Redox-Active Bis(phenolate) N-Heterocyclic Carbene [OCO] Pincer Ligands Support Cobalt Electron Transfer Series Spanning Four Oxidation States

Article abstract of DOI:10.1021/acs.inorgchem.7b01906

A new family of low-coordinate Co complexes supported by three redox-noninnocent tridentate [OCO] pincer-type bis(phenolate) N-heterocyclic carbene (NHC) ligands are described. Combined experimental and computational data suggest that the charge-neutral four-coordinate complexes are best formulated as Co(II) centers bound to closed-shell [OCO]^2- dianions, of the general formula [(OCO)Co^IIL] (where L is a solvent-derived MeCN or THF). Cyclic voltammograms of the [(OCO)Co^IIL] complexes reveal three oxidations accessible at potentials below 1.2 V vs Fc⁺/Fc, corresponding to generation of formally Co(V) species, but the true physical/spectroscopic oxidation states are much lower. Chemical oxidations afford the mono- and dications of the imidazoline NHC-derived complex, which were examined by computational and magnetic and spectroscopic methods, including single-crystal X-ray diffraction. The metal and ligand oxidation states of the monocationic complex are ambiguous; data are consistent with formulation as either [(^SOCO)Co^III(THF)₂]⁺ containing a closed-shell [^SOCO]^2- diphenolate ligand bound to a S = 1 Co(III) center, or [(^SOCO^?)Co^II(THF)₂]⁺ with a low-spin Co(II) ion ferromagnetically coupled to monoanionic [^SOCO^?]^- containing a single unpaired electron distributed across the [OCO] framework. The dication is best described as [(^SOCO⁰)Co^II(THF)₃]²⁺, with a single unpaired electron localized on the d⁷ Co(II) center and a doubly oxidized, charge-neutral, closed-shell ^SOCO⁰ ligand. The combined data provide for the first time unequivocal and structural evidence for [OCO] ligand redox activity. Notably, varying the degree of unsaturation in the NHC backbone shifts the ligand-based oxidation potentials by up to 400 mV. The possible chemical origins of this unexpected shift, along with the potential utility of the [OCO] pincer ligands for base-metal-mediated organometallic coupling catalysis, are discussed.

Full text of DOI:10.1021/acs.inorgchem.7b01906

Article
pubs.acs.org/IC
Redox-Active Bis(phenolate) N‑Heterocyclic Carbene [OCO] Pincer
Ligands Support Cobalt Electron Transfer Series Spanning Four
Oxidation States
Caleb F. Harris,† Michael B. Bayless,† Nicolaas P. van Leest,‡ Quinton J. Bruch,† Brooke N. Livesay,§
John Bacsa,†,∥ Kenneth I. Hardcastle,∥ Matthew P. Shores,*,§ Bas de Bruin,*,‡ and Jake D. Soper*,†
†School of Chemistry and Biochemistry, Georgia Institute of Technology, Atlanta, Georgia 30332-0400, United States
‡Van ’t Hoff Institute for Molecular Sciences (HIMS), University of Amsterdam (UvA), Science Park 904, 1098 XH Amsterdam, The
Netherlands
§Department of Chemistry, Colorado State University, Fort Collins, Colorado 80523-1872, United States
∥X-ray Crystallography Center, Department of Chemistry, Emory University, 1515 Dickey Drive, Atlanta, Georgia 30322, United
States
S
* Supporting Information
ABSTRACT: A new family of low-coordinate Co complexes
supported by three redox-noninnocent tridentate [OCO] pincer-
type bis(phenolate) N-heterocyclic carbene (NHC) ligands are
described. Combined experimental and computational data suggest
that the charge-neutral four-coordinate complexes are best
formulated as Co(II) centers bound to closed-shell [OCO]2−
dianions, of the general formula [(OCO)CoIIL] (where L is a
solvent-derived MeCN or THF). Cyclic voltammograms of the
[(OCO)CoIIL] complexes reveal three oxidations accessible at
potentials below 1.2 V vs Fc+/Fc, corresponding to generation of
formally Co(V) species, but the true physical/spectroscopic
oxidation states are much lower. Chemical oxidations afford the
mono- and dications of the imidazoline NHC-derived complex, which were examined by computational and magnetic and
spectroscopic methods, including single-crystal X-ray diffraction. The metal and ligand oxidation states of the monocationic
complex are ambiguous; data are consistent with formulation as either [(SOCO)CoIII(THF)2]+ containing a closed-shell
[SOCO]2− diphenolate ligand bound to a S = 1 Co(III) center, or [(SOCO•)CoII(THF)2]+ with a low-spin Co(II) ion
ferromagnetically coupled to monoanionic [SOCO•]− containing a single unpaired electron distributed across the [OCO]
framework. The dication is best described as [(SOCO0)CoII(THF)3]2+, with a single unpaired electron localized on the d7 Co(II)
S
center and a doubly oxidized, charge-neutral, closed-shell OCO0 ligand. The combined data provide for the first time
unequivocal and structural evidence for [OCO] ligand redox activity. Notably, varying the degree of unsaturation in the NHC
backbone shifts the ligand-based oxidation potentials by up to 400 mV. The possible chemical origins of this unexpected shift,
along with the potential utility of the [OCO] pincer ligands for base-metal-mediated organometallic coupling catalysis, are
discussed.
alternatives must address the propensity of later 3d metals to
exist in oxidation states that differ by only one electron.7
One potential solution is to expand the redox sphere beyond
the metal center by harnessing cooperative metal−ligand redox
processes for multielectron chemistry.3,8−12 In this approach,
the capacity of redox-active ligands to store and deliver charge
is a tool to bring about precious metal-like two-electron
organometallic reactivity at metal ions that are more commonly
prone to one-electron transfer or that are even redox inert. This
strategy has recently been applied to a wide array of bond-
INTRODUCTION
■
Cost, toxicity, and sustainability concerns motivate the
continued pursuit of base metals as alternatives to precious-
metal catalysts for small-molecule and fine-chemical trans-
formations, including hydrogenation, C−C and C−X cross-
coupling, and C−H bond functionalization protocols.1−3 The
utility of platinum-group metals for such processes derives from
their capacity to mediate the multielectron organometallic
bond-making and -breaking steps which comprise most
catalytic cycles.4,5 Selectivity in these systems is therefore a
consequence of their thermodynamic bias for concerted two-
electron transfer.4,6 Accordingly, efforts to develop base-metal
Received: July 26, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/acs.inorgchem.7b01906
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
making and -breaking reactions.13−17 However, while redox-
active ligand complexes have found some applications in base-
metal catalysis,18−42 in most cases advancements are still
needed to make these preformed complexes competitive with
catalysts generated in situ from base-metal salts, reductants, and
potential ligand additives.5,43−57 Successes in rational base-
metal catalyst design often begin with robust, tunable ligands.
For instance, the redox-active bis(imino)pyridine [NNN]
pincer ligands58,59 were termed “privileged” because of their
utility in Fe and Co catalysis.60 Accordingly, elaborating a
library of ligands that stabilize low-coordinate, later 3d metals
across two or more formal oxidation states is a first step toward
designed base-metal complexes for sustainable catalysis.
oxidized forms, and the increased electron density facilitates
oxidative addition at the coordinated metal center. Its
significant trans influence also enhances the lability of auxiliary
ligands,64,65 which is particularly beneficial for generating
catalytically relevant species at substitutionally inert metal
centers such as Co(III). Finally, the modular synthesis allows
both of the lateral moieties as well as the medial NHC core to
be easily modified, offering significant steric and electronic
tunability.66,67
This [OCO] ligand has been reported on metals including
Ti, Zr, Hf, V, Mn, Ir, Ni, Pd, Pt, and Al.63,68−72 Most of these
reports formulate the ligand as a bis(phenolate) dianion, but
two previous papers discussed an expanded redox capacity.70,72
Bercaw and co-workers described one- and two-electron
oxidations of an [(OCO)IrIIICl(PCy3)2] complex and sug-
gested the possibility of a ligand-centered radical in the oxidized
materials.72 However, the solution electron paramagnetic
resonance (EPR) data did not permit definitive assignment of
the oxidized species as metal-centered (OCO2−)IrIV/V or ligand
radical complexes (OCO•−)IrIII/IV. Ligand-centered oxidations
were similarly invoked to rationalize the electrochemical and
EPR data for oxidized homoleptic bis(OCO) ligated group 4
metals Ti, Zr, and Hf, but no structural data were reported and
the putative ligand radical complexes were not isolated.70
Reported herein are data for three new cobalt electron
transfer series, each spanning four formal oxidation states.
Structural data are presented that unambiguously show the
OCO pincer ligands are redox centers in the highly oxidized
complexes, and computational data support their assignments
as ligand-centered radicals. Furthermore, variations in the NHC
backbone saturation are shown to modulate the redox potential
of the ligand-centered oxidations by over 400 mV, providing a
convenient handle to tune these ligands for applications in
small-molecule redox catalysis.
Previous results from the Soper laboratory have explored
redox-active ligand aminophenol Co complexes for Negishi-
type cross-couplings of unactivated alkyl halides with alkyl- or
arylzinc halides (Scheme 1).61,62 Both the C−X oxidative
Scheme 1
RESULTS
■
Diphenolate NHC Ligands. The chloride salts of three
tridentate pincer proligands featuring two di-tert-butylphenolate
moieties on a central N,N′-disubstituted imidazole [H3(OCO)-
Cl], imidazoline [H3(SOCO)Cl], or benzimidazole
[H3(PhOCO)Cl] core, were synthesized via slight modifications
of previously reported procedures. The saturated variant
[H3(SOCO)Cl] was prepared in 82% yield from triethyl
orthoformate promoted cyclization of N,N′-bis(2-hydroxy-3,5-
di-tert-butylphenyl)ethylenediamine in the presence of con-
centrated hydrochloric acid (Scheme 3a).68,73 Analogous
addition and C−C reductive elimination steps apparently occur
via Pd-like concerted two-electron steps. However, catalytic
turnover is limited because C−C reductive elimination from the
putative high-valent dialkyl/aryl intermediate cobalt complexes
is inhibited by energetically unfavorable isomerization of the
square-planar bis(amidophenolate) ligand field. This precludes
access to the cis sites required for C−C elimination.
To address these issues, we sought a system that would
capture the redox-active amidophenolate fragments within a
mer pincer-type tridentate ligand scaffold. In this regard, the
previously reported diphenolate imidazolyl carbene appeared
promising.63 As shown in Scheme 2, the [OCO] pincer
contains two di-tert-butylphenolate moieties flanking a central
N-heterocyclic carbene (NHC) core. Beyond constraining the
coordination geometry to open cis coordination cites for
oxidative addition and reductive elimination transformations,
we envisioned the NHC conferring other advantages. The
strongly σ donating carbene makes the ligand less labile in its
Scheme 3
Scheme 2
B
DOI: 10.1021/acs.inorgchem.7b01906
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
reaction of the o-phenylenediamine precursor afforded
[H3(PhOCO)Cl] in 79% yield.74 The unsaturated NHC core,
[H3(OCO)Cl], was prepared according to a previously
reported method by treating the diimine precursor with
paraformaldehyde and TMSCl in ethyl acetate at 70 °C for 6
h (Scheme 3b).63,73
Synthesis and Structures of Reduced [OCO] Cobalt
Complexes. The cobalt complexes 1−3, [(SOCO)-
CoII(MeCN)], [(OCO)CoII(MeCN)], and [(PhOCO)-
CoII(MeCN)], respectively, were prepared by a general method
wherein the corresponding ligands were deprotonated with 3
equiv of NaOMe and subsequently treated with stoichiometric
CoCl2 (Scheme 4). As described below, the Co-pincer
imidazole backbone of 2 inhibit rotation along the C16−C17
bond, making the ligand coplanar with the Co center. In the
solid state, the complex resides on a crystallographic mirror
plane (Figure 1b, bottom). The benzimidazole complex 3
exhibits significant puckering about the Co center, causing both
phenolic oxygen atoms to bend 7.2° above the N3−Co−C1
plane. The benzimidazole backbone is lifted out of plane
(Figure 1c, bottom). The C−C bonding metrics about the
flanking phenolate groups are statistically indistinguishable
within the rings, across all complexes 1−3 (Figure 2),
consistent with formulation of the phenolate substituents as
fully reduced phenoxides,75 from which we conclude that the
NHC ligands are dianionic and the Co centers adopt a formal
+II oxidation state.
Scheme 4
The Co−O bond distances remain constant across the series,
averaging 1.811 0.005 Å, and are in line with Co−Ophenoxide
distances in low-spin square-planar Co(II) centers supported
by salen-type ligands.76−79 A search of the CCDC in June 2017
revealed that all three Co1−C3 bond lengths (1.811(2)−
1.830(8) Å) are substantially shorter than those in previously
reported Co−NHC complexes (1.9135(19)−2.152(6) Å),80−85
likely reflecting both the strong σ-donor ability of the NHC as
well as geometric constraints imposed by the chelating
phenolate arms. Notably, unsaturation in the backbone leads
to a statistically significant elongation of the Co−CNHC
distances along with a contraction of the Co−N bonds to the
MeCN ligands and lengthening of the nitrile CN bonds
(Figure 3). These bonding patterns can be rationalized by
invoking enhanced π back-donation from the Co center to the
saturated carbene in 1 vs 2 and 3, consistent with previous
suggestions that imidazoline-derived NHCs are better π
acceptors than their unsaturated imidazole counterparts.65,86
However, nitriles are poor π acceptors, and similar effects might
be expected only on the basis of the relative σ-donor abilities of
the varying NHCs.
All three complexes are paramagnetic, as evidenced by the
observation of broad, shifted 1H NMR spectra and the solution
magnetic moments (μeff) of 1.88, 1.82, and 1.90 for 1−3 in
THF-d8, respectively. These measured moments are slightly
higher than the spin-only values for an S = 1/2 center, but
square-planar Co(II) species are well-known to have significant
contributions from the angular momentum term, L, typically
leading to higher than spin-only values of μeff.87 The measured
moments match those previously reported for low-spin square-
planar Co(II) complexes.78,88 The observed magnetism is
therefore consistent with a d7 configuration with a single
unpaired electron.
complexes bind an additional solvent-derived ligand upon
workup, resulting in four-coordinate complexes. Samples
isolated from MeCN or THF solutions vary in color from
orange to red, and the isolated yields ranged from 88 to 93%.
Suspending THF-derived samples in MeCN and adding the
minimal amount of toluene required to completely dissolve the
sample afforded analytically pure needle crystals of the
corresponding MeCN complexes (Chart 1), implying that the
THF ligands are labile.
Recrystallization of the complexes from concentrated MeCN
solutions at −25 °C afforded crystals suitable for analysis by X-
ray diffraction. As shown in Figure 1, the gross geometric
features are consistent across the series. In each case, the four-
coordinate Co center has approximate square-planar geometry,
with the diphenolate carbene pincer ligand occupying three
meridional sites and a solvent-derived MeCN ligand bound
trans to the carbene.
Unsaturation in the ligand backbone leads to variations in the
Co−ligand bonding metrics as well as the overall planarity of
the complexes. Selected bond lengths are collected in Figure 2.
The saturated imidazoline backbone of 1 twists the ligand
backbone, as evidenced by the C2−N1−C3−N2 torsion angle
of 5.72(1)°. The phenolate arms of 1 are displaced below the
cobalt, causing the O1−Co−O2 and C3−Co−N3 bite angles
to slightly contract to 170.3(1) and 172.2(1)°, respectively
(Figure 1a, bottom). In contrast, the sp2 carbons in the
Electrochemistry. Cyclic voltammograms of 1−3 were
obtained in MeCN solutions containing 0.1 M tetra-n-
butylammonium hexafluorophosphate ([nBu4N][PF6]) as the
supporting electrolyte. As shown in Figure 4, all three
Chart 1. MeCN Adducts of Complexes 1−3
C
DOI: 10.1021/acs.inorgchem.7b01906
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 1. ORTEP plots of complexes (a) [(SOCO)Co(MeCN)] (1), (b) [(OCO)Co(MeCN)] (2), and (c) [(PhOCO)Co(MeCN)] (3), as viewed
from above (top) and down the Co1−C3 bond (bottom). Ellipsoids are drawn at 50% probability. Hydrogen atoms and noncoordinated acetonitrile
molecules are omitted for clarity.
Figure 3. Bond lengths (Å) for the NHC−Co−NCCH3 fragments in
(a) 1, (b) 2, and (c) 3.
Table 1. Substitutions in the NHC backbone minimally affect
the position of the first oxidation, E1/21, which varies less than
95 mV between the most (3) and the least easily oxidized (2).
2
However, the second and third oxidations, E1/2 and E1/23, are
Figure 2. Schematic of selected bond lengths (Å) for complexes (a) 1,
(b) 2, and (c) 3 drawn to correspond to Figure 1.
strongly influenced by (un)saturation of the NHC backbone.
For example, in the saturated NHC complex 1, the second
complexes show three quasi-reversible 1e oxidations between
oxidation event, E1/22, occurs nearly 300 mV more negative of
the unsaturated imidazole-derived complex 2 and nearly 400
−0.5 and +1.5 V vs Fc+/Fc. The potentials are collected in
D
DOI: 10.1021/acs.inorgchem.7b01906
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
prepared by salt metathesis with NaBPh4 followed by
recrystallization from solvent diffusion of pentane into a
concentrated THF solution at −25 °C. The structure of 1+ is
presented in Figure 5b. For comparison, an X-ray structure of a
Figure 4. Cyclic voltammograms of 1 (blue), 2 (red), and 3 (green) in
MeCN. Conditions: 5 mM 1−3, 0.1 M [Bu4N][PF6], 1.0 mm Pt
electrode, 100 mV s−1 scan rate.
Table 1. Redox Potentials (V vs Fc+/Fc) of 1−3
E1/2 (ΔEp) X/X+ E1/2 (ΔEp) X+/X2+ E1/2 (ΔEp) X2+/X3+
1
2
3
complex
couple
couple
couple
1
2
3
−0.322 (0.086)
−0.252 (0.208)
−0.344 (0.132)
0.299 (0.085)
0.596 (0.138)
0.694 (0.105)
0.773 (0.108)
0.963 (0.169)
1.187 (0.189)
mV below the benzimidazole NHC complex 3. Similar trends
3
are observed in the E1/2 values spanning a range of over 400
Figure 5. ORTEP plots of complexes (a) [(SOCO)Co(THF)] (1),
(b) [(SOCO)Co(THF)2][BPh4] (1+), and (c) [(SOCO)Co(THF)3]-
[PF6]2 (12+). Ellipsoids are drawn at 50% probability. Hydrogen
atoms, counteranions, and noncoordinated solvent molecules are
omitted for clarity.
2
mV, with 1 < 2 < 3. E1/2 and E1/23 represent the formation of
complexes two and three redox levels above Co(II). However,
as discussed below, the relative scarcity and presumed oxidizing
power of molecular Co(IV) and Co(V) complexes led us to
consider alternative formulations containing ligand-centered
radicals. In this context, the observed E1/22 and E1/23 values are
well within the range of reported ligand oxidations in
complexes containing aryl oxide chelates (0.33−1.04 V vs
Fc+/Fc) that generate phenoxyl radicals.75
Synthesis and Structures of the [(OCO)Co(THF)n]Xn‑1
(n = 1−3) Electron Transfer Series. To better understand
the structural properties of the high-valent Co complexes
observed in the cyclic voltammograms, chemical oxidations of 1
were performed in THF. Treating a dark orange, air-sensitive
THF solution of 1 with 1 equiv of AgOTf afforded a dark green
solution from which 1+ was isolated as an air-stable green solid
in 92% yield. A sample suitable for X-ray diffraction was
sample of 1 isolated from THF is shown in Figure 5a. Bond
length data in Figure 6a show that, with the exception of Co−
C1 bond contraction from 1.811(2) to 1.790(4) Å, the
structural perturbations in 1 are minimal on substitution of
MeCN for THF.
The Co center in 1+ is five-coordinate (Figure 5b), with the
pincer ligand occupying three mer sites. Two THF ligands
complete the pseudo-square-pyramidal geometry with a τ value
of 0.13 (cf. τ = 0 for a perfect square pyramid and τ = 1 for a
perfect trigonal bipyramid).89 Oxidation of 1 to 1+ occurs with
contractions of each Co−O and O−CAr bond by ca. 0.01 Å
E
DOI: 10.1021/acs.inorgchem.7b01906
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
upon workup. Crystals of 12+ suitable for single-crystal X-ray
analysis were obtained by diffusion of pentane into a
concentrated THF solution at −25 °C. Three THF molecules
complete the coordination sphere about a pseudo-octahedral
Co center (Figure 5c). Examination of the ligand metrical data
(Figure 6c) shows a clear quinoid-type pattern of four long and
two short C−C bonds in both phenoxide moieties. Addition-
ally, the CAr−O and CAr−N bond lengths are contracted by an
average of 0.044(4) and 0.022(5) Å, respectively, in
comparison to 1, in line with the contractions expected upon
oxidation of the phenoxide arms to phenoxyls. In total, the sum
of the ligand data clearly contrast with those in 1 or 1+ and are
most consistent with formulation of the ligand in 12+ as a
S
charge-neutral doubly oxidized OCO0 ligand. Charge balance
implies that the Co center is in the +II oxidation state;
therefore, the complex is formulated as [(SOCO0)-
CoII(THF)2]2+. The measured THF solution magnetic moment
(μeff) of 2.51 is substantially above that expected for an S = 1/2
ionsuggestive of substantial orbital contributions to the
magnetic momentbut well below the spin-only moment for
three or more unpaired spins. As discussed below, this could
reflect multiple contributors to the ground state of 12+.
However, 12+ is also prone to degradation to undefined Co
species in solution at ambient temperature, thus complicating
magnetic measurements, and hence this piece of magnetic data
should be interpreted with caution.
All attempts to synthesize the 13+ complex observed in the
CV were unsuccessful. The use of strong oxidants such as N(p-
C6H4Br)3 and NO+ salts provided only an intractable mixture
+
of products.
Solid-State Magnetic Properties. Magnetic susceptibility
data were collected for 1 and 1+ in the temperature range 2−
300 K (Figure 7). For 1, the χMT value is 0.55 cm3 K mol−1
Figure 6. Schematic of selected bond lengths (Å) for complexes (a) 1,
(b) 1+, and (c) 12+. Bond length changes greater than 0.01 Å in
comparison to the fully reduced species, 1, are indicated by colored
labels: red indicates bond contraction; blue indicates bond elongation.
each, and the Co−C1 bond to the carbene is lengthened from
1.789(2) Å in 1 to 1.849(3) Å (Figure 6b). All of the C−C
bonds within the phenolate rings are indistinguishable from 1
within the error of the X-ray measurement, but both rings in 1+
display distinctive quinoid-type bond alternations typical of
those expected for phenoxyl radicals,75 suggesting some degree
of ligand oxidation might be operative.
THF solutions of 1+ are paramagnetic with a solution
moment (μeff) of 2.88, consistent with the spin-only value for
an S = 1 center. The sum of these data are therefore consistent
with formulation of 1+ as either [(SOCO)CoIII(THF)2]+
containing a closed-shell [SOCO]2− bis(phenolate) ligand on
an intermediate-spin Co(III) center or [(SOCO•)-
CoII(THF)2]+ with a low-spin Co(II) ion ferromagnetically
coupled to an oxidized monoanionic [SOCO•]− fragment
containing a single unpaired electron distributed across the
[OCO] framework. The X-ray metrical data alone are
insufficient to distinguish these possibilities, but computational
and solid-state magnetism data discussed below are informative
in this regard.
Figure 7. Temperature dependence of magnetic susceptibility for
[(SOCO)Co(THF)] (1) and [(SOCO)Co(THF)2](OTf) (1+),
collected at 5 and 1 kOe, respectively. The best fits to the data were
determined using PHI.90
(μeff = 2.10 μB) at 300 K, slightly higher than the spin-only
value for a low-spin Co(II) species (S = 1/2, 0.375 cm3 K
mol−1, g = 2). When the temperature is decreased to 10 K, the
χMT product decreases monotonically down to 0.50 cm3 K
mol−1 (μeff = 2.00 μB), and upon further cooling to 2 K, χMT
decreases more sharply to 0.45 cm3 K mol−1 (μeff = 1.9 μB). The
susceptibility data were fit successfully to an S = 1/2 model
using PHI;90 the magnetic parameters are shown in Table 2.
The dication 12+ was prepared from 1 by treatment with 2.1
equiv of [N(p-C6H4Br)3][PF6] in THF, giving immediate
conversion to a dark green solution and a green-brown powder
F
DOI: 10.1021/acs.inorgchem.7b01906
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 2. Magnetic Parameters for [(SOCO)Co(THF)] (1) and [(SOCO)Co(THF)2](OTf) (1+)
a
compound
[(sOCO)Co(THF)]
[(sOCO)Co(THF)2](OTf)
g
Θ (cm−1
−0.11
)
D (E) (cm−1
)
J (cm−1
)
TIP (10−6 cm3 mol−1
150
)
R2
2.33
0.9998
model 1
2.56
2.24
2.18
−10.3
−252
600
400
0.9962
0.9976
0.9988
bc
,
model 2
62.1 (20.7)
c
model 3
−40.3 (13.4)
400
a
b
In PHI, temperature-independent paramagnetic susceptibility (TIP) is added to the data but is not a fitted parameter. Note that the signs of
anisotropy parameters are not reliably determined from fits to susceptibility data, but a positive D value is consistent with population of an MS = 0
c
microstate as T approaches zero, as observed in the experimental data. Anisotropy values were adjusted according to a literature procedure to ensure
that |E| ≤ |D|/3.93
The isotropic g value obtained from the fit is similar to that
found via EPR spectroscopy (vide infra) and is reasonable
considering the large intrinsic spin−orbit coupling in previously
characterized low-spin Co(II) species;91,92 free refinement of gx,
gy, and gz do not improve the overall fit quality and do not agree
with EPR simulations (Table S37 in the Supporting
Information). Inclusion of a mean field correction (Θ)
improved the fit, consistent with weak intermolecular
antiferromagnetic coupling, likely from π stacking through
[SOCO] ligands on adjacent species as shown in the crystal-
packing diagram (Figure S9 in the Supporting Information).
The field dependence of magnetization (Figure S10 in the
Supporting Information) shows near-saturation of magnet-
ization at ∼1.1 N μB, consistent with one unpaired electron in
the ground state.
Co(II) center antiferromagnetically coupled to a ligand radical.
To definitively differentiate models 2 and 3, much higher
measuring temperatures and/or high fields would be needed.
However, on comparison to fits to the high-temperature data
(Figure S12), the “intermediate spin” model (2) follows the
linear increase in χMT, whereas the “antiferromagnetic
coupling” model (3) shows slight curvature. Thus, given the
current data, model 2, a highly anisotropic triplet, offers the
best agreement with data. Whether this is localized on the
“Co(III)” ion or is the result of exceptionally strong
ferromagnetic coupling between low-spin Co(II) and a ligand
radical cation cannot be distinguished from the current data.
EPR Spectroscopy. The X-band EPR spectrum of a THF-
derived sample of 1 was obtained in toluene glass at 20 K
(Figure 8). It displays a rhombic signal with well-resolved
The magnetic properties of the mono-oxidized compound 1+
offer several interpretations. This complex displays a room-
temperature magnetic susceptibility value of 1.38 cm3 K mol−1
(μeff = 3.0), consistent with an S = 1 species (1.00 cm3 K mol−1
expected for g = 2). Upon cooling, a gradual linear decrease is
observed: at 50 K, the χMT product is 1.05 cm3 K mol−1 (μeff
=
2.90). A more pronounced decrease is observed upon further
cooling, and at 2 K χMT is 0.08 cm3 K mol−1 (μeff = 0.79). The
near zero magnetization at 2 K increases very slowly as the field
is increased to 5 kOe (Figure S11 in the Supporting
Information) and does not show any tendency toward
saturation, consistent with a relatively well isolated MS = 0
microstate at low temperatures. On the basis of these data,
several spin distribution scenarios are feasible: (1) low-spin
Co(II) antiferromagnetically coupled to a ligand radical, (2)
intermediate-spin Co(III) center or ferromagnetic coupling
between a low-spin Co(II) and a ligand radical, or (3) high-spin
Co(II) antiferromagnetically coupled to a ligand radical. The
first scenario would give a net S = 0 ground state, whereas the
other models would give net S = 1 ground states but with large
anisotropy such that the MS = 0 microstate is populated at
lowest temperatures. The data were fit to several models using
PHI; all fit parameters are collected in Table S38 in the
Supporting Information. Model 1 did not provide reasonable g
or coupling parameters; importantly, curvature in the simulated
χMT values does not agree with the linear decrease observed at
the lowest temperatures probed (Figure S12 in the Supporting
Information). In contrast, models 2 and 3 provide reasonable g
values as well as large zero-field splitting (D) values. The
presence of significant axial anisotropy is consistent with the
shape of the downturn at low temperatures. Model 3 also gave a
large negative J value, consistent with strong antiferromagnetic
coupling between a putative high-spin Co(II) center and a
ligand radical. On the basis of these fits, it is likely that 1+
contains an intermediate-spin Co(III) center or a high-spin
Figure 8. X-band EPR spectrum of a THF-derived sample of 1 in
toluene glass at 20 K. Conditions: microwave frequency, 9.388589
GHz; power, 0.6325 mW; modulation amplitude, 2.000 G.
cobalt hyperfine coupling interactions (HFIs) along two
directions of the g tensor. The spectrum is characteristic for a
metalloradical S = 1/2 system, with the unpaired electron being
localized mainly at the cobalt center. A satisfactory simulation
of the experimental spectrum was obtained with the EPR
parameters shown in Table 3. Spin-unrestricted (collinear)
spin−orbit DFT calculated EPR parameters are overall in
reasonable agreement with the experimental values (Table 3).
The DFT computed ACo and ACo HFI values are somewhat
x
y
too high though, while the gz value is underestimated. This is
G
DOI: 10.1021/acs.inorgchem.7b01906
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
spin density is delocalized over both phenoxide arms of the
OCO ligand, mainly on the oxygen atoms (total of 12.6%). A
small amount of spin-down density is observed on the NHC
carbon center (4.2%), which can be attributed to spin
polarization. The computed structure of 1 is therefore best
described as [(SOCO)CoII(THF)], with a closed-shell
[SOCO]2− dianion on a low-spin Co(II) center.
Table 3. Experimental (Simulation) and DFT Calculated
EPR Parameters of 1
a
sim
DFT
g Tensor
gx
gy
gz
2.061
2.056
2.102
2.302
2.105
1+ converged as a triplet (⟨s2⟩ = 2.0156). The involvement of
a quintet state (⟨s2⟩ = 6.0132) is highly unlikely, as this state is
computed to be +20 kcal mol−1 uphill vs the triplet state.
However, the singlet state is found to be only +5 kcal mol−1
above the triplet state, and the open-shell singlet state is only
+4 kcal mol−1 uphill from the triplet state. In the triplet state,
66% of the electron density is localized at Co; the remainder is
delocalized over the phenoxide moieties (Figure 10). These
2.500
Hyperfine Coupling (MHz)
ACo
ACo
ACo
45
+145
+82
x
y
z
NR (<30)
170
−176
a
ADF, B3LYP, TZ2P, unrestricted collinear spin−orbit calculations.
suggestive of a slightly overestimated DFT HOMO−SOMO
energy gap, perhaps caused by solvation effects and/or concrete
toluene solvent adduct formation at 20 K in the experimental
system.94
As can be expected for a metal-based electronic triplet system
(S = 1, with expected large zero-field splitting parameter D), the
X-band EPR spectrum of 1+ measured in toluene glass at 20 K
gave no signal in the range of 0−6000 G. The solution
instability of 12+ precluded measurement of an EPR spectrum.
Computed Electronic Structures of 1, 1+, and 12+. The
electronic structures of the THF adducts 1, 1+, and 12+ were
computed with unrestricted DFT-D3 calculations (BP86, def2-
TZVP, disp3), excluding counterions, and with starting spin
states as determined experimentally. Optimized geometries
were compared to the single-crystal X-ray structures and found
to have maximum bond length deviations of 0.023, 0.023, and
0.026 Å for 1, 1+ and 12+, respectively, suggesting that the spin
,
states, functional, and basis set (BP86, def2-TZVP, disp3) used
in the geometry optimizations accurately capture the bond
distances in the [SOCO] ligands across all three formal
oxidation states. The spin densities per atom in optimized
geometries were also calculated. Accordingly, complex 1
converged as a doublet (⟨s2⟩ = 0.7611), with 81% of the total
spin density being located at cobalt (Figure 9). Some residual
Figure 10. (top) Spin density plot of 1+ (S = 1), generated with
IQMol (isosurface value 0.005). (bottom) Spin density per atom for
the bis-THF adduct of 1+ (S = 1).
data are most consistent with description of 1+ as containing a
Co(II) center supported by a monooxidized [SOCO•]− ligand
radical. However, the accessibility of multiple low-lying excited
states suggests that other contributors to the ground state
might need to be considered (perhaps requiring multireference
calculations, beyond the scope of the present study).
Complex 12+ converged as a doublet (⟨s2⟩ = 0.8342). The
spin density of 12+ is almost entirely localized at the cobalt
center, with some spin polarization causing some negative spin
population on the NHC ligand atoms (and hence more than
100% positive spin at cobalt; Figure 11). Unrestricted natural
and corresponding orbital analysis of the α and β molecular
orbitals revealed the presence of only one α MO without
(large) overlap with a corresponding β MO (UCO overlap
values of all other α and β MO pairs >0.96). Accordingly, the
computational data suggest that the complex is best described
as [(SOCO0)CoII(THF)3]2+, with a single unpaired electron
localized on the d7 Co(II) center and a doubly oxidized, charge-
S
neutral, and closed-shell diamagnetic OCO0 ligand (Figure
Figure 9. (top) Spin density plot of 1 (S = 1/2), generated with
IQMol (isosurface value 0.005). (bottom) Spin density per atom for
the THF adduct of 1 (S = 1/2).
12a). While the OCO0 ligand could in principle be described
as a singlet diradical with very strong antiferromagnetic
S
H
DOI: 10.1021/acs.inorgchem.7b01906
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
commonly supported by tetradentate macrocycles, such as
porphyrins or bis(Schiff base) bis(phenolate) N2O2
salens.76−78,96,97 However, square-planar Co(II) complexes of
pincer [PNP] amidodiphosphine ligands have been recently
described,88 and the electronic structures of high-spin square-
planar Co(II) have been discussed.98 Here, the reluctance of
1−3 to bind additional ligands is not attributable to sterics, as
evidenced by the quasi-octahedral structure of [(SOCO0)-
CoII(THF)3]2+, which is formulated analogously as a low-spin
cobalt(II) ion. Rather, the preference for four-coordination in
the lower oxidation states is apparently electronic in origin. In
2
2
the absence of significant π bonding, the dx −y orbital that is
directed at the vacant coordination siteswhen the z axis is
colinear with the Co−CNHC bondis half-filled in the low-spin
d7 configuration. Oxidizing the ligand does not change the
electron count at Co; therefore, the preference for additional
THF ligands in 12+ reflects increased Coulombic attraction to
the dication (Figure S13 and Table S38 in the Supporting
Information). This also accounts for the nearly invariant Co−O
distances upon oxidation of 1 (Figure 6); increased charge
counterbalances the decreased O-donor strength/basicity in the
oxidized ligand.
Figure 11. (top) Spin density plot of 12+ (S = 1/2), generated with
IQMol (isosurface value 0.005). (bottom) Spin density per atom for
the tris-THF adduct of 12+ (S = 1/2).
Isolation of 12+ provides clear structural evidence for redox
activity within the [OCO] pincer ligand. Although such ligand
oxidations had been previously suggested,70,72 this is to our
knowledge the first isolated [OCO] pincer complex with an
oxidized ligand. X-ray data for the doubly oxidized ligand are
consistent with those expected for a diradical, but calculations
show it is better formulated as a closed-shell SOCO0 diamagnet,
implying the NHC mediates electronic communication across
the [OCO] framework.
The electronic structure of the intermediate oxidation state
1+ species defies simple assignment. As noted above, the solid-
state structural and magnetic data can be reasonably interpreted
as either [(SOCO)CoIII(THF)2]+ with a closed-shell [SOCO]2−
bis(phenolate) ligand on an intermediate-spin cobalt(III)
center, or [(SOCO•)CoII(THF)2]+ with a low-spin Co(II)
ion ferromagnetically coupled to a monoanionic [SOCO•]−
ligand radical, containing a single unpaired electron distributed
over the [OCO] π system. DFT calculations are more
consistent with the latter formulation but reveal closed-shell
singlet and open-shell singlet states at thermally accessible
energies relative to the triplet (S = 1) ground state. In reality,
the ground state of 1+ might have multiple contributors. If the
true electronic structure is indeed some combination of the
formulations listed above, then the physical oxidation states of
the metal and ligand should be highly susceptible to small
perturbations in the solvent, temperature, and coordination
environment, making the ligand truly noninnocent.99
Effects of NHC Unsaturation on Ligand-Centered
Oxidations. The oxidized variants of 2 and 3 were not
isolated, but given the structural homology in 1−3, it seems
reasonable to assume that 22+ and 32+ also contain ligand
radicals. So how do we rationalize the large (up to 400 mV)
anodic shifts in the ligand-centered oxidations upon incorpo-
ration of unsaturation into the NHC ligand backbone? Two
potential explanations were considered.
Figure 12. Key resonance structures of the closed-shell (a) and open-
shell singlet (b) configurations of the doubly oxidized, charge-neutral
SOCO0 ligand. The closed-shell representation (a) prevails according
to DFT.
coupling between the two conjugated ligand-radical electrons
(Figure 12b), there are actually no indications for such a
(broken-symmetry) electronic structure with multiple unpaired
electrons. Complex 12+ contains only one unpaired electron
according to DFT, which is located at cobalt.
DISCUSSION
■
Physical Oxidation States in the (OCO)Co Electron
Transfer Series. The bis(phenolate) NHC cobalt complexes
reported herein span a remarkably broad range of redox states.
Cyclic voltammograms of complexes 1−3 all show three quasi-
reversible one-electron oxidations at <1.2 V vs Fc+/Fc. These
formally correspond to the generation of Co(III), Co(IV), and
Co(V) in the mono-, di-, and trications, respectively. However,
the combined experimental and computational data suggest
that the physical oxidation states of cobalt are significantly
lower.95
Complex 1 is appropriately described as [(SOCO)-
CoII(THF)], with a closed-shell [SOCO]2− dianion supporting
a low-spin Co(II) center, and the unsaturated NHC variants 2
and 3 are formulated similarly. Exclusively four-coordinate
quasi-square-planar structures were obtained for crystals of 1
grown from either THF or MeCN and for 2 and 3 grown from
MeCN. Four-coordinate cobalt(II) complexes are most
(1) Wolczanski and co-workers have recently suggested that
the relative accessibility of ligand-centered oxidations can be
understood by simple π-electron counting and the application
of Huckel’s rule.100,101 By this logic, the [SOCO]2− ligand is a
̈
20e− π system that becomes aromatic upon 2e− oxidation to
SOCO0 (see Figure 12a). The unsaturated [OCO]2− and
I
DOI: 10.1021/acs.inorgchem.7b01906
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
[PhOCO]2− ligands have 22 and 26 π electrons, respectively,
making them aromatic in their reduced forms; they should have
a much lower propensity to be oxidized, which is consistent
with the CV data. This argument implies that the unsaturated
backbones in the imidazole- and benzimidazole-derived ligands
are capable of delocalizing their π electrons across the entire
[OCO] ligand. Previous literature has suggested that
unsaturated imidazole-derived NHCs may102 or may not103
be aromatic. If the backbone π electrons are electronically
decoupled from the rest of the [OCO] framework, then this
rationale fails. The DFT results strongly point to full
delocalization of the π system, in favor of aromaticity of the
SOCO0 ligand in 12+.
(2) As detailed above, the varying Co−C bond distance
across 1−3 might be ascribed to enhanced π back-donation
from the cobalt(II) center in 1 to the NHC in comparison to 2
and 3. The saturated imidazoline-derived NHC is known to be
a better π acceptor than the unsaturated (benz)imidazole
NHCs.65,86 Enhanced back-bonding increases π electron
density in the saturated [SOCO]2− ligand, thereby increasing
its susceptibility to oxidations that generate phenoxyl radicals
by removal of one or more π electrons.
factors that make this system particularly attractive as a
platform for new reaction chemistry include the capacity to
stabilize Co over three or more formal oxidation states at
relatively modest potentials, along with an impressive flexibility
in coordination number and geometry. All of these are
predicated on the capacity of the [OCO] ligand to persist in
oxidation states that span multiple electrons. Although this
redox capacity had been previously inferred, this report
describes, to our knowledge, the first structurally characterized
[OCO] ligand radicals and provides unequivocal proof for the
redox activity of this ligand scaffold.
The [OCO] pincer ligands additionally show a remarkable
and unexpected electronic tunability, with ligand-centered
oxidations spanning over 400 mV depending on the degree
of unsaturation in the NHC ligand backbone. Going forward,
we posit that these properties will be broadly beneficial for the
development of new methods for selective, organometallic-type
small-molecule redox transformations at later 3d metals. The
first efforts in our laboratory have focused on Co-mediated C−
C coupling, but extensions to other classes of reactions, and
other base metals, will be described in subsequent reports.
A definitive explanation for the observed shifts in the ligand-
centered oxidation potentials eludes us. It might in fact arise
from a combination of the factors listed above. In contrast, the
relative invariance in the first oxidation events across 1−3 is
reasonably attributed to significant Co(III) character in 1+−3+.
From the perspective of the cobalt(II) center, the coordination
environment is nearly uniform in all of 1−3, rendering a metal-
centered oxidation relatively invariant to unsaturation or
substitutions in the distal half of the NHC fragment.
EXPERIMENTAL DETAILS
■
General Considerations. Unless otherwise specified, all manip-
ulations were performed under anaerobic conditions using standard
vacuum line techniques or in an inert-atmosphere glovebox under
purified nitrogen. Routine NMR spectra were acquired on either a
1
Varian Mercury 300 spectrometer (300.323 MHz for H; 75.5 MHz
for 13C) or a Varian Mercury 400 spectrometer (399.94 MHz for H;
1
101.1 MHz for 13C). All chemical shifts are reported in parts per
million (ppm) relative to TMS, with the residual solvent peak serving
as an internal reference.104 Solution magnetic moments were obtained
by the Evans NMR method105,106 and are reported as the average of
three independent measurements, unless otherwise specified. UV−
visible absorption spectra were acquired using a Varian Cary 50
spectrophotometer. Unless otherwise noted, all electronic absorption
spectra were recorded at ambient temperatures in 1 cm quartz cells. IR
absorption spectra were obtained via attenuated total reflection (ATR)
with a diamond plate using a Bruker ALPHA Fourier-transform
infrared spectrophotometer. All mass spectra were recorded in the
Georgia Institute of Technology Bioanalytical Mass Spectrometry
Facility. Electrospray ionization mass spectrometry (ESI-MS) was
carried out with acetonitrile solutions using a Micromass Quattro LC
spectrometer. Electron impact mass spectra (EI-MS) were obtained
using a VG Instruments Model 70-SE spectrometer. The EPR
spectrum was recorded on a Bruker EMX X-band spectrometer
equipped with a He cryostat. The spectrum was analyzed and
simulated using the W95EPR program of Prof. F. Neese. Cyclic
voltammetry experiments were performed inside an N2-filled glovebox
in MeCN with 0.1 M [nBu4N][PF6] as the supporting electrolyte,
unless otherwise noted. The voltammograms were recorded with a CH
Instruments 620C potentiostat, using a 2.5 mm (o.d.) 1.0 mm (i.d.)
Pt-disk working electrode, Ag-wire quasi-reference electrode, and a Pt-
wire auxiliary electrode, at a scan rate of 0.1 V s−1, unless reported
otherwise. Reported potentials are referenced to the ferrocenium/
ferrocene (Fc+/Fc) redox couple, added as an internal standard at the
conclusion of each experiment. Elemental analyses were performed by
Atlantic Microlab, Inc., Norcross, GA. All analyses were performed in
duplicate, and the reported compositions are the average of two runs.
Full details of X-ray data collection and refinement are provided in the
Supporting Information.
[OCO] as a Platform for Multielectron Transforma-
tions at Co. The sum of the structural and electronic data
presented above presents an exciting opportunity to develop
new small-molecule reaction chemistry at [OCO] complexes.
For instance, a list of design criteria for a new Co cross-
coupling catalyst might include a low-coordinate complex in a
low formal oxidation state, with a multielectron capacity that
favors higher coordination numbers in higher oxidation states.
In this respect, the [OCO] ligand seems ideally suited for such
organometallic reactivity at later 3d metals. As shown herein, it
stabilizes low-coordinate Co in low formal oxidation states, and
higher formal oxidation states lead to higher coordination
complexes. The (OCO)Co core is robust, but labile ancillary
ligands afford multiple cis coordination sites for small-molecule
binding and activation. Most importantly, the [OCO] frame-
work gives access to oxidation states three levels above Co(II)
at modest potentials. Accordingly, oxidative addition to Co(II)
can be envisioned to occur without formation of high-energy
Co(IV) species. Furthermore, unsaturation in the NHC
backbone is shown to afford a high degree of electronic
tunability, allowing the reducing power of the metal complexes
to be tuned by up to 400 mV without significantly perturbing
the coordination environment or sterics about the Co centers.
These properties should be broadly transferrable to other
organometallic catalysis cycles that utilize precious metals for
multielectron bond activation and functionalization.
Methods and Materials. Anhydrous acetonitrile (MeCN),
dichloromethane, pentane, tetrahydrofuran (THF), and toluene
solvents for air- and moisture-sensitive manipulations were purchased
from Sigma-Aldrich and further dried by passage through columns of
activated alumina, degassed by at least three freeze−pump−thaw
cycles, and stored under N2 prior to use. Anhydrous methanol
CONCLUSIONS
■
Tridentate, mer pincer-type ligands are now ubiquitous in
catalysis, and the [OCO] pincer appears to be well positioned
to take a place alongside “privileged” ligands for base-metal
catalysis, such as bis(imino)pyridine [NNN] pincers.60 The
J
DOI: 10.1021/acs.inorgchem.7b01906
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(Drisolv) and benzene (Drisolv) were purchased from EMD Millapore
and used as received. Deuterated solvents were purchased from
Cambridge Isotope Laboratories. Acetonitrile-d3 was dried over excess
calcium hydride, vacuum-distilled to an oven-dried sealable flask, and
degassed by successive freeze−pump−thaw cycles. Tetrahydrofuran-d8
was used as received. The ligand precursors N,N′-bis(2-hydroxy-3,5-di-
tert-butylphenyl)ethylenediamine,73 N,N′-bis(2-hydroxy-3,5-di-tert-
butylphenyl)phenylenediamine,74 and the ligand H3[(OCO)]Cl63
were prepared according to published procedures. [N(p-C6H4Br)3]-
[PF6] was synthesized following a literature procedure.107 Sodium tert-
butoxide (5.4 M in MeOH; Alfa Aesar), triethyl orthoformate (Alfa
Aesar), paraformaldehyde (Alfa Aesar), formic acid (Alfa Aesar), and
anhydrous CoCl2 (Strem) were used as received.
Synthesis of [(OCO)Co(MeCN)] (2). The above representative
procedure, followed by recrystallization from MeCN−toluene at −25
°C, afforded 2 as X-ray-quality crystals in 93% yield. Anal. Calcd for
C33H45CoN3O2: C, 68.97; H, 7.89; N, 7.31. Found: C, 68.54; H, 7.78;
N, 7.28. UV−vis (MeCN; λmax, nm (ε, M−1 cm−1)): 308 (23,000), 327
(2200), 418 (5100). FTIR (ATR, cm−1): 2953(m), 2907(m),
2873(m), 1444(s), 1328(m), 1323(s), 1274(w), 1239(m), 1204(w),
1096(m), 914(w), 843(s), 774(m), 695(m), 663(s), 605(m), 587(m),
577(m), 550(m), 507(m), 422(m).
Synthesis of [(PhOCO)Co(MeCN)] (3). The above representative
procedure, followed by recrystallization from MeCN−toluene at −25
°C, afforded 3 as single crystals suitable for XRD in 89% yield. Anal.
Calcd for C37H47CoN3O2: C, 71.14; H, 7.58; N, 6.73. Found: C,
71.41; H, 7.51; N, 6.77. ESI-MS (m/z): 655.4 [M]+. UV−vis (MeCN;
λmax, nm (ε, M−1 cm−1)): 305 (19000), 427 (4900). FTIR (ATR,
cm−1): 2952(m), 2903(m), 2865(m), 1476(m), 1459(m), 1427(m),
1377(m), 1356(s), 1335(m), 1301(m), 1218(m), 1199(m), 1179(w),
1158(w), 921(w), 860(s), 770(m), 737(s), 719(m), 632(m), 553(m),
436(m), 421(m).
Synthesis of [H3(SOCO)]Cl. In a modified literature procedure,68
N,N′-bis(2-hydroxy-3,5-di-tert-butylphenyl)ethylenediamine (5.65 g,
12.1 mmol) was dissolved in (EtO)3CH (250 mL) and heated to 70
°C in air. Concentrated HCl (1.33 mL, 16.1 mmol) was added
dropwise over 5 min, resulting in a color change from golden yellow to
red-purple. After 15 min, a thick white precipitate formed and the
stirring was increased to ensure even mixing. After it was stirred for 1
h, the reaction mixture was cooled to room temperature and filtered.
The white solid was washed with copious Et2O (150 mL) and dried
under high vacuum at 60 °C overnight (4.93 g, 9.56 mmol, 79%). All
characterization data matched those previously reported.68
Synthesis of [(SOCO)Co(THF)2]+ (1+). A 20 mL scintillation vial
was charged with 1 (607.8 mg, 1.0 mmol) and AgOTf (256.9 mg, 1.0
mmol), and these were dissolved in THF (15 mL). The resulting
green suspension was stirred for 0.5 h and then filtered through a 2.0
mm pad of Celite. The solution was concentrated to ca. 5 mL, layered
with pentane, and stored at −25 °C for 18 h, producing green crystals
of [(SOCO)Co(THF)2][OTf] (762.6 mg, 0.92 mmol, 92% yield).
Single crystals suitable for X-ray analysis were generated from the salt
metathesis of [(SOCO)Co(THF)2][OTf] and Na[BPh4] in THF.
Subsequent layering of pentane over the concentrated THF solution
and storage at −25 °C for 18 h afforded green block X-ray-quality
crystals of [(SOCO)Co(THF)2][BPh4]. Anal. Calcd for
C40H60CoF3N2O7S: C, 57.96; H, 7.30; N, 3.38. Found: C, 57.77; H,
7.28; N, 3.28. UV−vis (THF; λmax, nm (ε, M−1 cm−1)): 434 (3100),
747 (2700), 878 (2000). FTIR (ATR, cm−1): 3054(m), 2953(m),
2902(m), 2865(m), 1588(m), 1556(m), 1478(m), 1440(m),
1388(m), 1359(m), 1311(m), 1241(m), 1200(m), 1124(m),
1071(m), 1026(m), 913(m), 843(m), 733(s), 700(s), 647(w),
625(m), 610(w), 580(w), 543(w), 518(w), 498(w), 471(w),
446(w), 400(w).
Synthesis of [H3(PhOCO)]Cl. In a modified literature method,74
N,N′-bis(2-hydroxy-3,5-di-tert-butylphenyl)phenylenediamine (3.52 g,
6.81 mmol) was dissolved in (EtO)3CH (150 mL) and heated to 70
°C in air. Concentrated HCl (0.84 mL, 10.2 mmol) was added
dropwise over ∼5 min, resulting in a color change from golden yellow
to red-purple. After 15 min, a thick white precipitate formed and the
stirring was increased to ensure even mixing. After it was stirred for 1
h, the reaction mixture was cooled to room temperature and filtered.
The white solid was washed with copious Et2O (150 mL) and dried
under high vacuum at 60 °C overnight (3.14 g, 5.58 mmol, 82%). All
characterization data matched those previously reported.74
Synthesis of 1−3. The same general procedure was followed for
all of 1−3. A 20 mL scintillation vial was charged with a solution of
[H3(SOCO)]Cl (515 mg, 1.0 mmol) in MeOH (8 mL). NaOMe (5.4
M in MeOH, 555 μL, 3.0 mmol) was added dropwise. The resulting
yellow solution was stirred for 5 min and then added dropwise to a
separate 20 mL vial containing anhydrous CoCl2 (129.8 mg, 1.0
mmol) in MeOH (10 mL). The solution became dark red, and a red-
orange precipitate formed within 0.5 h. The suspension was stirred for
an additional 6 h, and the solids were then separated by vacuum
filtration.
Synthesis of [(SOCO)Co(THF)3]2+ (12+). A 20 mL scintillation vial
was charged with 1 (54.7 mg, 0.090 mmol) in CH2Cl2 (3 mL),
affording a red-orange solution. Dropwise addition of a dark blue
solution of [N(p-C6H4Br)3][PF6] (118 mg, 0.188 mmol) in CH2Cl2
(5 mL) gave a green-brown suspension, which was stirred for 30 min
at room temperature, filtered through a pad of Celite, and washed with
THF (5 mL). The filtrate was concentrated in vacuo to ca. 2 mL and
layered with pentane. Standing at −20 °C overnight deposited dark
green crystals of [(SOCO)Co(THF)3][PF6]2 suitable for X-ray
analysis (561 mg, 0.073 mmol, 81%). The trans-disposed THF ligands
can be removed under high vacuum. Because samples for elemental
analysis were prepared by thorough drying in vacuo, the reported
analysis is for [(SOCO)Co(THF)][PF6]2. Anal. Calcd for
C35H52CoF12N2O3P2: C, 46.83; H, 5.84; N, 3.12. Found: C, 46.48;
H, 6.14; N, 3.34. UV−vis (THF; λmax, nm (ε, M−1 cm−1)): 436
(2700), 456 (2700), 585 (1000), 805 (1700), 874 (1600). FTIR
(ATR, cm−1): 3326 (w), 2956(m), 2905(m), 2872(m), 1681(w),
1627(w), 1592(m), 1531(w), 1484(s), 1352(m), 1309(m), 1284(m),
1266(m), 1239(m), 1195(m), 1159(m), 1117(m), 1059(m), 1024(w),
1007(w), 948(w), 902(w), 818(s), 739(m), 711(w), 694(w), 650(w),
556(s), 493(s), 436(w), 416(w).
Synthesis of [(SOCO)Co(MeCN)] (1). Following the above
representative procedure, the collected solids were suspended in
MeCN (5 mL) and a minimal amount of toluene was added to
completely dissolve the material. This solution was stored at −25 °C
for 18 h, producing orange crystals of 1 suitable for XRD analysis
(507.5 mg, 0.88 mmol, 88%). Satisfactory elemental analysis required
the inclusion of solvent-derived MeOH; the reported analysis is for
[(SOCO)Co(MeCN)]·1.5MeOH. Anal. Calcd for C34.5H53CoN3O3.5:
C, 66.33; H, 8.55; N, 6.73. Found: C, 66.84; H, 8.16; N, 6.23. UV−vis
(MeCN; λmax, nm (ε, M−1 cm−1)): 438 (5600). FTIR (ATR, cm−1):
2949(m), 2900(m), 2865(m), 1627(w), 1477(s), 1449(s), 1389(m),
1357(m), 1325(s), 1285(m), 1236(m), 1201(m), 1073(w), 982(w),
842(m), 760(m), 701(m), 678(m), 644(m), 579(m), 514(m),
434(m), 413(m).
Synthesis of [(SOCO)Co(THF)] (1). The THF adduct to 1 was
obtained analogously using THF solvent in place of MeCN.
Crystalline solids suitable for single-crystal X-ray diffraction were
obtained by slow diffusion of pentane into a saturated THF solution at
−20 °C. UV−vis (THF; λmax, nm (ε, M−1 cm−1)): 430 (5300). FTIR
(ATR, cm−1): 2947(m), 2899(m), 2866(m), 1506(s), 1478(m),
1447(m), 1389(m), 1358(m), 1325(s), 1283(m), 1270(m), 1240(m),
1200(m), 1159(m), 1076(m), 1051(m), 914(m), 894(m), 869(m),
836(m), 756(m), 699(m), 643(m), 605(m), 578(m), 545(m),
512(m), 466(m), 434(m), 419(m).
Solid-State Magnetic Measurements. Magnetic susceptibility
data were collected using a Quantum Design MPMS XL SQUID
magnetometer. All sample manipulations were performed inside a
dinitrogen-filled glovebox (MBRAUN Labmaster 130). Finely ground
polycrystalline samples were loaded into polyethylene bags and sealed
in the glovebox, and then the bags were inserted into drinking straws
and inserted into the instrument. Ferromagnetic impurities were
checked through a variable field analysis (0−10 kOe) of the
magnetization at 100 K; susceptibility data were collected where
linearity of the M vs H plots indicated quenching of any significant
K
DOI: 10.1021/acs.inorgchem.7b01906
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
ferromagnetic impurities (Figures SM1 and SM2 in the Supporting
Information). Magnetic susceptibility data were collected in the
temperature range 2−300 K for 1 and 1+. Magnetization data were
collected at applied fields between 0 and 50 kOe and at temperatures
of 1.8 and 2 K for 1 and 1+, respectively. Data were corrected for the
diamagnetic contributions of the sample holder and for the sample
holding bag by subtracting empty containers; diamagnetic corrections
for the samples were calculated from Pascal’s constants.108 Magnetic
susceptibility data were fit using PHI,90 according to the spin
Hamiltonians
Jake D. Soper: 0000-0002-1961-8076
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We gratefully acknowledge financial support of this work from
the National Science Foundation (Grant Nos. CHE 1464852 to
J.D.S. and CHE 1363274 to M.P.S.) and The Netherlands
Organization for Scientific Research (NWO-CW TOP grant
716.015.001 to B.d.B.). We thank David Bostwick and Lucas
Evans for mass spectrometry assistance.
⎡
1
2
̂
⎢
H = −2JS1·S2 + ∑ D Sz,i,1
−
i
⎢
⎣
3Si,1(Si,1 + 1)
⎤
2
2
REFERENCES
̌
⎥
+ EiD(Sx,i,1 − Sy,i,1) + gμ Sx,i,1
■
i
B
⎥
⎦
(1) Friend, C. M.; Sanford, M. S.; Abruna, H. D. Catalytic Chemistry
Workshop on Defining Critical Directions for the Future; National Science
Foundation: Washington, DC, 2011.
⎡
⎤
⎦
1
2
2
D⎢S2
−
+ EiD(Sx,i − Sy,i)⎥ + gμ Sx,i
̌
̂
H =
∑
i
z,i
i
(2) Chirik, P. J.; Jarvo, E. R. Report on NSF SusChem Workshop and
American Chemical Society Symposium; National Science Foundation:
Washington, DC, 2014.
(3) Chirik, P. J.; Wieghardt, K. Radical Ligands Confer Nobility on
Base-Metal Catalysts. Science 2010, 327, 794−795.
B
3Si(Si + 1)
⎣
Computational Studies. DFT geometry optimizations were
performed using TURBOMOLE109 coupled to the PQS Baker
optimizer110,111 via the BOpt package112 at the BP86,113,114 def2-
TZVP115,116 level of theory (m4 grid), on full models (including tBu
groups) in the gas phase, using Grimme’s version 3 (disp3, “zero
damping”) dispersion corrections.117 All minima (no imaginary
frequencies) were characterized by numerically calculating the Hessian
matrix.
(4) Negishi, E.-I., Fundamental Properties of Palladium and Patterns
of the Reactions of Palladium and Its Complexes. In Handbook of
Organopalladium Chemistry for Organic Synthesis; Negishi, E.-I., de
Meijere, A., Eds.; Wiley-Interscience: New York, 2002; pp 17−35.
(5) Tasker, S. Z.; Standley, E. A.; Jamison, T. F. Recent advances in
homogeneous nickel catalysis. Nature 2014, 509, 299−309.
(6) Meijere, A. d.; Diederich, F. o. Metal-catalyzed cross-coupling
reactions, 2nd ed.; Wiley-VCH: Weinheim, Germany, 2004.
(7) Greenwood, N. N.; Earnshaw, A. Chemistry of the Elements, 2nd
ed.; Butterworth Heinemann: London, 1997.
(8) Luca, O. R.; Crabtree, R. H. Redox-active ligands in catalysis.
Chem. Soc. Rev. 2013, 42, 1440−1459.
(9) Lyaskovskyy, V.; de Bruin, B. Redox Non-Innocent Ligands:
Versatile New Tools to Control Catalytic Reactions. ACS Catal. 2012,
2, 270−279.
EPR parameters were calculated with the ADF118−120 program
system at the B3LYP/TZ2P level, using the coordinates from the
structures optimized in Turbomole as input. ZORA basis sets as
supplied with the ADF program were used. Unrestricted SPINORBIT
ZORA COLLINEAR calculations were used to obtain the SOC
corrected HFI tensors and Zeeman corrected g tensors.
Spin density pictures were generated using IQMol (http://www.
iqmol.org/).
ASSOCIATED CONTENT
* Supporting Information
■
(10) Praneeth, V. K. K.; Ringenberg, M. R.; Ward, T. R. Redox-Active
Ligands in Catalysis. Angew. Chem., Int. Ed. 2012, 51, 10228−10234.
(11) Jacquet, J.; Desage-El Murr, M.; Fensterbank, L. Metal-
Promoted Coupling Reactions Implying Ligand-Based Redox
Changes. ChemCatChem 2016, 8, 3310−3316.
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.inorg-
chem.7b01906.
Complete X-ray structure reports for [(SOCO)Co-
(THF)], [(OCO)Co(MeCN)], [(SOCO)Co(MeCN)],
[(PhOCO)Co(MeCN)], [(SOCO)Co(THF)2][BPh4],
and [(SOCO)Co(THF)3][PF6]2 and Cartesian coordi-
nates of the DFT optimized geometries of 1−3 (PDF)
(12) Broere, D. L. J.; Plessius, R.; van der Vlugt, J. I. New avenues for
ligand-mediated processes - expanding metal reactivity by the use of
redox-active catechol, o-aminophenol and o-phenylenediamine ligands.
Chem. Soc. Rev. 2015, 44, 6886−6915.
(13) Haneline, M. R.; Heyduk, A. F. C-C bond-forming reductive
elimination from a zirconium(IV) redox-active ligand complex. J. Am.
Chem. Soc. 2006, 128, 8410.
Accession Codes
(14) Munha, R. F.; Zarkesh, R. A.; Heyduk, A. F. Group transfer
reactions of d0 transition metal complexes: redox-active ligands
provide a mechanism for expanded reactivity. Dalton Trans. 2013, 42,
3751−3766.
(15) Zarkesh, R. A.; Ziller, J. W.; Heyduk, A. F. Four-electron
oxidative formation of aryl diazenes using a tantalum redox-active
ligand complex. Angew. Chem., Int. Ed. 2008, 47, 4715.
(16) Myers, T. W.; Berben, L. A. Redox active aluminium(III)
complexes convert CO2 into MgCO3 or CaCO3 in a synthetic cycle
using Mg or Ca metal. Chem. Commun. 2013, 49, 4175−4177.
(17) Wong, J. L.; Sanchez, R. H.; Logan, J. G.; Zarkesh, R. A.; Ziller,
J. W.; Heyduk, A. F. Disulfide reductive elimination from an iron(III)
complex. Chem. Sci. 2013, 4, 1906−1910.
(18) Bouwkamp, M. W.; Bowman, A. C.; Lobkovsky, E.; Chirik, P. J.
Iron-Catalyzed [2pi + 2pi] Cycloaddition of alpha, w-Dienes: The
Importance of Redox-Active Supporting Ligands. J. Am. Chem. Soc.
2006, 128, 13340−13341.
(19) Sylvester, K. T.; Chirik, P. J. Iron-Catalyzed, Hydrogen-
Mediated Reductive Cyclization of 1,6-Enynes and Diynes: Evidence
CCDC 1563838−1563843 contain the supplementary crystal-
lographic data for this paper. These data can be obtained free of
charge via www.ccdc.cam.ac.uk/data_request/cif, or by email-
ing data_request@ccdc.cam.ac.uk, or by contacting The
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
AUTHOR INFORMATION
Corresponding Authors
*E-mail for M.P.S.: matthew.shores@colostate.edu.
*E-mail for B.d.B.: b.debruin@uva.nl.
*E-mail for J.D.S.: jake.soper@gatech.edu.
■
ORCID
Caleb F. Harris: 0000-0002-6844-8923
Matthew P. Shores: 0000-0002-9751-0490
Bas de Bruin: 0000-0002-3482-7669
L
DOI: 10.1021/acs.inorgchem.7b01906
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
for Bis(imino)pyridine Ligand Participation. J. Am. Chem. Soc. 2009,
131, 8772−8774.
(37) Corcos, A. R.; Villanueva, O.; Walroth, R. C.; Sharma, S. K.;
Bacsa, J.; Lancaster, K. M.; MacBeth, C. E.; Berry, J. F. Oxygen
Activation by Co(II) and a Redox Non-Innocent Ligand: Spectro-
scopic Characterization of a Radical-Co(II)-Superoxide Complex with
Divergent Catalytic Reactivity. J. Am. Chem. Soc. 2016, 138, 1796−
1799.
(38) Hoffman, J. M.; Oliver, A. G.; Brown, S. N. The Metal or the
Ligand? The Preferred Locus for Redox Changes in Oxygen Atom
Transfer Reactions of Rhenium Amidodiphenoxides. J. Am. Chem. Soc.
2017, 139, 4521−4531.
(39) Randolph, A. H.; Seewald, N. J.; Rickert, K.; Brown, S. N.
Tris(3,5-di-tert-butylcatecholato)molybdenum(VI): Lewis Acidity and
Nonclassical Oxygen Atom Transfer Reactions. Inorg. Chem. 2013, 52,
12587−12598.
(20) Russell, S. K.; Lobkovsky, E.; Chirik, P. J. Iron-Catalyzed
Intermolecular [2 pi+2 pi] Cycloaddition. J. Am. Chem. Soc. 2011, 133,
8858−8861.
(21) Darmon, J. M.; Stieber, S. C. E.; Sylvester, K. T.; Fernandez, I.;
Lobkovsky, E.; Semproni, S. P.; Bill, E.; Wieghardt, K.; DeBeer, S.;
Chirik, P. J. Oxidative addition of carbon-carbon bonds with a redox-
active bis(imino)pyridine iron complex. J. Am. Chem. Soc. 2012, 134,
17125−17137.
(22) Jacquet, J.; Blanchard, S.; Derat, E.; Desage-El Murr, M.;
Fensterbank, L. Redox-ligand sustains controlled generation of CF3
radicals by well-defined copper complex. Chem. Sci. 2016, 7, 2030−
2036.
(23) Tondreau, A. M.; Atienza, C. C. H.; Darmon, J. M.; Milsmann,
C.; Hoyt, H. M.; Weller, K. J.; Nye, S. A.; Lewis, K. M.; Boyer, J.; Delis,
J. G. P.; Lobkovsky, E.; Chirik, P. J. Synthesis, electronic structure, and
alkene hydrosilylation activity of terpyridine and bis(imino)pyridine
iron dialkyl complexes. Organometallics 2012, 31, 4886−4893.
(24) Yu, R. P.; Darmon, J. M.; Hoyt, J. M.; Margulieux, G. W.;
Turner, Z. R.; Chirik, P. J. High-activity iron catalysts for the
hydrogenation of hindered, unfunctionalized alkenes. ACS Catal. 2012,
2, 1760−1764.
(25) Hojilla Atienza, C. C.; Tondreau, A. M.; Weller, K. J.; Lewis, K.
M.; Cruse, R. W.; Nye, S. A.; Boyer, J. L.; Delis, J. G. P.; Chirik, P. J.
High-selectivity bis(imino)pyridine iron catalysts for the hydro-
silylation of 1,2,4-trivinylcyclohexane. ACS Catal. 2012, 2, 2169−2172.
(26) Tondreau, A. M.; Atienza, C. C. H.; Weller, K. J.; Nye, S. A.;
Lewis, K. M.; Delis, J. G. P.; Chirik, P. J. Iron Catalysts for Selective
Anti-Markovnikov Alkene Hydrosilylation Using Tertiary Silanes.
Science 2012, 335, 567−570.
(27) Villanueva, O.; Weldy, N. M.; Blakey, S. B.; MacBeth, C. E.
Cobalt catalyzed sp(3) C-H amination utilizing aryl azides. Chem. Sci.
2015, 6, 6672−6675.
(28) Bagh, B.; Broere, D. L. J.; Sinha, V.; Kuijpers, P. F.; van Leest, N.
P.; de Bruin, B.; Demeshko, S.; Siegler, M. A.; Vlugt, J. I. D. Catalytic
Synthesis of N-Heterocycles via Direct C(sp(3))-H Amination Using
an Air-Stable Iron(III) Species with a Redox-Active Ligand. J. Am.
Chem. Soc. 2017, 139, 5117−5124.
(29) Kuijpers, P. F.; Tiekink, M. J.; Breukelaar, W. B.; Broere, D. L. J.;
van Leest, N. P.; van der Vlugt, J. I.; Reek, J. N. H.; de Bruin, B.
Cobalt-Porphyrin-Catalysed Intramolecular Ring-Closing C-H Amina-
tion of Aliphatic Azides: A Nitrene-Radical Approach to Saturated
Heterocycles. Chem. - Eur. J. 2017, 23, 7945−7952.
(30) Lyaskovskyy, V.; Suarez, A. I. O.; Lu, H. J.; Jiang, H. L.; Zhang,
X. P.; de Bruin, B. Mechanism of Cobalt(II) Porphyrin-Catalyzed C-H
Amination with Organic Azides: Radical Nature and H-Atom
Abstraction Ability of the Key Cobalt(III)-Nitrene Intermediates. J.
Am. Chem. Soc. 2011, 133, 12264−12273.
(31) Olivos Suarez, A. I.; Jiang, H. L.; Zhang, X. P.; de Bruin, B. The
radical mechanism of cobalt(II) porphyrin-catalyzed olefin aziridina-
tion and the importance of cooperative H-bonding. Dalton Trans.
2011, 40, 5697−5705.
(32) Das, B. G.; Chirila, A.; Tromp, M.; Reek, J. N. H.; de Bruin, B.
Co-III-Carbene Radical Approach to Substituted 1H-Indenes. J. Am.
Chem. Soc. 2016, 138, 8968−8975.
(33) Paul, N. D.; Mandal, S.; Otte, M.; Cui, X.; Zhang, X. P.; de
Bruin, B. Metalloradical Approach to 2H-Chromenes. J. Am. Chem.
Soc. 2014, 136, 1090−1096.
(34) Paul, N. D.; Chirila, A.; Lu, H. J.; Zhang, X. P.; de Bruin, B.
Carbene Radicals in Cobalt(II)-Porphyrin-Catalysed Carbene Carbon-
ylation Reactions; A Catalytic Approach to Ketenes. Chem. - Eur. J.
2013, 19, 12953−12958.
(35) Dzik, W. I.; Zhang, X. P.; de Bruin, B. Redox Noninnocence of
Carbene Ligands: Carbene Radicals in (Catalytic) C-C Bond
Formation. Inorg. Chem. 2011, 50, 9896−9903.
(36) Dzik, W. I.; Xu, X.; Zhang, X. P.; Reek, J. N. H.; de Bruin, B.
’Carbene Radicals’ in Co-II(por)-Catalyzed Olefin Cyclopropanation.
J. Am. Chem. Soc. 2010, 132, 10891−10902.
(40) Lippert, C. A.; Arnstein, S. A.; Sherrill, C. D.; Soper, J. D.
Redox-Active Ligands Facilitate Bimetallic O2 Homolysis at Five-
Coordinate Oxorhenium(V) Centers. J. Am. Chem. Soc. 2010, 132,
3879−3892.
(41) Lippert, C. A.; Soper, J. D. Deoxygenation of Nitroxyl Radicals
by Oxorhenium(V) Complexes with Redox-Active Ligands. Inorg.
Chem. 2010, 49, 3682−3684.
(42) Lippert, C. A.; Hardcastle, K. I.; Soper, J. D. Harnessing Redox-
Active Ligands for Low-Barrier Radical Addition at Oxorhenium
Complexes. Inorg. Chem. 2011, 50, 9864−9878.
(43) Beletskaya, I. P.; Cheprakov, A. V. Copper in cross-coupling
reactions: The post-Ullmann chemistry. Coord. Chem. Rev. 2004, 248,
2337−2364.
(44) Phapale, V. B.; Cardenas, D. J. Nickel-catalysed Negishi cross-
coupling reactions: scope and mechanisms. Chem. Soc. Rev. 2009, 38,
1598−1607.
(45) Sherry, B. D.; Furstner, A. The Promise and Challenge of Iron-
̈
Catalyzed Cross Coupling. Acc. Chem. Res. 2008, 41, 1500−1511.
(46) Wendlandt, A. E.; Suess, A. M.; Stahl, S. S. Copper-Catalyzed
Aerobic Oxidative C−H Functionalizations: Trends and Mechanistic
Insights. Angew. Chem., Int. Ed. 2011, 50, 11062−11087.
(47) Hu, X. L. Nickel-catalyzed cross coupling of non-activated alkyl
halides: a mechanistic perspective. Chem. Sci. 2011, 2, 1867−1886.
(48) Hirano, K.; Miura, M. Copper-mediated oxidative direct C-C
(hetero)aromatic cross-coupling. Chem. Commun. 2012, 48, 10704−
10714.
(49) Truong, T.; Klimovica, K.; Daugulis, O. Copper-Catalyzed,
Directing Group-Assisted Fluorination of Arene and Heteroarene C-H
Bonds. J. Am. Chem. Soc. 2013, 135, 9342−9345.
(50) Yoshikai, N.; Matsumoto, A.; Norinder, J.; Nakamura, E. Iron-
Catalyzed Direct Arylation of Aryl Pyridines and Imines Using Oxygen
as an Oxidant. Synlett 2010, 2010, 313−316.
(51) Sarhan, A. A. O.; Bolm, C. Iron(III) chloride in oxidative C-C
coupling reactions. Chem. Soc. Rev. 2009, 38, 2730−2744.
(52) Mako, T. L.; Byers, J. A. Recent advances in iron-catalysed cross
coupling reactions and their mechanistic underpinning. Inorg. Chem.
Front. 2016, 3, 766−790.
(53) Bedford, R. B.; Brenner, P. B. The Development of Iron
Catalysts for Cross-Coupling Reactions. Top. Organomet. Chem. 2015,
50, 19−46.
(54) Itazaki, M.; Nakazawa, H. Iron-Catalyzed Cross-Dehydrogen-
ative-Coupling Reactions. Top. Organomet. Chem. 2015, 50, 47−81.
(55) Liu, C.; Zhang, H.; Shi, W.; Lei, A. W. Bond Formations
between Two Nucleophiles: Transition Metal Catalyzed Oxidative
Cross-Coupling Reactions. Chem. Rev. 2011, 111, 1780−1824.
(56) Seyferth, D. The Grignard Reagents. Organometallics 2009, 28,
1598−1605.
(57) Tan, G.; He, S.; Huang, X.; Liao, X.; Cheng, Y.; You, J. Cobalt-
Catalyzed Oxidative C-H/C-H Cross-Coupling between Two
Heteroarenes. Angew. Chem., Int. Ed. 2016, 55, 10414−10418.
(58) de Bruin, B.; Bill, E.; Bothe, E.; Weyhermuller, T.; Wieghardt, K.
Molecular and electronic structures of bis(pyridine-2,6-diimine)metal
complexes [ML2](PF6)(n) (n = 0, 1, 2, 3; M = Mn, Fe, Co, Ni, Cu,
Zn). Inorg. Chem. 2000, 39, 2936−2947.
M
DOI: 10.1021/acs.inorgchem.7b01906
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(59) Budzelaar, P. H. M.; de Bruin, B.; Gal, A. W.; Wieghardt, K.; van
Lenthe, J. H. Metal-to-ligand electron transfer in diiminopyridine
complexes of Mn-Zn. A theoretical study. Inorg. Chem. 2001, 40,
4649−4655.
(60) Chirik, P. J. Electronic Structures of Reduced Manganese, Iron,
and Cobalt Complexes Bearing Redox-Active Bis(imino) pyridine
Pincer Ligands. Pincer and Pincer-Type Complexes: Applications in
Organic Synthesis and Catalysis 2014, 189−212.
(61) Smith, A. L.; Hardcastle, K. I.; Soper, J. D. Redox-Active Ligand-
Mediated Oxidative Addition and Reductive Elimination at Square
Planar Cobalt(III): Multielectron Reactions for Cross-Coupling. J. Am.
Chem. Soc. 2010, 132, 14358−14360.
ketoaminato complexes and their cytotoxic activity towards human
tumor cell lines. J. Inorg. Biochem. 2011, 105, 858−866.
(77) Chiang, L.; Allan, L. E. N.; Alcantara, J.; Wang, M. C. P.; Storr,
T.; Shaver, M. P. Tuning ligand electronics and peripheral substitution
on cobalt salen complexes: structure and polymerisation activity.
Dalton Trans. 2014, 43, 4295−4304.
(78) Nguyen, A. I.; Hadt, R. G.; Solomon, E. I.; Tilley, T. D. Efficient
C-H bond activations via O-2 cleavage by a dianionic cobalt(II)
complex. Chem. Sci. 2014, 5, 2874−2878.
(79) Kochem, A.; Kanso, H.; Baptiste, B.; Arora, H.; Philouze, C.;
Jarjayes, O.; Vezin, H.; Luneau, D.; Orio, M.; Thomas, F. Ligand
Contributions to the Electronic Structures of the Oxidized Cobalt(II)
salen Complexes. Inorg. Chem. 2012, 51, 10557−10571.
(80) Meng, Y.-S.; Mo, Z.; Wang, B.-W.; Zhang, Y.-Q.; Deng, L.; Gao,
S. Observation of the single-ion magnet behavior of d8 ions on two-
coordinate Co(i)-NHC complexes. Chem. Sci. 2015, 6, 7156−7162.
(81) Massard, A.; Braunstein, P.; Danopoulos, A. A.; Choua, S.; Rabu,
P. Studies on Three-Coordinate [Co{N(SiMe3)2}2L] Complexes, L =
N-Heterocyclic Carbene. Organometallics 2015, 34, 2429−2438.
(82) Harrison, D. J.; Daniels, A. L.; Korobkov, I.; Baker, R. T.
Tetracarbonyl(trifluoromethyl)cobalt(I) [Co(CO)4(CF3)] as a Pre-
cursor to New Cobalt Trifluoromethyl and Difluorocarbene
Complexes. Organometallics 2015, 34, 4598−4604.
(83) Danopoulos, A. A.; Braunstein, P. Mono-N-heterocyclic carbene
amido and alkyl complexes. Cobalt-mediated C-H activation and C-C
coupling reactions involving benzyl ligands on a putative 3-coordinate
intermediate. Dalton Trans. 2013, 42, 7276−7280.
(62) Dzik, W. I.; van der Vlugt, J. I.; Reek, J. N. H.; de Bruin, B.
Ligands that Store and Release Electrons during Catalysis. Angew.
Chem., Int. Ed. 2011, 50, 3356−3358.
́
(63) Borre, E.; Dahm, G.; Aliprandi, A.; Mauro, M.; Dagorne, S.;
Bellemin-Laponnaz, S. Tridentate Complexes of Group 10 Bearing
Bis-Aryloxide N-Heterocyclic Carbene Ligands: Synthesis, Structural,
Spectroscopic, and Computational Characterization. Organometallics
2014, 33, 4374−4384.
(64) Lee, M.-T.; Hu, C.-H. Density Functional Study of N-
Heterocyclic and Diamino Carbene Complexes: Comparison with
Phosphines. Organometallics 2004, 23, 976−983.
(65) Lummiss, J. A. M.; Higman, C. S.; Fyson, D. L.; McDonald, R.;
Fogg, D. E. The divergent effects of strong NHC donation in catalysis.
Chem. Sci. 2015, 6, 6739−6746.
(66) Clavier, H.; Coutable, L.; Toupet, L.; Guillemin, J.-C.; Mauduit,
M. Design and synthesis of new bidentate alkoxy-NHC ligands for
enantioselective copper-catalyzed conjugate addition. J. Organomet.
Chem. 2005, 690, 5237−5254.
(67) Waltman, A. W.; Grubbs, R. H. A New Class of Chelating N-
Heterocyclic Carbene Ligands and Their Complexes with Palladium.
Organometallics 2004, 23, 3105−3107.
(68) Bellemin-Laponnaz, S.; Welter, R.; Brelot, L.; Dagorne, S.
Synthesis and structure of V(V) and Mn(III) NHC complexes
supported by a tridentate bis-aryloxide-N-heterocyclic carbene ligand.
J. Organomet. Chem. 2009, 694, 604−606.
(69) Dagorne, S.; Bellemin-Laponnaz, S.; Romain, C. Neutral and
Cationic N-Heterocyclic Carbene Zirconium and Hafnium Benzyl
Complexes: Highly Regioselective Oligomerization of 1-Hexene with a
Preference for Trimer Formation. Organometallics 2013, 32, 2736−
2743.
(70) Romain, C.; Choua, S.; Collin, J.-P.; Heinrich, M.; Bailly, C.;
Karmazin-Brelot, L.; Bellemin-Laponnaz, S.; Dagorne, S. Redox and
Luminescent Properties of Robust and Air-Stable N-Heterocyclic
Carbene Group 4 Metal Complexes. Inorg. Chem. 2014, 53, 7371−
7376.
(71) Romain, C.; Fliedel, C.; Bellemin-Laponnaz, S.; Dagorne, S.
NHC Bis-Phenolate Aluminum Chelates: Synthesis, Structure, and
Use in Lactide and Trimethylene Carbonate Polymerization. Organo-
metallics 2014, 33, 5730−5739.
(84) Gibson, S. E.; Johnstone, C.; Loch, J. A.; Steed, J. W.;
Stevenazzi, A. Novel Structures and Pauson−Khand Activities of N-
Heterocyclic Carbene Dicobalt Complexes. Organometallics 2003, 22,
5374−5377.
(85) Coleman, A. W.; Hitchcock, P. B.; Lappert, M. F.; Maskell, R.
K.; Muller, J. H. Routes to optically active electron-rich olefins (L*2
̈
and some derived carbenemetal complexes; x-ray structures of
[Co(CO)(L*)(NO(PPh3)] and cis-[Rh(Cl)(COD)(L′*)][L*= (S)-
CN(Me)CH(Me)CH2NMe. J. Organomet. Chem. 1983, 250, c9−c14.
(86) Liske, A.; Verlinden, K.; Buhl, H.; Schaper, K.; Ganter, C.
Determining the pi-Acceptor Properties of N-Heterocyclic Carbenes
by Measuring the Se-77 NMR Chemical Shifts of Their Selenium
Adducts. Organometallics 2013, 32, 5269−5272.
(87) Miessler, G. L.; Tarr, D. A. Inoganic Chemistry, 2nd ed.; Prentice
Hall: Englewood Cliffs, NJ, 1998.
(88) Lagaditis, P. O.; Schluschass, B.; Demeshko, S.; Wurtele, C.;
Schneider, S. Square-Planar Cobalt(III) Pincer Complex. Inorg. Chem.
2016, 55, 4529−4536.
(89) Addison, A. W.; Rao, T. N.; Reedijk, J.; van Rijn, J.; Verschoor,
G. C. Synthesis, structure, and spectroscopic properties of copper(II)
compounds containing nitrogen-sulphur donor ligands; the crystal and
molecular structure of aqua[1,7-bis(N-methylbenzimidazol-2[prime or
minute]-yl)-2,6-dithiaheptane]copper(II) perchlorate. J. Chem. Soc.,
Dalton Trans. 1984, 1349−1356.
(90) Chilton, N. F.; Anderson, R. P.; Turner, L. D.; Soncini, A.;
Murray, K. S. PHI: A powerful new program for the analysis of
anisotropic monomeric and exchange-coupled polynuclear d- and f-
block complexes. J. Comput. Chem. 2013, 34, 1164−1175.
(91) Cole, G. M.; Garrett, B. B. Atomic and Molecular Spin-Orbit
Coupling Constants for 3d Transition Metal Ions. Inorg. Chem. 1970,
9, 1898.
(92) Jenkins, D. M.; Di Bilio, A. J.; Allen, M. J.; Betley, T. A.; Peters,
J. C. Elucidation of a low spin cobalt(II) system in a distorted
tetrahedral geometry. J. Am. Chem. Soc. 2002, 124, 15336−15350.
(93) Shores, M. P.; Sokol, J. J.; Long, J. R. Nickel(II)-molybdenum-
(III)-cyanide clusters: Synthesis and magnetic behavior of species
incorporating [(Me3tacn)Mo(CN)3]. J. Am. Chem. Soc. 2002, 124,
2279−2292.
(94) Radonovich, L. J.; Klabunde, K. J.; Behrens, C. B.; McCollor, D.
P.; