Family of Cofacial Bimetallic Complexes of a Hexaanionic Carboxamide Cryptand

Article abstract of DOI:10.1021/ic200143b

A series of coordination compounds has been prepared comprising manganese, iron, nickel, and zinc bound by a hexaanionic cryptand where carboxamides are anionic N-donors. The metal complexes have been investigated by X-ray crystallography, and possess metal centers in trigonal monopyramidal geometries with intermetallic distances spanning d_Mn,avg = 6.080 A to d_Ni,avg = 6.495 A. All complexes featuring trigonal monopyramidal metal(II) ions crystallize in Cc, and feature extended three-dimensional networks composed of cryptate anions linked by bridging potassium countercations. We also report the first solid state structure of the free cryptand ligand, which features no guest in its cavity and which possesses an extended hydrogen-bonding network. SQuID magnetometry data of the metal complexes reveal weak antiferromagnetic coupling of the metal centers. Only the diiron(II) complex exhibits reversible electrochemistry, and correspondingly, its chemical oxidation yields a powder formulated as the diiron(III) congener. The insertion of cyanide into the intermetallic cleft of the diiron(II) complex has been achieved, and comparisons of its solid state structure to the recently reported dicobalt(II) analogue are made. The antiferromagnetic coupling between the diiron(II) and the dicobalt(II) centers when bridged by cyanide does not increase significantly relative to the unbridged congeners. A one-site model satisfactorily fits Moessbauer spectra of unbridged diiron(II) and diiron(III) complexes whereas a two site fit was needed to model the iron(II) centers that are bridged by cyanide.

Full text of DOI:10.1021/ic200143b

ARTICLE
pubs.acs.org/IC
Family of Cofacial Bimetallic Complexes of a Hexaanionic
Carboxamide Cryptand
Glen E. Alliger, Peter M€uller, Loi H. Do, Christopher C. Cummins,* and Daniel G. Nocera*
Department of Chemistry, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, Massachusetts 02139-4307,
United States
S Supporting Information
b
ABSTRACT: A series of coordination compounds has been prepared comprising manga-
nese, iron, nickel, and zinc bound by a hexaanionic cryptand where carboxamides are anionic
N-donors. The metal complexes have been investigated by X-ray crystallography, and possess
metal centers in trigonal monopyramidal geometries with intermetallic distances spanning
dMn,avg = 6.080 Å to dNi,avg = 6.495 Å. All complexes featuring trigonal monopyramidal
metal(II) ions crystallize in Cc, and feature extended three-dimensional networks composed
of cryptate anions linked by bridging potassium countercations. We also report the first solid
state structure of the free cryptand ligand, which features no guest in its cavity and which possesses an extended hydrogen-bonding
network. SQuID magnetometry data of the metal complexes reveal weak antiferromagnetic coupling of the metal centers. Only the
diiron(II) complex exhibits reversible electrochemistry, and correspondingly, its chemical oxidation yields a powder formulated as
the diiron(III) congener. The insertion of cyanide into the intermetallic cleft of the diiron(II) complex has been achieved, and
comparisons of its solid state structure to the recently reported dicobalt(II) analogue are made. The antiferromagnetic coupling
between the diiron(II) and the dicobalt(II) centers when bridged by cyanide does not increase significantly relative to the unbridged
congeners. A one-site model satisfactorily fits M€ossbauer spectra of unbridged diiron(II) and diiron(III) complexes whereas a two
site fit was needed to model the iron(II) centers that are bridged by cyanide.
’ INTRODUCTION
more stable octaaza counterparts, obtained from the borohydride
reduction of the hexaimino-cryptand, even more attractive as
ligands for binuclear complexation. To date, however, such
complexes have been based on the use of ligands featuring solely
charge neutral N-donors, despite the well-documented ability of
triply anionic TREN moieties to complex a wide range of
transition metals.15ꢀ19
Designed to be three-dimensional complements to crown
ethers, cryptands are unique macrocycles. Since Lehn’s seminal
work with polyethereal aza-cryptands,1 the hostꢀguest chemis-
try of cryptands has burgeoned, and now is prominent in the
chemistry of complex cation binding,2,3 siderophore modeling,4,5
and electride synthesis.6 Cryptands are particularly prominent as
cation-specific sequestration reagents with binding affinities that
are several orders of magnitude greater than monomacrocyclic
crown ethers.7 Size recognition properties of the cavities of
smaller cryptands engender the selective sequestration of alkali
and alkaline earth cations;8 the stability constant of the potassium
complex of the exemplar crypt[2.2.2] is more than an order of
magnitude greater than complexes of this crypt with other alkali
cations. Such selectivity finds its genesis in smaller energies of
complexation for selected metal ions of incompatible size.9
Protonated aza-cryptands have also seen use in anion sequestra-
tion.10ꢀ12
The advent of hexaimino-cryptands considerably expanded
the cryptand class of macrocycles from monometallic binding
constructs for alkali and alkaline earth cations to bimetallic
binding constructs for transition metal cations.13 Such ligands
are notable not only for their binucleating ability but also for their
ease of synthesis. In many cases, the condensation of 3 equiv of an
aromatic dialdehyde with 2 equiv of TREN (TREN = tris(2-
aminoethyl)-amine) furnishes the desired hexaiminocryptand in
good yields without the need for high-dilution reaction condi-
tions. The hydrolytic sensitivity of these ligands14 makes their
A factor accounting for the dearth of anionic octaaza-cryptand
complexes is the oxidative instability of secondary amines.20
Incorporation of carboxamide functionalities into the ligand
provides a potential means to circumvent this instability. When
used as anionic N-donors, carboxamide residues are known to
improve substantially the oxidative stability of ligands, and
accordingly, they have enjoyed success in the stabilization of
high-valent transition metal centers.21,22 The implementation of
neutral carboxamide cryptands has been explored for anion
sequestration, and inclusion complexes of halides and polyox-
oanions have been observed.21ꢀ25 Nonetheless, complexes of
deprotonated carboxamide-based cryptands are unusual.
We recently realized the first complex of a hexaanionic
N-donor cryptand (1, Chart 1) with dicobalt(II) within the cleft,
and demonstrated access to the intermetallic cleft through
reaction with cyanide anion.26 We now show that the method
developed for double insertion of cobalt(II) into 1 can be
generalized to other first-row transition metals (M = Mn through
Received: January 21, 2011
Published: March 29, 2011
r
2011 American Chemical Society
4107
dx.doi.org/10.1021/ic200143b Inorg. Chem. 2011, 50, 4107–4115
|

Inorganic Chemistry
ARTICLE
Chart 1
ligand can be found in the Supporting Information of previously
published work.26
K2(DMF)6Mn2C72H84N8O15 ([K2(DMF)6][Mn2L], 2). A slurry of
1 (234 mg, 179 μmol) and Mn(OAc)2 (62 mg, 360 μmol) was stirred in
1 mL of DMF for 30 min (Slurry 1). The mixture was frozen in the
glovebox cold well, as was a solution of KN(SiMe3)2 (218 mg, 1.09
mmol, Solution 2) in 1 mL of dimethylformamide (DMF). As the slurry
and solution thawed, Solution 2 was added to Slurry 1, and the mixture
was allowed to warm to glovebox temperature over the course of 2 h.
