr)2]-
(BPh4)2 was also reported.23 In this case, the structure matches
the structure of 14 reported here.
For several of the complexes, solution IR spectra were
recorded in acetonitrile. In previous work on complex 14, it was
found that in solution some of the carbonyl oxygens detach
from the iron center to create a vacant site23 to which either a
triflate ion or a solvent molecule could coordinate. For
[Fe(BMIPnPr)2](OTf)2 (14), two distinct CO stretching
vibrations were observed: one at 1737 cm−1 (noncoordinated)
and one at 1702 cm−1 (coordinated; Figure 4). When the
imidazole donors were substituted for benzimidazole donors as
in [Fe(BMBIPnPr)2](OTf)2 (17), only one CO vibration
corresponding to a coordinated carbonyl group was present at
1691 cm−1. However, this intense peak has small shoulders,
indicating the possible presence of other species in solution.
For [Fe(BMIPtBu)2](OTf)2 (15) and corresponding benzi-
midazole compound [Fe(BMBIPtBu)2](OTf)2 (18), similar IR
features were observed in acetonitrile (Figure 4, right).
Compound 15 again showed two distinct carbonyl peaks, and
compound 18 showed one sharp carbonyl peak. These
observations indicate that the presence of the benzimidazole
group hinders/prevents the decoordination of the carbonyl
groups in solution. Figure 4 also shows the triflate vibrations for
the four complexes in the 1000−1300 cm−1 region. These
vibrations are characteristic of noncoordinated triflates.32
Because of the low-yielding synthesis of the amide ligands, no
solution IR spectrum could be obtained for the corresponding
Fe compounds.
Structural Features of the Iron Triflate Complexes in
the Solid State (X-ray Crystal Structures). Crystals of
complexes 14−20 suitable for X-ray analysis were obtained by
slow vapor diffusion of diethyl ether into a solution of the
corresponding complex in acetonitrile. The crystal structures
for the first group of complexes with the imidazole backbone,
[Fe(BMIPnPr)2](OTf)2 (14), [Fe(BMIPtBu)2](OTf)2 (15),
and [Fe(BMIPAnPr)2](OTf)2 (16), are depicted in Figure 5.
In the crystal structures of complexes 14−16, the Fe atoms
of all molecules are located on inversion centers. The
complexes have a nearly ideal octahedral geometry. Both
ligands in these complexes are coordinated in a facial manner
around the iron center, with the two coordinating carbonyl
groups in the trans position with respect to each other. The two
ester complexes, 14 and 15, are very much alike. Their ester
Table 2 shows selected bond lengths and angles for
complexes 14−16, and for each complex, the data for all
independent residues in the asymmetric unit are shown. The
observed Fe−N bond lengths vary between 2.105(3) and
2.147(6) Å and are characteristic of iron(II) high-spin
complexes with imidazole-like ligands.33,34 Overall, the variation
of bond lengths in the individual complexes is very small, as can
be seen in the quadratic elongation ⟨λoct⟩.35 The intraligand N−
Fe−N angles are all smaller than 90°, reflecting the small “bite
angle” of the bis-imidazole methane moiety, whereas all
interligand N−Fe−N angles are larger than 90°. All complexes
35
2
have a significant angular variance σoct
.
A comparison of all
independent molecules in the crystal structures of 14−16
shows only minor differences in the bond lengths and angles.
The variation in the ester moiety in the first group of BMIP
ligands, therefore, does not have a large effect on the overall
structure of the corresponding [Fe(BAIP)2](OTf)2 complexes.
The structures of the three complexes of the benzimidazole
group [Fe(BMBIPnPr)2](OTf)2 (17), [Fe(BMBIPtBu)2](OTf)2
(18), and [Fe(BMBIPAnPr)2](OTf)2 (19) are shown in Figure
6. In contrast to centrosymmetric complexes 14−16, complex
17 is located on a 2-fold rotation axis with a cisoid coordination
of the ligands. The basal plane of 17 is formed by symmetry-
related nitrogen atoms N31 and N31a and oxygen atoms O21
and O21a (a = x, 0.5 − y, 0.5 − z). The symmetry-related
nitrogen atoms N11 and N11a are above and below this plane,
respectively. Both solid-state and solution-IR spectra of
complex 17 indicate the presence of more than one species,
which does not rule out the presence of an isomer of 17 in
which the oxygen donors are in the trans position. Most likely,
E
dx.doi.org/10.1021/ic400096e | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 5. Molecular structures of cationic complexes 14−16 in the crystals shown with displacement ellipsoid plots (30% probability). All C−H
hydrogen atoms, noncoordinated triflate anions, and noncoordinated acetonitrile molecules are omitted for clarity. In 14, only one out of four
independent molecules is shown, and only the major conformation of the disordered n-propyl moiety is displayed. In 15, only one out of two
independent molecules is depicted. Symmetry operations: (a) −x, −y, −z; (e) 1 − x, −y, 1 − z; (g) 1 − x, 1 − y, 1 − z.
the cis isomer crystallizes preferentially. This different
coordination behavior of benzimidazole complex 17 in
comparison to its imidazole derivative 14 might be caused by
the presence of the more bulky benzimidazole groups that
preclude the formation of a single thermodynamic conformer.
However, the structure of [Cu(BMBIP)2], containing the
analogous anionic benzimidazole propionate ligand, was earlier
found to crystallize in a transoid manner,25 which may indicate
an additional role for the ester moiety in determining the
overall structure.
Complexes 18 and 19 show the same coordination geometry
as that of complexes 14−16, a distorted octahedral
coordination with the carbonyl oxygens in the trans position.
Although complex 19 is again centrosymmetric, complex 18 is
located in a general position and has no molecular symmetry.
Nevertheless, the distortion of the octahedral geometry in 18 is
only slightly larger than that of the other complexes. The
N,N,O ligands in 18 are different from those in 17 in the sense
that the n-propylester moieties in 17 are exchanged for tert-
butyl ester moieties in 18. In this case, the steric bulk of these
tert-butyl groups seem to “suppress” or “overrule” the steric
effect of the benzimidazole groups. A cis orientation of the tert-
butyl ester groups most likely is unfavorable because of the
steric congestion of these groups in this orientation. The
introduction of an n-propyl amide group into BMBIPAnPr also
results in a trans-octahedral complex (19). In this case, steric
reasons are less likely to be at play because the n-propyl ester
moiety in BMBIPnPr allowed the formation of cis complexes. It
therefore seems more reasonable that electronic effects result in
the trans disposition of the amide-oxygen donors. The amides
are stronger donors than the esters and seem to overrule steric
arguments (vide infra). Selected bond lengths and angles for
complexes 17−19 are shown in Table 3.
F
dx.doi.org/10.1021/ic400096e | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
a
Table 2. Selected Bond Lengths (Angstroms) and Angles (Degrees) for Complexes 14−16
14
15
16
Res 1
(x = 1, y = a)
Res 2
(x = 2, y = b)
Res 3
(x = 3, y = c)
Res 4
(x = 4, y = d)
Res 1
(x = 1, y = e)
Res 2
(x = 2, y = f)
Res 1
(x = 1, y = g)
Bond Lengths
Fex−N1x
Fex−N4x
2.123(5)
2.124(4)
2.249(4)
1.006
2.144(5)
2.124(5)
2.174(4)
1.004
2.126(5)
2.136(5)
2.186(4)
1.004
2.130(5)
2.127(5)
2.229(4)
1.005
2.115(4)
2.137(4)
2.145(4)
1.005
2.140(3)
2.105(3)
2.167(3)
1.006
2.147(6)
2.107(7)
2.157(5)
1.005
Fex−O1x
b
⟨λoct
⟩
Bond Angles
85.21(18)
86.37(16)
94.79(18)
93.63(16)
87.49(17)
92.51(17)
15.57
N1x−Fex−N4x
N1x−Fex−O1x
N1x−Fex−N4xy
N1x−Fex−O1xy
N4x−Fex−O1x
N4x−Fex−O1xy
85.35(18)
87.26(17)
94.63(18)
92.74(17)
86.33(16)
93.67(16)
15.43
85.78(19)
85.88(17)
94.22(19)
94.12(17)
89.43(17)
90.56(17)
12.59
85.31(19)
88.82(17)
94.69(19)
91.18(17)
85.89(17)
94.11(17)
14.68
85.14(14)
86.85(14)
94.86(14)
93.15(14)
85.46(16)
94.54(16)
19.80
84.66(10)
85.12(10)
95.34(10)
94.88(10)
86.99(10)
93.01(10)
22.43
83.9(2)
86.6(2)
96.1(2)
93.4(2)
89.5(2)
90.5(2)
17.76
c
σoct (deg2)
2
a
Symmetry operations: (a) −x, −y, −z; (b) −x, 1 − y, −z; (c) 1 − x, −y, 1 − z; (d) 1 − x, 1 − y, −z; (e) 1 − x, −y, 1 − z; (f) − x, −y, 1 − z; (g) 1 −
b
c
x, 1 − y, 1 − z. ⟨λoct⟩ = Σi6= 1(li/lo)2/6 σoct = Σi1=2 1(θi−90°)2/11
2
Figure 6. Molecular structures of the cations of complexes 17−19 in the crystals are shown with displacement ellipsoid plots (30% probability). All
C−H hydrogen atoms, noncoordinated triflate anions, and noncoordinated acetonitrile molecules are omitted for clarity. Symmetry operations: (a)
x, 0.5 − y, 0.5 − z; (b) 1 − x, 1 − y, 1 − z.
