Metal-metal bonding in low-coordinate dicobalt complexes supported by phosphinoamide ligands

Article abstract of DOI:10.1021/ic3018375

Homobimetallic dicobalt complexes featuring metal centers in different coordination environments have been synthesized, and their multielectron redox chemistry has been investigated. Treatment of CoX₂ with MesNKP ⁱPr₂ leads to self-assembly of [(THF)Co(MesNP ⁱPr₂)₂(μ-X)CoX] [X = Cl (1), I (2)], with one Co center bound to two amide donors and the other bound to two phosphine donors. Upon two-electron reduction, a ligand rearrangement occurs to generate the symmetric species (PMe₃)Co(MesNPⁱPr₂) ₂Co(PMe₃) (3), where each Co has an identical mixed P/N donor set. One-electron oxidation of 3 to generate a mixed valence species promotes a ligand reararrangement back to an asymmetric configuration in [(THF)Co(MesNPⁱPr₂)₂Co(PMe₃)] [PF₆] (4). Complexes 1-4 have been structurally characterized, and their metal-metal interactions are discussed in the context of computational results.

Full text of DOI:10.1021/ic3018375

Article
pubs.acs.org/IC
Metal−Metal Bonding in Low-Coordinate Dicobalt Complexes
Supported by Phosphinoamide Ligands
Ramyaa Mathialagan, Subramaniam Kuppuswamy, Alexandra T. De Denko, Mark W. Bezpalko,
Bruce M. Foxman, and Christine M. Thomas*
†Department of Chemistry, Brandeis University, 415 South Street, Waltham, Massachusetts 02451, United States
S
* Supporting Information
ABSTRACT: Homobimetallic dicobalt complexes featuring metal
centers in different coordination environments have been synthesized,
and their multielectron redox chemistry has been investigated.
Treatment of CoX2 with MesNKPiPr2 leads to self-assembly of
[(THF)Co(MesNPiPr2)2(μ-X)CoX] [X = Cl (1), I (2)], with one Co
center bound to two amide donors and the other bound to two
phosphine donors. Upon two-electron reduction, a ligand rearrange-
ment occurs to generate the symmetric species (PMe3)Co-
(MesNPiPr2)2Co(PMe3) (3), where each Co has an identical mixed
P/N donor set. One-electron oxidation of 3 to generate a mixed valence
species promotes a ligand reararrangement back to an asymmetric
configuration in [(THF)Co(MesNPiPr2)2Co(PMe3)][PF6] (4). Com-
plexes 1−4 have been structurally characterized, and their metal−metal
interactions are discussed in the context of computational results.
Scheme 1
INTRODUCTION
■
Our group has recently been exploring the metal−metal
interactions in heterobimetallic Zr/Co complexes and their
effects on redox properties and reactivity.1 The vastly different
electronic natures of the two metals in these systems leads to
highly polar metal−metal interactions, shifting the CoI/0 and
Co0/−I potentials by ca. 1 V with respect to monometallic Co
analogues and facilitating reactivity with polar bonds in
substrates such as CO2.2,3 We were curious to ascertain
whether the same enhanced redox properties and reactivity
could be imparted by placing the two metals in homobimetallic
dicobalt complexes in disparate coordination environments.
Similarly, we have recently found that treatment of late metal
salts (FeX2, MnX2) with phosphinoamide ligands leads to self-
assembly of homobimetallic species with each metal center in a
different (polyamide vs polyphosphine) coordination environ-
ment.4 Herein, we extend this chemistry to dicobalt systems
and uncover some interesting low-coordinate geometries and
ligand rearrangements in metal−metal bonded species.
substituents (e.g., four distinct isopropyl-methyl resonances)
is indicative of asymmetric complexes (later confirmed by X-ray
diffraction). The solution magnetic moments of complexes 1
and 2 were determined using the Evans method [μeff = 3.0 (1)
and 3.3 μB (2)]5,6 and are lower than would be expected for
two high spin CoII centers (spin only value expected for two
uncoupled S = 3/2 CoII centers = 6.48 μB), implying some
degree of magnetic interaction or bonding between the two Co
centers leading to an intermediate spin state.