The reaction mixture, which turned slightly yellow over the course of the
reaction, was filtered to remove precipitated potassium acetate. To the
filtrate was added 10 mL of ether dropwise with rapid stirring. A white
powder precipitated, and it was collected by filtration. The precipitate
was washed with 12 mL of 5:1 ether/DMF and dried in vacuo. The
powder comprised 183 mg (94.6 μmol, 52.8%) of analytically pure
product. Elemental analysis confirms the presence of 6 DMF molecules
per K2Mn2C72H84N8O15 unit. Crystals suitable for X-ray diffraction
studies were grown by vapor diffusion of ether into a concentrated DMF
solution of the product. This product is NMR silent. Anal. Calcd.
(found) for C90H126N14O21K2Mn2: C, 56.06 (55.51); H, 6.59 (6.55); N
10.17 (10.45).
Zn, with the exception of Cu). The structural features and
spectroscopy of these complexes presage this bitopic cryptand
as a new motif to support bimetallic cooperativity.
’ EXPERIMENTAL SECTION
K2(DMF)6Fe2C72H84N8O15 (K2(DMF)6Fe2L, 3). Complex 3 was
synthesized in the same fashion as complex 2, using 1.051 g (803.8
μmol) of 1, 278 mg (1.60 mmol) of Fe(OAc)2, and 978 mg (4.90 mmol)
of KN(SiMe3)2. Yield: 839 mg (435 μmol, 54.1%) of a yellow powder.
1H NMR (300 MHz, DMSO-d6, δ, all signals paramagnetically broad-
ened): 77.34 (6 H), 62.08 (6 H), 38.60 (3 H), 7.99 (6 H), 2.87 (18 H),
2.76 (18 H), 2.26 (3 H), ꢀ0.72 (12 H) ꢀ1.06 (12 H), ꢀ1.25 (18 H),
ꢀ2.00 (6 H), ꢀ4.71 (6 H), ꢀ27.61 (6 H).
General Procedures. All manipulations were performed using
either Schlenk techniques or a nitrogen-atmosphere glovebox. Reagents
were purchased from Aldrich. 1 was prepared according to the pre-
viously reported synthesis by us.26 18-crown-6 was recrystallized from
dry acetonitrile. Solvents (EMD Chemicals) were purified on a Glass
Contour Solvent Purification System built by SG Water U.S.A. UVꢀvis
spectra were obtained on a Cary 5000 spectrophotometer. IR spectra
were obtained on a Perkin-Elmer Model 2000 FT-IR spectrometer.
Cyclic voltammetry was performed using a BAS CV-50W Voltammetric
Analyzer potentiostat. SQuID magnetometry was performed using a
Quantum Design AC and DC Magnetic Property Measurement System,
and data were fit using the program julX.27 M€ossbauer spectra were
recorded on an MSI spectrometer (WEB Research) and referenced to
metallic iron. The spectra were fit using the program WMOSS (WEB
Research). Electron paramagnetic resonance (EPR) spectra were ob-
tained on a Bruker EMX spectrometer equipped with an ER 4199HS
cavity and Gunn diode microwave source. Spectra were obtained using
X-band radiation. NMR solvents were obtained from Cambridge Iso-
tope Laboratories and 1H and 13C{1H} NMR spectra were obtained on
Varian 300 and 500 MHz spectrometers and were referenced to residual
protio-solvent signals. Elemental analyses were performed by Midwest
Microlabs.
X-ray Crystallographic Studies. Low-temperature diffraction
data were collected on a three-circle diffractometer coupled to a
Bruker-AXS Smart Apex CCD detector, performing φ-and ω-scans,
with graphite-monochromated Mo KR radiation (λ = 0.71073 Å) for the
structures of 3 and 6 and Cu KR radiation (λ = 1.54178 Å) for the
structures of 1, 2, 5, and 10. The structures were solved by direct
methods using SHELXS and refined against F2 on all data by full-matrix
least-squares with SHELXL-9728 using established methods.29 All non-
hydrogen atoms were refined anisotropically. All hydrogen atoms were
included into the model at geometrically calculated positions and refined
using a riding model. The isotropic displacement parameters of all
hydrogen atoms were fixed to 1.2 times the U value of the atoms they are
linked to (1.5 times for methyl groups). Disorders were refined with the
help of similarity restraints on 1,2- and 1,3-distances and displacement
parameters as well as rigid bond restraints for anisotropic displacement
parameters. With exception of the structure of the free ligand 1, all
structures contained voids filled with heavily disordered solvent mol-
ecules. The program SQUEEZE30 as implemented in Platon31 was used
to remove the contribution of the disordered solvent to the structure
factors. Detailed information about the other structures featuring 1 as a
K2(DMF)6Ni2C72H84N8O15 (K2(DMF)6Ni2L, 5). Complex 5 was
synthesized in the same fashion as complex 2, using 440 mg (336 μmol)
of 1, 119 mg (675 μmol) of Ni(OAc)2, and 409 mg (2.05 mmol) of
KN(SiMe3)2. Yield: 254 mg (131 μmol, 39.0%) of an orange-pink
powder.1H NMR (300 MHz, DMSO-d6, δ, all signals paramagnetically
broadened): 65.90 (6 H), 48.64 (6 H), 12.30 (3 H), 8.00 (6 H), 4.03 (6
H), 3.42 (3 H), 2.84 (18 H), 2.76 (18 H), 0.83 (6 H), 0.66 (12 H),
ꢀ0.24 (12 H), ꢀ0.61 (18 H), ꢀ18.10 (6 H). Anal. Calcd (found) for
C90H126N14O21K2Ni2: C, 55.85 (56.47); H, 6.56 (6.60); N 10.13
(10.23).
K2(DMF)6Zn2C72H84N8O15 (K2(DMF)6Zn2L, 6). Complex 6 was
synthesized in the same fashion as complex 2 using 124 mg (95.1 μmol)
of 1, 35 mg (190 μmol) of Zn(OAc)2, and 169 mg (579 μmol of
KN(SiMe3)2. Yield: 98 mg (50 μmol, 53%) of a white powder.1H NMR
(300 MHz, DMF-d7, δ): 7.90 (t, 3H), 6.63 (d, 6H), 6.49 (d, 6H), 6.17 (t,
3H), 4.00, (t, 12H), 3.90 (m, 6H), 2.95 (m, 6H), 2.73 (m, 6H), 2.56 (m,
6H), 1.72 (m, 12H), 0.98 (t, 18H).13C{1H} NMR (75 MHz, DMSO-d6,
δ): 173.33, 162.34, 160.37, 160.29, 155.60, 143.93, 121.88, 116.20,
99.29, 96.64, 69.14, 54.16, 41.78, 35.80, 30.77, 22.08, 10.42. Anal. Calcd.
(found) for C90H126N14O21K2Zn2: C, 55.46 (55.97); H, 6.52 (6.37); N
10.06 (9.26).
[K(C12H24O6)]2Fe2C72H84N8O15([K(18-crown-6)]2-Fe2L, 7).