G
dx.doi.org/10.1021/ic400096e | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Very similar trends in bond lengths and angles are found for
17−19 as compared to those of the first group, complexes 14−
16. All bond lengths are in agreement with high-spin (S = 2)
iron(II) complexes. As found for complexes 14−16, the
intraligand N−Fe−N angles in 17−19 are smaller than 90°,
reflecting the small bite angle of the bis-benzimidazole methane
moiety. In complex 19, the Fe−O bond length is considerably
shorter than those in 17 and 18, which indeed suggests that the
carbonyl oxygen atoms in this complex are more tightly bound
to the iron center. This effect is reflected among the series of
complexes reported here by the overall shorter Fe−O bond
lengths for the amide complexes (2.0886(13) and 2.157(5) Å)
compared to those of the ester complexes (2.145(4) and
2.249(4) Å).
Table 3. Selected Bond Lengths (Angstroms) and Angles
(Degrees) for Complexes 17−19.
a
Bond Lengths
17
(y = a)
19
(y = b)
18
Fe1−N11
Fe1−N31
Fe1−O21
2.115(3)
2.115(3)
2.180(3)
2.1426(15)
2.1584(15)
2.0886(13)
Fe1−N11
Fe1−N31
Fe1−O21
Fe1−N12
Fe1−N32
Fe1−O22
2.140(4)
2.128(4)
2.195(3)
2.132(4)
2.122(4)
2.203(3)
1.005
⟨λoct
⟩
1.010
1.004
Bond Angles
[Fe(BEiPrIPnPr)2](OTf)2 (20) crystallizes in the non-
centrosymmetric space group P1, with three independent
metal complexes in the unit cell in which two have tetrahedral
coordination geometry and one has octahedral geometry.
Figure 7 shows both the structure of the octahedral residue and
the structure of one of the tetrahedral residues. The structure of
the octahedral residue is reminiscent of the structure of
complex 17 (i.e., an overall octahedral coordination geometry
in which the oxygen donor atoms are found in cisoid positions).
The combined bond lengths and angles again show a deviation
from the ideal octahedral geometry, most likely caused by the
tripodal structure of the ligand (Table 4). The bond lengths are
in agreement with a high-spin (S = 2) iron(II) complex. In this
residue, all bond lengths around Fe are larger than those in 17.
In particular, the Fe−O bond lengths are remarkably longer
than those in 17 (2.252(3) and 2.303(3) vs 2.180(3) Å,
respectively). These elongated bonds most likely reflect the
overall steric demand of the BEIPrIPnPr ligand as compared to
that of the parent BMIPnPr and BMBIPnPr ligands.
17
(y = a)
19
(y = b)
18
N11−Fe1−
87.31(13)
86.22(12)
173.91(19)
96.48(13)
89.23(12)
103.4(2)
83.99(6)
89.57(6)
180
N11−Fe1−
83.54(15)
N31
N31
N11−Fe1−
N11−Fe1−
88.65(14)
95.51(15)
96.49(15)
84.47(15)
87.74(14)
87.24(14)
87.58(14)
178.91(13)
179.68(17)
178.40(16)
17.08
O21
O21
N11−Fe1−
N11−Fe1−
N11y
N32
N11−Fe1−
96.01(6)
90.43(6)
180
N12−Fe1−
N31y
N31
N11−Fe1−
N12−Fe1−
O21y
N32
N31−Fe1−
N12−Fe1−
N31y
O22
N31−Fe1−
86.81(13)
169.54(13)
83.12(17)
89.54(6)
90.46(6)
180
N31−Fe1−
O21
O21
N31−Fe1−
N32−Fe1−
O21y
O22
O21−Fe1−
O21−Fe1−
O21y
O22
N11−Fe1−
N12
In the tetrahedral residues of 20, the BEIPrIPnPr ligands act
as bidentate nitrogen ligands, whereas the propyl ester moieties
do not coordinate to iron but rather point away from the metal
center. Selected bond lengths and angles for the two tetrahedral
N31−Fe1−
N32
σoct (deg2)
34.12
13.25
2
a
Symmetry operations: (a) x, 0.5 − y, 0.5 − z; (b) 1 − x, 1 − y, 1 − z.
Figure 7. Molecular structures of the octahedral and the tetrahedral forms of the [Fe(BEiPrIPnPr)2]2+ cation in the crystal structure of 20 shown
with displacement ellipsoid plots (30% probability). All hydrogen atoms, noncoordinated triflate anions, and noncoordinated solvent molecules are
omitted for clarity. Only the major disordered components of the n-propyl groups are shown.
H
dx.doi.org/10.1021/ic400096e | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
geometries of 20 are present in the same unit cell, this indicates
that their energies are almost equal and that the transition
energy between the two geometries is low. To provide more
insight into the ground-state energy difference between both
geometries, DFT geometry optimizations were performed at
the bp86, SV(P) level. Improved energies were obtained with
single-point SCF energy calculations at the b3-lyp TZVP level.
In these calculations, simplified versions of the ligands were
used in which all alkyl tails were substituted for methyl groups
to limit computing time. The resulting calculated structures are
in agreement with the octahedral and tetrahedral geometries in
the crystal of 20 (Figure 8 and Table 6). Comparison of the
calculated structures with the experimental crystal structures
shows that the geometry around the iron center is the same and
that no clear differences in the overall structures are present.
The calculated bond lengths and angles are comparable to the
data obtained from the crystal structures of the two different
geometries (Tables 4 and 5). The substitution of all alkyls for
methyl groups could cause minor differences between the
geometry of the crystal structure and the calculated structures,
but because all alkyl tails are found on the outside of the overall
structures and all of them point away from the metal center,
only a small deviation is expected between the energy value of
the structure with alkyl chains in comparison to that of the
calculated structure with methyl groups. The difference in
energy between the two geometries was computed to be 5.6
kcal/mol, favoring the octahedral geometry. This small energy
difference points out that the octahedral and tetrahedral
structures may well coexist in solution and in the solid state.
The different coordination modes of BAIP-type ligands
around iron point out that the coordination geometry of this
ligand system is quite amendable and sensitive to changes in
the overall ligand structure. This could also mean that it may be
difficult to point out what the real structure of an active species
is during catalysis. It could be possible that multiple species are
present in solution and that some of these are kinetically more
competent than others. Earlier studies demonstrated the
coordination flexibility of BAIP ligands.23 A good example of
the coordination flexibility of these ligands is the character-
ization of [Fe(BMIPnPr)2(MeOH)2](OTf)2 in which two
MeOH molecules have replaced the ester donors and further
rearranged the ligands to coordinate in a mutual cis position.
The current study shows that IR spectroscopy provides an
insightful view of the coordination mode of the BAIP ligands. A
combination of solid-state and solution IR spectra indicated the
presence of different coordination conformers for complexes 17
and 20 on the basis of the coordinating/noncoordinating
nature of the carbonyl moieties. Furthermore, the IR studies
indicated that complexes with an imidazole backbone are more
prone to rearrangement in solution, involving both dissociation
of ester donor moieties and coordination of solvent molecules.
The difference in coordination flexibility between the imidazole
and benzimidazole ligands can be explained by the more
electron-deficient nature of the benzimidazole donor in
comparison to that of the imidazole donor and, accordingly,
a stronger interaction between the metal and the ester moieties
in the benzimidazole complexes.
Table 4. Selected Bond Lengths (Angstroms) and Angles
(Degrees) for the Octahedral Geometry of 20
Bond
Lengths
Bond Angles
Bond Angles
79.57(11)
Fe1−
N11
2.136(3) N11−Fe1−
168.80(13) N12−Fe1−
N12
O12
Fe1−
2.133(3) N11−Fe1−
92.31(13)
95.69(12)
79.98(12)
93.17(11)
93.78(13)
92.70(13)
90.55(12)
37.09
N31−Fe1−
98.45(13)
90.65(12)
170.40(12)
170.10(12)
88.84(12)
82.56(11)
N12
N31
N32
Fe1−
N31
2.154(3) N11−Fe1−
N31−Fe1−
O11
N32
Fe1−
N32
2.137(3) N11−Fe1−
N31−Fe1−
O12
O11
Fe1−
O11
2.252(3) N11−Fe1−
N32−Fe1−
O11
O12
Fe1−
O12
2.303(3) N12−Fe1−
N32−Fe1−
O12
N31
N12−Fe1−
N32
O11−Fe1−
O12
N12−Fe1−
O11
⟨λoct
⟩
1.012
σoct (deg2)
2
residues are shown in Table 5. Steric bulk caused by the ethyl
and isopropyl groups make the ester groups noncoordinated to
Table 5. Selected Bond Lengths (Angstroms) and Angles
(Degrees) for the Two Tetrahedral Residues in 20
Bond
Lengths
Bond Angles
Residue 1
Bond Angles
Fe2−
2.022(3) N13−Fe2−
134.35(14) N14−Fe2−
111.27(13)
95.89(13)
104.38(13)
N13
N14
N33
Fe2−
N14
2.025(3) N13−Fe2−
95.59(13)
N14−Fe2−
N34
N33
Fe2−
N33
2.048(3) N13−Fe2−
112.84(13) N33−Fe2−
N34
N34
Fe2−
N34
a
2.057(3)
⟨λtet
⟩
1.049
σtet (deg2)
2
207.57
Residue 2
Fe3−
2.041(3) N15−Fe3−
104.34(13) N16−Fe3−
110.98(13)
95.54(12)
135.33(13)
N15
N16
N35
Fe3−
N16
2.045(3) N15−Fe3−
95.42(13)
N16−Fe3−
N36
N35
Fe3−
N35
2.027(3) N15−Fe3−
112.51(13) N35−Fe3−
N36
N36
Fe3−
N36
b
2.026(3)
2
⟨λtet
⟩
1.052
σtet (deg2)
219.60
a
b
2
⟨λtet⟩ = Σi4= 1(li/lo)2/4 σtet = Σi6= 1(θi−109.47°)2/5
the metal center in the tetrahedral complex. In the octahedral
coordination mode, this bulk results in elongated Fe−O bonds.
