Interaction between anions and molybdenum allyl dicarbonyl complexes of 1,4,7-trithiacyclononane

Article abstract of DOI:10.1002/chem.201201327

The labile complex [MoCl(η³-methallyl)(CO) ₂(NCMe)₂] reacts with the ligand 1,4,7-trithiacyclononane ([9]aneS₃) and the salt NaBAr'₄ to afford [Mo(η³-methallyl)(CO)₂([9]aneS₃)][BAr' ₄] (1·BAr'₄). An analogous reaction of [MoBr(η³-allyl)(CO)₂(NCMe)₂] yields [Mo(η³-allyl)(CO)₂([9]aneS₃)][BAr' ₄] (2·BAr'₄). The new compounds 1·BAr' ₄ and 2·BAr'₄ were characterized by IR and NMR spectroscopic analysis and X-ray diffraction studies. Both compounds feature the cyclic thioether [9]aneS₃ coordinated as a tridentate ligand to the molybdenum center. The allyl ligand in 2·BAr'₄ is aligned with the middle of the OC-Mo-CO angle, which is acute. Both of these features are typical of most pseudo-octahedral allyl dicarbonyl molybdenum complexes. In contrast, the allyl group is rotated in 1·BAr'₄, which is attributed to steric hindrance between the methyl substituent and the ligated thioether, and the OC-Mo-CO angle is obtuse. Compound 1·BAr'₄ undergoes rapid substitution of [9]aneS₃ by either chloride and fluoride ions in dichloromethane, and the products include the known species [{Mo(η³-methallyl)(CO)₂}₂(μ-Cl) ₃]^- and a structurally similar new anionic complex with two fluoro and one hydroxo bridging ligands, respectively. Stable supramolecular adducts were formed in the reactions of 1·BAr'₄ and 2·BAr₄ with bromide, iodide, hydrogensulfate, and methanesulfonate compounds. The binding constants of these adducts in dichloromethane were calculated from ¹H NMR spectroscopic titration data, and the solid-state structures of the 1·Br, 1·HSO ₄, 1·I, and 2·I adducts were determined by X-ray diffraction studies. The surprising slightly higher stability of the iodide adduct relative to that of bromide was investigated theoretically, with the results pointing to an effect of the differential solvation of the halide ions.

Full text of DOI:10.1002/chem.201201327

DOI: 10.1002/chem.201201327
Interaction between Anions and Molybdenum Allyl Dicarbonyl Complexes
of 1,4,7-Trithiacyclononane
Dolores Morales,[a, b] Marcos Puerto,[a] Ignacio del Rꢀo,[a] Julio Pꢁrez,*[a] and
Ramꢂn Lꢂpez*[c]
Abstract:
The
labile
complex
are typical of most pseudo-octahedral
allyl dicarbonyl molybdenum com-
plexes. In contrast, the allyl group is ro-
tated in 1·BAr’4, which is attributed to
steric hindrance between the methyl
substituent and the ligated thioether,
and the OC-Mo-CO angle is obtuse.
Compound 1·BAr’4 undergoes rapid
substitution of [9]aneS3 by either chlor-
ide and fluoride ions in dichlorome-
thane, and the products include the
with two fluoro and one hydroxo bridg-
ing ligands, respectively. Stable supra-
molecular adducts were formed in the
reactions of 1·BAr’4 and 2·BAr4 with
bromide, iodide, hydrogensulfate, and
methanesulfonate compounds. The
binding constants of these adducts in
dichloromethane were calculated from
1H NMR spectroscopic titration data,
and the solid-state structures of the
1·Br, 1·HSO4, 1·I, and 2·I adducts were
determined by X-ray diffraction stud-
ies. The surprising slightly higher stabil-
ity of the iodide adduct relative to that
of bromide was investigated theoreti-
cally, with the results pointing to an
effect of the differential solvation of
the halide ions.
[MoCl(h3-methallyl)(CO)
CTHUNGTRENNUNG
reacts with the ligand 1,4,7-trithiacyclo-
nonane ([9]aneS3) and the salt NaBAr’4
to
methallyl)(CO)2([9]aneS3)]
(1·BAr’4). An analogous reaction of
[MoBr 2A(NCMe)2] yields
(h3-allyl)(CO)
[Mo [BAr’4]
(h3-allyl)(CO)2([9]aneS3)]
afford
[Mo(h3-
AHCTUNGERTG[NNUN BAr’4]
T
CHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
(2·BAr’4). The new compounds 1·BAr’4
and 2·BAr’4 were characterized by IR
and NMR spectroscopic analysis and
X-ray diffraction studies. Both com-
pounds feature the cyclic thioether
[9]aneS3 coordinated as a tridentate
ligand to the molybdenum center. The
allyl ligand in 2·BAr’4 is aligned with
the middle of the OC-Mo-CO angle,
which is acute. Both of these features
known
methallyl)(CO)2}
turally similar new anionic complex
species
[{Mo(h3-
CTHUNGTRENNUNG
Keywords: host–guest systems
·
hydrogen bonds · metal complexes ·
receptors · thioethers
Introduction
cause not long ago their very existence was under discus-
sion.[4] The incorporation of metal centers as part of the
structure of anion hosts has been found to serve several pur-
poses and is a fast developing area.[5]
Hydrogen bonds are the most useful interactions in nonco-
valent synthesis[1] and in anion-coordination chemistry.[2]
ꢀ
Although NH groups are most often employed, C H
Bedford, Tucker, and co-workers reported in 2008 the
first metal-based anion host that employed aliphatic CH
groups as the sole hydrogen-bond donor groups.[6] In their
host, a penta-coordinate palladium complex (Figure 1), the
metal center provides the positive charge and enforces a
conformation of the 1,4,7-trithiacyclononane ([9]aneS3)
ligand in which some of the CH groups converge toward an
anionic guest. Unfortunately, chloride and, to a lesser extent
bromide, ions were nucleophilic enough to displace the
[9]aneS3 ligand. As [9]aneS3 is a fac-capping tridentate
ligand, we reasoned that a metal fragment with three mutu-
groups, especially in aromatic systems, as hydrogen-bond
donors, are receiving growing attention.[3] The theoretical
study of these weak interactions is a challenging area be-
[a] Dr. D. Morales, M. Puerto, Dr. I. del Rꢀo, Dr. J. Pꢁrez
Departamento de Quꢀmica Orgꢂnica e Inorgꢂnica – IUQOEM
Facultad de Quꢀmica, Universidad de Oviedo – CSIC
33006 Oviedo (Spain)
E-mail: japm@uniovi.es
[b] Dr. D. Morales
Instituto Tecnolꢃgico de Materiales (ITMA)
Centro de Investigaciꢃn en Nanomateriales y Nanotecnologꢀa
(CINN) – CSIC
Parque Tecnolꢃgico de Asturias, 33428 Llanera (Spain)
[c] Dr. R. Lꢃpez
Departamento de Quꢀmica Fꢀsica y Analꢀtica
Universidad de Oviedo, C/Juliꢂn Claverꢀa, no. 8
33006 Oviedo (Spain)
E-mail: rlopez@uniovi.es
ꢀ
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201201327.
Figure 1. The first metal-based anion host that employed aliphatic C H
groups as hydrogen-bond donor groups.[6]
16186
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16186 – 16195

FULL PAPER
ally fac, open coordination sites should be an ideal match to
the OC-Mo-CO angle is acute, and accordingly the n˜CO IR
band is the most intense at higher frequencies. Compound
1·BAr’4 is a rare example in which the OC-Mo-CO angle is
obtuse (98.1(3)8). In the IR spectrum of 1·BAr’4 in CH2Cl2,[9]
the most intense n˜CO band occurs at a lower frequency, thus
indicating that the rare, wide OC-Mo-CO angle persists in
solution.
form a stable complex.