A 1 mL methylene chloride solution of 22 mg (0.083 mmol, 2.9 equiv)
of 18-crown-6 was added to solid stirring 56 mg (0.029 mmol, 1.0 equiv)
3. This reaction mixture was allowed to stir for 1 h. To the solution was
added 2 mL of ether, and the reaction mixture was filtered and stored in a
ꢀ35 ꢀC freezer. After several weeks, large pale yellow blocks formed.
Yield: 48 mg (0.024 mmol, 82%). 1H NMR (300 MHz, DMSO-d6, δ, all
signals paramagnetically broadened): 77.34 (6 H), 62.08 (6 H), 38.60 (3
H), 3.38 (48 H), 2.26 (3 H), ꢀ0.72 (12 H) ꢀ1.06 (12 H), ꢀ1.25 (18
H), ꢀ2.00 (6 H), ꢀ4.71 (6 H), ꢀ27.61 (6 H). Anal. Calcd. (found) for
C96H132N8O27K2Fe2: C, 57.08 (57.03); H, 6.59 (6.66); N, 5.55 (5.52).
Fe2C72H84N8O15 (Fe2L, 8). In a 20 mL scintillation vial, 41 mg
(0.16 mmol, 2.1 equiv) of silver triflate was dissolved in 2 mL of
tetrahydrofuran (THF). This solution was frozen. Upon thawing, 144
4108
dx.doi.org/10.1021/ic200143b |Inorg. Chem. 2011, 50, 4107–4115

Inorganic Chemistry
ARTICLE
Scheme 1
’ RESULTS AND DISCUSSION
Figure 1. Solid state structure of 1. Thermal ellipsoids at 50% prob-
ability level. Non-hydrogen bonding H atoms and dipropoxyphenoxyl
substituents omitted for clarity. This view of the ligand shows both the
intramolecular and intermolecular hydrogen bonding observed in the
solid state.
Synthesis. A multistep synthesis of a hexacarboxamide cryp-
tand featuring pendant polyether moieties (1, Chart 1) was
recently reported by us.26 X-ray quality crystals of the cryptand 1
can be grown by vapor diffusion of ether into a THF solution.
The solid-state structure is displayed in Figure 1. Coordinates for
the carboxamide hydrogen atoms were taken from the difference
Fourier synthesis, and the hydrogen atoms were subsequently
refined semifreely, restraining the NꢀH distances to 0.88 Å,
while constraining their Uiso values to 1.2 times the Ueq of the
respective nitrogen atoms. This structure illustrates a rare
example of a hexacarboxamide-cryptand that does not possess
a guest molecule inside the cryptand cavity. Most “guestless”
hexacarboxamide cryptands have their hydrogen bonding net-
works interrupted by the presence of water molecules in the
crystal; a search of the Cambridge Structural Database32,33
reveals only two such structures where this is not the case.34,35
In contrast to metal complexes of 1, which possess an approx-
imate C3 axis of symmetry (vide infra), the free ligand folds upon
itself to engage in intramolecular hydrogen bonding. These
hydrogen bonds from H101 and H201 to O301 (2.262(19) Å
and 2.125(19) Å, respectively) and from H202 to O102
(1.925(19) Å) break the C3 symmetry in the solid state, though
this symmetry is restored on the NMR time scale for the
compound in solution at room temperature. A complex variable
temperature 1H NMR spectrum indicates that the high symmetry is
lost as a solvated sample of the cryptand is cooled to ꢀ85 ꢀC.
The cryptand forms an extended network owing to intermo-
lecular hydrogen bonding. Each cryptand unit in the crystal
engages in four hydrogen bonds: two originate from H102 and
H301, which bind to O202 and O101, respectively, in two
neighboring cryptand molecules. Likewise, carboxamide oxygen
atoms O201 and O102 act as receptors for hydrogen bonds from
other cryptand units. H102 and H301 are 2.00(2) Å and
1.982(19) Å from their nearest acceptors, respectively. The
extended network arising from intermolecular hydrogen bonding
gives rise to infinite two-dimensional sheets of cryptands, with the
crystallographic c axis oriented normal to the extended planes.
Metalation of H6L 1 proceeds by treatment of a DMF slurry of
1 and a divalent metal acetate with a slight excess of potassium
hexamethyldisilazide at low temperatures followed by warming
the mixture to 25 ꢀC over 2 h (Scheme 1). This procedure is a
modification of a reported metalation protocol,36ꢀ38 wherein
metalation proceeds by deprotonation of the ligand in the
presence of the metal source. This allows for introduction of
mg (0.0746 mmol, 1.00 equiv) of 3 that was chilled to 77 K was added
to the solution as a solid, and the remnants of the solid were washed
in with 2 mL of THF. The reaction mixture was allowed to stir for 1.5
h, during which time it turned deep red. The reaction mixture was
filtered through Celite and the filter cake was washed with THF until
the washings were colorless. The filtrate was taken to dryness in
vacuo, and triturated twice with hexane. The brown solid was then
dissolved in 2 mL of methylene chloride, and this solution was
filtered through Celite. The filtrate was taken to dryness in vacuo,
and the solid so obtained was triturated twice with 2 mL of hexane.
The brown solid was again dissolved in 1 mL of methylene chloride,
and the solution was filtered through Celite. This was the first
filtration for which no solids were observed to be removed from
solution. A brown solid was precipitated from the filtrate with
hexane, and it was collected. This material was subjected to
M€ossbauer analysis, but did not pass elemental analysis. A CDCl3
solution of this solid was NMR silent.
Reduction of Fe2L (8) with Cobaltocene. In 0.5 mL of DMSO-
d6 was dissolved 22 mg of 8. This solution was added to a slurry of 6 mg
of cobaltocene stirring in 0.5 mL of DMSO-d6. The reaction mixture was
allowed to stir for 30 min, during which time it became homogeneous.
The presence of 3 was confirmed spectroscopically by 1H NMR.
[K(C12H24O6)]3Fe2(μ-CN)C72H84N8O15 ([K(18-crown-6)]3-
Fe2(μ-CN)L, 10). In 3 mL of DMF was dissolved 187 mg (92.6 μmol)
of [K(18-crown-6)]2Fe2L and 50 mg (190 μmol) of 18-crown-6. To this
mixture was added 20 mg (0.30 mmol) of potassium cyanide. This
reaction mixture was sealed in a glass thick-walled vessel and heated at
75 ꢀC with stirring for 48 h. At this point, the reaction mixture was
filtered, and the filtrate was taken to dryness. The solid product so
obtained was crystallized by vapor diffusion of ether into a concentrated
THF solution overnight. The resulting crystalline material was washed
with 5 mL of 2:1 ether/THF and dried in vacuo. The product comprised
151 mg (64.1 μmol, 69.2%) of bright yellow crystals. Crystals suitable for
X-ray analysis were grown by layering a THF solution with pentane. 1H
NMR (300 MHz, CDCl3, δ, all signals paramagnetically broadened):
34.55 (3 H), 28.74 (3 H), 17.55 (3 H), 9.12 (3 H), 4.66 (3 H), 2.38 (12
H), 2.23 (72 H), 0.91, (12 H), 0.41 (18 H), ꢀ34.86 (3 H), ꢀ44.36 (3H),
ꢀ56.13 (3 H), ꢀ71.55 (3 H). IR (Nujol, cmꢀ1): 2109 (CN). Anal.