The tetrahedral N−Fe−N angles vary between 95.4 and 135.3°
and are far from the ideal geometry of 109°. The bond lengths
in the tetrahedral residues are the shortest among all structures
presented in this study. Steric factors seem to play a decisive
role in this. Because the enhanced overall steric bulk of the
BEIPrIPnPr ligand prevents the coordination of the oxygen
donors, the nitrogen-donor moieties can approach the iron(II)
ion more closely. The remarkable presence of two different
coordination geometries of 20 in the solid state explains the
presence of the three carbonyl peaks in its IR spectrum (vide
infra).
Magnetic Susceptibility Measurements. The magnetic
moments of all reported iron(II) complexes were determined in
acetonitrile at 298 K using Evans’ NMR method.37,38 All
complexes showed magnetic moments consistent with high-
spin (S = 2) iron(II) centers. For complexes 16 and 19, no
magnetic moments were determined, but according to the bond
We are not aware of many examples in the literature in which
two different coordination geometries of the same molecular
metal complex are present within one crystal.36 Because both
I
dx.doi.org/10.1021/ic400096e | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 8. Calculated structures (DFT) of the two geometries of complex 20.
Table 6. Bond Lengths (Angstroms) and Angles (Degrees) of the Calculated DFT Structures of Complex 20
octahedral
tetrahedral
Bond Lengths
Bond Angles
Bond Lengths
Bond Angles
Fe−N1
Fe−N2
Fe−N3
Fe−N4
Fe−O1
Fe−O2
2.186
2.155
2.187
2.154
2.191
2.192
O1−Fe−N1
88.0
87.5
Fe−N1
Fe−N2
Fe−N3
Fe−N4
2.050
2.051
2.048
2.047
N1−Fe−N2
94.9
111.6
94.8
N1−Fe−N2
N2−Fe−N3
N3−Fe−O1
O1−Fe−O2
O1−Fe−N2
O2−Fe−N4
N2−Fe−N3
N3−Fe−N4
N1−Fe−N4
100.4
83.0
130.4
85.6
171.8
171.9
Table 7. Properties of the Synthesized Complexes
a
complex
CO position
ν(CO)(cm−1
)
Δν(CO)(cm−1
)
μEFF (μB)
[Fe(BMIPnPr)2](OTf)2 (14)
[Fe(BMIPtBu)2](OTf)2 (15)
[Fe(BMIPAnPr)2](OTF)2 (16)
[Fe(BMBIPnPr)2](OTf)2 (17)
trans
1691
36
5.1
5.1
trans
1692
33
b
trans
n.a.
n.a.
n.a. (HS)
cis, (trans)
trans
1709, 1694
1665
25, 40
55
5.1
5.3
[Fe(BMBIPtBu) ](OTf)2 (18)
2
b
[Fe(BMBIPAnPr)2](OTf)2 (19)
trans
1628
28
n.a. (HS)
c
d
c
c
[Fe(BEiPrIPnPr)2](OTf)2 (20)
cis , (trans)
1736 ,1705 , 1691
2, 29, 43
5.1
a
b
c
d
Compared to the free ligand. Indicated by Fe−N bond lengths. Octahedral. Tetrahedral.
lengths in the crystal structures of 16 and 19, a high-spin
configuration was also assigned to these complexes. An
overview of the spectroscopic properties of complexes 14−20
is summarized in Table 7.
CV Measurements. The oxidation potentials of Fe(II)
complexes 14, 18, and 19 were investigated by means of cyclic
voltammetry. These complexes showed (quasi)reversible,
single-wave features that indicate clean one-electron Fe(II)/
Fe(III) oxidation−reduction processes (Table 8 and Figure 9).
Table 8. Summarized Cyclic Voltammetry Data for
Complexes 14, 18, and 19
complex
E1/2 (V)
ΔEp (mV)
Figure 9. Cyclic voltammograms of complexes 18 and 19.
[Fe(BMIPnPr)2](OTf)2 (14)
0.735
0.910
0.595
175
80
[Fe(BMBIPtBu) ](OTf)2 (18)
2
The cyclic voltammetry measurements showed that amide
complex 19 is oxidized at a much lower potential (Ep,a = 0.63
[Fe(BMBIPAnPr)2](OTf)2 (19)
60
J
dx.doi.org/10.1021/ic400096e | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
V) in comparison to that of the ester complexes 14 and 18 (Ep,a
= 0.82 and 0.95 V), making complex 19 more sensitive to
oxidation. Complex 19 was at first isolated as an off-white
crystalline solid just like the other complexes, but the color of
the isolated complex turned red after 1 day of storage. This
change in color points to the oxidation of the Fe(II) complex to
an Fe(III) complex. The same observation was also made
during catalysis where the reaction mixture turned red when 19
was used as the catalyst. The lower oxidation potential of 19
likely results from a combination of (more) electron-deficient
benzimidazole donors and strong amide donors (as shown by
the short Fe−O distance in the crystal). This combination is
also expected to lead to a complex with a diminished
coordination flexibility and high conformational integrity. The
latter aspect may be reflected in the close-to-Nernstian ΔEp in
CV for the redox change between Fe(II) and Fe(III) in 19. The
observation that complex 14 is oxidized at a lower potential
than complex 18 seems to reflect the relative electron-donating
abilities of the imidazole donor with respect to the
benzimidazole donor groupings.
In Figure 9, the cyclic voltammogram of the benzimidazole
amide complex 19 is combined with the voltammogram of the
benzimidazole ester complex 18. Unfortunately, no CV
measurements could be carried out on amide complex 16
because of to the low yield of the complex.
Oxidation Catalysis. The ability of complexes 14−20 to
catalyze the oxidation of olefins was investigated by using
cyclooctene as a benchmark substrate (Scheme 4). The
overall catalytic characteristics are lower, and only minor traces
of diol are formed. The overall catalytic activity of 14 is the
highest among the complexes tested here.
The reactivity of the group of benzimidazole complexes
(17−19) is comparable to the activity of complex 16. The most
interesting result from this group of complexes was obtained
with amide complex 19. In this case, equal amounts of epoxide
and cis-diol were formed at a catalyst/oxidant/substrate ratio of
1:20:1000. Although the amount of cis-diol product is lower
than the amount produced by complex 14, a larger amount of
epoxide was formed in the latter case. During the catalytic
reactions with complex 19, a change in color from off-white to
red was observed upon the addition of the oxidant (vide supra).
For all of the other complexes, the color of the solution
changed to dark-yellow/orange after the addition of hydrogen
peroxide. Because of its crystal structure, complex 17
[Fe(BMBIPnPr)2](OTf)2 was expected to be a good candidate
for a cis-dihydroxylation catalyst as a result of the cis
coordination of the two carbonyl groups. However, 17 gave a
3:1 ratio of epoxide/cis-diol product at best. This observation
adds to the notion that the cisoid structure, as observed by X-
ray crystallography, is one of several conformers of 17 and
apparently does not stand out in its kinetic competence during
catalysis. Complex 20 was among the least active complexes in
this study. A small amount of epoxide was formed, and, in some
cases, a small trace of diol was observed with an H2O2
conversion of between 3 and 7%.
Few other Fe-based catalysts, in particular, those based on a
mixed N,N,O ligand manifold, are known to perform cis-
dihydroxylations of alkenes using H2O2, and cyclooctene is not
always included as a substrate in these studies. The Ph-DPAH-
based system reported by Que et al. shows a selectivity toward
the diol in the oxidation of cyclooctene of 14:1 with a TON
value of 7.0 toward the diol product.16 Later, this system was
optimized through ligand derivatization to yield exclusively the
diol product in styrene oxidation (TON = 9.4) and 1-octene
oxidation (TON = 7.7).39 Cyclooctene was not included as a
substrate in this study. The same group also reported on a
polydentate mixed N,O ligand derived from Kemp’s acid in
which the corresponding Fe(II) complex is able to oxidize 1-
octene to the epoxide/diol in a 1:6 ratio, albeit with a TON
value of lower than 1 per iron.40 These data show that turnover
values in cis-dihydroxylation reactions using mixed N,O ligands
in combination with iron are, in general, low. The cis-
dihydroxylation activity of nonheme iron complexes derived
from all-nitrogen ligands has been described somewhat more
extensively (vide supra). The most active systems reported so
far are derived from the PyTACN ligand system reported by
the group of Costas. These systems show TON values for diols
of up to 140 and diol/epoxide ratios of up to 4.9 in the
oxidation of cyclooctene (300 equiv of H2O2 was used). All of
these studies used large excesses (∼1000 equiv) of alkene
substrate per iron.41
Scheme 4. General Representation of the Oxidation of
Cyclooctene
oxidations were carried out in acetonitrile at ambient
temperature and under a nitrogen atmosphere, using H2O2 as
the sacrificial oxidant. The oxidant was added dropwise over a
period of 20 min to minimize the chance for peroxide
disproportionation. The reactions were carried out with
different ratios among the catalyst, oxidant, and substrate. In
all cases, the substrate was present in a large excess (oxidant
limiting conditions). The reactions were monitored by GC, and
samples were taken 1 and 3 h after the first drop of oxidant was
added to the reaction mixture as well as after running the
reaction for 1 night (Table 9).