The {Mo
because previous studies have shown that [Mo
allyl)(CO)2L3][BAr’4] derivatives, where L3 represents three
ACHTUNGTRENNUNG
(h3-allyl)(CO)2} metal fragments were selected
AHCTUNGTRENNUNG
(h3-
ACHTUNGTRENNUNG
monodentate ligands or a tridentate ligand that occupies
three fac-capping positions, could be straightforwardly pre-
pared and are relatively stable.[7] Tetraarylborate BAr’4
(Ar’=3,5-bis(trifluoromethyl)phenyl) was chosen as the
counteranion of the cationic complex hosts to minimize the
competition with the anionic guests.[5e,f]
In 2·BAr’4, as in most pseudo-octahedral [Mo
AHCTUNTGRENGUN(N CO)2L3] complexes, the allyl open mouth points toward the
(h3-allyl)-
ACHTUNGTRENNUNG
middle of the OC-Mo-CO angle. In contrast, the h3-methall-
yl ligand is severely rotated in 1·BAr’4 (Figure 2). The inter-
dependence between the OC-M-CO angle and the orienta-
tion of the allyl ligand in tris(pyrazolyl)borate tungsten allyl
dicarbonyl species has been studied by Templeton and co-
workers by using extended Hꢅckel molecular orbital
(EHMO) calculations on a model complex.[10] The adoption
of the allyl-rotated configuration in 1·BAr’4 can be attribut-
Results and Discussion
Compounds
(1·BAr’4) and [Mo
[Mo(h3-methallyl)(CO)2([9]aneS3)]
(h3-allyl)(CO)2([9]aneS3)]
[BAr’4] (2·BAr’4)
ACHTUNGERTN[NUNG BAr’4]
N
ACHTUNGTRENNUNG
were prepared from [9]aneS3, NaBAr’4, and either
[MoCl(h3-methallyl)(CO) (h3-
2A(MeCN)2] or [MoBr
allyl)(CO)2A(MeCN)2], respectively, and were characterized
by IR and NMR spectroscopic analysis in solution and
single-crystal X-ray diffraction studies (see Figure 2 and the
Supporting Information) in the solid state. No pseudo-octa-
ed to the steric hindrance between the methyl group of the
3
ꢀ
G
N
h -methallyl group and the C H groups of [9]aneS3, which
E
would occur in a normal, nonrotated structure such as in
2·BAr’4. Steric interactions have been deemed responsible of
the observed allyl orientations in complexes with substituted
allyl groups.[11]
1
hedral compounds with [Mo
N
In the room-temperature H NMR spectra of 1·BAr’4 and
dentate S-donor ligand) have been previously reported.
The two intense n˜CO IR bands of the new compounds (n˜ =
1983 and 1909 cmꢀ1 for 1·BAr’4 and n˜ =1980 and 1910 cmꢀ1
for 2·BAr’4 in CH2Cl2) consistent with a cis-{Mo(CO)2} ge-
ometry occur at wavenumbers significantly higher than
2·BAr’4, the six exo hydrogen atoms give rise to a single
multiplet, and the same is true for the six endo hydrogen
atoms, thus indicating the occurrence of a dynamic process
that lends to the complex an apparent C3 axis. Trigonal twist
rearrangements are generally proposed to account for a sim-
ilar dynamic behavior in pseudo-octahedral [Mo(h3-methal-
lyl)(CO)2L3] complexes.[12] In low-temperature (ꢀ608C)
NMR spectra, each of these multiplets splits into two signals
with integrals in a 2:4 ratio, thus suggesting a Cs-symmetric
instantaneous structure such as that found in the solid-state
structure of 2·BAr’4.
those for [Mo(h3-methallyl)(CO)2N3]
ACHTNUGTRENUNG[BAr’4] (N3 =three
monodentate N-donor ligands or a tridentate N-donor
ligand) compounds, thus indicating that the cyclic thioether
is less electron donating than the N-donor ligands. The
[9]aneS3 ligand has been found to be a relatively strong
p acceptor.[8]
1
In the cationic complex of 2·BAr’4, as in the vast majority
The H NMR spectrum of 1·BAr’4 in solution displays a
of pseudo-octahedral [Mo(h3-methallyl)(CO)2L3] complexes,
singlet for the two syn allyl hydrogen atoms and another sin-
glet for the anti allyl hydrogen
atoms, thus reflecting the exis-
tence of a molecular mirror
plane. However, such a plane
is absent in the solid-state
structure, in which the allyl
group is far from an endo or
exo orientation; furthermore, it
should be noted that the solu-
tion IR spectrum suggests a
solution instantaneous struc-
ture such as that found in the
solid state. Therefore, a dy-
namic process must make the
two allyl termini equivalent.
The most obvious process that
could achieve this state is a ro-
tation of the allyl group
ꢀ
around the Mo allyl bond.
This process would rapidly
Figure 2. a) Thermal-ellipsoid (30%) plot of the cation in 1·BAr’4; b) a simplified view that shows the relative
orientation of the allyl and carbonyl groups.
Chem. Eur. J. 2012, 18, 16186 – 16195
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16187

J. Pꢁrez, R. Lꢃpez et al.
equilibrate a structure such as that displayed in Figure 2 and
and associated errors (also given in Figure 3; see below for
the fluoride and chloride ions).[13] These results reflect re-
ceptor–anion interactions of significant strength. The bind-
ing constants for all the anions were larger for 2·BAr’4 than
for 1·BAr’4 (Table 1). Density-functional calculations in
its enantiomer with a h3-C4H7 ligand rotated 1808 about the
ꢀ
Mo allyl axis. Such a process, with the mentioned enantiom-
ers as energy minima, would explain the observed solution
spectra. IR spectroscopic analysis (a fast technique in which
timescale fluxional processes appear to be slow) affords two
bands of those enantiomeric species (therefore, with identi-
cal spectral properties), whereas average signals are seen in
the 1H NMR spectra; therefore, at low temperature, at
which the proposed trigonal twist slows down, this second
process remains operative.
Table 1. Binding constants Ka (mꢀ1) for the 1:1 adducts of 1·BAr’4 and
2·BAr’4 with various anions.[a]
Anion
Br
I
CH3SO3
HSO4
1·BAr’4
2·BAr’4
2320
11252
4699
14408
2933
15990
3839
–
The behavior of 1·BAr’4 and 2·BAr’4 toward tetrabutylam-
monium fluoride, chloride, bromide, iodide, hydrogensulfate,
and methanesulfonate salts was studied by NMR and IR
spectroscopy in solution with dichloromethane. As previous-
ly found by Bedford, Tucker, and co-workers[6] one of the
[9]aneS3 multiplets in the 1H NMR spectrum that corre-
sponds to the exo hydrogen atoms shifted significantly
downfield upon the addition of the tetrabutylammonium
salts. The variation in the chemical shift of this signal as a
response to the addition of the salts was used to construct
the titration profiles shown in Figure 3. The presence of a
single set of signals for these hydrogen atoms along the ti-
tration indicated fast anion exchange. The data were treated
with the WinEQ-NMR program to obtain binding constants
[a] Errors are estimated <20%.
CH2Cl2 on the cationic complexes in 1·BAr’4 and 2·BAr’4
reveal that the areas of the triangles defined by the three
hydrogen atoms oriented toward the inside of the macrocy-
cle are 2.316 and 2.239 ꢆ2, respectively (see Figure S1 in the
Supporting Information). These values are in line with the
areas calculated from the experimentally determined solid-
state structures of 1·BAr’4 and 2·BAr’4 (2.248 and 2.190 ꢆ2,
respectively). The interaction of anions with both hosts in-
creased these areas; however, this increase was larger in the
allyl host (for instance, an increase of 1.8% in the interac-
Figure 3. 1H NMR titration profiles of 1·BAr’4 with tetrabutylammonium salts in CD2Cl2.
16188
www.chemeurj.org
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16186 – 16195

Molybdenum Complexes as Anion Receptors
FULL PAPER
tion with the bromide ion). Therefore, the lower binding
constants for the methallyl complex can be traced to its
higher rigidity because it is more difficult to expand the size
of the binding cavity of the macrocycle to accommodate
guest species in the case of 1·BAr’4.
The binding constants are somewhat larger for iodide
than bromide ions. Although the differences were small,
these results, which were reproducible and occur with both
1·BAr’4 and 2·BAr’4, are contrary to our expectations for a
noncyclic receptor; therefore, this aspect was theoretically
investigated (see below).
ing Information). Depending on the crystallization condi-
tions, a salt with either the Bu4N+ or [Mo(h3-methallyl)-
ACHUTNGRENNUG ACHTUNGTRENNUNG
(CO)2([9]aneS3)]+ ion was isolated. The analogous [{Mo(h3-
allyl)(CO)2}2ACTHNUTRGNEUNG
(m-Cl)3]ꢀ ion has been previously reported.[14]
Moreover, the presence of free [9]aneS3, in an amount that
increased upon the addition of chloride ions, was established
1
by H and 13C NMR spectroscopic analysis of the 1·BAr’4/
Bu4NCl mixtures. These results indicate that, besides an ini-
tial formation of a 1:Cl supramolecular adduct, displacement
of the [9]aneS3 ligand by chloride ions occurred to a signifi-
cant extent. With the IR and 1H NMR spectra of the
From the same NMR data, Job plots were constructed to
confirm the 1:1 stoichiometry. Figure 4 shows the results for
the bromide ions and 1·BAr’4.
2ACTHNGUTERNNUG
[{Mo(h3-methallyl)(CO)2} (m-Cl)3]ꢀ ion in hand, it could be
established that over 1–2 hours, the time required to carry
out an NMR titration experiment, this dimeric complex was
the major Mo-containing product of the reaction of 1·BAr’4
and Bu4NCl. These results can be taken as a warning that an
NMR titration experiment with a well-behaved appearance
alone cannot be taken as evidence of host stability.
A different cyclic tridentate ligand, namely, b-2,4,6-tri-
methyl-s-trithiane, was labile when coordinated to {M(CO)3}
(M=Mo, W) fragments,[15] which is attributed to the strain
due to the small size of the ring. However, we could not
find previous reports of the labile behavior of [9]aneS3 other
than its displacement by halide ions from its palladium com-
plex in the seminal studies by Bedford, Tucker, and co-
workers, which inspired the present study.[6] The [9]aneS3
ligand has formed stable complexes with several metal cati-
ons,[16] and the soft S donors were expected to be a good
match for the organometallic molybdenum allyl dicarbonyl
fragments. In fact, none of the features in the structures of
1·BAr’4 or 2·BAr’4 suggested that the [9]aneS3 ligand should
ꢀ
be labile: the Mo S distances (2.5308(16), 2.5309(16), and
2.4584(15) ꢆ in 1·BAr’4 and 2.5120(14), 2.5433(16), and
2.5350(13) in 2·BAr’4) and the average torsion angles (S-C-
C’-S’: 49.2, C’-S-C-C’: 132.78 in 1·BAr’4 and S-C-C’-S’: 49.9,
C’-S-C-C’: 132.18 in 2·BAr’4) are similar to those found in
the Mo0 complex [Mo(CO)3([9]aneS3)][17] and in the MoII
hepta-coordinated ionic complex [MoI(CO)3([9]aneS3)]+.[18]
Moreover, in [9]aneS3, unlike in other cyclic thioethers, the
ligand is quite rigid and the sulfur atoms are endodentate,[19]
so that the conformation of the free ligand is quite similar
to that needed for facial coordination, that is, metal coordi-
nation requires minimal rearrangement of the ligand. The
displacement of the tridentate S-donor ligand by chloride
ions is therefore surprising and illustrates the highly nucleo-
philic character of the chloride ion in low-solvating
media.[20] Although unstable toward the chloride ion, com-
plexes 1·BAr’4 and 2·BAr’4 were stable in the presence of
excess bromide ions and therefore more stable than the pal-
ladium–[9]aneS3 host.
Figure 4. Representation Job plot for the titration of 1·BAr’4 with
[Bu4N]Br in CD2Cl2.
Titration profiles with an apparently correct shape, and
that afforded large (larger than for the bromide ions) bind-
ing constants were obtained for chloride ions (see the Sup-
porting Information). However, repeated attempts failed to
afford correct 1:1 Job plots for this anion. The inspection of
the whole spectrum of each 1·BAr’4/Bu4NCl mixture along
the titration series revealed that a new, NMR-distinct, h3-
methallyl-containing species gradually formed as tetrabuty-
lammonium chloride was added. Accordingly, monitoring
the reaction of 1·BAr’4 and successive additions of Bu4NCl
by means of IR spectroscopic analysis of the n˜CO bands
showed the rapid formation of a new species with n˜CO IR
bands substantially different from those of 1·BAr’4. Analo-
gous results were obtained for 2·BAr’4, and bis(triphenyl-
phosphine)iminium chloride ([PPN]Cl) as the chloride
source.
The addition of an equimolar amount of Bu4NF·3H2O to
solutions of 1·BAr’4 in dichloromethane lowered the n˜CO
bands from n˜ =1982 and 1908 cmꢀ1 to n˜ =1912 and
1809 cmꢀ1, thus suggesting, again, major changes in the mo-
lybdenum first coordination sphere. Attempts to isolate an
adduct between the cationic complex 1 and fluoride ion by
crystallization from 1/Bu4NF·3H2O mixtures led to products
Fortunately, the new species could be isolated by fraction-
al crystallization and was found to be a salt containing the
2ACTHNUTRGNEUNG
binuclear anion [{Mo(h3-methallyl)(CO)2} (m-Cl)3]ꢀ (deter-
mined by IR and NMR spectroscopic analysis and X-ray dif-
fraction studies; see the Experimental Section and Support-
Chem. Eur. J. 2012, 18, 16186 – 16195
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16189

J. Pꢁrez, R. Lꢃpez et al.
that did not contain [9]aneS3. Crystals of X-ray quality iso-
lated from these mixtures were used for structural determi-
nation and revealed the formation of an anionic complex
with two fluoro and one hydroxo ligands that bridge two
{Mo(h3-methallyl)(CO)2} fragments, thereby structurally
water molecules likely present in the highly hygroscopic tet-
rabutylammonium salts. In all the examples, the [9]aneS3
ligand maintains its tridentate coordination to the molybde-
num center, and the orientation of the h3-allyl ligand and
the OC-Mo-CO angles are similar to those in the parent
compounds 1·BAr’4 and 2·BAr’4. The structure of 1·HSO4
shows the typical hydrogen bonding between hydrogensul-
fate anions, the presence of dichloromethane, and two dif-
ferent cationic complexes, in one of which three of the exo
2ACTHNUTRGNEUNG
similar to [{Mo(h3-methallyl)(CO)2} (m-Cl)3]ꢀ (see above).
The presence of the hydroxo ligand is attributed to water
(the tetrabutylammonium fluoride used is a hydrate) depro-
tonation by the basic fluoride ion. In solution, the most in-
formative spectroscopic signal of this compound is a triplet
ꢀ
C H groups of the [9]aneS3 ligand converge toward one hy-
that corresponds to the Mo CO ligands in the 13C NMR
drogensulfate anion (Figure 5). This latter feature is also en-
countered in one out of the three independent cationic com-
plexes found in 1·Br (Figure 5), but not in 1·I (five inde-
pendent cations) or 2·I, in which only single-point hydrogen
ꢀ
spectrum and results from coupling to the two 19F bridges.
In the crude reaction solutions, this compound was mixed
with a species that displays a quadruplet for the CO ligands
in the 13C NMR spectra (the ratio of both signals could be
altered by changing the Mo/fluoride ion ratio), thus suggest-
ing the presence of a structurally related complex with three
bridging fluoride ligands; however, this species has not been
isolated so far. A plethora of binuclear and trinuclear com-
plexes containing {Mo(h3-methallyl)(CO)2} fragments linked
by two or three hydroxo or fluoro bridging ligands (includ-
ꢀ
bonds exist between each iodide ion and the C H groups of
the [9]aneS3 ligand of a given complex. This lack of a struc-
ture with hydrogen bonds between each iodide ion and
ꢀ
three C H groups of a metal complex must be attributed to
packing factors because binding constants (see Table 1) indi-
cate that the interaction between the [9]aneS3 complexes
(both allyl and methallyl) and iodide ions is even stronger
than with bromide ions. Also, NMR spectroscopic studies
ing a salt of the [{Mo(h3-methallyl)(CO)2}
have been isolated by Limberg and co-workers from the re-
actions of either the labile [Mo(methallyl)(CO)2A(NCMe)2-
(THF)]BF4 precursor, its triflate analogue, or
2A
(m-F)3]ꢀ complex)
2ACHTUNGTRENNUNG
ꢀ
show that the [9]aneS3 C H signals undergo larger down-
R
field shifts as a response to the presence of anions in the
ACHTUNGTRENNUNG
1H NMR spectra, thus indicating that contacts between the
ꢀ
[MoCl(methallyl)(CO)
G
anions and allyl C H groups in the solid state do not reflect
or fluoride in the presence of [18]crown-6.[21]
significant interactions in solution.
Equimolar amounts of either 1·BAr’4 or 2·BAr’4 and tetra-
butylammonium bromide, iodide, hydrogensulfate, and
methanesulfonate salts were dissolved in CH2Cl2, the solvent
was evaporated, and the solid residue was washed with di-
To understand why binding is stronger with iodide than
with bromide ions, which is difficult to rationalize for a non-
rigid host, we carried out a theoretical study on the forma-
tion of 1·Br and 1·I. In the gas phase, first, we investigated
the process (1) and then, with the aim of taking into account
the discrete nature of the dicholoromethane solvent mole-
cules, the reactions (2) and (3). In the three cases X= Br, I.
ethyl ether to remove the very soluble salt [Bu4N]ACTHNUTRGEN[UNG BAr’4].
The remaining solid was dried under vacuum and dissolved
in CH2Cl2, and these solutions were layered with hexane
and allowed to slowly diffuse
at ꢀ208C. In the cases of 1·Br,
1·HSO4, 1·I, and 2·I, single
crystals were obtained that
were used for structural deter-
mination by X-ray diffraction
studies. The IR n˜CO bands (see
the Experimental Section) of
these compounds occur at
slightly lower values than those
of 1·BAr’4 and 2·BAr’4, as ex-
pected for weak interactions
that slightly increase the elec-
tron density at the metal cen-
ters, and therefore back dona-
tion to the CO ligands. The re-
sults confirm the stability of
the molybdenum complex host
in the presence of these anions
and show complex networks of
hydrogen bonds involving not
only the added anions and the
cationic complex, but also
Figure 5. Thermal-ellipsoid (30%) plot of a) 1·Br and b) 1·HSO4.
16190
www.chemeurj.org
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16186 – 16195

Molybdenum Complexes as Anion Receptors
FULL PAPER
ꢀ
ꢀ
Finally, we studied all the above-mentioned reactive proc-
esses in bulk dichloromethane modeled by a continuum-sol-
vation model (see Computational Methods). The BLYP[22]
density-functional method along with the basis set 6-31+G-
is affected more by the solvent than the C H···I interac-
tion. Concerning the energetics of the formation of the Mo
complexes, our results for the gas-phase Gibbs energy
(DGgas) for reaction (1) shows that the formation of [Mo]+
Brꢀ is 5.6 kcalmolꢀ1 more favored than [Mo]+Iꢀ. The pres-
ence of one and two CH2Cl2 molecules [reactions (2) and
(3)] reduces this value to 3.5 and 2.0 kcalmolꢀ1, respectively.
The inclusion of a continuum medium in the calculations de-
creases the difference in stability between the bromide- and
iodide-containing Mo complexes even more, on the basis of
our values of the solution Gibbs energy (DGsolution) for reac-
tions (1)–(3). For instance, in the absence of discrete CH2Cl2
molecules, the relative greater stability of [Mo]+Brꢀ is just
1.0 kcalmolꢀ1. This value can mainly be attributed to a
ACHTUNGTRENNUNG
(d,p)[23] (LANL2DZ for Mo[24] and LANL2DZpd for bro-
mide and iodide ions[25]) were chosen as the computational
level to investigate reactions (1)–(3) (see Computational
Methods for details).
½Moꢁþ þ Xꢀ ! ½MoꢁþXꢀ
½Moꢁþ þ XꢀðCH2Cl2Þ ! ½MoꢁþXꢀ þ CH2Cl2
½Moꢁþ þ XꢀðCH2Cl2Þ2 ! ½MoꢁþXꢀðCH2Cl2Þ þ CH2Cl2
ð1Þ
ð2Þ
ð3Þ
The optimized geometries
obtained for complexes [Mo]+
Xꢀ
ACHTUNGTRENNUNG(CH2Cl2)n (X=Br, I; n=0,
1) are collected in Figure 6,
whereas the remaining species
involved in reactions (1)–(3)
are displayed in Figure S2 (see
the Supporting Information).
Tables S1 and S2 (see the Sup-
porting Information) list the
energy data for these reactions.
The geometrical parameters
for the analogous structures in-
volving the bromide and
iodide ions are very similar,
except those that imply the in-
teraction of a halide ion with a
hydrogen atom. In complexes
[Mo]+Xꢀ
ACHTUNGTRENNUNG(CH2Cl2)n (X=Br, I;
n=0, 1), in which the halide
ion always establishes a simul-
taneous interaction with one
hydrogen atom of each ethyl-
ene group of the [9]aneS3
ligand, the C-H···Brꢀ distances
are about 0.21–0.24 ꢆ shorter
than the H···Iꢀ distances both
in the gas phase and in solu-
tion with dichloromethane. In
solution, however, the inclu-
sion of the continuum-solva-
tion model in the gas-phase
calculations provokes a length-
ꢀ
ꢀ
ening of the average C H···X
distances by about 6.3 and
5.7% for the bromide and
iodide ions in [Mo]+Xꢀ, re-
spectively, and about 5.1 and
4.6% for the bromide and
iodide ions in [Mo]+Xꢀ-
Figure 6. BLYP/6-31+GACTHNUGTRNEUNG(d,p) (LANL2DZ for Mo and LANL2DZpd for Br and I) optimized geometries in di-
chloromethane of the complexes [Mo]+X
stroms.
ꢀ(CH2Cl2)n (X=Br, I; n=0, 1). The main distances are given in ang-
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG(CH2Cl2), respectively. There-
ꢀ
ꢀ
fore, the C H···Br interaction
Chem. Eur. J. 2012, 18, 16186 – 16195
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16191

J. Pꢁrez, R. Lꢃpez et al.
greater stabilization of bromide over iodide ions when going
from the gas phase to solution in dichloromethane owing to
the smaller size of the bromide over iodide ions. Indeed, the
Gibbs energy of solvation is ꢀ50.2 and ꢀ55.4 kcalmolꢀ1 for
the bromide over iodide ions, respectively. Consequently, in
the presence of two discrete CH2Cl2 molecules, the forma-
tion of the iodide complex is just 0.1 kcalmolꢀ1 more fa-
vored relative to the bromide complex, as it could be ex-
pected from the DGgas values obtained for reactions (1)–(3).
This small difference in Gibbs energy is very close to that
obtained (0.4 kcalmolꢀ1) from the equilibrium constants of
formation for complexes [Mo]+Iꢀ and [Mo]+Brꢀ (4.7ꢇ103
and 2.3ꢇ103 mꢀ1, respectively) determined experimentally.
Therefore, the solvent plays a crucial role in making the
iodide complex slightly more stable than the bromide com-
plex.
were prepared by simple substitution reactions from free
[9]aneS3 and suitable labile sources of the {MoACHTUNTRGNE(UNG allyl)(CO)2}
moieties and fully characterized in solution and the solid
state. Their behavior toward several anions (as their tetrabu-
tylammonium salts) was studied spectroscopically in di-
chloromethane by using NMR and IR spectroscopic analy-
sis. Fluoride and chloride ions were found to displace the tri-
dentate thioether ligand. For other anions, supramolecular
ion-pair adducts were formed of sufficient stability to allow
their characterization in solution and, in several examples,
also in the solid state by single-crystal X-ray diffraction
studies. Solution NMR spectroscopic data indicate that the
ꢀ
C H groups of coordinated [9]aneS3 are the main hydrogen-
bond donors. X-ray data for some anions show that those
ꢀ
hydrogen bonds occur simultaneously with three C H
groups of each [9]aneS3 ligand. Binding constants between
To check whether the strength of the C-H···Xꢀ interac-
tions could also influence the relative stabilities of the
[Mo]+Brꢀ and [Mo]+Iꢀ complexes, we performed a thor-
ough theoretical analysis of these interactions at the [Mo]+
the compounds 1·BAr’4 and 2·BAr’4 and several anions were
1
calculated from H NMR titration experiments. Iodide ions
were found to form adducts more stable than those of bro-
mide ions, which is an unexpected result that, according to
theoretical calculations, can be traced to the stronger solva-
tion of bromide ions by CH2Cl2.
Brꢀ(CH2Cl2) and [Mo]+Iꢀ
ACHTUNGTRENNUNG ACHTUNGTRNE(NUGN CH2Cl2) optimized geometries in
dichloromethane solution (Figure 6). Taking the van der
Waals radii (r) for hydrogen, bromine, and iodine atoms
(1.20, 1.95, and 2.15 ꢆ, respectively)[26] as a reference, the
ꢀ
ꢀ
average C H···Br distance relative to the sum of r(H)+
Experimental Section
r(Br) is shortened by 13.9%. To a lesser extent (12.5%), the
same trend was found for the iodide adduct. A natural-bond
order (NBO) analysis indicates that the donor–acceptor in-
teraction from the lone pairs of electrons on the bromide
ꢀ
All manipulations were carried out at room temperature in a dinitrogen
atmosphere employing Schlenk techniques. CH2Cl2 was dried over CaH2,
diethyl ether over Na/benzophenone, and hexane and toluene over
sodium. IR and NMR spectra were recorded on Perkin–Elmer FT1720-X
and Bruker AV-400 or AV-300 spectrometers, respectively. Deuterated
solvents were degassed by three freeze/pump/thaw cycles, dried over mo-
lecular sieves (4 ꢆ), and stored in the dark. [MoCl(h3-methallyl)(CO)2-
ion to the three antibonding s*ACTHUNRTGNEUNG(C H) bonds of the [9]aneS3
ligand is about 1.0 kcalmolꢀ1 more stabilizing than that
found for the iodide anion (see Table S3 in the Supporting
ACTHNUTRGNEUNG
Information). For both the [Mo]+Brꢀ(CH2Cl2) and [Mo]+Iꢀ-
ACHTUNGTRENNUNG(CH2Cl2) complexes, we also located bond critical points of
the electron density between the halide ion that act as the
(NCMe)2],[28] (methallyl=C3H4-Me-2), [MoBr
A
2ACHTUNGTRENNUNG(NCMe)2],
[29]
and NaBAr’4
Figure 7). All other chemicals were purchased and used without further
purification.
ꢀ
hydrogen-bond acceptor and the C H donor (see the Table
[Mo(h3-methallyl)(CO)2([9]aneS3)]
[BAr’4] (1·BAr’4):
A
mixture of
S3 in the Supporting Information) and those points that cor-
respond to the bromide ion have an average electron density
(0.0130 eꢆꢀ3) slightly higher than those for the iodide ion
(0.0112 eꢆꢀ3). Also, our topological analysis of the electron
density reveals positive values for both the Laplacian of the
[MoCl(h3-methallyl)(CO)
CTHUNGTRENNUNG
(0.056 g, 0.308 mmol), and NaBAr’4 (0.273 g, 0.308 mmol) in CH2Cl2
(15 mL) was stirred for 2 h. After filtration through diatomaceous earth,
the yellow solution was evaporated to dryness and the resulting solid was
2
electron density (5 1(r)) and the total energy density (H(r))
ꢀ
ꢀ
at the C H···X hydrogen-bond critical points, thus making
them weak hydrogen bonds according to the recent classifi-
cation by Rozas et al.[27] In summary, C H···X hydrogen
ꢀ
ꢀ
bonds in [Mo]+Brꢀ
(CH2Cl2) are stronger than those in
[Mo]+Iꢀ
ACHTUNGTRENNUNG(CH2Cl2), as expected based on the relative sizes of
Figure 7. Labeling scheme for BAr’4, methallyl, and exo hydrogen atoms
bromine and iodine atoms. However, the stronger solvation
of the bromide ions by dichloromethane overcomes that dif-
ference, thus leading to the experimentally encountered
slightly higher stability of the iodide adduct.
of [9]aneS3.
washed with hexane (3ꢇ5 mL). By slow diffusion of hexane into a con-
centrated solution of 1·BAr’4 in CH2Cl2 at ꢀ208C yellow crystals were
obtained (0.319 g, 83%), one of which was used for X-ray analysis.
1H NMR (CD2Cl2): d=7.84 (m, 8H; Ho of BAr’4), 7.68 (s, 4H; Hp of
BAr’4), 3.24 (s, 2H; Hsyn), 3.14 (m, 6H; [9]aneS3), 2.77 (m, 6H; [9]aneS3),
2.36 (s, 3H; CH3 of methallyl), 2.10 ppm (s, 2H; Hanti); 13C{1H} NMR
Conclusion
(CD2Cl2): d=219.6 (s; CO), 161.5 (q, (J
(s; Co of BAr’4), 128.5 (m; Cm of BAr’4), 124.6 (q, (J
BAr’4), 117.5 (s; Cp of BAr’4), 103.9 (s; Cc of methallyl), 57.8 (s; Ct of
ACHTUNGTRENNUNG
Compounds
(1·BAr’4) and [Mo
[Mo(h3-methallyl)(CO)2([9]aneS3)]
(h3-allyl)(CO)2([9]aneS3)]
[BAr’4] (2·BAr’4)
ACHTUNGERTN[NUNG BAr’4]
AHCTUNGTRENNUNG
N
ACHTUNGTRENNUNG
16192
www.chemeurj.org
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16186 – 16195

Molybdenum Complexes as Anion Receptors
FULL PAPER
methallyl), 33.8 (s; [9]aneS3), 23.1 ppm (s; CH3 of methallyl); IR
(CH2Cl2): n˜ =1982 s, 1908 vs (nCO) cmꢀ1 ; elemental analysis calcd (%)
for C44H31BF24MoO2S3: C 42.26, H 2.50; found: C 43.24, H 2.43.
allyl), 2.00 ppm (s, 2H; Hanti); 13C{1H} NMR (CD2Cl2): d=221.3 (s; CO),
102.2 (s; Cc of methallyl), 57.0 (s; Ct of methallyl), 39.5 (s; CH3SO3), 34.7
(s; [9]aneS3), 23.4 ppm (s CH3 of methallyl); IR (CH2Cl2): n˜ =1975 s,
1898 vs (nCO) cmꢀ1; elemental analysis calcd (%) for C13H22MoO5S4: C
32.36, H 4.56; found: C 34.38, H 4.77.
Reaction of 1·BAr’4 with [nBu4N]Cl: A solution of 1·BAr’4 (0.100 g,
0.080 mmol) and [nBu4N]Cl (0.022 g, 0.080 mmol) in CH2Cl2 (15 mL) was
stirred for 15 min. The solvent was removed under vacuum and the
yellow residue was washed with diethyl ether (5ꢇ5 mL). The collected
washings of diethyl ether were dried in vacuum and the resulting solid
was dissolved in dichloromethane (10 mL). This solution was layered
with hexane (20 mL), and yellow crystals of [Mo(h3-methallyl)-
[Mo
N
ACHTUGNRTEN[NUGN BAr’4] (2·BAr’4): A mixture of [MoBr-
A
CHTUNGTRENNUNG
0.422 mmol), and NaBAr4’ (0.374 g, 0.422 mmol) in CH2Cl2 (15 mL) was
stirred for 2 h. After filtration through diatomaceous earth, the solvent
was removed in vacuo. The yellow residue was dissolved in CH2Cl2
(5 mL), layered with hexane (20 mL), and placed at ꢀ208C until forma-
tion of yellow crystals (ca. 3 days; 0.427 g, 82%). One of these crystals
was employed for the solid-state structure determination of 2·BAr’4 by
X-ray diffraction studies. 1H NMR (CD2Cl2): d=7.77 (m, 8H; Ho of
BAr’4), 7.61 (s, 4H; Hp of BAr’4), 4.17 (m, 1H; Hc), 3.33 (d, J=6.5 Hz;
2H; Hsyn), 3.03 (m, 6H; [9]aneS3), 2.74 (m, 6H; [9]aneS3), 1.97 ppm (d,
J=9.6 Hz; 2H; Hanti); 13C{1H} NMR (CD2Cl2): d=218.3 (s; CO), 161.7 (q,
ACHTUNGTRENNUNG 2ACHTUNGTERN(NUGN m-Cl)3] were obtained from
(CO)2([9]aneS3)][{Mo(h3-methallyl)(CO)2}
slow diffusion (0.011 g, 15%). 1H NMR (CD2Cl2): d=3.46 (m, 6H;
[9]aneS3), 3.20 (s, 4H; Hsyn), 3.16 (s, 2H; Hsyn), 3.15 (m, 6H; [9]aneS3),
2.38 (s, 6H; CH3 of methallyl), 2.12 (s, 3H; CH3 of methallyl), 2.05 (s,
2H; Hanti), 0.89 ppm (s, 4H; Hanti); IR (CH2Cl2): n˜ =1976 s, 1942 s,
1933 sh, 1900 vs, 1839 s (nCO) cmꢀ1; elemental analysis calcd (%) for
C24H33Cl3Mo3O6S3: C 31.61, H 3.66; found: C 31.72, H 3.74.
JACHTUNGTRNE(GNU C,B)=49.8 Hz; Ci of BAr’4), 134.8 (s; Co of BAr’4), 128.8 (m, Cm of
BAr’4 ), 124.6 (q, JACTHUNRTGNEUNG(C,F)=272.5 Hz; CF3 of BAr’4), 117.5 (s; Cp of BAr’4),
The residue that did not dissolve in diethyl ether (see above) was crystal-
ꢈ
lized similarly, thus affording [nBu4N][{Mo(h3-methallyl)(CO)2}
2ACTHNUTRGNE(NUG m-Cl)3]
(0.013 g, 21%;). 1H NMR (CD2Cl2): d=3.23 (m, 12H; Bu), 3.13 (s, 4H;
Hsyn), 2.12 (s, 6H; CH3 of methallyl), 1.71 (m; Bu), 1.54 (m; Bu), 1.09 (t,
J=7.2 Hz, 12H; Bu), 0.85 ppm (s, 4H; Hanti); 13C{1H} NMR (CD2Cl2): d=
229.5 (s; CO), 84.0 (s; Cc of methallyl), 59.2 (Bu), 53.8 (s; Ct of methall-
yl), 23.9 (Bu), 22.2 (s CH3 of methallyl), 19.7 and 13.4 ppm (Bu); IR
(CH2Cl2): n˜ =1979 s, 1942 s, 1933 sh, 1903 vs, 1840 s (nCO) cmꢀ1; elemen-
tal analysis calcd (%) for C28H50Cl3Mo2NO4: C 31.61, H 3.66; found: C
31.72, H 3.74.
71.9 (s; Cc of allyl), 55.3 (s; Ct of allyl), 33.8 ppm (s; [9]aneS3); IR
(CH2Cl2): n˜ =1980 vs, 1910 s (nCO) cmꢀ1; elemental analysis calcd (%)
for C43H29BF24MoO2S3: C 41.76, H 2.36; found: C 41.87, H 2.33.
(h3-allyl)(CO)2([9]aneS3)]Br (2·Br): A mixture of 2·BAr’4 (0.040 g,
[MoACHTNUTGRNEUNG
0.032 mmol) and [Bu4N]Br (0.010 g, 0.032 mmol) in CH2Cl2 (15 mL) was
stirred for 1 h. After drying under vacuum, the yellow residue was
washed with diethyl ether (3ꢇ10 mL) and dried in vacuo (0.012 g, 83%).
1H NMR (CD2Cl2): d=4.11 (m, 1H; Hc), 3.81 (m, 6H; [9]aneS3), 3.28 (d,
J=6.5 Hz; 2H; Hsyn), 2.94 (m, 6H; [9]aneS3), 1.85 ppm (d, J=9.5 Hz;
2H; Hanti); IR (CH2Cl2): n˜ =1972 vs, 1899 s (nCO) cmꢀ1; elemental analy-
sis calcd (%) for C11H17BrMoO2S3: C 29.15, H 3.78; found: C 29.05, H
3.76.
[Mo(h3-methallyl)(CO)2([9]aneS3)]Br (1·Br):
A solution of 1·BAr’4
(0.050 g, 0.040 mmol) and [Bu4N]Br (0.013 g, 0.040 mmol) in CH2Cl2
(15 mL) was stirred for 30 min. After evaporation of the solvent, the
yellow residue was washed with diethyl ether (3ꢇ10 mL) (to remove the
[Mo
added to
(h3-allyl)(CO)2([9]aneS3)]I (2·I): [Bu4N]I (0.018 g, 0.048 mmol) was
[Bu4N]ACHTUNGTRENNUNG[BAr’4] salt) and dried in vacuo. By slow diffusion of hexane into
a concentrated solution of 1·Br in CH2Cl2 at ꢀ208C, yellow crystals of
good quality (0.015 g, 78%) for the determination of the structure by X-
ray analysis were obtained. 1H NMR (CD2Cl2): d=3.79 (m, 6H;
[9]aneS3), 3.16 (s, 2H; Hsyn), 3.06 (m, 6H; [9]aneS3), 2.35 (s, 3H; CH3 of
methallyl), 2.01 ppm (s, 2H; Hanti); 13C{1H} NMR (CD2Cl2): d=221.3 (s;
CO), 102.2 (s; Cc of methallyl), 57.1 (s; Ct of methallyl), 35.5 (s;
[9]aneS3), 19.5 ppm (s; CH3 of methallyl); IR (CH2Cl2): n˜ =1975 s,
a
solution of 2·BAr’4 (0.060 g, 0.048 mmol) in CH2Cl2
(15 mmol), and the reaction mixture was stirred for 1 h. The rest of the
procedure is similar to that described for 1·Br, including the solid-state
determination by means of X-ray analysis. Yield: 0.020 g, 83%. 1H NMR
(CD2Cl2): d=4.38 (m, 1H; Hc), 3.60 (m, 6H; [9]aneS3), 3.31 (d, J=
6.9 Hz; 2H; Hsyn), 3.01 (m, 6H; [9]aneS3), 1.83 ppm (d, J=9.5 Hz; 2H;
Hanti); 13C{1H} NMR (CD2Cl2): d=220.3 (s; CO), 70.9 (s; Ct of allyl), 55.1
(Ct of allyl), 35.3 ppm (s; [9]aneS3); IR (CH2Cl2): n˜ =1974 vs, 1901 s
(nCO) cmꢀ1; elemental analysis calcd (%) for C11H17IMoO2S3: C 26.41, H
3.42; found: C 26.50, H 3.47.
1898 vs
(nCO) cmꢀ1
;
elemental
analysis
calcd
(%)
for
C12H19BrMoO2S3·H2O: C 29.69, H 4.36; found: C 29.82, H 4.44.
[Mo(h3-methallyl)(CO)2([9]aneS3)]I (1·I): This compound was prepared
in a similar way to 1·Br starting from 1·BAr’4 (0.050 g, 0.040 mmol) and
[Mo
similar to that described for the other adducts, but starting from 2·BAr’4
(h3-allyl)(CO)2([9]aneS3)]
ACHTUNGTRNEUNG ACHTUNGTERN[NNGU CH3SO3] (2·CH3SO3): The procedure is
1
[Bu4N]I (0.015 g, 0.040 mmol) to yield 0.017 g (83%). H NMR (CD2Cl2):
d=3.61 (m, 6H; [9]aneS3), 3.21 (s, 2H; Hsyn), 3.08 (m, 6H; [9]aneS3),
2.26 (s, 3H; CH3 of methallyl), 2.04 ppm (s, 2H; Hanti); 13C{1H} NMR
(CD2Cl2): d=220.9 (s; CO), 102.7 (s; Cc of methallyl), 57.5 (s; Ct of meth-
allyl), 35.3 (s; [9]aneS3), 24.2 ppm (s CH3 of methallyl); IR (CH2Cl2): n˜ =
(0.040 g, 0.032 mmol) and [Bu4N]ACHTNUGTRENUNG[CH3SO3] (0.011 g, 0.032 mmol). Yield:
0.012 g, 83%. 1H NMR (CD2Cl2): d=4.17 (m, 1H; Hc), 3.50 (m, 6H;
[9]aneS3), 3.29 (d, J=6.6 Hz; 2H; Hsyn), 2.94 (m, 6H; [9]aneS3), 2.56 (s,
3H; CH3SO3), 1.85 ppm (d, J=9.5 Hz; 2H; Hanti); IR (CH2Cl2): n˜ =
1976 s, 1900 vs (nCO) cmꢀ1
C61H97Cl2I5Mo5O14S15: C 26.93, H 3.59; found: C 26.70, H 3.77.
[Mo(h3-methallyl)(CO)2([9]aneS3)]
[HSO4] (1·HSO4): This compound
was prepared as described for 1·Br starting from 1·BAr’4 (0.060 g,
0.048 mmol) and [Bu4N][HSO4] (0.016 g, 0.048 mmol) in CH2Cl2 (15 mL).
;
elemental analysis calcd (%) for
1972 vs, 1900 s (nCO) cmꢀ1
;
elemental analysis calcd (%) for
C12H20MoO5S4: C 30.77, H 4.30; found: C 30.53, H 4.18.
AHCTUNGTRENNUNG
X-ray diffraction analysis: Selected crystal and refinement data are given
in Table 2. The diffraction data were collected on an Oxford Diffraction
Xcalibur Nova single-crystal diffractometer with CuKa radiation. Empiri-
cal absorption corrections were applied using the SCALE3 ABSPACK
algorithm as implemented in the program CrysAlis Pro RED (Oxford
Diffraction Ltd.).[30] Structures were solved by Patterson interpretation
by using the program DIRDIF.[31] Isotropic and full-matrix anisotropic
least-square refinements were carried out using SHELXL.[32] Three
chemically identical but crystallographically independent ion pairs of
1·Br were found in its unit cell, one of which appears in Figure 5.