Calcd. (found) for C109H156N9O33K3Fe2: C, 55.72 (55.45); H, 6.69
(6.58); N, 5.37 (5.38).
4109
dx.doi.org/10.1021/ic200143b |Inorg. Chem. 2011, 50, 4107–4115

Inorganic Chemistry
ARTICLE
solutions of the complexes. Crystallographic studies show that
compounds 2ꢀ6 are isostructural; they all crystallize in the same
space group, Cc, and utilization of the same crystallographic
model for all structures yields satisfactory results. In all cases, the
values of R1 are best when the identity of the metal is the same as
the metal used in the synthesis. In Figure 2, the view of the anion
of 2 looking down the NapꢀNap axis reveals the dipropoxyphe-
noxyl substituents to be splayed out to the periphery of the
cryptate. Substantial disorder is observed in the polyether
component of this structure, as in the structures of 1 and other
metalated cryptates. Nonetheless, the cores, defined by the
TREN motifs and the phenylene spacers that span them, are
ordered as shown by the representations of the anions in
Figure 3.
No electron density was observed in the difference map in the
void between the metal centers, ruling out the presence of apical
ligands occupying the fifth coordination site of either metal
center. Thus, the two metal centers are coordinated in the rare
trigonal monopyramidal geometry.15,41ꢀ50 Omission of the
solubilizing substituents, potassium counterions, and solvent
molecules of crystallization in Figure 3 allows for easy viewing
of the intermetallic space. The metal centers are disposed in a
cofacial orientation, such that the vacant coordination sites are
directed toward one another. The intermetallic distance varies
depending on the identity of the metal complexed, covering a
range of 0.415 Å.
Figure 2. View of one of the cryptate units in the asymmetric unit of 2,
looking down the NapꢀNap vector. Thermal ellipsoids at 50% prob-
ability level. H atoms, Kþ counterions, and solvents of crystallization
omitted for clarity. Color scheme: black, carbon; blue, nitrogen; red,
oxygen.
Figure 4 presents an overlay line drawing of the cores of
complexes 2 and 6 that highlights the similarities and differences
between these complexes. The metalꢀmetal distance in 2 is the
shortest of the complexes at davg = 6.080 Å. The line drawing
shows that this is a result of the manganese centers puckering out
of the planes defined by their respective equatorial nitrogen
donors slightly. This is in contrast to 6, which has a metalꢀmetal
distance of davg = 6.495 Å. Here, the metal center is relaxed into
its TREN binding pocket. This difference of the metal residency
may be due to the better size match of Zn2þ ion for the TREN
pocket as opposed to a poorer match for the larger Mn2þ ion
(ionic radius 0.80 Å for Mn2þ vs 0.74 Å for Zn2þ).51 The
intermetallic distance tracks with the ionic radius of the com-
plexed metal ion (see Supporting Information). Table 1 com-
pares and contrasts some important metrical parameters for the
bimetallic crypts.
The extended structure of compounds 2, 3, 5, and 6 is
complex. As recently reported for isomorphous compound 4,26
infinite one-dimensional chains are formed by anionic cryptate
units bridged by potassium cations. These one-dimensional
chains are further bridged to another set of one-dimensional
chains. An infinite three-dimensional extended network results
from the crossing of the one-dimensional chains. This extended
network may be responsible for the poor solubility properties of
these materials. Compounds 2ꢀ6 are only soluble in highly polar
organic solvents, such as DMF and dimethylsulfoxide (DMSO).
Solubility in less polar solvents such as THF and methylene
chloride may be imparted by introduction of 2 equiv of 18-
crown-6 as previously reported for [K(18-crown-6)]2Co2L.26
Analogous compounds, such as [K(18-crown-6)]2Fe2L (7), can
be crystallized by cooling a methylene chloride/ether solution of
the compound.
the metal to the ligand without the need to form a discrete
hexaanionic species. After removal of precipitated potassium
acetate by filtration, analytically pure material of the formula
[K2(DMF)6][M2L] (2ꢀ6) is precipitated from the DMF solu-
tion by addition of ether. NMR spectroscopy confirms the
consumption of starting material, as does the lack of a stretch
representing an NꢀH oscillator in the infrared spectra of these
materials.
1
Dizinc complex 6 shows a complex pattern in its H NMR
spectrum for the protons featured on the TREN methylene
residues. The complex appears to be helical and rigid enough that
each of the two methylene residues on any given ethylene arm of
the TREN moiety are rendered diastereotopic on the NMR time
scale.39 Variable temperature NMR confirms that this behavior is
maintained until at least 150 ꢀC, though the peaks do begin to
broaden at elevated temperature. Although complex 2 is NMR
silent, paramagnetic species 3ꢀ5 exhibit well-behaved NMR
spectra, with the broadening and shifting of resonances that is
typical of paramagnetic compounds.40 Protons located distal
with respect to the metal centers, such as those featured on the
dipropoxyphenoxyl substituents, resonate closer to the typical
“diamagnetic region” of 0ꢀ10 ppm, while protons proximate to
the metal center including those from the arms of the TREN
moiety exhibit dramatically shifted signals. Compounds 3 and 5
display 10 paramagnetically shifted and broadened resonances
(discounting those arising from DMF) in their 1H NMR spectra;
9 signals should be observed based on a hypothetical D3h
symmetry for protons on a given TREN methylene unit equiva-
lent. The fact that only 10 resonances are observed instead of the
11 predicted for a C3h-symmetric compound is probably due to
either a broadening of one resonance into the baseline or the
overlap of one peak with another. The 1H NMR spectrum of 4
does in fact display the expected 11 resonances.
The cyclic voltammogram of 3 (vide infra) suggests that the
þ3 oxidation state of iron is accessible via chemical oxidation.
This indeed is the case. Treatment of 3 in DMF with silver triflate
led to a darkening of the reaction mixture from golden brown to
Crystals of complexes 2ꢀ6 suitable for X-ray diffraction
studies were grown by vapor diffusion of ether into DMF
4110
dx.doi.org/10.1021/ic200143b |Inorg. Chem. 2011, 50, 4107–4115

Inorganic Chemistry
ARTICLE
Figure 3. Solid state structure of the cores of compounds 2, 3, 5, and 6. Thermal ellipsoids at 50% probability level. H atoms, Kþ counterions,
dipropoxyphenoxyl substituents, and solvents of crystallization omitted for clarity. The atom label for N102 has been omitted in all cases. See Table 1 for
notable metrical parameters.
Table 1. Comparison of Metrical Parameters for Bimetallic
Cryptatesa
MꢀM MꢀNap ∑(NeqꢀMꢀNeq) NapꢀNap distance to eq.
metal (Å)
(Å)
(deg)
(Å)
plane (Å)
Mn (2) 6.080 2.246
Fe (3) 6.298 2.180
Co (4)b 6.408 2.142
Ni (5) 6.495 2.064
Zn (6) 6.423 2.162
353.20
356.17
357.35
358.73
358.12
10.567
10.654
10.689
10.616
10.745
0.315
0.229
0.187
0.128
0.157
a All measurements are averages over the two cryptate units in the
asymmetric unit. b From ref 26.
Figure 4. Overlay line drawing of 2 (black) and 6 (red).
the intermetallic region of 3 to give the bridging cyanide complex,
[K(18-crown-6)]3Fe2(μ-CN)L, 10. As with 9, compound 10
displays more resonances in its 1H NMR spectrum than that of its
unbridged analogue 3; this is in keeping with the breaking of
mirror plane symmetry upon addition of cyanide. A single crystal
of 10 grown from a THF solution layered with pentane was
subjected to X-ray analysis, and the solid-state structure shown in
Figure 5 was deduced. Crystals of 10 are isomorphous with
crystals of 9, crystallizing in P1 and having unit cells that differ by
only 9 Å3. The cyano ligand is also disordered end over end; the
ratio of cryptate units possessing a cyano ligand that is C-bound
to Fe1 versus N-bound is 52%, which is comparable to the value
found for the cobalt congener (63%). Complex 10 displays a
longer M2ꢀC3 bond than 9 (2.13(2) Å in 10 vs 2.074(18) Å in
9), and similar M1ꢀN3 (2.12(2) Å in 10 vs 2.100(14) Å in 9)
and C3ꢀN3 bond distances (1.149(12) Å in 10 vs 1.159(1) Å in
9). A CꢀN stretch appears at 2109 cmꢀ1 in the infrared
spectrum of 10 as compared to 2129 cmꢀ1 for 9. Although the
C3ꢀN3 distances in 9 and 10 are essentially unchanged relative
to the CꢀN distance of 1.16 Å in free cyanide,52 the observed
infrared stretches are higher in energy than that of free cyanide
anion (2080 cmꢀ1).53 This observation suggests minimal
deep crimson. Removal of the solvent and subsequent workup
led to the isolation of a brown powder that we have formulated as
the impure diiron(III) cryptate, 8. This species would be
expected to have no extended network of anions bridged by
potassium counter-cations, and it is freely soluble in solvents
such as THF and methylene chloride. This complex does not
pass elemental analysis, and attempts to crystallize it have not yet
met with success. While it is not clear what has prevented the
isolation of this compound in pure form, it should be noted that
difficulty in the characterization of oxidation products of trigonal
monopyramidal iron(II) has been observed previously.43 Anal-
ysis of the worked-up material by M€ossbauer spectroscopy (vide
infra) suggests the presence of one high spin iron(III) environ-
ment. Treatment of this as-isolated material with two equiv of
cobaltocene results in reduction of 8 to the dicobaltocenium
analogue of 3, as judged by 1H NMR spectroscopy, wherein the
potassium counter-cations have been replaced with cobaltoce-
nium cations without any loss of solubility.
Cyanide ion may be included between the metals of the
cryptand cavity. As reported for the conversion of 4 to [K(18-
crown-6)]3Co2(μ-CN)L (9), cyanide ion can be inserted into
4111
dx.doi.org/10.1021/ic200143b |Inorg. Chem. 2011, 50, 4107–4115

Inorganic Chemistry
ARTICLE
Figure 5. Solid state structure of the core of compound 10. Thermal
ellipsoids at 50% probability level. H atoms, dipropoxyphenoxyl sub-
stituents, crown ethers, and solvents of crystallization omitted for clarity.
One phenylene spacer has been grayed for ease of viewing. Notable
metrics: Fe1ꢀN3: 2.12(2) Å, Fe2ꢀC3: 2.13(2) Å, C3ꢀN3: 1.149(12)
Å, Fe1ꢀFe2: 5.3869(1) Å, ∑(NeqꢀFe1ꢀNeq)
∑(NeqꢀFe2ꢀNeq) = 344.10(29)ꢀ.
= 340.83(26)ꢀ;
backbonding between the metals and the bridging cyanide. The
depletion of electron density from the CNꢀ σ* orbital is likely
responsible for the observed increase in the CN frequency of 9
and 10.54
€
Mossbauer Studies. Low-temperature M€ossbauer spectra of
iron complexes 3, 8, and 10 are presented in Figure 6. The
M€ossbauer spectrum of 3 was obtained at 80 K, and this
spectrum was fit by simulation using a single site, in accordance
with the symmetry of the complex. The spectrum consists of a
single quadrupole doublet centered at δ = 0.80 mm/s, with a
quadrupole splitting of ΔEQ = 1.09 mm/s. These values fall in the
standard range for high spin iron(II) complexes,39 though they
are notably different than the values for the trigonal monopyr-
amidal iron(II) complex [{N(CH2CONiPr)3}Fe]ꢀ (δ = 1.05
mm/s and ΔEQ = 3.31 mm/s).42 The main features in the
spectrum of 3, as well as spectra obtained for the other iron
complexes discussed here, are quite broad (line width = ca. 0.7
mm/s) when obtained at liquid nitrogen temperatures. The
origin of the observed broadening is not immediately clear, and
there is little information on the M€ossbauer spectra of trigonal
monopyramidal iron(II) to serve as a guide for what to expect.
One potentially germane benchmark is trigonal monopyramidal
mononuclear iron(II) species of the tris(phenylamido)amine
scaffold.50 In the M€ossbauer spectrum of this compound, two
species are observed, both of which appear to be high spin
iron(II) (δ = 0.75 mm/s and ΔEQ = 0.91 mm/s (Site 1) and δ =
0.76 mm/s and ΔEQ = 1.43 mm/s (Site 2)). Minor asymmetries
in the ligand field about the metal centers, as the compound
crystallizes as a potassium-bridged dimer, are invoked to explain
the presence of two sites in the M€ossbauer spectrum. Given that
3 displays a broad spectrum at low T, and possesses four iron(II)
sites (two cryptate units per unit cell) that are slightly inequi-
valent in the solid state, we considered the possibility of a
multiple site fit for the M€ossbauer data for this compound.
However, a simple one site fit does, in fact, yield a satisfactory
Figure 6. M€ossbauer spectra of (a) 3, (b) 8, and (c) 10. Samples
collected on polycrystalline samples at 80 K (3 and 8) or 5 K (10).
model, excepting the unusually large linewidths. The spectrum
does sharpen considerably upon acquisition at higher tempera-
tures (250 K), with line widths dropping to about 0.45 mm/s;
these data are also satisfactorily fit with a one site model.
The M€ossbauer spectrum of 8 (Figure 6b) at 80 K displays a
single quadrupole doublet with an isomer shift of 0.41 mm/s, and
a quadrupolar splitting of 1.07 mm/s, indicating that the sample
is pure in iron, though it does not pass elemental analysis. While
the linewidths are again quite broad (line width = ca. 0.8 mm/s),
the parameters obtained by a single-site fit suggest that the
complex indeed contains high-spin iron(III). At 5 K, this
compound exhibits a complex spectrum that is difficult to
interpret but is characteristic of a magnetic sample with slow
electronic relaxation.39 Again, M€ossbauer spectra of trigonal
monopyramidal iron(III) have not been reported, but the spectra
of related compounds, [N{CH2CH2NCO-NHtBu}3FeO]2ꢀ and
[N{CH2CH2NC(O)NHtBu}3FeOH]ꢀ, also display complex
spectra at liquid helium temperature, presumably because of
slow electronic relaxation. At 77 K, the signals collapse to
quadrupole doublets with δ = 0.30 mm/s, ΔEQ = 0.71 mm/s
and δ = 0.32 mm/s, ΔEQ = 0.92 mm/s, respectively.55 Another
related trigonal iron(III) compound, Fe[N(SiMe3)2]3,56 which
has been structurally characterized,57 displays parameters of δ =
4112
dx.doi.org/10.1021/ic200143b |Inorg. Chem. 2011, 50, 4107–4115

Inorganic Chemistry
ARTICLE
Figure 7. Cyclic voltammogram of compound 3, referenced to ferro-
cenium/ferrocene.
0.30 mm/s and ΔEQ = 5.12 mm/s.58,59 Of note is the fact that the
spectrum of this compound is broadened significantly because of
slow electronic relaxation, though in contrast to the low tem-
perature spectrum of 8, only one peak of the quadrupole doublet
experiences this effect.
The spectrum of the bridging cyanide complex 10, shown in
Figure 6c, is best modeled by a superposition of two signals,
indicating that 10 possesses two inequivalent sites. The two
signals overlap and display very similar isomer shifts: the para-
meters for the two sites are δ = 0.97 mm/s, ΔEQ = 2.03 mm/s
and δ = 0.99 mm/s, ΔEQ = 2.61 mm/s. The similarity of the
isomer shifts can be ascribed to the qualitative similarity of the
iron environments. Both are tetrahedral iron(II), differing only in
the atom, C or N, of the cyanide ligand that coordinates to the
metal center. Because the magnitude of the quadrupole splitting
decreases with increasing covalency of a given iron center’s
coordination sphere, we assign the signal of smaller quadrupole
splitting at δ = 0.97 mm/s to the iron center that is C-bound to
the cyanide ligand.60 As such, the signal with the larger quadru-
pole splitting at δ = 0.99 mm/s is assigned to the iron that is
N-bound to the cyanide.
Electrochemistry. Whereas complexes 2, 4, and 5 display
irreversible oxidations at potentials positive of ferrocene (see
Supporting Information, Figure S1), 3 displays reversible elec-
trochemical behavior. The cyclic voltammogram in Figure 7
exhibits two reversible one electron waves at ꢀ148 mV
and ꢀ309 mV relative to ferrocene. The difference between
the two oxidation potentials, ΔEox, is equal to 161 mV, which
equates to a comproportionation constant61 of 536. The separa-
tion of the electrochemical waves suggests that a mixed valent
Fe(II)/Fe(III) species may be accessible.
Figure 8. EPR spectra at 4.2 K for (a) 2 and (b) 4. Spectra obtained as
frozen DMF solutions at X-band frequency.
is observed. Manganese and cobalt complexes of trigonal mono-
pyramidal geometry are very rare. No EPR spectra are reported
for the few examples of trigonal monopyramidal manganese(II)
complexes,15 and in only one case has a trigonal monopyramidal
cobalt(II) complex been characterized by EPR.43 The signal ob-
served in the case of this compound is broad, with a g value of 4.17.
Conversely, trigonal bipyramidal cobalt(II) complexes are
more common, and they exhibit EPR spectra that qualitatively
resemble that of 4. The X-band EPR spectrum of pentakis-
(picoline-N-oxide)cobalt(II) perchlorate displays a broad spec-
trum having g1 = 5.67, g2 = 3.53 and g3 = 1.86.62 In this case, g1
and g3 match the spectrum of 4 fairly well, though the spectrum
of 4 lacks an analogous g2 feature. The EPR spectrum of
[CoBr(Me6TREN)]Br qualitatively matches that of 4 nicely,
and has parameters of g|| = 2.27 and g^ = 4.30.63 Note that the
assignment of g|| and g^ are opposite to those assigned in 4. This
is due to the fact that for [CoBr(Me6TREN)]Br, the magnitude
of the feature at g = 2.27 is larger than that at g = 4.30. This is the
reverse of what is observed for 4; further experiments would be
required to definitively assign the identities of the g-values. The
EPR spectra of trigonal bipyramidal manganese(II) complexes
contain, in general, features that are broad and difficult to
interpret. For example, [Mn2(TREN)2(NCO)2](BPh4)2 dis-
plays a very broad signal covering a large magnetic field range,
with several features that could not be confidently assigned.64
Even the EPR spectrum of manganese in pseudotetrahedral fields
is difficult to interpret and simulate.65
The electrochemistry of 10 (Supporting Information, Figure
S1) shows that the reversible electrochemistry observed for 3 is
lost upon insertion of the bridging cyanide. An irreversible
reduction event is observed at ꢀ3 V vs Fc/Fcþ, which feature
remains irreversible with increasing scan rate. Scanning anodi-
cally, irreversible oxidation events are observed at ꢀ970 and
620 mV vs Fc/Fcþ. These events also remain irreversible with
increasing scan rate. Scanning anodically before scanning cath-
odically reveals that the oxidation peaks are not observed until
the potential is swept through the reduction event.
Magnetism. SQuID magnetometric data for the unbridged
bimetallic cryptates is provided in Figure 9. In each case, a fit of
the magnetic data to the Hamiltonian,
EPR Studies. Figure 8 presents the low temperature (4.2 K)
X-band EPR spectra for complexes 2 and 4 as frozen DMF
solutions. The broad features in these spectra are not believed to
be due to aggregation, as less concentrated samples do not
display sharper spectra. The cobalt species displays an axial
spectrum with g|| = 4.61 and g^ = 1.70, and no hyperfine coupling
^
^
^
^
^
^
^
^
H ¼ ꢀ JSA SB þ SA DA SA þ SB DB SB þ βðSA gA
3
3
3
3
3
3
^
þ SB gBÞ B
3
3
using the program julX,27 supports the formulation of the
complexes as high-spin at high temperature. The first term
4113
dx.doi.org/10.1021/ic200143b |Inorg. Chem. 2011, 50, 4107–4115

Inorganic Chemistry
ARTICLE
ineffective overlap between the metal d orbitals and the orbitals
of the cyanide ligand.71
’ CONCLUSIONS
The hexacarboxamide ligand and their metal complexes
reported herein expand the chemistry of cryptands. The sextuply
anionic cryptand can support two metals from the first row
within its cleft of þ2 oxidation state. The diiron(II) complex
exhibits reversible electrochemistry, and we have reported here
initial investigations into the isolation of the analogous diiron-
(III) complex; work in this area is ongoing. Complexation of
cyanide anion to the iron centers in the diiron(II) complex
results in a bridging cyano species, in analogy to the previously
reported dicobalt(II) bridging cyanide. Magnetic analysis of
these unbridged divalent metal complexes and the bridging
cyanide complexes show that they are very weakly antiferromag-
netically coupled. The studies reported herein establish the
hexacarboxamide ligand as a general scaffold for the cofacial
positioning of trigonal metal sites and opens avenues for
cooperative redox and substrate binding.
Figure 9. SQuID data for bimetallic cryptates 2, 3, 4, 5, 9, and 10.