Overall, all tested complexes are able to oxidize cyclooctene,
albeit with rather low TONs and low productive H2O2
consumption. This observation is in line with earlier studies
on the oxidation of cyclooctene with [Fe(BMIPnPr)2](OTf)2
and [Fe(BMIPnPr)2](BPh4)2.23 In the present case, productive
H2O2 conversion varies between 5 and 25% after 1 night of
reaction time, which corresponds to a range of 0−10 turnovers
per iron. For the first group of complexes (14−16), complex 14
was the most active. The use of 20 equiv of H2O2 per iron
resulted in a productive conversion of H2O2 from 23 to 25%
after 1 night with different amounts of substrate (500 and 1000
equiv, respectively). These conditions resulted in epoxide to
diol ratios of 1.5:1 and 1.6:1. An increase in the oxidant loading
to 100 equiv resulted in an increase in the turnover number to
10 (10% productive H2O2 consumption) and a change in the
epoxide/cis-diol ratio to 5.5:1. For complexes 15 and 16, the
The overall low activity of the complexes presented here
poses a challenge for their further improvement or even
practical use. Issues that could be at play here, in addition to
metal release from the ligand manifold, are the structural
integrity of the complexes and diol product inhibition. Our
investigation of the structure of iron complexes based on the
BAIP-ligand family has shown the coordination flexibility of
these ligands in solution as well as in the solid state.
Accordingly, this may lead to multiple species under catalytic
conditions, with a variation in activity and specificity. For the
K
dx.doi.org/10.1021/ic400096e | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 9. Overview of the Catalytic Results
a
a
epoxide
3 h
cis-diol
3 h
b
c
complex
eq. H2O2
eq. substrate
1 h
1 night
1 h
1 night
conversion (%)
epoxide/diol ratio
14
20
20
500
1000
1000
500
3.3
3.7
7.1
2.8
3.4
5.1
1.0
1.3
2.5
1.6
1.7
3.4
1.1
1.4
2.3
1.0
1.2
2.9
1.0
1.3
2.0
3.0
3.1
6.6
2.7
3.1
5.3
1.1
1.5
3.4
1.8
1.6
4.0
1.0
1.1
2.8
1.1
1.3
3.3
1.0
1.3
2.3
2.7
3.1
8.2
2.4
2.4
5.7
0.9
1.0
4.8
1.7
1.5
6.0
0.8
0.9
3.3
0.7
0.8
4.3
1.2
1.6
2.9
1.4
1.5
1.0
1.2
1.3
1.0
1.8
1.9
1.5
23
25
10
12
12
6
1.5:1
1.6:1
100
20
5.5:1
15
16
17
18
19
20
> 100:1
> 100:1
> 100:1
> 100:1
10:1
20
1000
1000
500
100
20
5
20
1000
1000
500
0.1
0.1
0.1
0.1
0.4
0.5
0.1
5
100
20
5
48:1
0.1
0.3
0.1
0.2
0.3
0.1
11
10
6
4.3:1
20
1000
1000
500
3:1
100
20
60:1
4
> 100:1
> 100:1
> 100:1
1.4:1
20
1000
1000
500
0.1
5
100
20
3
0.8
1.4
0.4
0.1
0.1
0.7
1.3
0.6
0.1
0.1
0.5
0.8
0.1
0.1
0.1
0.1
6
20
1000
1000
500
8
1:1
100
20
4
43:1
7
12:1
20
1000
1000
9
16:1
100
3
29:1
a
b
c
Yields expressed as turnover numbers (TON = mol product/mol catalyst). Conversion of H2O2 into epoxide and cis-diol after 1 night. After 1
night.
BAIP ligands, our investigations show that a change from
imidazole to benzimidazole donors prevents ester dissociation
from the metal. On the one hand, this reduces the number of
solution species, but on the other hand the creation of vacant
sites on the metal is hampered. In effect, this has led to a lower
activity for the benzimidazole complexes. Another factor that
influences product formation and turnover, which is less
discussed, is the possibility of product inhibition. Although
reaction mixtures involving H2O2 tend to be acidic, the
formation of stable iron-diolate complexes cannot be excluded.
Such complexes seem to lack sufficient open coordination sites
to accommodate a cis-dihydroxylation pathway and are
proposed to be inactive. We have previously reported on Fe-
catecholate complexes derived from BAIP ligands and found
that these are rather stable or show catechol-based chemistry.22
The release of diols from Fe-diolate complexes, leading to
increased TONs, may be accomplished by the use of a proper
reaction medium. This matter is currently under investigation
in our laboratories
reactivity of such complexes in catalysis. In particular, the role
of benzimidazole donors in providing a well-defined tripodal
coordination mode for the ligands in solution, the electron-rich
nature of amide donors that may help to stabilize the higher
oxidation states of the metal, and the subtle role of steric factors
in the balance between site isolation and coordination-mode
integrity are of interest. These considerations are currently
included in the design of new ligands and their corresponding
nonheme iron complexes, with the aim of arriving at site-
isolated mononuclear complexes “capped” by a single N,N,O
ligand that feature enhanced activities in the catalytic cis-
dihydroxylation of olefins.
EXPERIMENTAL SECTION
■
General. Air-sensitive reactions were carried out under an inert N2
atmosphere using standard Schlenk techniques. The solvents were
dried and distilled before use. The chemicals were commercially
obtained and used as received or were reproduced from the literature.
1H and 13C NMR spectra were recorded with a Varian 400
spectrometer at 400 and 100 MHz, respectively, operating at 25 °C.
Infrared spectra were recorded with a Perkin-Elmer Spectrum One
FTIR instrument. Solution IR measurements were recorded with a
Mettler Toledo ReactIR 1000 spectrometer with a SiComp probe
placed in a Schlenk tube under a N2 atmosphere. GC-MS
measurements were recorded with a Perkin-Elmer Autosystem XL
gas chromatograph with an attached Perkin-Elmer Turbomass
Upgrade mass spectrometer. High-resolution ESI-MS data were
acquired with a Waters LCT Premier XE machine. Solution magnetic
moments were determined by the Evans NMR method in acetone-d6/
cyclohexane (95/5 v/v) or in acetonitrile-d3/cyclohexane (95/5 v/v)
at 25 °C.37,38 GC analyses were performed with a Perkin-Elmer Clarus
500 GC (30 m, Econo-Cap EC-5) with an FID detector. Elemental
microanalyses were carried out by the Mikroanalytisches Laboratorium
CONCLUSIONS
■
We have presented an extension of the BAIP ester ligand family
and have studied the coordination chemistry of the new ligands
toward Fe(OTf)2. These studies have shown the intrinsic
property of these ligands to bind to iron in a facial N,N,O
manner, thereby mimicking the 2-His-1-carboxylate facial triad
found in mononuclear nonheme iron enzymes. The coordina-
tion flexibility of the ligands also manifested itself in these
studies, where in addition to trans bis-ligand arrangements, cis
arrangements are also possible and where tetrahedral, all-
nitrogen coordination can be enforced by the steric bulk.
Although these studies have not led to the development of an
improved olefin cis-dihydroxylation catalyst, they do provide
design tools for the further development of biomimetic
mononuclear nonheme iron complexes and for steering the
KOLBE, Mulheim an der Ruhr, Germany. Cyclic voltammograms
̈
were recorded in a single-compartment cell under a dry N2
atmosphere. The cell was equipped with a Pt working electrode, a
Pt wire counter electrode, and a Ag/AgCl reference electrode.
L
dx.doi.org/10.1021/ic400096e | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Potential control was achieved with a PAR model 263A potentiostat.
Bis(1-methylimidazol-2-yl)methane (1),26,27 bis(1-methylbenzimida-
zol-2-yl)methane (2),28 propyl 3,3-bis(1-methyl(benz)imidazol-2-yl)-
propionate (BMIPnPr (3), BMBIPnPr (4)),25 3,3-bis(1-methyl(benz)-
imidazol-2-yl)propionic acid (7,10),25 propyl 3,3-bis(1-ethyl-4-iso-
125.9, 145.8, 173.2, 174.1. ESI-MS: m/z = 332.135 ([M + H]+, calcd
332.136).
3,3-Bis(1-methylimidazol-2-yl)-N-propylpropanamide, BMIPAnPr
(9). Propylamine (112 μL, 1.38 mmol) was added to a solution of 8
(450 mg, 1.36 mmol) in dichloromethane (20 mL). The colorless
solution was stirred for 48 h at room temperature. The reaction
mixture was washed with a saturated NaHCO3 solution (2 × 15 mL)
and H2O (2 × 15 mL). The combined aqueous layers were extracted
with dichloromethane (1 × 15 mL), and the combined organic layers
were dried over magnesium sulfate, filtered, and evaporated in vacuo,
yielding a white solid that was recrystallized from a dichloromethane/
propylimidazol-2-yl)propionate (BEiPrIPn P r
,
(13)),30
Fe(OTf)2·2MeCN,31 and [Fe(BMIPnPr)2](OTf)2 (14)23 were pre-
pared according to published procedures.
tert-Butyl 3,3-Bis(1-methylimidazol-2-yl)propionate, BMIPtBu (5).