Although they are not depicted, hydrogen bonding between the chlorine
ACHTUNGTRENNUNG
By slow diffusion of toluene into a concentrated solution of 1·HSO4 in
CH2Cl2 at ꢀ208C, yellow crystals were obtained (0.019 g, 78%), one of
which was used for an X ray analysis. 1H NMR (CD2Cl2): d=7.56 (sbr,
1H; HSO4), 3.43 (m, 6H; [9]aneS3), 3.16 (s, 2H; Hsyn), 3.11 (m, 6H;
[9]aneS3), 2.36 (s, 3H; CH3 of methallyl), 1.99 ppm (s, 2H; Hanti);
13C{1H} NMR (CD2Cl2): d=221.5 (s; CO), 102.0 (s; Cc of methallyl), 57.0
(s; Ct of methallyl), 34.7 (s; [9]aneS3), 23.4 ppm (s; CH3 of methallyl); IR
(CH2Cl2): n˜ =1974 s, 1897 vs (nCO) cmꢀ1; elemental analysis calcd (%)
for C49H82Cl2Mo4O24S16: C 29.10, H 4.09; found: C 29.21, H 4.26.
ꢀ
atoms, some C H bonds of the other cationic moieties, and solvent water
[Mo(h3-methallyl)(CO)2([9]aneS3)]
pound was prepared as described for 1·Br starting from 1·BAr’4 (0.050 g,
0.040 mmol) and [Bu4N][CH3SO3] (0.013 g, 0.040 mmol) in CH2Cl2
ACHTUNGNERT[UNNG CH3SO3] (1·CH3SO3): This com-
molecules were also observed. The unit cell of 1·HSO4 contains two
chemically identical but crystallographically independent molecules. Hy-
ꢀ
G
drogen bonding between both HSO4 ions was observed. Only the
(15 mL) and was obtained as a yellow microcrystalline solid (0.014 g,
75%). 1H NMR (CD2Cl2): d=3.53 (m, 6H; [9]aneS3), 3.16 (s, 2H; Hsyn),
3.05 (m, 6H; [9]aneS3), 2.60 (s, 3H; CH3SO3), 2.36 (s, 3H; CH3 of meth-
carbon and chlorine atoms of a disordered CH2Cl2 solvent molecule were
included in the final refinement model. Hydrogen-atom positions were
located in the corresponding Fourier difference maps or set in calculated
Chem. Eur. J. 2012, 18, 16186 – 16195
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16193

J. Pꢁrez, R. Lꢃpez et al.
Table 2. Crystal data and refinement details for complexes 1·BAr’4, 1·Br, 1·HSO4, 1·I, 2·BAr’4, and 2·I.
1·BAr’4 1·Br 1·HSO4 1·I
C44H31BF24MoO2S3 C12H21BrMoO3S3
2·BAr’4
2·I
empirical formula
formula weight
crystal system
space group
a [ꢆ]
C49H80Cl2Mo4O24S16 C61H97Cl2I5Mo5O14S15 C43H29BF24MoO2S3 C11H17IMoO2S3
1250.62
triclinic
P1
485.34
monoclinic
P21/c
2020.91
monoclinic
P21/c
2720.54
triclinic
P1
1236.62
orthorhombic
Pbca
17.7460(1)
19.0202(2)
28.7621(2)
90
400.30
orthorhombic
Pn/21
16.7367(4)
10.7752(2)
8.9779(2)
90
¯
¯
10.7550(6)
15.0119(9)
15.5219(5)
92.694(4)
95.589(4)
103.645(5)
2417.4(2)
2
13.22390(10)
29.9235(2)
14.60240(10)
90
115.8150(10)
90
18.4126(2)
12.41590(10)
17.6312(2)
90
112.8410(10)
90
16.3675(4)
17.9234(6)
19.5620(7)
115.853(3)
101.074(2)
99.778(3)
4854.8(3)
2
b [ꢆ]
c [ꢆ]
a [8]
b [8]
90
90
90
90
g [8]
V [ꢆ3]
5201.61(8)
12
3714.59(7)
2
9708.14(13)
8
1619.09(6)
4
Z
T [K]
100(2)
1.718
1244
100(2)
1.859
2904
100(2)
1.807
2048
100(2)
1.875
2668
293(2)
1.692
4912.0
100(2)
2.052
968
1calcd [gcmꢀ3
F [000]
]
l (Ka) [ꢆ]
1.5418
1.5418
1.5418
1.5418
1.5418
1.5418
crystal size [mm]
m (Ka) [mmꢀ1
collection range [8]
0.29ꢇ0.12ꢇ0.07
4.640
2.86 to 73.81
0.089ꢇ0.045ꢇ0.021 0.22ꢇ0.20ꢇ0.01
0.07ꢇ0.09ꢇ0.23
22.031
2.79 to 74.51
0.21/0.14
18963
0.40ꢇ0.30ꢇ0.10
4.614
3.07–74.72
0.608/0.231
9953
0.08ꢇ0.04ꢇ0.03
25.154
4.09–73.78
0.46/0.11
4742
2652ACTHNGUTERNN[UG 0.0458]
1.050
0.0610, 0.1572
0.0564, 0.1572
]
12.300
10.884
2.94 to 71.84
0.77/0.53
10014
2.60 to 73.79
1/0.43
7517
max/min trans. factors 0.72/0.55
refl. collected
ind. refl. [R
9425
8060
(int)]
N
8879
G
6880
N
15812
1.091
N
7949
N
goodness-of-fit on F2
R, wR2 (all data)
R, wR2 (I>2s(I))
0.999
0.0825, 0.2492
0.0750, 0.2298
1.000
1.003
0.973
0.0894, 0.2490
0.0803, 0.2331
0.0330, 0.0740
0.0283, 0.0717
0.0337, 0.0801
0.0305, 0.0779
0.0949, 0.24728
0.0859, 0.2346
positions and refined riding on their parent atoms. The molecular plots
were made with the PLATON program package.[33] The WINGX pro-
gram system[34] was used throughout the structure determinations.
CCDC-871589 (1·BAr’4), CCDC-871590 (3), CCDC-871591 (1·Br),
CCDC-871592 (4), CCDC-871593 (1·HSO4), CCDC-871594 (1·I), CCDC-
871595 (2·BAr’4), and CCDC-871596 (2·I) contain the supplementary
crystallographic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
[3] a) Y. Hua, A. H. Flood, Chem. Soc. Rev. 2010, 39, 1262–1271;
b) B. P. Hay, V. S. Bryantsev, Chem. Commun. 2008, 2417–2428;
c) S. Zhu, H. Staats, K. Brandhorst, J. Grunenberg, F. Gruppi, E.
Dalcanale, A. Luetzen, K. Rissanen, C. A. Schalley, Angew. Chem.
2008, 120, 800–804; Angew. Chem. Int. Ed. 2008, 47, 788–792;
d) W. J. Belcher, M. Fabre, T. Farhan, J. W. Steed, Org. Biomol.
Chem. 2006, 4, 781–786; e) Z. R. Laughrey, T. G. Upton, B. C. Gibb,
Chem. Commun. 2006, 970–972; f) J. Yoon, S. K. Kim, N. J. Singh,
K. S. Kim, Chem. Soc. Rev. 2006, 35, 355–360; g) V. S. Bryantsev,
B. P. Hay, J. Am. Chem. Soc. 2005, 127, 8282–8283; h) K. Chellap-
pan, N. J. Singh, I-Ch. Hwang, J. W. Lee, K. S. Kim, Angew. Chem.
2005, 117, 2959–2963; Angew. Chem. Int. Ed. 2005, 44, 2899–2903;
Angew. Chem. 2005, 117, 2959–2963; i) H. Maeda, Y. Kusunose,
Chem. Eur. J. 2005, 11, 5661–5666; j) C. A. Ilioudis, D. A. Tocher,
J. W. Steed, J. Am. Chem. Soc. 2004, 126, 12395–12402; k) C. L. D.
Gibb, E. D. Stevens, B. C. Gibb, J. Am. Chem. Soc. 2001, 123, 5849–
5850.
[4] a) L. Pedzisa, B. P. Hay, J. Org. Chem. 2009, 74, 2554–2560; b) G. R.
Desiraju, Chem. Commun. 2005, 2995–3001; c) C. B. Aakerçy, T. M.
Evans, K. R. Seddon, I. Pꢂlinkꢃ, New. J. Chem. 1999, 23, 145–152;
d) R. Taylor, O. Kennard, J. Am. Chem. Soc. 1982, 104, 5063–5070.