RESULTS AND DISCUSSION
The solid state structures of both 1 and 2 were obtained via
X-ray structure analysis of single crystals (Figure 1 and the
Supporting Information). Interestingly, the dicobalt complex
adopts an asymmetric structure in which one CoII center is
coordinated by two amide donors, while the other is bound by
two phosphines. This preference was also observed in several
other phosphinoamide-linked bimetallic complexes that we
■
Synthesis and Characterization. The construction of the
bimetallic framework was achieved via treatment of cobalt
halide salts CoX2 (X = Cl, I) with an equimolar equivalent of
the phosphinoamide potassium salt MesNKPiPr2. In this
manner, the green complexes [(THF)Co(MesNPiPr2)2(μ-
X)CoX] [X = Cl (1), I (2)] were obtained in high yield
(Scheme 1). Complexes 1 and 2 have similar paramagnetically
1
shifted H NMR spectra with 11 distinct resonances. The
Received: August 22, 2012
Published: January 8, 2013
apparent inequivalence of many of the phosphinoamide
© 2013 American Chemical Society
701
dx.doi.org/10.1021/ic3018375 | Inorg. Chem. 2013, 52, 701−706

Inorganic Chemistry
Article
Figure 1. Displacement ellipsoid (50%) representations of 2, 3, and 4. For clarity, only one of two independent molecules in the asymmetric unit of
2 is shown, only one of two disordered PF6− positions in 4 is shown, and hydrogen atoms have been omitted. Relevant interatomic distances (Å) and
angles (°): 2 (parameters for one of two independent molecules are listed): Co1−Co2, 2.5939(4); Co1−N1, 1.9124(18); Co1−N2, 1.9254(19);
Co2−P1, 2.3298(6); Co2−P2, 2.3429(7); Co1−I1, 2.7294(3); Co2−I1, 2.6768(4); Co2−I2, 2.5264(4); Co1−O1, 2.0540(16), Co1−I1−Co2,
57.335(10). 3: Co1−Co1, 2.5536(3); Co1−P1, 2.1763(3); Co1−P2, 2.2699(3); Co1−N1, 1.9569(8); Co1−Co1−P2, 160.245(11). 4: Co1−Co2,
2.4864(6); Co1−N1, 1.889(2); Co1−N2, 1.883(2); Co2−P1, 2.2255(9); Co2−P2, 2.2283(9); Co2−P3, 2.2554(9); Co1−O1, 2.063(2).
have recently reported.4,7 The geometry at each Co center in 2
is distorted pseudotetradrahedral. As a result of the asymmetric
coordination environments of the two Co centers, the iodide
bridges somewhat asymmetrically, with a slightly longer
distance to the amide-bound Co center [2.7294(3) Å vs
2.6768(4) Å]. The short Co−Co distance in 2 is suggestive of a
metal−metal bond [2.5939(4) Å and 2.6142(4) Å in two
independent molecules in the asymmetric unit of 2]. Other
nonorganometallic CoIICoII complexes with similar metal−
metal distances include Co2(NR2)4 [R = SiMe3, 2.583(1) Å; R
= Ph, 2.566(3) Å],8,9 while the 3- and 4-fold symmetric
The cyclic voltammogram of complex 2 revealed a well-defined
quasi-reversible one-electron reduction at −1.2 V, followed by a
series of broad irreversible features at −2.0 and −2.3 V (vs
ferrocene). While the reduction at −1.2 V appears fully
reversible at scan rates greater than 0.4 V/s, the relative
intensity of the return oxidative wave decreases as scan rate is
decreased, and at a scan rate of 0.01 V/s, this reduction wave
appears entirely irreversible (see the Supporting Information).
The quasi-reversible nature of this reductive feature suggests a
substantial chemical change in the composition of complex 2
upon reduction.
To investigate these redox processes further, complex 2 was
treated with excess Na/Hg amalgam in THF. While this
reaction led to an intractable mixture of products, treatment of
2 with Na/Hg in the presence of excess PMe3 resulted in clean
formation of a new red complex, later identified as (PMe3)-
Co(MesNPiPr2)2Co(PMe3) (3), in 63% yield (Scheme 1). The
1H NMR spectrum of 3 remained broad and paramagnetically
shifted, but in this case, only six resonances were observed,
suggesting a more symmetric structure. X-ray diffraction
structure analysis of single crystals of 3 confirmed this
observation and revealed that a ligand rearrangement had
occurred to generate a complex in which both CoI centers are
in identical coordination environments with one amide and one
phosphine donor from the two bridging phosphinoamides
(Figure 1). Each Co center is also coordinated by a PMe3
ligand, and the six-membered Co2N2P2 core is rigorously
planar. On the basis of the isolated product 3, ligand
rearrangement as well as halide loss may be responsible for
the quasi-reversibility of the first reductive feature observed in
the CV of 3. Although it remains unclear how this ligand
rearrangement occurs, it seems reasonable that the preferences
for a reduced CoI center for soft phosphine donors might lead
to this phenomenon. Indeed, the requirement of excess PMe3
to promote clean product formation may imply that PMe3
coordination plays a key role in the ligand rearrangement
process.
+
amidinato- and triazenato-bridged complexes Co2(amidinato)3
[2.885(1) Å],10 Co2(amidinato)4 [2.3735(9) Å],11 and
Co2(triazenato)4 [2.265(2) Å]12 complexes have much longer
and shorter interatomic distances, respectively. Notably, these
complexes were also reported to have room temperature
magnetic moments indicative of low or intermediate spin states
via antiferromagnetic exchange, although no further explanation
of the electronic structure was provided.8−13
The redox behavior of 2 was investigated using cyclic
voltammetry (see Figure 2), revealing multiple reductive events.
While the Co2N2P2 core of the molecule is planar, the
terminal PMe3 ligands lie 0.93 Å out of this plane. Upon closer
inspection, a weak agostic interaction is observed between the
Figure 2. CVs of complexes 2 and 3 (2 mM analyte in 0.4 M
[nBu4N][PF6] in THF, scan rate = 100 mV/s).
702
dx.doi.org/10.1021/ic3018375 | Inorg. Chem. 2013, 52, 701−706

Inorganic Chemistry
Article
Co center and one of the mesityl-methyl hydrogen atoms, and
it is likely this interaction that leads to distortion. The hydrogen
atom was located in the Fourier difference map and refined,
revealing a Co1−H153 distance = 2.327(17) Å and a Co−C−
H angle of 120.8°, both indicative of an agostic interaction.14
Further support for this weak Co−C−H interaction comes
from a medium intensity μ(C−H) stretch at 2714 cm−1 in the
solid state IR spectrum of 3 (see the Supporting Information).
Similar weak agostic interactions have been documented in at
least one other low-coordinate cobalt complex15 and are an
indicator of the coordinative unsaturation of the two Co centers
of 3.
further to 2.4864(6) Å, in line with other nonorganometallic
CoIICoI complexes in the literature: Co2(amidinato)3 [2.385(1)
Å, 2.3201(9) Å],10 [Co2(SAr)5]− [2.511(4) Å],19 and
Co2(μ-tBu2P)2Cl(PMe3)2 [2.508(2) Å].20 As might be expected
based on the magnetic moment of precursor complex 3, the
solution magnetic moment of complex 4 is sufficiently low to
indicate distribution of the metal d electrons throughout a
metal−metal bonding manifold (μeff = 1.4 μB). Furthermore,
the solution EPR spectrum of 4 (77 K, X-band) is indicative of
an S = 1/2 system (see the Supporting Information). Again, it is
difficult to speculate on how the ligand rearrangement occurs
upon oxidation from 3 to 4, but the driving force may be that
the oxidized CoII center would be more likely to prefer π-
donating amide donors than a CoI center. This implies an
assignment of the amide-ligated Co center as CoII in this mixed
valence CoIICoI complex, and such an assignment is consistent
with computational predictions (vide infra).
The metal−metal distance in 3 [2.5536(3) Å] is shorter than
that in complexes 1 and 2, despite the absence of constraints
imposed by a monatomic bridging ligand. The Co−Co distance
is, however, longer than that observed in the limited number of
nonorganometallic CoICoI complexes in the literature,
including Co2(PMe3)4(μ-SPh)2 [2.3997(5) Å],16 the amidinate
To further address the metal−metal interactions in complex
4, UV−vis-NIR data were collected. As shown in Figure 3, the
t
and guanidate dimers Co2((ArN)2CR)2 [R = Bu, 2.1404(10)
Å; R = NCy2, 2.1345(7) Å],17 and (PNP)2Co2 [2.254(1) Å].18
Much like 1 and 2, the solution magnetic moment of complex 3
is also significantly lower than would be expected for two high
spin CoI centers (μeff = 2.9 μB vs the expected value of 5.66 μB
for two noninteracting CoI centers or 4.90 for a delocalized S =
2 system). This implies that the two CoI centers are either
antiferromagnetically coupled or that the electrons on the two
metal centers in this highly symmetric complex are delocalized
throughout a metal−metal bonded d orbital manifold. The
latter explanation is supported by computational results (vide
infra).
The cyclic voltammogram of complex 3 (Figure 2) revealed
multiple irreversible one-electron redox events, including an
oxidation at −1.1 V and a further reduction at −2.7 V (vs Fc/
Fc+). On the basis of the mild oxidation potential of 3, attempts
were made to carefully oxidize this complex to obtain a mixed
valence complex. Treatment of 3 with 1 equiv of [Cp2Fe][PF6]
resulted in a mixed valence complex [(THF)Co(MesNPiPr2)2-
Co(PMe3)][PF6] (4) in which the phosphinoamide had again
unexpectedly rearranged to adopt a structure in which one Co
center was bound to two phosphines and the other was ligated
by two amides (Scheme 2). Notably, treatment of 3 with 2
Figure 3. UV−vis-NIR spectra of 3 and 4 (in THF solution). The
inset shows an expanded view of the near-infrared region of the
spectrum.
Scheme 2
UV−vis-NIR spectrum of 4 has a number of low intensity
transitions in the 650−950 nm range, including a somewhat
broad low intensity band at 949 nm (ε = 140 M−1 cm−1).
Because complex 4 is formally a mixed valence CoIICoI system,
this band could reasonably be assigned as an intervalence
charge transfer (IVCT) band, and the asymmetric coordination
environments of the two cobalt centers might be expected to
increase the energy of this transition into the visible range.
Given the metal−metal bonding and the degree of orbital
mixing in complex 4, however (vide infra), such an
interpretation may be oversimplified in this case. For
comparison, the UV−vis-NIR spectrum of the symmetric
CoICoI complex 3 is also shown in Figure 3. This spectrum is
less complex than that of 4, with a very broad low intensity
feature centered at 1476 nm (ε = 64 M−1 cm−1), which can
likely be attributed to a d−d transition within the metal−metal
bonding orbital manifold.
equiv of [Cp2Fe][PF6] in attempts to generate a halide-free
dicobalt(II) complex led to isolation of the monometallic CoII
complex [Co(MesNPiPr2)(PMe3)3]PF6 (5, see the Supporting
Information).
1
The paramagnetically shifted H NMR of 4 revealed nine
distinct resonances, suggesting a relatively symmetric structure.
The solid state structure of 4 (Figure 1) reveals that PMe3
remains coordinated to the phosphine-bound Co center, while
THF coordinates to the amide-ligated Co ion. In this mixed
valence complex, the Co−Co separation is contracted even
Computational Investigation. To further investigate the
electronic structure and metal−metal bonding in the planar
dicobalt complexes 3 and 4, a computational investigation was
conducted using density functional theory (DFT) methods
703
dx.doi.org/10.1021/ic3018375 | Inorg. Chem. 2013, 52, 701−706

Inorganic Chemistry
Article
Figure 4. Frontier molecular orbital diagram of (A) 3 and (B) 4 [BP86/LANL2TZ(f)/6-311+G(d)/D95V].
[BP86/LANL2TZ(f)/6-311+G(d)/D95 V]. Solution magnetic
moment data for complexes 3 and 4 indicate triplet and doublet
configurations, respectively. Nonetheless, geometry optimiza-
tions were performed on both molecules in all possible spin
states (triplet and quintet for 3; doublet, quartet, and sextet for
4). Comparison of the optimized geometries computed for
these various spin states with the solid state optimized
geometries obtained for these two complexes using X-ray
crystallography reveals that the best prediction is obtained with
the S = 1 (3) and S = 1/2 (4) configurations (see the
Supporting Information).
The frontier molecular orbital diagram of 3 reveals both
metal−metal σ and π interactions, as shown in Figure 4A. As a
result of substantial phosphinoamide ligand contributions, 12
MOs with metal orbital contributions are shown in the figure.
There are three orbitals comprised of metal−metal σ
interactions, including a low-lying metal−metal σ bonding
orbital (−5.06 eV) that is also σ bonding with respect to the
metal−amide bonds, a metal−metal σ bond that is σ* with
respect to the metal−amide bonds (−4.27 eV), and the LUMO
(−1.64 eV), which is σ* with respect to the metal−metal
interaction as well as the metal−ligand interactions. Additional
δ and π interactions between the two Co centers are also
present, with the symmetry partially disrupted by the two
different ligand donors on each Co ion. The LUMO is
sufficiently higher in energy than the remaining occupied
orbitals in the metal−metal bonding manifold, resulting in an
intermediate spin (S = 1) ground state. On the basis of the MO
diagram shown in Figure 4, the Co−Co bond order is
estimated to be ca. 1.
Table 1. Natural Charges and Wiberg Bond Indices (WBIs)
Calculated for 3 and 4 Using NBO Calculations and
Calculated Mayer Bond Orders (MBOs)
Nat charge
Co1
−0.16
0.62 (CoN)
Co2
−0.16
−0.23 (CoP)
Co−Co WBI Co−Co (MBO)
3
4
0.54
0.51
0.98
0.80
to interpret, comparison of the Co−Co bond orders calculated
for these two molecules is a useful exercise and indicates that
the metal−metal bond in 3 is stronger than that in 4.
NBO analysis reveals that the Co−Co σ bond in 3 is covalent
in nature, with 50% orbital contributions from each Co center
(for both α and β NBOs, see the Supporting Information). This
is consistent with the identical coordination environments and
equivalent natural charges of the two metal centers (Table 1).
In contrast, the Co centers in the asymmetric mixed valence
molecule 4 have vastly different natural charges, with the
amide-bound Co center significantly more positively charged
than either the phosphine bound Co atom or the two Co
centers in starting material 3. In this case, there is poorer orbital
overlap between the two different Co centers because of
mismatched orbital energies, as illustrated in the calculated
frontier MO diagram shown in Figure 4B. There are no
apparent δ interactions, and the π interactions are weaker and
more polarized. Nonetheless, the Co−Co σ bond remains
relatively covalent (α NBO: 29% CoN, 71% CoP; β NBO: 75%
CoN, 28% CoP). As a result, the overall Co−Co bond order in 4
remains ∼1, although the bond may be slightly weaker than
that in 3 as a result of mismatched orbital energies.
More information about the bonding between the cobalt
centers was obtained via natural bond orbital (NBO)21 analysis
and Mayer population analysis.22 The Co−Co Wiberg bond
indices calculated for 3 and 4 (0.54 and 0.51, respectively) are
nearly identical (Table 1), while calculations suggest that the
Mayer bond order for 3 is slightly higher than that of 4. While
the absolute values of these computed bond orders are difficult
CONCLUSION
■
In summary, the [MesNPiPr2]− ligand supports low-coordinate
dicobalt dimers with significant metal−metal bonding. The
orientation of the ligands in these complexes is highly
dependent on the overall redox state of the dicobalt unit.
704
dx.doi.org/10.1021/ic3018375 | Inorg. Chem. 2013, 52, 701−706

Inorganic Chemistry
Article
impurities. Upon standing at room temperature, the concentrated
Future studies will focus on the reactivity of the reduced
complex 3 toward small molecule activation.
ether solution of 1 yielded analytically pure 1 as green blocks (0.36 g,
1
95%). H NMR (400 MHz, C6D6): δ 26.3 (6H, iPr-Me), 17.0 (2H,
Mes), 12.3 (2H, Mes), 9.3 (6H, Mes-Me), 6.7 (6H, Mes-Me), 4.3 (6H,
iPr-Me), 2.9 (6H, Mes-Me), −4.2 (6H, iPr-Me), −4.9 (4H, THF),
−8.0 (6H, iPr-Me), −11.4 (4H, THF) (isopropyl-methine proton is
not observed because of its close proximity to the paramagnetic Co
center, tentative assignments based on relative integration). UV−vis
(C6H6) λmax, nm (ε, L mol−1 cm−1): 450 (390), 506 (230), 365 (910),
670 (730). Evans’ method (C6D6): 2.98 μB. Anal. calcd for
C34H58Co2N2P2OCl2: C, 53.62; H, 7.68; N, 3.69. Found: C, 53.53;
H, 7.79; N, 3.75.
EXPERIMENTAL SECTION
■
General Considerations. Unless specified otherwise, all manip-
ulations were performed under an inert atmosphere using standard
Schlenk or glovebox techniques. Glassware was oven-dried before use.
Benzene, pentane, diethyl ether, tetrahydrofuran, and toluene were
dried using a Glass Contours solvent purification system. All solvents
were stored over 3 Å molecular sieves prior to use. Benzene-d6
(Cambridge Isotopes) was degassed via repeated freeze−pump−
thaw cycles and dried over 3 Å molecular sieves. THF-d8 was dried
over CaH2, vacuum-transferred, and degassed via repeated freeze−
pump−thaw cycles. MesNKPiPr2 was synthesized using literature
procedures.2,4 Anhydrous CoCl2 and CoI2 were purchased from Strem
Chemicals and used after 12 h of drying at 100 °C under vacuum.
NMR spectra were recorded at ambient temperature on a Varian Inova
400 MHz instrument. Chemical shifts are reported in δ (ppm). For 1H
and 13C{1H} NMR spectra, the solvent resonance was used as an
internal reference, and for 31P{1H} NMR spectra, 85% H3PO4 was
referenced as an external standard (0 ppm). IR spectra were recorded
on a Varian 640-IR spectrometer controlled by Resolutions Pro
software. UV−vis spectra were recorded on either a Cary 50 UV−vis
or Cary 5000 UV−vis-NIR spectrophotometer using Cary WinUV
software. Elemental analyses were performed at Complete Analysis
Laboratory Inc. (Parsippany, NJ). Solution magnetic moments were
measured using Evans’ method.5,6
X-ray Crystallography. All operations were performed on a
Bruker-Nonius Kappa Apex2 diffractometer, using graphite mono-
chromated Mo Kα radiation. All diffractometer manipulations,
including data collection, integration, scaling, and absorption
corrections, were carried out using the Bruker Apex2 software.23
Preliminary cell constants were obtained from three sets of 12 frames.
Fully labeled diagrams and data collection and refinement details are
included in Table S1 and on pages S18−S31 in the Supporting
Information.
Computational Details. All calculations were performed using
Gaussian09, Revision A.02, for the Linux operating system.24 Density
functional theory calculations were carried out using a combination of
Becke’s 1988 gradient-corrected exchange functional25 and Perdew’s
1986 electron correlation functional26 (BP86). A mixed basis set was
employed, using the LANL2TZ(f) triple-ζ basis set with effective core
potentials for cobalt,27,28 Gaussian09’s internal 6-311+G(d) for
heteroatoms (nitrogen, oxygen, phosphorus), and Gaussian09’s
internal LANL2DZ basis set (equivalent to D95 V29) for carbon
and hydrogen. Using crystallographically determined geometries as a
starting point, the geometries were optimized to a minimum, followed
by analytical frequency calculations to confirm that no imaginary
frequencies were present. Mayer bond analysis was performed with the
routines included in the Gaussian09 software package,22 and Wiberg
bond indices and NBO calculations were carried out using Gaussian
NBO Version 3.1.21
Synthesis of (THF)Co(MesNPiPr2)2(μ-I)CoI (2). A solution of
MesNKPiPr2 (0.85 g, 3.0 mmol) was cooled to −32 °C in THF (15
mL) and this was added to CoI2 (0.94 g, 3.0 mmol) in THF (10 mL)
dropwise over 5 min. The resulting mixture was gradually allowed to
warm to room temperature and continuously stirred for 12 h. The
insoluble materials were removed by filtration through Celite, and all
volatiles were subsequently removed from the filtrate in vacuo. The
resulting green material was extracted with diethyl ether (4 × 5 mL)
and filtered to remove the byproduct, KI, and other insoluble
impurities. Concentration of this diethyl ether solution and storage at
−32 °C afforded analytically pure 2 as green blocks (0.92 g, 66%). 1H
NMR (400 MHz, C6D6): δ 24.5 (6H, iPr-Me), 15.8 (2H, Mes), 11.7
(2H, Mes), 10.0 (6H, Mes-Me), 7.9 (6H, Mes-Me), 5.3 (6H, iPr-Me),
1.9 (6H, Mes-Me), −2.9 (6H, iPr-Me), −5.7 (4H, THF), −10.6 (6H,
iPr-Me), −11.7 (4H, THF) (isopropyl-methine proton is not observed
because of its close proximity to the paramagnetic Co center, tentative
assignments based on relative integration). UV−vis (C6H6) λmax, nm
(ε, L mol−1 cm−1): 364 (440), 608 (640), 688 (860), 746 (610).
Evans’ method (C6D6): 3.29 μB. Anal. calcd for C34H58Co2N2P2I2: C,
43.24; H, 6.19; N, 2.97. Found: C, 43.30; H, 6.29; N, 3.04.
Synthesis of (PMe3)Co(MesNPiPr2)2Co(PMe3) (3). A 0.5% Na/
Hg amalgam was prepared from 0.003 g of Na (0.1 mmol) and 0.6 g of
Hg. To this vigorously stirred amalgam in 10 mL of THF was added a
cold (−32 °C) solution of 2 (0.047 g, 0.050 mmol) in THF (5 mL).
Neat PMe3 (26 μL, 0.20 mmol) was added to the reaction mixture
immediately, and the solution rapidly changed from green to brick red
in color. After it was stirred for 2.5 h, the resulting red solution was
decanted from the amalgam and filtered through Celite. Volatiles were
removed from the filtrate in vacuo. The resulting red material was
extracted with diethyl ether (4 × 2 mL) to remove NaI and other
insoluble impurities. Upon concentration of this diethyl ether solution
of 3 followed by storage at room temperature for 12 h, dark reddish
1
brown single crystals of 3 were obtained (0.024 g, 63%). H NMR
(400 MHz, C6D6): δ 10.6 (6H, Mes-Me), 7.9 (18H, PMe3), 4.2 (12H,
Mes-Me), 0.3 (4H, Mes), −6.2 (12H, iPr-Me), −9.0 (12H, iPr-Me)
(isopropyl-methine proton is not observed because of its close
proximity to the paramagnetic Co center, tentative assignments based
on relative integration). UV−vis (C6H6) λmax, nm (ε, L mol−1 cm−1):
471 (2500), 623 (470), 1476 (64). Evans’ method (C6D6): 2.86 μB.
Anal. calcd for C36H68Co2N2P4: C, 56.10; H, 8.89; N, 3.63. Found: C,
56.11; H, 8.94; N, 3.73.
Synthesis of [(THF)Co(MesNPiPr2)2Co(PMe3)]PF6 (4). A
solution of 3 (0.039 g, 0.050 mmol) in THF (3 mL) was cooled to
−32 °C for 30 min, and this solution was added to a THF (2 mL)
solution of FcPF6 (0.017 g, 0.050 mmol). The reaction progress was
Electrochemistry. CV measurements were carried out in a
glovebox under a dinitrogen atmosphere in a one-compartment cell
using a CH Instruments electrochemical analyzer. A glassy carbon
electrode and platinum wire were used as the working and auxiliary
electrodes, respectively. The reference electrode was Ag/AgNO3 in
THF. Solutions of electrolyte (0.40 M [nBu4N][PF6] in THF) and
analyte (2 mM) were also prepared in the glovebox. All potentials are
reported versus an internal ferrocene/ferrocenium reference.
1
monitored by H NMR spectroscopy, and after 2 h of stirring the
mixture at rt, the starting material had cleanly converted to a new
compound. At this point, the solution was filtered through Celite to
remove insoluble materials. The volatiles were removed from the
filtrate in vacuo, and this material was washed with pentane to remove
ferrocene and other byproducts. The remaining orange material was
redissolved in THF (2 mL), layered with pentane (3 mL), and stored
at room temperature, resulting in orange blocks of 4 along with a small
amount of [Co(PMe3)4]PF6 as a minor byproduct. Complex 4 was
Synthesis of [(THF)Co(MesNPiPr2)2(μ-Cl)CoCl] (1). A solution of
MesNKPiPr2 (0.28 g, 1.0 mmol) in THF (3 mL) was cooled to −32
°C and this was added to CoCl2 (0.130 g, 1.00 mmol) in THF (2 mL)
dropwise over 5 min. The resulting mixture was gradually allowed to
warm to room temperature and continuously stirred for 12 h. The
insoluble materials were removed by filtration through Celite, and all
volatiles were subsequently removed from the filtrate in vacuo. The
resulting green material was extracted with diethyl ether (4 × 2 mL)
and filtered to remove the byproduct, KCl, and other insoluble
1
isolated by manual separation from the mixture (0.039 g, 29%). H
NMR (400 MHz, C6D6): δ 80.5 (4H, THF), 32.3 (Mes-Me), 20.3
(4H, THF), 17.0 (9H, PMe3), 1.2 (12H, Mes-Me), −7.6 (12H, iPr-
Me), −10.0 (12H, very broad, Mes), −17.4 (iPr-Me) (isopropyl-
705
dx.doi.org/10.1021/ic3018375 | Inorg. Chem. 2013, 52, 701−706

Inorganic Chemistry
Article
methine proton is not observed because of its close proximity to the
paramagnetic Co center, tentative assignments based on relative
integration). UV−vis (C6H6) λmax, nm (ε, L mol−1 cm−1): 454 (1300),
682 (150), 728 (140), 949 (140). Evans’ method (C6D6): 1.45 μB.
Complex 4 is thermally unstable in both solution and the solid state
it is likely for this reason that the solution magnetic data are artificially
low and satisfactory combustion analysis data could not be obtained.
Synthesis of [Co(MesNPiPr2)(PMe3)3]PF6 (5). A solution of 3
(0.039 g, 0.050 mmol) in THF (3 mL) was cooled to −32 °C for 30
min, and this was added to a THF (2 mL) solution of FcPF6 (0.033 g,
0.10 mmol). The reaction progress was monitored by 1H NMR
spectroscopy; the starting materials were completely converted to a
new compound after 2 h of stirring the reaction mixture at rt. The
solution was filtered through Celite to remove insoluble materials.
Volatiles were removed from the filtrate, and the orange material was
subsequently washed with pentane to remove ferrocene and other
byproducts. The remaining orange material was redissolved in THF (2
mL), layered with pentane (3 mL), and stored at room temperature,
(10) Cotton, F. A.; Daniels, L. M.; Maloney, D. J.; Matonic, J. H.;
Murillo, C. A. Inorg. Chim. Acta 1997, 256, 283−289.
(11) Cotton, F. A.; Daniels, L. M.; Feng, X.; Maloney, D. J.; Matonic,
J. H.; Murilio, C. A. Inorg. Chim. Acta 1997, 256, 291−301.
(12) Cotton, F. A.; Poli, R. Inorg. Chem. 1987, 26, 3652−3653.
(13) Theopold, K. H.; Silvestre, J.; Byrne, E. K.; Richeson, D. S.
Organometallics 1989, 8, 2001−2009.
(14) Brookhart, M.; Green, M. L. H.; Parkin, G. Proc. Natl. Acad. Sci.
U.S.A. 2007, 104, 6908−6914.
(15) Olmstead, M. M.; Power, P. P.; Sigel, G. Inorg. Chem. 1986, 25,
1027−1033.
(16) Jiao, G.; Li, X.; Sun, H.; Xu, X. J. Organomet. Chem. 2007, 692,
4251−4258.
(17) Jones, C.; Schulten, C.; Rose, R. P.; Stasch, A.; Aldridge, S.;
Woodul, W. D.; Murray, K. S.; Moubaraki, B.; Brynda, M.; La;
Macchia, G.; Gagliardi, L. Angew. Chem., Int. Ed. 2009, 48, 7406−7410.
(18) Fout, A. R.; Basuli, F.; Fan, H.; Tomaszewski, J.; Huffman, J. C.;
Baik, M.-H.; Mindiola, D. J. Angew. Chem., Int. Ed. 2006, 45, 3291−
3295.
1
resulting in orange blocks of 5 (0.038 g, 40%). H NMR (400 MHz,
C6D6): δ 18.9 (2H, Mes), 1.8 (27H, PMe3), 1.3 (3H, Mes-Me), −1.7
(6H, iPr-Me), −2.4 (6H, Mes-Me), −8.0 (6H, iPr-Me) (isopropyl-
methine proton is not observed because of its close proximity to the
paramagnetic Co center, tentative assignments based on relative
integration). UV−vis (C6H6) λmax, nm (ε, L mol−1 cm−1): 450 (1400),
671 (380). Evans’ method (C6D6): 3.53 μB. Anal. calcd for
C37H67Co2N2P5F12: C, 42.70; H, 6.49; N, 2.69. Found: C, 43.23; H,
7.96; N, 1.85.
(19) Ruhlandt-Senge, K.; Power, P. P. J. Chem. Soc., Dalton Trans.
1993, 649−650.
(20) Jones, R. A.; Stuart, A. L.; Atwood, J. L.; Hunter, W. E.; Rogers,
R. D. Organometallics 1982, 1, 1721−1723.
(21) Glendening, E. D.; Reed, A. E.; Carpenter, J. E. NBO Version 3.1.
(22) Mayer, I. Int. J. Quantum Chem. 1986, 29, 477−483.
(23) Apex 2: Version 2 User Manual, M86-E01078; Bruker Analytical
X-ray Systems: Madison, WI, 2006.
(24) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.
P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.;
Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,
K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.;
Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega,
N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.;
Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;
Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.;
Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.;
Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.;
Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09,
revision A.2; Gaussian, Inc.: Wallingford, CT, 2009.
ASSOCIATED CONTENT
■
S
* Supporting Information
Spectroscopic and cyclic voltammetry data, crystallographic and
computational details, and crystallographic data in CIF format.
This material is available free of charge via the Internet at
http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: thomasc@brandeis.edu.
Notes
The authors declare no competing financial interest.
(25) Becke, A. D. Phys. Rev. A 1988, 38, 3098−3100.
(26) Perdew, J. P. Phys. Rev. B 1986, 33, 8822−8824.
ACKNOWLEDGMENTS
■
(27) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299−310.
(28) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 270−283.
(29) Dunning, T. H.; Hay, P. J. In Modern Theoretical Chemistry;
Schaefer, H. F., Ed.; Plenum: New York, 1976; Vol. 3, pp 1−28.
This manuscript is based on work supported by the U.S.
Department of Energy (DE-SC0004019). C.M.T. is grateful for
a 2011 Alfred P. Sloan Fellowship. We thank Jeremy P.
Krogman for assistance with X-ray structure determination and
Dr. Casey R. Wade for assistance with UV−vis-NIR measure-
ments.
REFERENCES
■
(1) Thomas, C. M. Comments Inorg. Chem. 2011, 32, 14−38.
(2) Greenwood, B. P.; Forman, S. I.; Rowe, G. T.; Chen, C.-H.;
Foxman, B. M.; Thomas, C. M. Inorg. Chem. 2009, 48, 6251−6260.
(3) Krogman, J. P.; Foxman, B. M.; Thomas, C. M. J. Am. Chem. Soc.
2011, 133, 14582−14585.
(4) Kuppuswamy, S.; Bezpalko, M. W.; Powers, T. M.; Turnbull, M.
M.; Foxman, B. M.; Thomas, C. M. Inorg. Chem. 2012, 51, 8225−
8240.
(5) Sur, S. K. J. Magn. Reson. 1989, 82, 169−173.
(6) Evans, D. F. J. Chem. Soc. 1959, 2003−2005.
(7) Kuppuswamy, S.; Cooper, B. G.; Bezpalko, M. W.; Foxman, B.
M.; Powers, T. M.; Thomas, C. M. Inorg. Chem. 2012, 51, 1866−1873.
(8) Murray, B. D.; Power, P. P. Inorg. Chem. 1984, 23, 4584−4588.
(9) Hope, H.; Olmstead, M. M.; Murray, B. D.; Power, P. P. J. Am.
Chem. Soc. 1985, 107, 712−713.
706
dx.doi.org/10.1021/ic3018375 | Inorg. Chem. 2013, 52, 701−706