A solution of n-butyllithium (0.48 mL, 0.77 mmol, 1.6 M in hexanes)
was added dropwise to a stirring solution of bis(1-methylimidazol-2-
yl)methane (1) (134 mg, 761 μmol) in THF (4 mL) at −78 °C. The
greenish solution was stirred for 1 h at −78 °C, followed by the
dropwise addition of tert-butyl bromoacetate (115 μL, 779 μmol). The
temperature of the mixture was allowed to rise to room temperature,
and the mixture was quenched with H2O (10 mL). All volatiles were
evaporated in vacuo, the water layer was extracted with ethyl acetate (4
× 10 mL), and the combined organic layers were dried over
magnesium sulfate, filtered, and concentrated in vacuo. The product
was obtained as a clear yellow oil (216 mg, 90%). 1H NMR (400 MHz,
CDCl3, 25 °C): δ = 1.16 (s, 9H, OC(CH3)3), 3.03 (d, 2H, 3JHH = 7.6
1
diethyl ether mixture (95 mg, 25%). H NMR (400 MHz, CDCl3, 25
3
°C): δ = 0.81 (t, 3H, JHH = 7.4 Hz, NHCH2CH2CH3), 1.40 (sextet,
3
2H, JHH
= 7.2 Hz, NHCH2CH2CH3), 3.11 (m, 2H,
3
NHCH2CH2CH3), 3.14 (d, 2H, JHH = 7.6 Hz, CHCH2), 3.50, (s,
3
6H, NCH3), 4.83 (t, 1H, JHH = 7.2 Hz, CHCH2), 6.53 (m, 1H,
NHCH2CH2CH3), 6.77 (s, 2H, Him), 6.93 (s, 2H, Him). 13C {1H}
NMR (100 MHz, CDCl3, 25 °C): δ = 11.5, 22.8, 33.3, 33.8, 39.1, 41.5,
122.1, 126.8, 145.8, 170.4. IR (solid) ν (cm−1): 3231.9, 3050.7, 2970.1,
2937.1, 2875.7, 1650.6, 1562.8, 1489.0, 1473.9, 1278.7, 1237.9, 1133.3,
1084.1, 971.1, 924.7, 843.0, 761.0, 738.2, 694.3. ESI-MS: m/z =
276.187 ([M + H]+, calcd 276.182).
3
Hz, CHCH2), 3.32 (s, 6H, NCH3), 4.74 (t, 1H, JHH = 8.0 Hz,
NHS 3,3-Bis(1-methylbenzimidazol-2-yl)propionate (11). N-Hy-
droxysuccinimide (166 mg, 1.44 mmol) and N,N′-dicyclohexyl-
carbodiimide (303 mg, 1.47 mmol) were added to a stirring
suspension of 10 (480 mg, 1.44 mmol) in dry THF (200 mL),
followed by the addition of pyridine (1.5 mL, 18.5 mmol). A reflux
condenser was put on top of the flask, and the reaction mixture was
stirred in an oil bath at 35 °C for 3 h, followed by stirring at room
temperature overnight. The reaction mixture became a clear yellow
solution upon heating to 35 °C again, and stirring at 35 °C was
continued for another 8 h, followed by stirring at room temperature
for 3 days. Some white precipitate (N,N′-dicyclohexylurea) was
present that was filtered off, followed by the evaporation of the solvent
in vacuo, yielding a yellow-white solid that was dissolved in
dichloromethane. The precipitated urea was filtered off, and the
dichloromethane was evaporated in vacuo. The remaining solid was
dissolved again in dichloromethane, and the precipitate was filtered off.
This cycle was repeated until no more precipitate was formed (four
cycles). The yellowish solid was recrystallized from a dichloro-
methane/diethyl ether mixture at −30 °C overnight, yielding a slightly
yellow solid (529 mg, 85%). 1H NMR (400 MHz, CDCl3, 25 °C): δ =
3
3
CHCH2), 6.58 (d, 2H, JHH = 0.8 Hz, Himid), 6.71 (d, 2H, JHH = 0.8
Hz, Himid). 13C {1H} NMR (100 MHz, CDCl3, 25 °C): δ = 28.0, 32.9,
34.7, 37.6, 80.9, 122.1, 127.0, 145.4, 170.3. IR (solid) ν (cm−1):
3107.0, 2977.4, 2932.3, 1725.5, 1520.5, 1491.0, 1455.9, 1411.4, 1392.9,
1366.0, 1310.9, 1280.4, 1249.7, 1151.1, 1133.7, 857.6, 763.7, 738.3.
ESI-MS: m/z = 291.182 ([M + H]+, calcd 291.182).
tert-Butyl 3,3-Bis(1-methylbenzimidazol-2-yl)propionate,
BMBIPtBu (6). A solution of n-butyllithium (0.75 mL, 1.2 mmol, 1.6
M in hexanes) was added dropwise to a stirring solution of 2 (308 mg,
1.11 mmol) in THF (20 mL) at −78 °C. The greenish solution was
stirred for 1 h at −78 °C, followed by the dropwise addition of tert-
butyl bromoacetate (165 μL, 1.11 mmol). The temperature of the
mixture was allowed to rise to room temperature, and the mixture was
quenched with H2O (15 mL). All volatiles were evaporated in vacuo,
the water layer was extracted with ethyl acetate (2 × 25 mL), and the
combined organic layers were dried over magnesium sulfate, filtered,
and concentrated in vacuo. The product was obtained as an off-white
1
solid (407 mg, 94%). H NMR (400 MHz, CDCl3, 25 °C): δ = 1.33
(s, 9H, OC(CH3)3), 3.50 (d, 2H, 3JHH = 7.6 Hz, CHCH2), 3.81 (s, 6H,
NCH3), 5.48 (m, 1H, CHCH2), 7.27 (m, 6H, Hbenzimid), 7.76 (m, 2H,
Hbenzimid). 13C {1H} NMR (100 MHz, CDCl3, 25 °C): δ = 28.2, 30.7,
36.1, 37.4, 81.7, 109.6, 119.7, 122.7, 123.3, 136.5, 151.2, 169.8. IR
(solid) ν (cm−1): 3046.3, 2934.1, 2871.2, 1719.6, 1614.6, 1467.3,
1438.8, 1359.7, 1331.5, 1145.2, 981.2, 741.9. ESI-MS: m/z = 391.215
([M + H]+, calcd 391.213).
3
2.76 (s, 4H, C(O)CH2), 3.73 (s, 6H, NCH3), 3.94 (d, 2H, JHH = 7.6
3
Hz, CHCH2), 5.50 (t, 1H, JHH = 7.6 Hz, CHCH2), 7.27 (m, 6H,
Hbenzimid), 7.79 (m, 2H, Hbenzimid). 13C {1H} NMR (100 MHz, CDCl3,
25 °C): δ = 25.7, 30.5, 33.5, 35.8, 109.6, 120.2, 122.7, 123.4, 136.6,
142.0, 150.1, 166.8, 168.8. IR (solid) ν (cm−1): 2936.7, 1813.9, 1782.0,
1728.3, 1502.5, 1469.4, 1438.2, 1363.4, 1287.3, 1197.2, 1156.2, 1094.8,
1062.6, 980.3, 869.0, 809.2, 743.4. ESI-MS: m/z = 432.163 ([M + H]+,
calcd 432.167).
NHS 3,3-Bis(1-methylimidazol-2-yl)propionate (8). N-Hydroxy-
succinimide (373 mg, 3.25 mmol) and N,N′-dicyclohexylcarbodiimide
(670 mg, 3.25 mmol) were added to a stirring suspension of 7 (759
mg, 3.24 mmol) in dry THF (250 mL), followed by the addition of
pyridine (1.5 mL, 18.5 mmol). A reflux condenser was put on top of
the flask, and the reaction mixture was stirred in an oil bath at 40 °C
overnight after which time the reaction mixture was still a white
suspension. After the reaction mixture was stirred at 45 °C for an
additional 3 h, the white solid particles were filtered off and the
evaporation of the solvent in vacuo yielded a yellow/white solid that
was dissolved in dichloromethane. The precipitated urea was filtered
off, and the dichloromethane was evaporated in vacuo. The remaining
solid was dissolved again in dichloromethane, and the precipitate was
filtered off. This cycle was repeated until no further precipitate was
formed (three cycles). The yellowish solid was recrystallized from a
dichloromethane/diethyl ether mixture at −30 °C overnight, yielding
an off-white solid (450 mg, 43%). This preparation was not entirely
3,3-Bis(1-methylbenzimidazol-2-yl)-N-propylpropanamide,
BMBIPAnPr (12). Propylamine (48 μL, 580 μmol) was added to a
solution of 11 (250 mg, 579 μmol) in dichloromethane (10 mL). The
colorless solution was stirred overnight at room temperature. The
reaction mixture was washed with a saturated NaHCO3 solution (2 ×
5 mL) and H2O (1 × 5 mL). The combined aqueous layers were
extracted with dichloromethane (1 × 5 mL), and the combined
organic layers were dried over magnesium sulfate, filtered, and
1
evaporated in vacuo, yielding a white solid (195 mg, 90%). H NMR
3
(400 MHz, CDCl3, 25 °C): δ = 0.77 (t, 3H, JHH = 7.4 Hz,
NHCH2CH2CH3), 1.37 (sextet, 2H, 3JHH = 7.2 Hz, NHCH2CH2CH3),
3.12 (q, 2H, 3JHH = 6.7 Hz, NHCH2CH2CH3), 3.40 (d, 2H, 3JHH = 6.8
3
Hz, CHCH2), 3.73 (s, 6H, NCH3), 5.32 (t, 1H, JHH = 7.2 Hz,
CHCH2), 6.38 (m, 1H, NHCH2CH2CH3), 7.29 (m, 6H, Hbenzimid),
7.73 (m, 2H, Hbenzimid). 13C {1H} NMR (100 MHz, CDCl3, 25 °C): δ
= 11.4, 22.8, 30.3, 31.1, 35.4, 39.3, 41.6, 109.5, 120.0, 122.4, 123.0,
136.4, 142.3, 152.1, 170.3. IR (solid) ν (cm−1): 3235.1, 3057.2, 2930.0,
1656.2, 1555.9, 1507.8, 1464.0, 1439.9, 1395.5, 1281.9, 1262.4, 1240.0,
1092.7, 773.6, 740.3. Anal. Calcd for C22H25N5O (375.47): C 70.38, H
6.71, N 18.65; found C 70.59, H 7.05, N 18.43.