[5] a) D. J. Mercer, S. J. Loeb, Chem. Soc. Rev. 2010, 39, 3612–3620;
b) M. G. Fisher, P. A. Gale, M. E. Light, S. J. Loeb, Chem. Commun.
2008, 5695–5697; c) S. R. Bayly, P. D. Beer, Struct. Bonding (Berlin,
Ger.) 2008, 129, 45–94; d) R. Vilar, Eur. J. Inorg. Chem. 2008, 357–
367; e) J. Perez, L. Riera, Chem. Soc. Rev. 2008, 37, 2658–2667; f) J.
Perez, L. Riera, Chem. Commun. 2008, 533–543; g) C. Rice, Coord.
Chem. Rev. 2006, 250, 3190–3199; h) M. H. Filby, J. W. Steed,
Coord. Chem. Rev. 2006, 250, 3200–3218; i) J. W. Steed, Chem.
Commun. 2006, 2637–2649; j) P. D. Beer, S. R. Bayly, Top. Curr.
Chem. 2005, 255, 125–162; k) I. D. Vega, P. A. Gale, M. E. Light,
S. J. Loeb, Chem. Commun. 2005, 4913–4915; l) C. R. Bondy, P. A.
Gale, S. J. Loeb, J. Am. Chem. Soc. 2004, 126, 5030–5031; m) P. D.
Beer, E. J. Hayes, Coord. Chem. Rev. 2003, 240, 167–189; n) P. D.
Beer, P. A. Gale, Angew. Chem. 2001, 113, 502–532; Angew. Chem.
Int. Ed. 2001, 40, 486–516; o) C. R. Bondy, P. A. Gale, S. J. Loeb,
Chem. Commun. 2001, 729–730.
Acknowledgements
An anonymous referee is thanked for helpful suggestions. We thank Prin-
cipado de Asturias (Grants IB05–069 and IB08–104 administered by
FICYT), FEDER funds, Ministerio de Ciencia y Tecnologꢀa (CTQ2006–
07036/BQU), and Ministerio de Ciencia e Innovaciꢃn (CTQ2009–12366/
BQU).
[1] L. J. Prins, D. N. Reinhoudt, P. Timmerman, Angew. Chem. 2001,
113, 2446–2492; Angew. Chem. Int. Ed. 2001, 40, 2382–2426.
[2] a) K. Bowman-James, A: Bianchi, E. Garcꢀa-EspaÇa, Anion Coordi-
nation Chemistry, Wiley-VCH, Weinheim, Germany, 2011; b) V.
Amendola, L. Fabbrizzi, L. Mosca, Chem. Soc. Rev. 2010, 39, 3889–
3915; c) J. L Sessler, P. Gale, W.-S. Cho in Anion Coordination
Chemistry (Ed.: J. F. Stoddart), RSC Cambridge, 2006; d) P. A. Gale,
Acc. Chem. Res. 2006, 39, 465–475; e) V. Amendola, D. Esteban-
Gꢃmez, L. Fabbrizzi, M. Licchelli, Acc. Chem. Res. 2006, 39, 343–
353; f) F. P. Schmidtchen, Top. Curr. Chem. 2005, 255, 1–29; g) K.
Choi, A. D. Hamilton, Coord. Chem. Rev. 2003, 240, 101–110;
h) J. W. Steed, J. L. Atwood, Supramolecular Chemistry, Wiley, Chi-
chester, UK, 2000; i) A. Bianchi, K. Bowman-James, E. Garcꢀa-
EspaÇa, Supramolecular Chemistry of Anions, Wiley-VCH, New
York, NY, 1997; j) F. P. Schmidtchen, M. Berger, Chem. Rev. 1997,
97, 1609–1646; k) J. M. Lehn, Supramolecular Chemistry, 1st ed.,
VCH-Weinheim, Germany, 1995.
16194
www.chemeurj.org
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16186 – 16195

Molybdenum Complexes as Anion Receptors
FULL PAPER
[6] R. D. Bedford, M. Betham, C. P. Butts, S. J. Coles, M. B. Hursthouse,
P. N. Scully, J. H. R. Tucker, J. Wilkie, Y. Willener, Chem. Commun.
2008, 2429–2431.
[7] J. Perez, D. Morales, S. Nieto, L. Riera, V. Riera, D. Miguel, Dalton
Trans. 2005, 884–888.
[8] G. E. D. Mullen, T. F. Fꢉssler, M. J. Went, K. Howland, B. Stein, P. J.
Blower, J. Chem. Soc. Dalton Trans. 1999, 3759–3766, and referen-
ces therein.
[9] R. H. Crabtree, The Organometallic Chemistry of the Transition
Metals, 4th ed., Wiley, New York, 2005, pp. 297–298.
[10] D. S. Frohnapfel, P. S. White, J. L. Templeton, H. Rueegger, P. S. Pre-
gosin, Organometallics 1997, 16, 3737–3750.
[22] a) A. D. Becke, Phys. Rev. A 1988, 38, 3098–3100; b) C. Lee, W.
Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785–789.
[23] W. J. Hehre, L. Radom, J. A. Pople, P. v. R. Schleyer, Ab Initio Mo-
lecular Orbital Theory, Wiley, New York, 1986.
[24] J. Hay, W. R. Wadt, J. Chem. Phys. 1985, 82, 299–310.
[25] C. E. Check, T. O. Faust, J. M. Bailey, B. J. Wright, T. M. Gilbert,
L. S. Sunderlin, J. Phys. Chem. A 2001, 105, 8111–8116.
[26] C. A. Ilioudis, K. S. B. Hancock, D. G. Georganopoulou, J. W. Steed,
New. J. Chem. 2000, 24, 787–798.
[27] I. Rozas, I. Alkorta, J. Elguero, J. Am. Chem. Soc. 2000, 122, 11154–
11161.
[28] H. T. Dieck, H. Friedel, J. Organomet. Chem. 1968, 14, 375–385.
[29] M. Brookhart, B. Grant, A. F. Volpe Jr, Organometallics 1992, 11,
3920–3922.
[11] V. S. Joshi, K. M. Sathe, M. Nandi, P. Chakrabarti, A: Sarkar, J. Or-
ganomet. Chem. 1995, 485, C1–C5.
[12] P. Espinet, R. Hernando, G. Iturbe, F. Villafane, G. A. Orpen, I.
Pascual, Eur. J. Inorg. Chem. 2000, 1031–1038.
[30] CrysAlisPro RED, Version 1.171.31.7; Oxford Diffraction Ltd.,
Oxford, UK, 2009.
[13] M. J. Hynes, J. Chem. Soc. Dalton Trans. 1993, 311–312.
[14] H. D. Murdock, J. Organomet. Chem. 1965, 4, 119–126.
[15] W. A. Schenk, M. Schmidt, J. Organomet. Chem. 1975, 96, 375–81.
[16] S. R. Cooper, S. C. Rawle, Struct. Bonding (Berlin, Ger.) 1990, 72,
1–72.
[17] M. T. Ashby, D. L. Lichtenberger, Inorg. Chem. 1985, 24, 636–638.
[18] P. K. Baker, S. J. Coles, M. C. Durrant, S. D. Harris, D. L. Hughes,
M. B. Hursthouse, R. L. Richards, J. Chem. Soc. Dalton Trans. 1996,
4003–4010.
[31] P. T. Beurskens, G. Beurskens, R. de Gelder, J. M. M. Smits, S.
Garcꢀa-Granda, R. O. Gould, The DIRDIF Program System, Ver-
sion 2008.3; Crystallography Laboratory, University of Nijmegen,
Nijmegen, The Netherlands, 2008.
[32] G. M. Sheldrick, SHELXL, Version 2008; G. M. Sheldrick, Acta
Crystallogr. Sect. A 2008, 64, 112.
[33] A. L. Spek, PLATON version 1.15: A Multipurpose Crystallograph-
ic Tool; University of Utrecht, Utrecht, The Netherlands, 2008.
[34] L. J. Farrugia, J. Appl. Crystallogr. 1999, 32, 837–838.
[19] R. S. Glass, G. S. Wilson, W. N. Setzer, J. Am. Chem. Soc. 1980, 102,
5068–5069.
[20] A. Maia, Pure Appl. Chem. 1995, 67, 697–702.
[21] C. Borgmann, C. Limberg, E. Kaifer, H. Pritzkow, L. Zsolnai, J. Or-
ganomet. Chem. 1999, 580, 214–224.
Received: April 19, 2012
Revised: July 4, 2012
Published online: October 23, 2012
Chem. Eur. J. 2012, 18, 16186 – 16195
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16195