Synthesis, spectroscopic characterization, and determination of the solution association energy of the dimer [Co{N(SiMe3) 2}2]2: Magnetic studies of low-coordinate Co(II) silylamides [Co{N(SiMe3)2}2L] (L = PMe3, pyridine, and THF) and related species that reveal evidence of very large zero-field splittings

Article abstract of DOI:10.1021/ic402019w

The synthesis, magnetic, and spectroscopic characteristics of the synthetically useful dimeric cobalt(II) silylamide complex [Co{N(SiMe ₃)₂}₂]₂ (1) and several of its Lewis base complexes have been investigated. Variable-temperature nuclear magnetic resonance (NMR) spectroscopy of 1 showed that it exists in a monomer-dimer equilibrium in benzene solution and has an association energy (ΔG _reacn) of -0.30(20) kcal mol^-1 at 300 K. Magnetic data for the polycrystalline, red-brown [Co{N(SiMe₃)₂} ₂]₂ (1) showed that it displays strong antiferromagnetic exchange coupling, expressed as -2J_exS₁S₂, between the two S = ³/₂ cobalt(II) centers with a J _ex value of -215(5) cm^-1, which is consistent with its bridged dimeric structure in the solid state. The electronic spectrum of 1 in solution is reported for the first time, and it is shown that earlier reports of the melting point, synthesis, electronic spectrum, and magnetic studies of the monomer Co{N(SiMe₃)₂}₂ are consistent with those of the bright green-colored tetrahydrofuran (THF) complex [Co{N(SiMe₃)₂}₂(THF)] (4). Treatment of 1 with various Lewis bases yielded monomeric three-coordinated species - [Co{N(SiMe₃)₂}₂(PMe₃)] (2), and [Co{N(SiMe₃)₂}₂(THF)] (4), as well as the previously reported [Co{N(SiMe₃)₂}₂(py)] (3) - and the four-coordinated species [Co{N(SiMe₃)₂} ₂(py)₂] (5) in good yields. The paramagnetic complexes 2-4 were characterized by electronic and ¹H NMR spectroscopy, and by X-ray crystallography in the case of 2 and 4. Magnetic studies of 2-5 and of the known three-coordinated cobalt(II) species [Na(12-crown-4)₂] [Co{N(SiMe₃)₂}₃] (6) showed that they have considerably larger χ_MT products and, hence, magnetic moments, than the spin-only values of 1.875 emu K mol^-1 and 3.87 μ_B, which is indicative of a significant zero-field splitting and g-tensor anisotropy resulting from the pseudo-trigonal crystal field. A fit of χ_MT for 2-6 yields a large g-tensor anisotropy, large negative D-values (between -62 cm^-1 and -82 cm^-1), and E-values between ±10 cm^-1 and ±21 cm^-1.

Full text of DOI:10.1021/ic402019w

Article
pubs.acs.org/IC
Synthesis, Spectroscopic Characterization, and Determination of the
Solution Association Energy of the Dimer [Co{N(SiMe3)2}2]2: Magnetic
Studies of Low-Coordinate Co(II) Silylamides [Co{N(SiMe3)2}2L] (L =
PMe3, Pyridine, and THF) and Related Species That Reveal Evidence
of Very Large Zero-Field Splittings
Aimee M. Bryan,† Gary J. Long,*,‡ Fernande Grandjean,*,‡ and Philip P. Power*,†
†Department of Chemistry, University of California, Davis, One Shields Avenue, Davis, California 95616, United States
‡Department of Chemistry, Missouri University of Science and Technology, University of Missouri, Rolla, Missouri 65409, United
States
S
* Supporting Information
ABSTRACT: The synthesis, magnetic, and spectroscopic
characteristics of the synthetically useful dimeric cobalt(II)
silylamide complex [Co{N(SiMe3)2}2]2 (1) and several of its
Lewis base complexes have been investigated. Variable-
temperature nuclear magnetic resonance (NMR) spectroscopy
of 1 showed that it exists in a monomer−dimer equilibrium in
benzene solution and has an association energy (ΔGreacn) of
−0.30(20) kcal mol−1 at 300 K. Magnetic data for the
polycrystalline, red-brown [Co{N(SiMe3)2}2]2 (1) showed
that it displays strong antiferromagnetic exchange coupling,
3
expressed as −2JexS1S2, between the two S = /2 cobalt(II)
centers with a Jex value of −215(5) cm−1, which is consistent
with its bridged dimeric structure in the solid state. The electronic spectrum of 1 in solution is reported for the first time, and it is
shown that earlier reports of the melting point, synthesis, electronic spectrum, and magnetic studies of the monomer
“Co{N(SiMe3)2}2” are consistent with those of the bright green-colored tetrahydrofuran (THF) complex [Co{N-
(SiMe3)2}2(THF)] (4). Treatment of 1 with various Lewis bases yielded monomeric three-coordinated species[Co{N-
(SiMe3)2}2(PMe3)] (2), and [Co{N(SiMe3)2}2(THF)] (4), as well as the previously reported [Co{N(SiMe3)2}2(py)] (3)and
the four-coordinated species [Co{N(SiMe3)2}2(py)2] (5) in good yields. The paramagnetic complexes 2−4 were characterized
by electronic and 1H NMR spectroscopy, and by X-ray crystallography in the case of 2 and 4. Magnetic studies of 2−5 and of the
known three-coordinated cobalt(II) species [Na(12-crown-4)2][Co{N(SiMe3)2}3] (6) showed that they have considerably larger
χMT products and, hence, magnetic moments, than the spin-only values of 1.875 emu K mol−1 and 3.87 μB, which is indicative of
a significant zero-field splitting and g-tensor anisotropy resulting from the pseudo-trigonal crystal field. A fit of χMT for 2−6 yields
a large g-tensor anisotropy, large negative D-values (between −62 cm−1 and −82 cm−1), and E-values between 10 cm−1 and
21 cm−1.
Mn, Fe, Co) as starting materials for many syntheses.3,4 The
INTRODUCTION
■
trivalent metal series also has been extended to include the M3+
Over 50 years ago, Burger and Wannagat used the silylamido
ligand −N(SiMe3)2 to obtain a series of stable, low-coordinate
(coordination numbers of two or three) first-row transition-
derivatives of scandium, titanium, vanadium, manganese, and
cobalt (i.e., M{N(SiMe3)2}3, where M = Sc,5,6 Ti,5−7 V,5−7 Mn,8
and Co8) and they were shown to be three-coordinate9
monomers either in the gas phase by electron diffraction,10 or
in the solid state by X-ray crystallography9,11−14 or by
spectroscopy.15−18 The divalent compounds [M{N(SiMe3)2}2]
(M = Mn, Fe, or Co) were also reported to be monomeric in
the gas phase by electron diffraction3 and in hydrocarbon
solution by cryoscopy and spectroscopy;19,20 however, as solids,
metal complexes.1a The synthesis and (partial) characterization
of the divalent M(II) derivatives [Mn{N(SiMe3)2}2],1b [Co-
{N(SiMe3)2}2],1a [Ni{N(SiMe3)2}2] (unstable),1b and the
trivalent M(III) complexes [Cr{N(SiMe3)2}3]1b and [Fe{N-
(SiMe3)2}3]1a were described. Their solubility in hydrocarbons
and high volatility suggested that they existed as unassociated
molecules. Later work extended the divalent series to include
the iron(II) species [Fe{N(SiMe3)2}2].2 This series has proven
to be a very useful group of compounds, because they are
widely used as hydrocarbon soluble sources of M2+ ions (M =
Received: August 5, 2013
Published: October 10, 2013
© 2013 American Chemical Society
12152
dx.doi.org/10.1021/ic402019w | Inorg. Chem. 2013, 52, 12152−12160

Inorganic Chemistry
Article
(SiMe3)2}3] (6)30 were prepared according to literature (2, 5, 6) or
modified procedures (1).
they have been shown, via X-ray diffraction (XRD), to have
dimeric structures in which the metal ions were bridged by
amido ligands.20−22 A feature of the investigations of the
divalent [M{N(SiMe3)2}2]2 (M = Mn, Fe, or Co) complexes
and their derivatives has been that their characterization has not
generally included quantitative investigations of their magnetic
properties. For example, details of the magnetic behavior of
[Mn{N(SiMe3)2}2]2 and [Fe{N(SiMe3)2}2]2 remain unpub-
lished. An investigation of the magnetism of solid “Co{N-
(SiMe3)2}2” in the range of 80−300 K by Fisher and Bradley in
1971 produced a magnetic moment of 4.83 μB that is
independent of temperature in the range of 80−300 K (cf.