Table 2. Magnetic Parameters for Bimetallic Cryptates
J12 (cmꢀ1
)
g
μeff
μeff
a
b
compound
Mn (2)
ꢀ0.1
ꢀ0.3
ꢀ0.3
ꢀ0.7
ꢀ0.7
ꢀ0.4
2.0
2.2
2.5
2.4
2.1
2.1
8.57
7.52
6.69
4.77
5.76
7.26
8.37
6.93
5.47
4.00
5.47
6.93
Fe (3)
Co (4)
’ ASSOCIATED CONTENT
Ni (5)
Co-μ-CN (9)
S
Supporting Information. Crystallographic details and cif
b
Fe-μ-CN (10)
files, CV data for compounds 2, 5, and 10, and UV/vis data for
compounds 2, 3, and 5. This material is available free of charge via
the Internet at http://pubs.acs.org.
a Room temperature, observed. b Expected spin-only value.
represents the exchange Hamiltonian, the second and third terms
introduce the effect of the local anisotropy of the metal centers,
and the last term represents the Zeeman splitting.66 Because of
the approximate C3h symmetry of the complexes, SA was con-
strained to be equal to SB in all cases. The anisotropy tensors for
both centers were also constrained to be equivalent, as were the
g-tensors for all fits.
A summary of the findings gleaned from the magnetic data is
compiled in Table 2. The magnetic measurements confirm
formulations of the bimetallic cryptates as high spin complexes
in all cases. The magnetic coupling between the metal centers is
antiferromagnetic and extremely weak (0 > J > ꢀ1 cmꢀ1). This
result is in line with expectations, given that the complexes
feature through-space metalꢀmetal distances in excess of 6 Å, an
exceedingly long distance for strong coupling.67,68 The shortest
through-bond pathway in the complexes occurs over a 7-atom
pathway; couplings of about 1 cmꢀ1 are consistent with super-
exchange over 7-atom pathways.69,70
SQuID magnetic data for 9 and 10, as well as the fits of the data
(obtained in the same fashion as those for the unbridged species
discussed above), appear in Figure 9; information obtained from
these fits is presented in Table 2. The metal centers in these
complexes are antiferromagnetically coupled, as they are in the
parent complexes. The exchange coupling might be expected to
significantly increase for two metal centers bridged by a two atom
linear bridge.54 However, this is not the case for the cyanide
complexes. The peculiarly long distance between the metal
centers in both 9 and 10, most likely enforced by the phenylene
spacers, may be playing a role in attenuating the coupling. The
metalꢀmetal distances in these complexes of 5.3869(10) Å and
5.3263(9) Å, respectively, are longer than any FeꢀCtNꢀFe or
CoꢀCtNꢀCo distances reported in the Cambridge Structural
Database.32,33 Moreover, the observation of weak metal-cyanide
π-backbonding in the infrared spectra (vide supra) points to
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: ccummins@mit.edu (C.C.C.); nocera@mit.edu (D.G.N.).
’ ACKNOWLEDGMENT
The compounds were discovered under the sole sponsorship
of Eni S.p.A under the Eni-MIT Alliance Solar Frontiers Pro-
gram. Characterization work was performed under grants from
the NSF (CHE-0533150 and CHE-0802907). The authors thank
Prof. S. J. Lippard for the use of his M€ossbauer spectrometer.
’ REFERENCES
(1) Dietrich, B.; Lehn, J. M.; Sauvage, J. P. Tetrahedron Lett. 1969,
10, 2885–2888.
(2) Chekhlov, A. N. J. Struct. Chem. 2003, 44, 335–339.
(3) Allwood, B. L.; Kohnke, F. H.; Stoddart, J. F.; Williams, D. J.
Angew. Chem., Int. Ed. Engl. 1985, 24, 581–584.
(4) Cohen, S. M.; Petoud, S.; Raymond, K. N. Inorg. Chem. 1999,
38, 4522–4529.
(5) McMurry, T. J.; Hosseini, M. W.; Garrett, T. M.; Hahn, F. E.;
Reyes, Z. E.; Raymond, K. N. J. Am. Chem. Soc. 1987, 109, 7196–7198.
(6) Dye, J. L. Acc. Chem. Res. 2009, 42, 1564–1572.
(7) Steed, J. W.; Atwood, J. L. Supramolecular Chemistry; Wiley:
Chichester, 2009.
(8) Dietrich, B.; Lehn, J. M.; Sauvage, J. P.; Blanzat, J. Tetrahedron
1973, 29, 1629–1645.
(9) Kauffmann, E.; Lehn, J. M. Helv. Chim. Acta 1976,
59, 1099–1111.
(10) Kintzinger, J. P.; Lehn, J. M.; Kauffmann, E.; Dye, J. L.; Popov,
A. I. J. Am. Chem. Soc. 1983, 105, 7549–7553.
(11) Nelson, J.; Nieuwenhuyzen, M.; Pal, I.; Town, R. M. Dalton
Trans. 2004, 229–235.
4114
dx.doi.org/10.1021/ic200143b |Inorg. Chem. 2011, 50, 4107–4115

Inorganic Chemistry
ARTICLE
(12) Ravikumar, I.; Lakshminarayanan, P. S.; Suresh, E.; Ghosh, P.
Inorg. Chem. 2008, 47, 7992–7999.
(49) Jones, M. B.; MacBeth, C. E. Inorg. Chem. 2007, 46, 8117–8119.
(50) Paraskevopoulou, P.; Ai, L.; Wang, Q.; Pinnapareddy, D.;
Acharyya, R.; Dinda, R.; Das, P.; C-elenligil-C-etin, R.; Floros, G.; Sanakis,
Y.; Choudhury, A.; Rath, N. P.; Stavropoulos, P. Inorg. Chem. 2009,
49, 108–122.
(51) Cotton, F. A.; Wilkinson, G.; Murillo, C. A.; Bochmann, M.
Advanced Inorganic Chemistry, 6th ed.; Wiley: New York, 1999.
(52) Elliott, N.; Hastings, J. Acta Crystallogr. 1961, 14, 1018.
(53) Nakamoto, K. Infrared and Raman Spectra of Inorganic and
Coordination Compounds, Part A: Theory and Applications in Inorganic
Chemistry; Wiley: New York, 1997.
(13) MacDowell, D.; Nelson, J. Tetrahedron Lett. 1988, 29, 385–386.
(14) Harding, C. J.; Lu, Q.; Malone, J. F.; Marrs, D. J.; Martin, N.;
McKee, V.; Nelson, J. Dalton Trans. 1995, 1739–1747.
(15) Cummins, C. C.; Lee, J.; Schrock, R. R.; Davis, W. D. Angew.
Chem., Int. Ed. Engl. 1992, 31, 1501–1503.
(16) Cummins, C. C.; Schrock, R. R. Inorg. Chem. 1994,
33, 395–396.
(17) Cummins, C. C.; Schrock, R. R.; Davis, W. M. Organometallics
1992, 11, 1452–1454.
(18) Cummins, C. C.; Schrock, R. R.; Davis, W. M. Angew. Chem.,
Int. Ed. Engl. 1993, 32, 756–759.
(19) Cummins, C. C.; Schrock, R. R.; Davis, W. M. Inorg. Chem.
1994, 33, 1448–1457.