1
pure, as some dicyclohexylurea was present according to NMR. H
NMR (400 MHz, CDCl3, 25 °C): δ = 2.63 (s, 4H, C(O)CH2), 3.25
(d, 2H, 3JHH = 6.4 Hz, CHCH2), 3.48 (s, 6H, NCH3), 4.98 (t, 1H, 3JHH
= 8.0 Hz, CHCH2), 6.72 (s, 2H, Him), 6.87 (s, 2H, Him). 13C {1H}
NMR (100 MHz, DMSO-d6, 25 °C): δ = 25.7, 32.3, 33.1, 38.8, 123.1,
M
dx.doi.org/10.1021/ic400096e | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 10. Experimental Details of the Crystal Structures for 14−16
14
15
16
formula
fw
[C28H40FeN8O4](CF3O3S)2·1.95(CH3CN)
[C30H44FeN8O4](CF3O3S)2 + disordered acetonitrile
[C28H42FeN10O2](CF3O3S)2
904.71
a
986.73
colorless
934.72
cryst color
cryst size (mm3)
cryst syst
space group
a (Å)
colorless
0.18 × 0.16 × 0.12
monoclinic
P21/n (No. 14)
17.2165(5)
13.1126(4)
20.2556(13)
90
colorless
0.63 × 0.42 × 0.04
triclinic
0.90 × 0.21 × 0.03
triclinic
P1 (No. 2)
̅
P1 (No. 2)
̅
14.3823(2)
14.6648(4)
22.2031(7)
87.999(1)
88.800(1)
89.251(1)
4678.7(2)
4
8.6294(5)
10.7097(8)
11.4184(6)
65.829(4)
88.557(3)
81.683(2)
951.95(10)
1
b (Å)
c (Å)
α (deg)
β (deg)
γ (deg)
V (Å3)
91.338(2)
90
4571.4(4)
4
Z
a
Dcalc (g/cm3)
1.401
1.358
1.578
θ/λ
(sin
)
(Å−1
)
0.55
0.53
0.48
max
reflns. measured/unique
parameters/restraints
R1/wR2 [I > 2σ(I)]
R1/wR2 (all reflns.)
S
47 185/12 969
1227/469
0.0680/0.1657
0.1102/0.1909
1.078
46 455/5600
545/195
6170/1681
334/187
0.0488/0.1275
0.0628/0.1355
1.148
0.0665/0.1626
0.0827/0.1741
1.106
ρ(min/max) (eÅ−3
)
−0.61/1.37
−0.85/0.68
−0.50/0.70
a
Derived values do not contain the contribution of the disordered solvent.
[Fe(BMIPtBu)2](OTf)2 (15). To a solution of BMIPtBu (5) (216 mg,
yellowish reaction mixture was stirred for 45 min at room temperature.
The solvent was evaporated in vacuo, and the remaining off-white solid
was recrystallized from an acetonitrile/diethyl ether mixture at −30 °C
overnight to optimize the yield. The solvent, which had a brownish
color, was removed with a cannula, leaving the white crystalline solid
(198 mg, 86%). Single crystals suitable for X-ray diffraction were
obtained by the slow vapor diffusion of diethyl ether into a solution of
18 in acetonitrile. IR (solid) ν (cm−1): 2981.9, 1665.1, 1490.5, 1454.0,
1418.2, 1398.0, 1373.2, 1260.1, 1226.1, 1149.8, 1029.5, 1010.5, 995.1,
744 μmol) in acetonitrile (2 mL) was added a solution of
Fe(OTf)2·2MeCN (171 mg, 370 μmol) in acetonitrile (2 mL), and
the yellowish reaction mixture was stirred for 1 h at room temperature.
The solvent was evaporated in vacuo, and the remaining brown solid
was recrystallized from an acetonitrile/diethyl ether mixture at −30 °C
overnight to optimize the yield. The product was obtained as a brown
crystalline solid (270 mg, 95%). Single crystals suitable for X-ray
diffraction were obtained by the slow vapor diffusion of diethyl ether
into a solution of 15 in acetonitrile. IR (solid) ν (cm−1): 2983.2,
1692.2, 1506.7, 1396.5, 1370.3, 1258.1, 1223.5, 1144.1, 1029.4, 988.7,
962.1, 948.2, 852.7, 831.8, 754.3, 734.4. Solution magnetic moment
(Evans’ method): μeff = 5.1μB. ESI-MS: m/z = 785.235 ([M-OTf]+,
calcd 785.236).
844.5, 745.9. Solution magnetic moment (Evans’ method): μeff
=
5.3μB. ESI-MS: m/z = 985.294 ([M − OTf]+, calcd 985.298).
[Fe(BMBIPAnPr)2](OTf)2 (19). To a solution of BMBIPAnPr (12) (154
mg, 410 μmol) in methanol (7 mL) was added a solution of
Fe(OTf)2·2MeCN (90 mg, 205 μmol) in methanol (4 mL), and the
reaction mixture was stirred for 30 min at room temperature. The
solvent was evaporated in vacuo, and the remaining off-white solid was
recrystallized from an acetonitrile/diethyl ether mixture at −30 °C
overnight to optimize the yield. The solvent was removed with a
cannula, leaving the off-white crystalline solid (178 mg, 79%). Single
crystals suitable for X-ray diffraction were obtained by the slow vapor
diffusion of diethyl ether into a solution of 19 in acetonitrile. The off-
white crystalline solid turned red after 1 day (in a Schlenk tube under
an N2 atmosphere). IR (solid) ν (cm−1): 3297.7, 3104.9, 2965.1,
1628.3, 1565.0, 1504.9, 1481.0, 1454.8, 1277.0, 1244.4, 1223.3, 1151.7,
1028.1, 741.0. ESI-MS: m/z = 955.292 ([M − OTf]+, calcd 955.299).
[Fe(BEiPrIPnPr)2](OTf)2 (20). Fe(OTf)2·2MeCN (86 mg, 197 μmol)
was added to a stirring solution of BEiPrIPnPr (13) (153 mg, 394
μmol) in methanol (4 mL), and the dark-brown solution was stirred
for 30 min at room temperature. The solvent was evaporated in vacuo,
and the remaining dark oil was recrystallized twice from an
acetonitrile/diethyl ether mixture at −30 °C, leaving a yellow
crystalline solid that was washed with diethyl ether (50 mg, 34%).
Crystals suitable for X-ray diffraction were obtained by the slow vapor
diffusion of diethyl ether into a solution of 20 in acetonitrile. IR (solid)
ν (cm−1): 3109.1, 2968.3, 2879.0, 1736.0, 1704.8, 1691.5, 1583.1,
1497.1, 1495.2, 1388.1, 1365.3, 1258.6, 1223.6, 1147.5, 1087.4, 1030.6,
[Fe(BMIPAnPr)2](OTf)2 (16). Compound 9 (13 mg, 48 μmol) was
mixed with Fe(OTf)2·2MeCN (11 mg, 24 μmol) in acetonitrile (16
mL) prior to using it in catalysis. Single crystals were grown from the
remaining solution by the slow vapor diffusion of diethyl ether into it
for X-ray analysis, and no further analysis was performed on the
crystals.
[Fe(BMBIPnPr)2](OTf)2 (17). To a solution of BMBIPnPr (4) (195
mg, 518 μmol) in methanol (5 mL) was added a solution of
Fe(OTf)2·2MeCN (114 mg, 259 μmol) in methanol (4 mL), and the
reaction mixture was stirred for 40 min at room temperature. The
solvent was evaporated in vacuo, and the remaining white solid was
recrystallized from an acetonitrile/diethyl ether mixture at −30 °C
overnight to optimize the yield. The solvent, which had a deep-red
color, was removed with a cannula, and the remaining crystalline solid
was washed with diethyl ether and dried in vacuo. The product was
obtained as an off-white crystalline solid (254 mg, 89%). Single crystals
suitable for X-ray diffraction were obtained by the slow vapor diffusion
of diethyl ether into a solution of 17 in acetonitrile. IR (solid) ν
(cm−1): 3510.2, 3058.2, 2967.6, 1708.8, 1693.7, 1618.4, 1493.7,
1455.6, 1401.0, 1258.5, 1208.1, 1160.5, 1030.6, 927.1, 739.1. Solution
magnetic moment (Evans’ method): μeff = 5.1μB. Anal. Calcd for
C46H48F6FeN8O10S2 (1106.89): C 49.91, H 4.37, N 10.12; found C
49.39, H 4.53, 10.03.