[Co{N(SiMe3)2}2]2 (1). A diethyl ether suspension of LiN-
33,34
(SiMe3)2
was synthesized in situ by adding n-BuLi (18 mL, 2.5
M solution in hexanes, 0.044 mol) dropwise to HN(SiMe3)2 (8.4 mL,
0.040 mol) in diethyl ether (40 mL), cooled in an ice bath. The
solution was allowed to come to room temperature and stirring was
continued for 12 h. The resulting opaque suspension was added
dropwise via cannula to a diethyl ether (40 mL) slurry of CoCl2 (2.86
g, 22.0 mmol) chilled in an ice bath. An immediate color change of the
slurry from blue to dark green was observed. When the addition was
complete, the suspension was warmed to ca. 35 °C and stirred for 12
h. The ether was removed under reduced pressure and the resulting
dark green solids were extracted with hexanes (40 mL), which resulted
in a dark green solution with a gray precipitate. The solution was then
filtered through a Celite padded filter-stick to afford a clear dark-green
solution. The hexanes were removed under reduced pressure to give a
dark green oil. The oil was distilled as a dark green vapor at ca. 100 °C
(5 × 10−2 Torr) using a short-path distillation apparatus. Upon
cooling, the vapor solidified to a brown-dark red solid. The solid was
dissolved in hexanes (ca. 30 mL) at 65 °C, which afforded an olive-
green solution. Cooling slowly to 0 °C gave 1, in the form of red-olive
dichroic crystals. Yield: 5.70 g (ca. 75%). Melting point (Mp): 89−90
°C.1 Anal. calcd. for 1: C, 37.96; H, 9.56; N, 7.38. Found: C, 38.89; H,
10.04; N, 7.17. UV-vis/NIR (hexane, nm [ε, M−1 cm−1]: 209 [3000],
223 [11000], 281 [3400], 324 [6500], 604 [140], 668 [200]. IR in
Nujol mull (cm−1) with CsI plates: 3140, 2890, 2710, 2650, 1450,
1368, 1357, 1340, 1290, 1250, 1239, 1150, 1070, 1010, 955, 918, 880,
840, 828, 810, 794, 726, 710, 657, 600, 348, 265. μB in C6D6 solution =
4.7(2) (Evans’ method). 1H NMR (400 MHz, C6D6, 295 K): 8.97 (br
s, −SiMe3, [Co{N(SiMe3)2}2], 0.51 (s, −SiMe3 (terminal), [Co{N-
(SiMe3)2}2]2), −4.22 (br s, −SiMe3 (bridging), [Co{N(SiMe3)2}2]2)
ppm.
3
3.87 μB for a spin-only moment, S = /2), and they proposed
that the high value of the magnetic moment was due to spin−
orbit coupling.19 However, this interpretation was based on the
assumption of a monomeric two-coordinate structure that is at
variance with the later X-ray crystallographic data, which
showed that “Co{N(SiMe3)2}2” exists as the dimer
[(Me3Si)2NCo{μ-N(SiMe3)2}2CoN(SiMe3)2] (1), in which
two three-coordinate Co2+ ions are bridged by two −N-
(SiMe3)2 ligands, and are also bonded to a terminal
−N(SiMe3)2 amido group.22 The burgeoning interest in low-
coordinate transition-metal derivatives, especially those of iron
and cobalt as potential candidates for single molecule
magnets,4,23−26 has added new urgency to the investigation of
their magnetic properties. Because of the low number of
ligands, these compounds often display high magnetic mo-
ments, due to either an orbital contribution to the mo-
ment21−24 or zero-field splitting and g-tensor anisotropy
resulting from the uniaxial ligand field.25−29 The position of
the divalent silylamides as key synthons for low-coordinate
transition-metal derivatives, as well as the apparently contra-
dictory results of the previous data for “Co{N(SiMe3)2}2”,
show that further work was warranted. We now describe our
investigations of the magnetic properties of [Co{N(SiMe3)2}2]2
(1) as well as spectroscopic and magnetic properties of its
Lewis base adducts [Co{N(SiMe3)2}2(PMe3)] (2), [Co{N-
(SiMe3)2}2(py)] (3), and [Co{N(SiMe3)2}2(THF)] (4), [Co-
{N(SiMe3)2}2(py)2] (5), and [Na(12-crown-4)2][Co{N-
(SiMe3)2}3] (6).30 The results indicate that the previous
synthesis,1a melting point,1a electronic spectrum,19 and
magnetic data19 reported for monomeric “Co{N(SiMe3)2}2”
[Co{N(SiMe3)2}2(PMe3)] (2). A diethyl ether solution of
trimethylphosphine (PMe3) (0.36 mL, 3.5 mmol) (30 mL) was
added dropwise via cannula at 298 K to a 30 mL hexane solution of
[Co{N(SiMe3)2}2]2 (1) (1.09 g, 1.44 mmol), which produced an
immediate color change of the solution from dark olive-green to blue-
green. The solution was concentrated under reduced pressure until
blue-green solids began to form, at which point the solution was
warmed to redissolve the solids. The solution was then placed in a −18
°C freezer overnight which afforded 2 as blue-green crystals. Yield:
1.14 g (ca. 86%). Mp: 97−99 °C. UV-vis/NIR (hexane, nm [ε, M−1
cm−1]): 605 [60], 653 [30], 715 [90], 888 [8], 1420 [30]. IR in Nujol
mull (cm−1) with CsI plates: 2945, 2910, 2840, 1450, 1368, 1240,
1
1160, 920, 875, 834, 710, 655, 600, 350. H NMR (400 MHz, C6D6,
298 K): 189.34 (br s, PMe3), −24.65 (br s, −SiMe3) ppm. Anal. Calcd.
for 2: C, 39.53; H, 9.95; N, 6.15. Found: C, 39.78; H, 10.38; N, 6.00.
[Co{N(SiMe3)2}2(py)] (3). This compound was prepared according
to literature procedures.32 UV-vis/NIR (hexane, nm [ε, M−1 cm−1]):
647 [100], 694 [100], 812 [6], 1461 [20]. 1H NMR (400 MHz, C6D6,
298 K): 342.6 (br, p-C6H5N), 139.9 (br, m/o-C6H5N), 67.8 (br, m/o-
C6H5N), −19.2 (br s, −SiMe3) ppm. Anal. Calcd. for 3: C, 44.50; H,
9.01; N, 9.16. Found: C, 44.57; H, 9.23; N, 9.04.
[Co{N(SiMe3)2}2(THF)] (4). THF (15 mL, ca. 180 mmol) was
syringed onto solid [Co{N(SiMe3)2}2]2 (1) (1.28 g, 1.69 mmol) in a
Schlenk flask. After stirring for several minutes, the bright green
solution was decanted from a small amount of undissolved solids. The
THF solution was concentrated under reduced pressure until green
solids began to form. The solution was warmed to redissolve the solids
and then stored at room temperature overnight, which afforded a mass
of 4 in the form of bright green crystals. Yield: 1.20 g (ca. 78%). Mp:
71−73 °C. UV-vis/NIR (hexane, nm [ε, M−1 cm−1]): 593 [6],
680 [90], 1565 [10]. IR in Nujol mull (cm−1) with CsI plates: 2980,
2950, 2910, 2900, 2680, 1450, 1375, 1340, 1250, 1080, 1030, 990, 830,
750, 670, 635, 615, 360. 1H NMR (400 MHz, C6D6, 296 K): 166.9 (br,
−CH2CH2O− or −CH2CH2O−), 99.5 (br, −CH2CH2O− or
−CH2CH2O−), −17.3 (br s, −SiMe3) ppm. Anal. Calcd. for 4: C,
42.53; H, 9.82; N, 6.20. Found: C, 42.84; H, 10.35; N, 6.10.
most probably were for the bright green (giftgrun)1a complex
̈
[{Co{N(SiMe3)2}2(THF)] (4), which is readily produced
when the synthesis of “Co{N(SiMe3)2}2” is conducted in THF
solvent.1a
EXPERIMENTAL SECTION
■
All manipulations were performed with the use of modified Schlenk
techniques or in a Vacuum Atmospheres drybox under N2 or argon.
Solvents were dried and collected using a Grubbs-type solvent
purification system31 (Glass Contour) and degassed by using the
freeze−pump−thaw method. All physical measurements were
obtained under strictly anaerobic and anhydrous conditions. IR
spectra were recorded as Nujol mulls between CsI plates on a Perkin−
Elmer 1430 spectrophotometer. UV−visible spectra were recorded as
dilute hexane solutions in 3.5 mL quartz cuvettes using an Olis 17
Modernized Cary 14 UV/vis/NIR spectrophotometer. Melting points
were determined on a Meltemp II apparatus using glass capillaries
sealed with vacuum grease and are uncorrected. Unless otherwise
stated, all materials were obtained from commercial sources and used
as received. [Co{N(SiMe3)2}2]2 (1),1a [Co{N(SiMe3)2}2(py)] (2),32
[Co{N(SiMe3)2}2(py)2] (5),32 and [Na(12-crown-4)2][Co{N-
12153
dx.doi.org/10.1021/ic402019w | Inorg. Chem. 2013, 52, 12152−12160

Inorganic Chemistry
Article
X-ray Crystallography. Crystals for XRD studies were removed
from the Schlenk tube under a stream of nitrogen and immediately
covered with hydrocarbon oil (Paratone-N). A suitable crystal was
selected, attached to a glass fiber on a mounting pin, and quickly
placed in a low-temperature stream of nitrogen (ca. 90 K for 2, ca. 150
K for 4).35 Data for compound 4 were collected at 150 K due to a
phase transition that occurred at ca. 140 K. Data for compounds 2 and
4 were obtained on a APEX-II DUO system, using Mo Kα radiation (λ
= 0.71073 Å) in conjunction with a CCD detector. A multiscan
absorption correction was applied with the program SADABS.36,37 The
structures were solved by direct methods and refined with the
SHELXTL (2013) software package, and the thermal ellipsoid plots
were drawn using OLEX2 software.37,38 Refinement was by full-matrix
least-squares procedures with all carbon-bound hydrogen atoms
included in calculated positions and treated as riding atoms. A
summary of crystallographic and data collection parameters for 2 and 4
is given in Table 1.
cm and its contribution was thus negligible. Diamagnetic corrections
of −0.000505, −0.000271, −0.000301, −0.000254, −0.000315, and
−0.000610 emu/mol, obtained from tables of Pascal’s constants, were
applied to the measured molar magnetic susceptibilities of 1−6,
respectively.39 Statistical fitting errors are reported below; the actual
errors may be as much as twice as large. The 5 K magnetizations of 1−
4 were subsequently measured in a 0−7 T applied field. No eicosane
or diamagnetic corrections were applied to the 5 K magnetization
results.
RESULTS AND DISCUSSION
■
Synthesis. The dimer [Co{N(SiMe3)2}2]2 (1) was
synthesized by a modified version of that described earlier by
Burger and Wannagat1a via the addition of LiN(SiMe3)233,34 to
̈
a suspension of CoCl2 in diethyl ether (originally the addition
of NaN(SiMe3)2 to CoCl2 in THF), which afforded a dark
olive-green solution. Removal of the solvent followed by
extraction with hexanes, filtration, concentration, and distil-
lation of the residue afforded [Co{N(SiMe3)2}2]2 as an
intensely colored dark green vapor, which solidified to a red-
brown solid (Mp 89−90 °C, cf. 73 °C in ref 1a). The same
synthesis in THF instead of ether affords the THF complex 4,
in preference to 1, which can be distilled intact (provided
temperatures do not exceed ca. 100 °C), as a green oil, which
solidifies to give 4 as bright green crystals that melt in the range
of 71−73 °C (described as “Co{N(SiMe3)2}2” in ref 1a). This
behavior is analogous to that of its manganese and iron
analogues [M{N(SiMe3)2}2(THF)], M = Mn21,40 and Fe,20
which can be distilled or sublimed at ≤100 °C without
significant dissociation of THF. The THF complex [Co{N-
(SiMe3)2}2(THF)] (4) can also be obtained by simply
dissolving 1 in THF and isolating crystals of the product by
concentrating the solution. Complexes 2 and 5 were made by
adding the Lewis bases in slight excess to 1 with stirring for
several hours. The monopyridine complex 3 was obtained by
subliming 5 at ca. 100 °C under reduced pressure (ca. 5−7
mTorr).32 Complex 6 was synthesized as described earlier.30
Recrystallization of 1−3, 5, and 6 at low temperature from
hexane produced good yields of the pure products. The PMe3
complex (2) resembles the PPh3 complex [Co{N-
(SiMe3)2}2(PPh3)] of Bradley, Hursthouse, and coworkers,41
which was synthesized by a different route involving the
reaction of 2 equiv LiN(SiMe3)2 with CoCl2(PPh3)2.
Structures. The previously published structure of 1 showed
that it featured two three-coordinated cobalt(II) ions bridged
by two amido ligands and each terminally bonded to one amido
ligand.22 The Co···Co separation was 2.583(1) Å and the
terminal and bridging Co−N distances were 1.922(5) and
2.062(4) Å, respectively, with a planar Co2N2 core structure as
well as trigonal planar coordination at each Co2+ ion.22 A new
crystal structure (see Supporting Information) afforded Co···
Co = 2.5865(5) Å, a terminal Co−N = 1.9135(12) Å, and an
identical Co−N bridging bond length. The Lewis base complex
2 features trigonal planar geometry at Co (Σ0 = 359.98(7))
with somewhat different P−Co−N angles of 108.03(6)° and
113.80(6)°. The Co−N distances are equal at 1.9159(18) and
1.9160(18) Å. The Co−P distance is 2.3976(7) Å. The crystal
structure of 4 shows that it possesses a 2-fold axis of symmetry
along the Co−O bond. It features a Co−N distance of
1.9000(15) Å and a Co−O distance of 2.0358(19) Å, with a
N−Co−N angle of 141.89(9)°. Its structure resembles that of
its iron analogue, [Fe{N(SiMe3)2}2(THF)],20 which has Fe−N
and Fe−O distances of 1.916(5) and 2.071(6) Å, respectively
(these bond distances are longer than those in 4, because of the
Table 1. Selected Crystallographic and Data Collection
Parameters for Complexes 2 and 4
[Co{N(SiMe3)2}2(PMe3)] [Co{N(SiMe3)2}2(THF)]
(2)
(4)
formula
C15H45CoN2PSi4
455.79
C16H44CoN2OSi4
451.82
formula weight,
Fw
color, habit
crystal system
space group
a (Å)
blue-green, needle
monoclinic
P21/c
bright green, needle
orthorhombic
Pbcn
15.1133(19)
10.0248(13)
17.973(2)
90.00
13.5395(3)
11.1656(3)
17.7402(4)
90.00
b (Å)
c (Å)
α (deg)
β (deg)
γ (deg)
V (Å3)
91.745(2)
90.00
90.00
90.00
2721.8(6)
4
2681.90(11)
4
Z
crystal dimensions
(mm)
0.32 × 0.27 × 0.14
0.40 × 0.22 × 0.016
T (K)
90
150
dcalc (g/cm3)
1.112
0.867
1.119
0.825
abs. coefficient μ
(mm−1
)
θ range (deg)
R(int)
2.27−27.55
0.0331
2.30−27.48
0.0322
obs reflections
[I > 2σ(I)]
5772
2601
data/restraints/
parameters
6259/0/223
0.0360
3080/46/156
0.0402
R1, observed
reflections
Magnetic Studies. The powdered samples of 1−6 used for
magnetic measurements were sealed under vacuum in 6-mm outer
diameter (OD) and 4-mm inner diameter (ID) quartz tubes with a
thin shelf and the sample moment was measured using a Quantum
Design MPMSXL7 superconducting quantum interference magneto-
meter. To prevent crystallite reorientation by the applied field, each
sample was anchored with eicosane. For each compound, the sample
was zero-field-cooled to 2 K and the moment was measured upon
warming to 300 K in an applied field of 0.01 T. In order to ensure
thermal equilibrium between the powdered sample sealed under
vacuum in the quartz tube and the temperature sensor, the moment
was measured at a given sensor temperature until a constant value
moment was observed; ca. 14 h were required for the measurements
between 2 and 300 K. The measured moments were corrected for the
presence of eicosane; no quartz tube correction was necessary because
the quartz tube extended equally above and below the sample by ca. 5
12154
dx.doi.org/10.1021/ic402019w | Inorg. Chem. 2013, 52, 12152−12160

Inorganic Chemistry
Article
larger size of the Fe2+ ion).42 Oddly, the N−Co−N angle in 4,
141.89(9)°, is ca. 2° narrower than that of its iron congener
(N−Fe−N = 144.0(3)°), despite the smaller size of cobalt. The
Co−N distances in 2 are longer than those in 4, possibly
because of the strong σ-donor characteristics of PMe3, which
increase the electron density at the Co ion and reduces the
attraction of the metal for the electronegative −N(SiMe3)2
ligand. However, the Co−P bond length in 2 is shorter than
that of the complex [Co{N(SiMe3)2}2(PPh3)].41 This may be
partly a result of the greater steric crowding in the latter species.
The structures of 2 and 4 are illustrated in Figures 1 and 2,
previously reported three-coordinated cobalt(II) anion (6) has
the longest M−N bonds,30 as a result of interelectronic
repulsion produced by the negative charge. Consistent with
previous observations,32 neutral three-coordinated complexes
have shorter M−N bonds than those in four coordinate
complexes by a margin of 0.06−0.09 Å, except in the case of 6,
which is the most crowded of the compounds, because of the
three large −N(SiMe3)2 ligands and the negative charge, which
lengthens the Co−N bonds.
Spectroscopy. The ground state of the free Co2+ ion is
4F9/2 and the splitting of the free-ion terms for a three-
4
coordinated compound in a D3h ligand field is A2′ + (4A1″,
17
4
4
4A2″) + E″ + E′. In idealized D3h symmetry, four d−d
transitions are expected. If the symmetry is lowered further, as
it is in the three-coordinated Lewis base complexes (2−4),
4
further splitting of the degenerate E states is expected so that
six bands are predicted. The electronic spectrum of 1 in
solution was assigned by Fisher and Bradley on the basis of the
monomeric formula Co{N(SiMe3)2}2 and revealed four bands
at 409, 585, 685, and 1538 nm with much more intense bands
at shorter wavelengths.19 The electronic spectrum of 1 obtained
from a sample synthesized by the method described in the
Experimental Section had absorptions at wavelengths of 209,
223, 281, 324, 604, and 668 nm, which are different from those
in ref. 19. However, the spectrum of the THF complex 4 in
hexane afforded three transitions (nm [ε, M−1 cm−1]) observed
at at 593 [10], 680 [90], and 1565 [10] over a spectral range of
500−2600 nm, which correspond closely to the three longest
wavelength values reported in ref 19. For 2, in hexane, five
transitions (nm [ε, M−1 cm−1]) were observed, at 605 [60],
653 [30], 715 [90], 888.8 [8], and 1420 [30]; for 3, in hexane,
four transitions were observed, at 647 [100], 694 [100],
812 [6], and 1461 [20].32 These absorptions agree with the
colors of the series: 1, red-brown;1 2, blue-green; 4, bright
green; 3, blue-green;32 and 5, aquamarine.32 Additional
absorptions are also observed for 2−5 in the shorter-
wavelength region (<400 nm).32 The absorptions below ca.
500 nm in 1 and its complexes have high intensities and are
most probably due to ligand-to-metal electron transfer from
nitrogen lone pairs to metal d-orbitals. The lower intensities (ε
= ca. 10−100 M−1 cm−1) of bands in the 500−2600 nm range
indicate that they are d−d transitions.
Figure 1. Thermal ellipsoid (50%) drawing of [Co{N-
(SiMe3)2}2(PMe3)] (2). H atoms are not shown for clarity. Selected
bond lengths and angles are given in Table 2.
The IR spectra are consistent with previous work.19,32 The
νas(MN2) bands identified in the compounds “Co{N-
(SiMe3)2}2” (362 cm−1),19 [Na(12-crown-4)2][Co{N-
(SiMe3)2}3] (368 cm−1),30 and [Co{N(SiMe3)2}2(py)2] (5)
(360 cm−1)32 are in agreement with our measurements (2, 350
cm−1 and 4, 360 cm−1). The Co−L stretching modes probably
appear at lower energies beyond our instrument’s lower limit of
250 cm−1, because of their longer and weaker bonds. Other
than this basic analysis, full spectral assignments would require
full-molecule density functional theory calculations on the
series in order to fully interpret the electronic spectra.
Figure 2. Thermal ellipsoid (50%) drawing of [Co{N-
(SiMe3)2}2(THF)] (4). H atoms and disorder are not shown for
clarity. Selected bond lengths and angles are given in Table 2.
NMR Spectroscopy. At ambient temperature, the 1H NMR
spectra of 2−4 display large paramagnetic shifts of the Lewis
base molecule hydrogens and −SiMe3 signals. Signals due to the
latter all appear at upfield shifts in the same spectral region:
−24.7 ppm for 2, −19.2 ppm for 3, and −17.3 ppm for 4. The
1H NMR spectrum for the PMe3 complex 2 displays one other
signal at a chemical shift of 189.3 ppm, which integrates to a 1:4
ratio with respect to the −SiMe3 signal and can thus be readily
assigned to the −PMe3 hydrogens. For 3, based on the
respectively, and some important bond distances and angles are
given in Table 2, along with data for the bispyridine complex 5,
the three-coordinated anion of the ionic complex [Na(12-
crown-4)2][Co{N(SiMe3)2}3] (6),30 and the related species
[Co{N(SiMe3)2}2(PPh3)].41
It can be seen that the three-coordinated Lewis base
complexes (2−4) all have Co−N distances near 1.9 Å. The
12155
dx.doi.org/10.1021/ic402019w | Inorg. Chem. 2013, 52, 12152−12160

Inorganic Chemistry
Article
Table 2. Selected Interatomic Distances (Å) and Angles (deg) for Compounds 2−6 and [Co{N(SiMe3)2}2(PPh3)]
[Co{N(SiMe3)2}2
(PMe3)] (2)
[Co{N(SiMe3) }2
a 2
[Co{N(SiMe3)2}2
(THF)] (4)
[Co{N(SiMe3) }2
a2
[Na(12-crown-4)2]
[Co{N(SiMe3)2}3] (6)
[Co{N(SiMe )2}2
b
c3
(py)] (3)
(py)2] (5)
(PPh3)]
Bond Distances (Å)
Co−N(1)
Co−N(2)
Co−N(3)
Co−L
1.9159(18)
1.9160(18)
1.904(3)
1.904(3)
1.9000(15)
1.984(1)
1.983(1)
1.975(2)
1.981(2)
1.973(2)
1.931(14)
1.924(13)
2.3976(7)
138.15(8)
2.055(5)
140.7(2)
2.0358(19)
Bond Angles (deg)
141.89(9)
2.1135(1) 2.115(1)
2.479(5)
130.7(7)
N(1)−Co−N(1A,2)
123.17(4)
90.1(4)
120.50(8), 120.37(8),
119.13(8)
N(1)−Co−L
N(2)−Co−L
108.03(6)
113.80(6)
109.7(1)
109.7(1)
109.05(4)
109.05(4)
118.5(4)
110.6(2)
Σ0 Co
359.98
360.10
359.99
360.00
359.80
a
b
c
Data taken from ref 32. Data taken from ref 30. Data taken from ref 41.
1
Figure 3. H NMR spectra recorded in C6D6 for the monomer−dimer equilibrium of 1 in the temperature range 289−324 K.
integrated intensities, the para hydrogen has the furthest
downfield shift at 342.6 ppm and the meta and ortho hydrogens
are due to absorptions at 139.9 and 67.8 ppm (assignments
uncertain). The two methylene −CH2− resonances of 4 can be
assigned to absorptions at 166.9 and 99.5 ppm, based on their
equal peak intensities and integration ratios, with respect to the
−N(SiMe3)2 signal.
dimer equilibrium in solution (Figure 3). An earlier study of the
iron(II) analogue Fe{N(SiMe3)2}2 showed that it was
essentially monomeric in solution at room temperature.20
1
This is not the case for 1. At 299 K, the H spectrum clearly
shows three signals in addition to those of the solvent: a broad
signal at 9.09 ppm, which can be assigned to the paramagneti-
cally shifted −SiMe3 protons of the Co{N(SiMe3)2}2 monomer;
a narrower signal at 0.51 ppm; and a broad signal at −4.20
ppm, which can be assigned to the terminal and bridging
The variable-temperature 1H NMR (800 MHz, C6D6)
spectrum of 1 clearly demonstrates evidence of a monomer−
12156
dx.doi.org/10.1021/ic402019w | Inorg. Chem. 2013, 52, 12152−12160

Inorganic Chemistry
Article
−SiMe3 hydrogens of the dimer [Co{N(SiMe3)2}2]2. Upon
cooling, the dimer signals intensify, with the terminal −SiMe3
signal shifting slightly downfield to 0.52 ppm (289 K) and the
bridging −SiMe3 signal shifting upfield to −4.27 ppm (289 K).
Simultaneously, the monomer resonance shifts upfield to 8.55
ppm and decreases in intensity. Upon warming to 324 K, the
dimer signals essentially disappear and the monomer signal
greatly intensifies and is shifted further downfield to 10.18 ppm.
The association equilibrium constants for eq 1 using the Van’t
Hoff equation over the range of 289−324 K and the relative
concentrations of monomer and dimer derived from peak
integration of a 0.12(1) M solution of the monomer
Co{N(SiMe3)2}2 in C6D6 gave values of −20(2) kcal mol−1
for ΔHr0eacn and −66(6) cal mol−1 K−1 for ΔS0reacn
2Co{N(SiMe3)2}2 ⇄ [Co{N(SiMe3)2}2]2
.
(1)
Figure 4. Temperature dependence of χMT obtained at 0.01 T for 1,
black points, and a fit, black line, obtained between 2 and 300 K, for
3d7 cobalt(II) and S1 = S2 = 3/2 and g = 2 with Jex = −215(5) cm−1, Nα
= 0.00157(5) emu/mol, and 2.5 mol % of an impurity of Co(II) with S
Using these values at 300 K, the ΔGreacn is −0.30(20) kcal
mol−1. Compared to the association constants for the iron(II)
analogue (−19 kcal mol−1 for ΔHr0eacn, −75 cal mol−1 K−1 for
ΔS0reacn and 3 kcal mol−1 for ΔGreacn).20 The relatively weak
association observed for 1 and [Fe{N(SiMe3)2}2]2 is supported
by the fact that monomeric structures are observed in the
crystal phase when one or two of the methyl substituents on
each −SiMe3 group are replaced by phenyl groups.42
3
= /2 and g = 2. Inset shows the magnetization of 1 obtained at 5 K.
the 2−300 K temperature dependence of χMT of 1 has been fit
with the Heisenberg isotropic exchange coupling Hamiltonian:
H = −2J S1S2
ex
The conclusion from the variable-temperature 1H NMR
investigation is that, in concentrations of ca. 0.1 M of
Co{N(SiMe3)2}2 in C6D6, the ratio of monomer to dimer is
ca. 4:1 at room temperature. Thus, the majority of the sample is
monomeric but there is a significant percentage of dimer also
present. The free energy of the reaction slightly favors (by ca.
−0.30(20) kcal mol−1) the dimeric structure at room
temperature in contrast to the corresponding iron(II) amide
equilibrium in which the monomer is favored by ca. 3 kcal
mol−1 at room temperature.20 Because of the smaller radius of
the Co2+ ion, and therefore greater steric crowding produced by
the −SiMe3 groups, it might be expected that the association
energy of the cobalt(II) dimer would be lower than that of the
iron analogue, but this is not the case. The stronger association
of 1 is supported by the fact that it has a higher melting point
than that of its iron congener and because our experience
indicates that it distills under reduced pressure at a significantly
higher temperature.
the result of this fit is shown as the line in Figure 4. For 3d7
3
cobalt(II) with S1 = S2 = /2 and g = 2, the fit yields Jex
=
−215(5) cm−1, Nα = 0.00157(5) emu/mol, and 2.5 mol % of
an impurity of Co(II) (the possibility that this impurity is
monomeric Co{N(SiMe3)2}2 suggests itself, but no evidence
for its presence was apparent from the X-ray data) with S = 3/2
and g = 2. Because of the strong correlation between the
parameters, g was not varied but fixed at g = 2. Furthermore,
there is a correlation between Jex and Nα, such that probably
the best conclusion is that (i) at the lower temperatures, as a
result of strong antiferromagnetic exchange, only the J = 0 state
is populated and (ii) the second-order Zeeman contribution
(Nα) is substantial. These conclusions are fully consistent with
the structure of 1 summarized above. The magnetization of 1
has been measured at 5 K and its initial slope agrees well with
the 5 K value of χM obtained at 0.01 T. As shown in the inset
for Figure 4, the 5 K magnetization of 1 is only 0.066 Nβ at 7 T
and arises mainly from the 2.5 mol % impurity plus possibly a
small contribution from canting of the antiferromagnetically
coupled moments. The lower magnetic moment for 1 is
consistent with the low magnetic moment (1.72 μB),
determined for the dimer [{Co(NPh2)2}2] via the Evans’
method.50
Magnetism. It should be remarked that these studies of
three- and four-coordinated cobalt(II) complexes are paralleled
by magnetic studies on related iron(II) complexes involving
[Fe{N(SiMe3)2}2] complexed by carbene,44,45 phosphine,46
pyridine,32 and β-diketiminate47,48 ligands. The magnetic
moment for the cobalt(II) amide Co{N(SiMe3)2}2 (which is
assumed to be monomeric, based on the electronic and mass
spectroscopy and molecular weight data in cyclohexane
solution) was reported to be 4.83 μB.19 This value falls within
the range measured for several two-coordinated Co(II) amides
(4.1−6.3 μB).4,28,43,49 We investigated the magnetic properties
of a polycrystalline sample of [Co{N(SiMe3)2}2]2 (shown to
have a dimeric structure by X-ray crystallography).22 The
magnetic properties of [Co{N(SiMe3)2}2]2 (1) have been
measured, after zero-field cooling, upon warming from 2 K to
300 K in a 0.01 T applied magnetic field. The results in terms of
χMT are shown as the points in Figure 4. The magnitude of
χMT is very low, in comparison to that in ref 19, and is
indicative of extensive antiferromagnetic exchange between two
We also investigated the magnetism of 1 in C6D6 solution via
the Evans’ method. At a concentration of 0.033 M, a magnetic
moment of 4.7(3) μB was observed, which is close to the 4.83
μB value measured for the putative “Co{N(SiMe3)2}2”
monomer in the solid state.19 However, this apparent
agreement is probably fortuitous. A magnetic moment of
5.883(3) μB is reported herein for [Co{N(SiMe3)2}2(THF)]
(4). If the 4.83 μB value of ref 19 is corrected for the difference
between the molecular weights of the putative monomer
“Co{N(SiMe3)2}2” (mol wt = 379.70 g/mol) and 4 (mol wt =
451.82 g/mol), a value of 5.75 μB is obtained, which is
consistent with the 5.883(3) μB reported herein. The solution
magnetic moment of 1, 4.7(3) μB, is consistent with the
existence of 1 as dissociated Co{N(SiMe3)2}2 monomers in
solution. The solution magnetic moment is larger than the
3
3d7 cobalt(II) ions with S = /2 with a second-order Zeeman
contribution, Nα, to the molar magnetic susceptibility. Thus,
12157
dx.doi.org/10.1021/ic402019w | Inorg. Chem. 2013, 52, 12152−12160

Inorganic Chemistry
Article
expected spin-only magnetic moment of 3.87 μB, and is equal to
the 4.7 μB of the bent geometry two-coordinate species
the χMT observed for 2−6 between 2 and 300 K and the 5 K
magnetization of 2−4 have been simultaneously fit with the
PHI code,52 with a model that assumes the presence of an axial
component (D) and a nonaxial component (E) to the zero-field
splitting. In these fits, D, E, gz, and gx = gy have been varied and
the best fit of χMT and, when available, the 5 K magnetization
obtained. The results of the 5 K magnetization calculated for
random powder samples are shown in the inset for Figure 5,
which shows good agreement for compounds 2 and 3, but
poorer agreement for compound 4. This poorer agreement for
(4) may be an indication of some nonrandom orientation of
the crystallites in the powder sample used for the magnetic
measurements. The resulting fits are shown as the solid lines in
Figure 5, and the resulting best fit parameters are given in the
lower portion of Table 3. The moments for 5 and 3 were
previously reported at 300 K as 4.65 and 4.69 μB, respectively.