(54) Shatruk, M.; Avendano, C.; Dunbar, K. R. In Prog. Inorg. Chem.;
Karlin, K., Ed.; Wiley: Hoboken, NJ, 2009; Vol. 56, pp 155ꢀ334.
(55) MacBeth, C. E.; Golombek, A. P.; Young, V. G.; Yang, C.;
Kuczera, K.; Hendrich, M. P.; Borovik, A. S. Science 2000, 289, 938–941.
(56) B€urger, H.; Wannagat, U. Monatsh. Chem. 1963,
94, 1007–1012.
(20) Keene, R. F. Coord. Chem. Rev. 1999, 187, 121–149.
(21) Chanda, A.; Popescu, D.-L.; de Oliveira, F. T.; Bominaar, E. L.;
Ryabov, A. D.;M€unck, E.;Collins, T. J.J. Inorg. Biochem. 2006, 100, 606–619.
(22) Miller, C. G.; Gordon-Wylie, S. W.; Horwitz, C. P.; Strazisar,
S. A.; Peraino, D. K.; Clark, G. R.; Weintraub, S. T.; Collins, T. J. J. Am.
Chem. Soc. 1998, 120, 11540–11541.
(23) Bowman-James, K. Acc. Chem. Res. 2005, 38, 671–678.
(24) Ghosh, S.; Roehm, B.; Begum, R. A.; Kut, J.; Hossain, M. A.;
Day, V. W.; Bowman-James, K. Inorg. Chem. 2007, 46, 9519–9521.
(25) Kang, S. O.; Llinares, J. M.; Powell, D.; VanderVelde, D.;
Bowman-James, K. J. Am. Chem. Soc. 2003, 125, 10152–10153.
(26) Alliger, G. E.; M€uller, P.; Cummins, C. C.; Nocera, D. G. Inorg.
Chem. 2010, 49, 3697–3699.
(57) Bradley, D. C.; Hursthouse, M. B.; Rodesiler, P. F. J. Chem. Soc.
D, Chem. Commun. 1969, 14–15.
(58) Alyea, E. C.; Bradley, D. C.; Copperthwaite, R. G.; Sales, K. D.;
Fitzsimmons, B. W.; E, J. C. J. Chem. Soc. D, Chem. Commun.
1970, 1715–1716.
(59) Fitzsimmons, B. W.; Johnson, C. E. Chem. Phys. Lett. 1974,
24, 422–424.
(60) Vꢀertes, A.; Korecz, L.; Burger, K. In M€ossbauer Spectroscopy;
Elsevier: Amsterdam, The Netherlands, 1979, p 139.
(61) Richardson, D. E.; Taube, H. Inorg. Chem. 1981,
20, 1278–1285.
(27) julX is a freely available program created by E. Bill for the
simulation and analysis of magnetic data. It may be downloaded at
http://ewww.mpi-muelheim.mpg.de/bac/ logins/bill/julX_en.php.
(28) Sheldrick, G. M. Acta Crystallogr. 2008, A64, 112–122.
(29) M€uller, P. Crystallogr. Rev. 2009, 15, 57–83.
(62) Bencini, A.; Benelli, C.; Gatteschi, D.; Zanchini, C. Inorg. Chem.
1980, 19, 3839–3841.
(63) Benelli, C.; Gatteschi, D. Inorg. Chem. 1982, 21, 1788–1790.
(64) Laskowski, E. J.; Hendrickson, D. N. Inorg. Chem. 1978,
17, 457–470.
(30) van der Sluis, P.; Spek, A. L. Acta Crystallogr. 1990,
A46, 194–201.
(65) King, E. R.; Betley, T. A. J. Am. Chem. Soc. 2009,
131, 14374–14380.
(31) Spek, A. Acta Crystallogr. 2009, D65, 148–155.
(32) Allen, F. H. Acta Crystallogr. 2002, B58, 380–388.
(33) Search of Cambridge Structural Database performed on July
12, 2010.
(34) Lai, J.-L.; Leung, M.-k.; Lee, G. H. J. Org. Chem. 1996,
61, 8364–8365.
(66) Kahn, O. Molecular Magnetism; VCH Publishers: New York,
1993.
(67) Verdaguer, M. Polyhedron 2001, 20, 1115–1128.
(68) Coffman, R. E.; Buettner, G. R. J. Phys. Chem. 1979,
83, 2387–2392.
(69) Itoh, K.; Minoru, K. Molecular Magnetism; Kodansha: Tokyo,
(35) Colin, J. C.; Mallah, T.; Journaux, Y.; Mollar, M.; Lloret, F.;
Julve, M.; Boubekeur, K. Inorg. Chim. Acta 1996, 246, 249–258.
(36) Yandulov, D. V.; Schrock, R. R.; Rheingold, A. L.; Ceccarelli, C.;
Davis, W. M. Inorg. Chem. 2003, 42, 796–813.
2000.
(70) Papoutsakis, D.; Kirby, J. P.; Jackson, J. E.; Nocera, D. G.
Chem.ꢀEur. J. 1999, 5, 1474–1480.
(71) Beltran, L. M. C.; Long, J. R. Acc. Chem. Res. 2005, 38, 325–334.
(37) Weare, W. W.; Schrock, R. R.; Hock, A. S.; M€uller, P. Inorg.
Chem. 2006, 45, 9185–9196.
(38) Wampler, K. M.; Schrock, R. R. Inorg. Chem. 2007, 46, 8463–8465.
(39) Drago, R. S. Physical Methods for Chemists, 2nd ed.; Surfside
Scientific: Gainesville, FL, 1992.
(40) NMR of Paramagnetic Molecules: Principles and Applications; La
Mar, G. N., Horrocks, D. W., Holm, R. H., Eds.; Academic Press: New
York, 1973.
(41) Ray, M.; Yap, G. P. A.; Rheingold, A. L.; Borovik, A. S. J. Chem.
Soc., Chem. Commun. 1995, 1777–1778.
(42) Ray, M.; Golombek, A. P.; Hendrich, M. P.; Young, V. G.;
Borovik, A. S. J. Am. Chem. Soc. 1996, 118, 6084–6085.
(43) Ray, M.; Hammes, B. S.; Yap, G. P. A.; Rheingold, A. L.;
Borovik, A. S. Inorg. Chem. 1998, 37, 1527–1532.
(44) Borovik, A. S. Acc. Chem. Res. 2004, 38, 54–61.
(45) Filippou, A. C.; Schneider, S.; Schnakenburg, G. Inorg. Chem.
2003, 42, 6974–6976.
(46) Greenwood, B. P.; Rowe, G. T.; Chen, C.-H.; Foxman, B. M.;
Thomas, C. M. J. Am. Chem. Soc. 2009, 132, 44–45.
(47) Schrock, R. R. Acc. Chem. Res. 1997, 30, 9–16.
(48) Das, U. K.; Bobak, J.; Fowler, C.; Hann, S. E.; Petten, C. F.;
Dawe, L. N.; Decken, A.; Kerton, F. M.; Kozak, C. M. Dalton Trans.
2010, 39, 5462–5477.
4115
dx.doi.org/10.1021/ic200143b |Inorg. Chem. 2011, 50, 4107–4115