975.4, 803.0. Solution magnetic moment (Evans’ method): μeff
=
[Fe(BMBIPtBu)2](OTf)2 (18). To a solution of BMBIPtBu (6) (158
mg, 405 μmol) in methanol (5 mL) was added a solution of
Fe(OTf)2·2MeCN (89 mg, 203 μmol) in methanol (5 mL), and the
5.1μB. ESI-MS: m/z = 981.455 ([M − OTf]+, calcd 981.454).
Catalysis Protocol. To a solution of catalyst (3 μmol) in
acetonitrile (2 mL) was added cis-cyclooctene (500 or 1000 equiv) and
N
dx.doi.org/10.1021/ic400096e | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 11. Experimental Details of the Crystal Structures for 17−19
17
18
19
formula
fw
[C44H48FeN8O4](CF3O3S)2·2(CH3CN)
[C46H52FeN8O4](CF3O3S)2·5(CH3CN)
[C44H50FeN10O2](CF3O3S)2·2(CH3CN)
1189.00
colorless
1340.22
colorless
1187.04
yellow
cryst color
cryst size (mm3)
cryst syst
space group
a (Å)
0.36 × 0.09 × 0.09
orthorhombic
Pnna (No. 52)
16.6451(2)
18.8376(3)
17.5683(3)
90
0.42 × 0.10 × 0.08
monoclinic
P21/c (No. 14)
11.7689(2)
26.0832(6)
22.7901(6)
90
0.57 × 0.30 × 0.12
monoclinic
P21/c (No. 14)
14.9769(1)
11.8607(1)
17.6525(2)
90
b (Å)
c (Å)
α (deg)
β (deg)
γ (deg)
V (Å3)
90
112.8675(8)
90
122.6929(5)
90
90
5508.61(14)
4
6446.1(3)
4
2638.96(4)
2
Z
Dcalc (g/cm3)
1.434
1.381
1.494
θ/λ
(sin
)
(Å−1
)
0.55
0.56
0.65
max
reflns. measured/unique
parameters/restraints
R1/wR2 [I > 2σ(I)]
R1/wR2 (all reflns.)
S
44 453/3850
388/81
52 160/9198
858/267
45 526/6063
366/0
0.0597/0.1721
0.0833/0.1897
1.057
0.0632/0.1562
0.1127/0.1851
1.077
0.0418/0.1173
0.0552/0.1275
1.093
ρ(min/max) (eÅ−3
)
−0.34/1.30
−0.38/0.53
−0.60/0.64
acetonitrile (to bring the total volume to 2.5 mL). 1,2-Dibromo-
benzene (10 μL) was added as internal standard, and subsequently 0.5
mL of oxidant solution (20 or 100 equiv, diluted from 35% aqueous
H2O2 with acetonitrile) was added dropwise over 20 min. The reaction
mixture was stirred at room temperature, and after 1 h the first sample
was taken. Diethyl ether was added to the sample to precipitate the
iron complex, after which it was analyzed by GC.
Table 12. Experimental Details of the Crystal Structure for
20
20
formula
[C44H72FeN8O4](CF3O3S)2 + disordered
acetonitrile
a
fw
cryst color
cryst size (mm3)
cryst syst
space group
a (Å)
1131.09
X-ray Crystal Structure Determinations. Reflections were
recorded with a Nonius KappaCCD diffractometer with a rotating
anode and a graphite monochromator (λ = 0.71073 Å) at a
temperature of 150(2) K. Intensity data were integrated with
Eval1442 (compounds 14 and 16), Eval1543 (compound 15), or
HKL200044 (compounds 17−20) software. Absorption correction and
scaling was performed on the basis of multiple measured reflections
with SADABS45 (14−16) or SORTAV46 (18−20). In compound 17,
no absorption correction was considered necessary. The structures
were solved by Direct Methods using the SHELXS-9747 (14−17, 19,
and 20) or SIR-9748 (18) program. Least-squares refinement was
performed with SHELXL-9747 against F2 for all reflections. Non-
hydrogen atoms were refined freely with anisotropic displacement
parameters. Hydrogen atoms were introduced in calculated positions
(14−18 and 20) or located in difference Fourier maps (19). Hydrogen
atoms were refined with a riding model, and the N−H hydrogen atom
of 19 was refined freely with isotropic displacement parameters.
Restraints for distances and angles and for the approximation of
isotropic behavior were used for triflate and acetonitrile. In 14, 16−18,
and 20, triflate anions were refined with disorder models. In 14, the
acetonitrile was refined with partial occupancy, and the propyl ester
moiety, with a disorder model. The crystal structures of 15 and 20
contain voids [594 (compound 15) and 282 Å3/unit cell (compound
20)] filled with disordered acetonitrile solvent molecules. Their
contribution to the structural factors was secured by back-Fourier
transformation with the SQUEEZE routine of PLATON,49 resulting in
112 (compound 15) and 70 electrons/unit cell (compound 20).
Geometry calculations and checking for higher symmetry was
performed with the PLATON program.49
yellow
0.26 × 0.24 × 0.06
triclinic
P1 (No. 1)
12.7766(1)
13.9229(1)
27.2566(4)
80.8220(3)
84.8511(4)
67.7105(7)
4426.61(8)
3
b (Å)
c (Å)
α (deg)
β (deg)
γ (deg)
V (Å3)
Z
a
Dcalc (g/cm3)
1.273
θ/λ
(sin
)
(Å−1
)
0.65
max
reflns. measured/unique
parameters/restraints
R1/wR2 [I > 2σ(I)]
R1/wR2 (all reflns.)
S
88 288/39 932
2142/2229
0.0496/0.1116
0.0786/0.1264
0.999
Flack x50
0.537(9)
ρ(min/max) (eÅ−3
)
−0.32/0.57
a
Derived values do not contain the contribution of the disordered
solvent.
Further details of the crystal structure determinations are given in
ri-DFT BP8653 level using the Turbomole SV(P) basis set51c,d on all
atoms. Improved energies were obtained with single-point energy
calculations at the DFT b3-lyp,54 def-TZVP51c−f level of theory.
Tables 10−12.
DFT Geometry Optimizations. The geometry optimizations were
carried out with the Turbomole program51a,b coupled to the PQS
Baker optimizer.52 Geometries were fully optimized as minima at the
O
dx.doi.org/10.1021/ic400096e | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(22) Bruijnincx, P. C. A.; Lutz, M.; Spek, Hagen, W. R.; Weckhuysen,
B. M.; van Koten, G.; Klein Gebbink, R. J. M. J. Am. Chem. Soc. 2007,
129, 2275−2286.
(23) Bruijnincx, P. C. A.; Buurmans, I. L. C.; Goisiewska, S.;
Moelands, M. A. H.; Lutz, M.; Spek, A. L.; van Koten, G.; Klein
Gebbink, R. J. M. Chem.Eur. J. 2008, 14, 1228−1237.
(24) Other studies using this ligand family were reported by Burzlaff
ASSOCIATED CONTENT
■
S
* Supporting Information
CIF file for the X-ray crystal structures of compounds 14−20.
This material is available free of charge via the Internet at
http://pubs.acs.org.
and Holland and co-workers. See (a) Peters, L.; Hubner, E.; Burzlaff,
̈
AUTHOR INFORMATION
■
N. J. Organomet. Chem. 2005, 690, 2009−2016. (b) Rocks, S. S.;
Brennessel, W. W.; Machonkin, T. E.; Holland, P. L. Inorg. Chim. Acta
2009, 362, 1387−1390.
Corresponding Author
*E-mail: r.j.m.kleingebbink@uu.nl.
(25) Bruijnincx, P. C. A.; Lutz, M.; Spek, A. L.; van Faassen, E. E.;
Weckhuysen, B. M.; van Koten, G.; Klein Gebbink, R. J. M. Eur. J.
Inorg. Chem. 2005, 779−787.
Notes
The authors declare no competing financial interest.
(26) Braussaud, N.; Ruther, T.; Cavell, K. J.; Skelton, B. W.; White,
̈
A. H. Synthesis 2001, 4, 626−632.
ACKNOWLEDGMENTS
■
(27) Lucas, P.; El Mehdi, N.; Ho, H. A.; Belanger, D.; Breau, L.
Synthesis 2000, 3, 1253−1258.
This work was financially supported by the National Research
School Combination-Catalysis (NRSC-C).
(28) Elgafi, S.; Field, L. D.; Messerle, B. A.; Turner, P.; Hambley, T.
W. J. Organomet. Chem. 1999, 588, 69−77.
REFERENCES
(29) Suijkerbuijk, B. M. J. M.; Slagt, M. Q.; Klein Gebbink, R. J. M.;
Lutz, M.; Spek, A. L.; van Koten, G. Tetrahedron Lett. 2002, 43, 6565−
6568.
■
(1) Tanase, S.; Bouwman, E. Adv. Inorg. Chem. 2006, 58, 29−75.
(2) Bataille, C. J. R.; Donohoe, T. J. Chem. Soc. Rev. 2011, 40, 114−
128.
(30) Bruijnincx, P. C. A.; Lutz, M.; Spek, A. L.; Hagen, W. R.;
Weckhuysen, B. M.; van Koten, G.; Klein Gebbink, R. J. M. J. Am.
Chem. Soc. 2007, 129, 2275−2286.
(3) Schroeder, M. Chem. Rev. 1980, 80, 187−213.
(4) Sharpless, K. B.; Amberg, W.; Bennani, Y. L.; Crispino, G. A.;
Hartung, J.; Jeong, K.-S.; Kwong, H.-L.; Morikawa, K.; Wang, Z.-M.;
Xu, D.; Zhang, X.-L. J. Org. Chem. 1992, 57, 2768−2771.