The value for 5 is in excellent agreement with the 4.642(4) μB
reported in Table 3, but the value for 3 is smaller than the
5.269(3) μB reported in Table 3. The earlier measurements
may have included larger errors due to smaller sample sizes.32
As is well-known, the fit of the temperature dependence of
[Co{N(H)ArMe }2] (Ar
= C6H3-2,6(C6H2-2,4,6-Me3)2),28
Me6
6
but is smaller than that of the linearly coordinated species
i
Pri6
28
[Co{N(H)ArPr }2] (Ar = C6H3-2,6(C6H2-2,4,6-Pri6)2).
6
The magnetic properties of [Co{N(SiMe3)2}2(L)] (2−5)
and [Na(12-crown-4)2][Co{N(SiMe3)2}3] (6) have been
measured, after zero-field cooling, upon warming from 2 K to
300 K in an applied magnetic field of 0.01 T; the resulting χMT
values are shown as the points in Figure 5. As would be
3
χMT with the zero-field splitting Hamiltonian for S = /2 is
insensitive to the sign of D. The fits reported herein, when D
was chosen to be positive, gz refined to a smaller value than gx =
gy, a g-tensor anisotropy that is not possible25 for cobalt(II),
which has a 3d7 electronic configuration and thus a negative
spin-orbit coupling parameter.
In view of their similar structures, perhaps the most
unexpected magnetic property observed for 2−4 is the
difference in the magnitude of their χMT values (see Figure
5), differences that are reflected in both the differing D-values
and g-values found in Table 3 and reproduced by separate
sample preparations. If the ordering of the 3d orbitals in the
electrostatic trigonal crystal field, with the z-axis oriented
normal to the trigonal plane and passing through the cobalt(II)
Figure 5. Temperature dependence of χMT obtained at 0.01 T (data
points), and the best fits obtained between 2 K and 300 K for 2−6
3
with S = /2 in the presence of zero-field splitting of the electronic
ground state. Inset shows the 5 K magnetization for 2−4 and its fit
obtained simultaneously with the χMT fit. The resulting best-fit
parameters are given in Table 3.
2
2
2
ion, is 3dxz ≈ 3dyz < 3dz < 3dx −y ≈ 3dxy, then the axial zero-
field splitting parameter, D, is predicted53 to be large and
positive, in disagreement with the observed negative D-values
(see Table 3). In order to achieve a large negative D value
through spin−orbit coupling between the electronic ground
state and a low-lying excited state, a mixing by spin−orbit
coupling54 must take place between orbitals that have the same
ml value. Such a process is impossible with the above D3h
expected for these complexes with no intermolecular exchange,
1/χM is almost linear with temperature and Curie−Weiss law
fits between 50 and 300 K yield the parameters at the top of
Table 3; these are parameters that are indicative of para-
3
magnetic cobalt(II) complexes with S = /2 in the presence of
zero-field splitting. The deviation from Curie−Weiss law
behavior below 50 K is indicative of the presence of zero-
field splitting of the electronic ground state of the cobalt(II)
cations in 2−6. Indeed, very similar magnetic results have been
observed for a tetrahedral complex, {Co(SPh)4}2−, that exhibits
slow magnetic relaxation in a zero applied field.51 As a result,
trigonal symmetry d-orbital ordering. However, if the 3d orbital
47,55
2
2
2
ordering is 3dz < 3dyz < 3dxz < 3dx −y < 3dxy, as is predicted
for the approximately C2 symmetry of compounds 2−4, the
mixing by spin−orbit coupling between the 3d7 cobalt(II)
Table 3. Magnetic Properties of the [Co{N(SiMe3)2}2(L)] Complexes
[Co{N(SiMe3)2}2
(PMe3)] (2)
[Co{N(SiMe3)2}2
[Co{N(SiMe3)2}2
(THF)] (4)
[Co{N(SiMe3)2}2
[Na(12-crown-4)2][Co{N(Si
(py)] (3)
(py)2] (5)
Me3)2}3] (6)
50−300 K, Curie−Weiss Law Fit
θ, deg
C,
emu K/mol
−2.3(4)
2.778(6)
−13.8(3)
3.471(7)
−9.6(2)
4.328(4)
−9.7(2)
2.695(5)
−19.8(4)
4.131(5)
μeff, μB
4.714(5)
2.435(3)
5.269(5)
2.722(3)
5.883(3)
3.038(1)
4.642(4)
2.397(2)
5.748(5)
2.968(3)
3
g for S = /2
2−300 K Fit of χMT
−73(2)
D, cm−1
E, cm−1
gz
−74(2)
9.6(5)
−82(2)
21.0(5)
2.84(1)
2.57(1)
−66(2)
11.0(5)
2.54(1)
2.22(1)
−62(1)
10(1)
14.6(5)
2.82(1)
2.10(1)
3.28(1)
2.75(1)
2.97(1)
gx = gy
2.78(1)
12158
dx.doi.org/10.1021/ic402019w | Inorg. Chem. 2013, 52, 12152−12160

Inorganic Chemistry
Article
22
2
2
1
3dz 3dyz 3dxz13dx −y 13dxy ground-state electronic configura-
transition-metal compounds, using AC magnetic measurements
for the investigation of their potential use as single molecule
magnets.