(5) Yip, W.-P.; Ho, C.-M.; Zhu, N.; Lau, T.-C.; Che, C.-M. Chem.
Asian J. 2008, 3, 70−77.
(31) Hagen, K. S. Inorg. Chem. 2000, 39, 5867−5869.
(32) Bergstrom, P.Å.; French, R. J. Phys. Chem. 1995, 99, 12603−
̈
12611.
(33) Ben
2005, 999−1002.
(34) Benisvy, L.; Halut, S.; Donnadieu, B.; Tuchagues, J.; Chottard,
́
isvy, L.; Chottard, J.; Marrot, J.; Li, Y. Eur. J. Inorg. Chem.
(6) Shing, T. K. M.; Tam, E. K. W.; Tai, V.W.-F.; Chung, I. H. F.;
Jiang, Q. Chem.Eur. J. 1996, 2, 50−57.
́
J.; Li, Y. Inorg. Chem. 2006, 45, 2403−2405.
(7) Yip, W.-P.; Yu, W.-Y.; Zhu, N.; Che, C.-M. J. Am. Chem. Soc.
2005, 127, 14239−14249.
(35) Robinson, K.; Gibbs, G. V.; Ribbe, P. H. Science 1971, 172,
567−570.
(8) Zimmer, R.; Collas, M.; Czerwonka, R.; Hain, U.; Reissig, H.
Synthesis 2008, 2, 237−244.
(36) A recent example features a dinuclear Fe(II) complex in which
one of the iron centers is found to be either six- or seven-coordinate
within the crystal as the result of a coordination shift of an exogeneous
carboxylate ligand. See Burger, B.; Dechert, S.; Große, C.; Demeshko,
S.; Meyer, F. Chem. Commun. 2011, 47, 10428−10430.
(37) Britovsek, G. J. P.; Gibson, V. C.; Spitzmesser, S. K.; Tellmann,
K. P.; White, A. J. P.; Williams, D. J. J. Chem. Soc., Dalton Trans. 2002,
1159−1171.
(9) Taylor, J. E. Can. J. Chem. 1984, 62, 2641−2645.
(10) de Boer, J. W.; Browne, W. R.; Harutyunyan, S. R.; Bine, L.;
Tiemersma-Wegman, T. D.; Alsters, P. L.; Hage, R.; Feringa, B. L.
Chem. Commun. 2008, 3747−3749.
(11) Chow, T. W.-S.; Liu, Y.; Che, C.-M. Chem. Commun. 2011, 47,
11204−11206.
(12) Bruijnincx, P. C. A.; van Koten, G.; Klein Gebbink, R. J. M.
Chem. Soc. Rev. 2008, 37, 2716−2744.
(38) Evans, D. F. J. Chem. Soc. 1959, 2003−2005.
(39) Oldenburg, P. D.; Feng, Y.; Pryjomska-Ray, I.; Ness, D.; Que, L.,
Jr. J. Am. Chem. Soc. 2010, 132, 17713−17723.
(13) Costas, M.; Mehn, M. P.; Jensen, M. P.; Que, L., Jr. Chem. Rev.
2004, 104, 939−986.
(40) Oldenburg, P. D.; Ke, C.-Y.; Tipton, A.; Shteinman, A. A.; Que,
L., Jr. Angew. Chem., Int. Ed. 2006, 45, 7975−7978.
(14) Hegg, E. L.; Que, L., Jr. Eur. J. Biochem. 1997, 250, 625−629.
(15) Chen, K.; Que, L., Jr. Angew. Chem., Int. Ed. 1999, 38, 2227−
2229.
́
(41) Company, A.; Gomez, L.; Fontrodona, X.; Ribas, X.; Costas, M.
Chem.Eur. J. 2008, 14, 5727−5731.
(16) Oldenburg, P. D.; Shteinman, A. A.; Que, L., Jr. J. Am. Chem.
(42) Duisenberg, A. J. M.; Kroon-Batenburg, L. M. J.; Schreurs, A. M.
M. J. Appl. Crystallogr. 2003, 36, 220−229.
Soc. 2005, 127, 15672−15673.
(17) Suzuki, K.; Oldenburg, P. D.; Que, L., Jr. Angew. Chem., Int. Ed.
2008, 47, 1887−1889.
(43) Schreurs, A. M. M.; Xian, X.; Kroon-Batenburg, L. M. J. J. Appl.
Crystallogr. 2010, 43, 70−82.
(18) (a) Bukowski, M. R.; Comba, P.; Lienke, A.; Limberg, C.; de
(44) Otwinowski, Z.; Minor, W. In Methods in Enzymology; Carter, C.
W., Jr., Sweet, R. M., Eds.; Academic Press: San Diego, 1997; Vol. 276,
pp 307−326.
́
Laorden, C. L.; Mas-Balleste, R.; Merz, M.; Que, L., Jr. Angew. Chem.,
Int. Ed. 2006, 45, 3446−3449. (b) Chow, T.W.-S.; Wong, E.L.-M.;
Guo, Z.; Liu, Y.; Huang, J.-S.; Che, C.-M. J. Am. Chem. Soc. 2010, 132,
13229−13239.
(45) Sheldrick, G. M. SADABS: Area-Detector Absorption Correction,
version 2.10; Universitat Gottingen: Gottingen, Germany, 1999.
̈
̈
̈
(19) (a) Rodriguez, M.-C.; Morgenstern-Badarau, I.; Cesario, M.;
Guilhem, J.; Keita, B.; Nadjo, L. Inorg. Chem. 1996, 35, 7804−7810.
(b) Beck, A.; Weibert, B.; Burzlaff, N. Eur. J. Inorg. Chem. 2001, 521−
(46) Blessing, R. H. Acta Crystallogr. 1995, A51, 33−38.
(47) Sheldrick, G. M. Acta Crystallogr. 2008, A64, 112−122.
(48) Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, G. L.;
Giacovazzo, C.; Guagliardi, A.; Moliterni, A. G. G.; Polidori, G.;
Spagna, R. J. Appl. Crystallogr. 1999, 32, 115−119.
527. (c) Beck, A.; Barth, A.; Hubner, E.; Burzlaff, N. Inorg. Chem.
̈
2003, 42, 7182−7188.
(20) (a) Dungan, V. J.; Wong, S. M.; Barry, S. M.; Rutledge, P. J.
Tetrahedron 2012, 68, 3231−3236. (b) Paria, S.; Halder, P.; Paine, T.
K. Inorg. Chem. 2010, 49, 4518−4523.
(49) Spek, A. L. Acta Crystallogr. 2009, D65, 148−155.
(50) Flack, H. D. Acta Crystallogr. 1983, A39, 876−881.
(51) (a) Ahlrichs, R.; Bar, M.; Baron, H.-P.; Bauernschmitt, R.;
(21) Jones, M. W.; Baldwin, J. E.; Cowley, A. R.; Dilworth, J. R.;
Karpov, A.; Smiljanic, N.; Thompson, A. L.; Adlington, R. M. Dalton
Trans. 2012, 41, 14068−14086.
̈
Bo
Haser, M.; Hattig, C.; Horn, H.; Huber, C.; Huniar, U.; Kattannek, M.;
cker, S.; Ehrig, M.; Eichkorn, K.; Elliott, S.; Furche, F.; Haase, F.;
̈
̈
̈
P
dx.doi.org/10.1021/ic400096e | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
̈
Kohn, A.; Kolmel, C.; Kollwitz, M.; May, K.; Ochsenfeld, C.; Ohm, H.;
̈
̈
Schafer, A.; Schneider, U.; Treutler, O.; Tsereteli, K.; Unterreiner, B.;
̈
von Arnim, M.; Weigend, F.; Weis, P.; Weiss, H. Turbomole, version
5.8; Theoretical Chemistry Group, University of Karlsruhe: Karlsruhe,
Germany, 2002. (b) Treutler, O.; Ahlrichs, R. J. Chem. Phys. 1995, 102,
346−354. (c) Turbomole basisset library, Turbomole, version 5.8; see
a. (d) Schafer, A.; Horn, H.; Ahlrichs, R. J. Chem. Phys. 1992, 97,
̈
2571−2577. (e) Andrae, D.; Haeussermann, U.; Dolg, M.; Stoll, H.;
Preuss, H. Theor. Chim. Acta 1990, 77, 123−141. (f) Schafer, A.;
̈
Huber, C.; Ahlrichs, R. J. Chem. Phys. 1994, 100, 5829−5835.
(g) Ahlrichs, R.; May, K. Phys. Chem. Chem. Phys. 2000, 2, 943−945.
(52) (a) PQS, version 2.4; Parallel Quantum Solutions: Fayetteville,
AR, 2001. (The Baker optimizer is available separately from PQS
upon request.) (b) Baker, J. J. Comput. Chem. 1986, 7, 385−395.
(53) (a) Becke, A. D. Phys. Rev. A 1988, 38, 3098−3100. (b) Perdew,
J. P. Phys. Rev. B 1986, 33, 8822−8824.
(54) (a) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785−
789. (b) Becke, A. D. J. Chem. Phys. 1993, 98, 1372−1377. (c) Becke,
A. D. J. Chem. Phys. 1993, 98, 5648−5652. (d) Calculations were
performed using the Turbomole functional b3-lyp, which is not
identical to the Gaussian B3LYP functional.
Q
dx.doi.org/10.1021/ic400096e | Inorg. Chem. XXXX, XXX, XXX−XXX