2
tion and the first excited state 3dz 3dyz13dxz23dx −y 13dxy
22
2
1
2
configuration is allowed because the 3dyz and 3dxz orbitals,
orbitals that have the same ml = 1 values, transform with the
same B irreducible representation. Then, the mixing of the
ground and excited states through the spin−orbit coupling will
stabilize54 the MS = 3/2 state of the cobalt(II) ion in a
compound and lead to a negative D-value. The absolute |D|
value is inversely proportional53 to the energy splitting between
the 3dyz and 3dxz orbitals. Hence, a small (or large) splitting will
lead to a large (or small) absolute |D| value. We propose that
the differences in D-values observed for compounds 2−4
originate in different energy splitting between the 3dyz and 3dxz
orbitals, a splitting that is related to the extent of the structural
and electric distortion from ideal C3 symmetry at the cobalt(II)
ion. In compound 3, where the cobalt(II) ion is bonded to
three nitrogens, the splitting is expected to be smaller than in
compound 2 or 4, where the cobalt(II) ion is bonded to two
nitrogens and one phosphorus or one oxygen, respectively.
Hence, a more negative D-value is expected for compound 3
than for compound 2 or 4, in agreement with observation.
At this point, it seems that full ab initio studies will be
required for a deeper understanding of the mixing by spin−
orbit coupling of the various excited states into the electronic
ground state to yield the observed differences in χMT and,
hence, moments. Furthermore, in view of the interesting
magnetic properties of 2−6 and their similarity to the
dianion{Co(SPh)4}2−,51 AC magnetic susceptibility studies
are underway.
ASSOCIATED CONTENT
* Supporting Information
■
S
1H NMR spectra of 1−4; Evans’ method 1H NMR spectra of 1;
calculations of solution association energy of 1; CIFs of 1, 2,
and 4; IR UV-vis spectra of 1, 2, and 4. Photographs of crystals
of 1−4. This material is available free of charge via the Internet
at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*E-mail addresses: pppower@ucdavis.edu (P.P.P.), glong@mst.
edu (G.J.L.).
Notes
■
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This research was supported by National Science Foundation
(NSF) (through Grant Nos. CHE-1263760 and DBIO
722538). A.M.B. thanks the NSF-Graduate Research Fellow-
ship Program for their financial support (through Grant No.
DGE-1148897), Professors M. M. Olmstead and H. Hope for
their invaluable crystallographic support, and Professor R. A.
Andersen for useful discussion on the magnetism and
properties of 1. We thank a reviewer for drawing our attention
to similarities in the electronic spectral data for 4, also, and
those in ref 19, as well as Jerry Dallas and Ping Yu for help with
the NMR spectroscopic measurements. Compound 6 and the
magnetic data were provided by Chris Melton.
CONCLUSIONS
■
Our detailed study of a representative group of two- and three-
coordinated cobalt(II) compounds, which includes two new
three-coordinated cobalt(II) silylamides, as well as the
synthetically important precursor [Co{N(SiMe3)2}2]2 (1),
provides ample evidence of their interesting magnetic behavior.
We have spectroscopically characterized the silylamide [Co{N-
(SiMe3)2}2]2 species for the first time and determined its
association energy. The magnetic data for [Co{N(SiMe3)2}2]2
(1) show that it displays strong antiferromagnetic coupling and
the moments for the monomeric three-coordinated species 2−
4 and 6 are considerably larger than the spin-only χMT product
of 1.875 emu K mol−1, and a magnetic moment of 3.87 μB.
Moreover, these compounds display substantial negative axial
zero-field splittings (D) that may be associated with substantial
spin reversal barriers.56 However, the significant rhombicity (|
E/D|) observed for these compounds may substantially reduce
the magnetic anisotropy and the barriers expected, based on the
D-values. Although the structural and magnetic data for 1 show
that it is a dimer in the solid phase, the solution 1H NMR data
show that, in ca. 0.12 M benzene solution, it is extensively
dissociated to the monomeric form Co{N(SiMe3)2}2, which is
in agreement with the mass spectral data reported in ref 19.
However, a comparison of the electronic spectra of 1 and its
THF complex [Co{N(SiMe3)2}2(THF)] (4) with those
reported in ref 19 show that the electronic spectrum given in
that paper corresponds closely with those that we have
measured for 4. Moreover, the color of 4 that we observe
(bright green) corresponds to that originally reported in ref 1a
REFERENCES
■
(1) (a) Burger, H.; Wannagat, U. Monatsh. Chem. 1963, 94, 1007−
̈
1012. (b) Burger, H.; Wannagat, U. Monatsh. Chem. 1964, 95, 1099−
̈
1102.
(2) Andersen, R. A.; Faegri, K.; Green, J. C.; Haaland, A.; Lappert, M.
F.; Leung, W.-P.; Rypdal, K. Inorg. Chem. 1988, 27, 1782−1786.
(3) Power, P. P. Chemtracts 1994, 6, 181−195.
(4) Power, P. P. Chem. Rev. 2012, 112, 3482−3507.
(5) Alyea, E. C.; Bradley, D. C.; Copperthwaite, R. G. J. Chem. Soc.,
Dalton Trans. 1972, 1580−1584.
(6) Bradley, D. C.; Copperthwaithe, R. G. Inorg. Synth. 1978, 18,
112−120.
(7) Bradley, D. C.; Copperthwaite, R. G. J. Chem. Soc. D. 1971, 764−
764.
(8) Ellison, J. J.; Power, P. P.; Shoner, S. C. J. Am. Chem. Soc. 1989,
111, 8044−8046.
(9) (a) Bradley, D. C.; Chisholm, M. H. Acc. Chem. Res. 1976, 9,
273−280. (b) Eller, P. G.; Bradley, D. C.; Hursthouse, M. B.; Meek, D.
W. Coord. Chem. Rev. 1977, 24, 1−95. (c) Cummins, C. C. Prog. Inorg.
Chem. 1998, 47, 885−936. (d) Alvarez, S. Coord. Chem. Rev. 1999,
193−5, 13−41.
(10) Fjeldberg, T.; Andersen, R. A. J. Mol. Struct. 1985, 128, 49−57.
(11) Bradley, D. C.; Hursthouse, M. B.; Rodesiler, P. F. J. Chem. Soc.
D. 1969, 14−15.
(12) Ghotra, J. S.; Hursthouse, M. B.; Welch, A. J. J. Chem. Soc.,
Chem. Commun. 1973, 669−670.
(13) Kohn, R. D.; Kociok-Kohn, G.; Haufe, M. Chem. Ber. 1996, 129,
̈
̈
25−27.
(giftgrun) for Co{N(SiMe ) } . Thus, it seems that bis-
̈
3
2 2
(14) Putzer, M. A.; Magull, J.; Goesmann, H.; Neumuller, B.;
̈
(trimethylsilyl)amido cobalt has been misidentified57 from the
beginning.1a Further studies are in progress to examine the
spin-relaxation barriers of these and other low-coordinate
Dehnicke, K. Eur. J. Inorg. Chem. 1996, 129, 1401−1405.
(15) Alyea, E. C.; Basi, J. S.; Bradley, D. C.; Chisholm, M. H. Chem.
Commun. (London). 1968, 495−495.
12159
dx.doi.org/10.1021/ic402019w | Inorg. Chem. 2013, 52, 12152−12160

Inorganic Chemistry
Article
(16) Bradley, D. C.; Copperthwaite, R. G.; Cotton, S. A.; Sales, K. D.;
Gibson, J. F. J. Chem. Soc., Dalton Trans. 1973, 191−194.
(17) Alyea, E. C.; Bradley, D. C.; Copperthwaite, R. G.; Sales, K. D. J.
Chem. Soc., Dalton Trans. 1973, 185−191.
(48) Panda, A.; Stender, M.; Wright, R. J.; Olmstead, M. M.; Klavins,
P.; Power, P. P. Inorg. Chem. 2002, 41, 3909−3916.
(49) (a) Chen, H.; Bartlett, R. A.; Olmstead, M.; Power, P. P.;
Shoner, S. C. J. Am. Chem. Soc. 1990, 112, 1048−1055. (b) Ni, C.;
Fettinger, J. C.; Long, G. J.; Power, P. P. Inorg. Chem. 2009, 48, 2443−
2448. (c) Ni, C.; Stich, T. A.; Long, G. J.; Power, P. P. Chem. Commun.
2010, 4466−4468.
(18) Lappert, M. F.; Pedley, J. B.; Sharp, G. J.; Bradley, D. C. J. Chem.
Soc., Dalton Trans. 1976, 1737−1740.
(19) Bradley, D. C.; Fisher, K. J. J. Am. Chem. Soc. 1971, 93, 2058−
2059.
(20) Olmstead, M. M.; Power, P. P.; Shoner, S. C. Inorg. Chem. 1991,
(50) Hope, H.; Olmstead, M. M.; Murray, B. D.; Power, P. P. J. Am.
Chem. Soc. 1985, 107, 712−713.
(51) Zadrozny, J. M.; Long, J. R. J. Am. Chem. Soc. 2011, 133,
20732−20734.
30, 2547−2551.
(21) Bradley, D. C.; Hursthouse, M. B.; Malik, K. M. A.; Moseler, R.
̈
(52) Chilton, N. F.; Anderson, R. P.; Turner, L. D.; Soncini, A.;
Murray, K. S. J. Comput. Chem. 2013, 34, 1164−1175.
(53) Gomez-Coca, S.; Cremades, E.; Aliaga-Alcade, N.; Ruiz, E. J.
Am. Chem. Soc. 2013, 135, 7010−7018.
Transition Met. Chem. 1978, 3, 253−254.
(22) Murray, B. D.; Power, P. P. Inorg. Chem. 1984, 23, 4584−4588.
(23) (a) Reiff, W. M.; LaPointe, A. M.; Witten, E. H. J. Am. Chem.
Soc. 2004, 126, 10206−10207. (b) Reiff, W. M.; Schulz, C. E.;
Whangbo, M.-H.; Seo, J. I.; Lee, Y. S.; Potratz, G. R.; Spicer, C. W.;
Girolami, G. S. J. Am. Chem. Soc. 2009, 131, 404−405.
(24) (a) Zadrozny, J. M.; Atanasov, M.; Bryan, A. M.; Lin, C.-Y.;
Rekken, B. D.; Power, P. P.; Neese, F.; Long, J. R. Chem. Sci. 2013, 4,
125−138. (b) Zadrozny, J. M.; Xiao, D. J.; Atanasov, M.; Long, G. J.;
Grandjean, F.; Neese, F.; Long, J. R. Nat. Chem. 2013, 5, 577−581.
(25) Kahn, O. Molecular Magnetism; VCH−Wiley: New York, 1993;
p 12.
(54) Ruamps, R.; Batchelor, L.; Maurice, R.; Gogoi, N.; Jimen
́
ez-
Lozano, P.; Guihery, N.; de Graaf, C.; Barra, A.-L.; Sutter, J.-P.; Mallah,
T. Chem.Eur. J. 2013, 19, 950−956.
́
(55) Campanion, A. L.; Komarynsky, M. A. J. Chem. Educ. 1964, 41,
257−262.
(56) Neese, F.; Pantazis, D. A. Faraday Discuss. 2011, 148, 229−238.
(57) The mischaracterization of “Co{N(SiMe3)2}2” as its THF
adduct 4 is partly understandable on the basis of its very high air
sensitivity, and the fact that no elemental analysis was obtained
because of this characteristic.1a Furthermore, the isolation and
characterization of stable THF-metal complexes was not common at
the time the original report was published. In addition, the recording
and processing of X-ray crystallographic data were difficult and time-
consuming tasks in the early 1960s. There is irony in the fact that,
since “Co{N(SiMe3)2}2” was first reported,1a the THF-free complex 1
has been synthesized numerous times and its structure determined in
the vapor phase by ged,2 and in the solid state by X-ray
crystallography.22
(26) Caneschi, A.; Gatteschi, D.; Sessoli, R.; Barra, A. L.; Brunel, L.
C.; Guillot, M. J. Am. Chem. Soc. 1991, 113, 5873−5874.
(27) Ishikawa, N.; Sugita, M.; Ishikawa, T.; Koshihara, S.; Kaizu, Y. J.
Am. Chem. Soc. 2003, 125, 8694−8695.
(28) Bryan, A. M.; Merrill, W. A.; Reiff, W. M.; Fettinger, J. C.;
Power, P. P. Inorg. Chem. 2012, 51, 3366−3373.
(29) Merrill, W. A.; Stich, T. A.; Brynda, M.; Yeagle, G. J.; Fettinger,
J. C.; De Hont, R.; Reiff, W. M.; Schulz, C. E.; Britt, R. D.; Power, P. P.
J. Am. Chem. Soc. 2009, 131, 12693−12702.
(30) Putzer, M. A.; Neumuller, B.; Dehnicke, K.; Magull, J. Chem.
̈
Ber. 1996, 129, 715−719.
NOTE ADDED AFTER ASAP PUBLICATION
■
(31) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.;
Timmens, F. J. Organometallics 1996, 15, 1518−1520.
(32) Panda, A.; Stender, M.; Olmstead, M. M.; Klavins, P.; Power, P.
P. Polyhedron 2003, 22, 67−73.
This paper was published on the Web on October 10, 2013,
with minor text errors throughout the paper. The corrected
version was reposted on October 21, 2013.
(33) Wannagat, U.; Niederprum, H. Chem. Ber. 1961, 94, 1540−
1547.
(34) Amonoo-Neizer, E. H.; Shaw, R. A.; Skovlin, D. O.; Smith, B. C.
Inorg. Synth. 1966, 8, 19−22.
(35) Hope, H. Prog. Inorg. Chem. 1995, 41, 1−19.
(36) Sheldrick, G. M. SADABS, Siemens Area Detector Absorption
Correction; Universitat Gottingen: Gottingen, Germany, 2008.
̈
̈
̈
(37) Sheldrick, G. M. SHELXTL-2013 and SHELXL-2013;
Universitat Gottingen: Gottingen, Germany, 1997.
̈
̈
̈
(38) Dolomanov, O. V.; Bourhis, L. J.; Gildea, R. J.; Howard, J. A. K.;
Puschmann, H. Appl. Crystallogr. 2009, 42, 339−341.
(39) Bain, G. A.; Berry, J. F. J. Chem. Educ. 2008, 85, 532−536.
(40) Horvath, B.; Moseler, R.; Horvath, E. G. Z. Anorg. Allg. Chem.
̈
1979, 450, 165−171.
(41) Bradley, D. C.; Hursthouse, M. B.; Smallwood, R. J.; Welch, A. J.
J. Chem. Soc., Chem. Commun. 1972, 872−873.
(42) (a) Bartlett, R. A.; Power, P. P. J. Am. Chem. Soc. 1987, 109,
7563−7564. (b) Chen, H.; Bartlett, R. A.; Dias, H. V. R.; Olmstead, M.
M.; Power, P. P. J. Am. Chem. Soc. 1989, 111, 4338−4345.
(43) Shannon, R. D. Acta Crystallogr., Sect. A: Cryst. Phys., Diffr.,
Theor. Gen. Crystallogr. 1976, A32, 751−768.
(44) Layfield, R. A.; McDouall, J. J. W.; Scheer, M.; Schwarzmaier,
C.; Tuna, F. Chem. Commun. 2011, 47, 10623−10625.
(45) Ingleson, M. J.; Layfield, R. A. Chem. Commun. 2012, 48, 3579−
3589.
(46) Lin, P. -H.; Smythe, N. C.; Gorelsky, S. I.; Maguire, S.; Henson,
N. J.; Korobkov, I.; Scott, B. L.; Gordon, J. C.; Baker, R. T.; Murugesu,
M. J. Am. Chem. Soc. 2011, 133, 15806−15809.
(47) Andres, H.; Bominaar, E. L.; Smith, J. M.; Eckert, N. A.;
Holland, P. L.; Munck, E. J. Am. Chem. Soc. 2002, 124, 3012−3025.
̈
12160
dx.doi.org/10.1021/ic402019w | Inorg. Chem. 2013, 52, 12152−12160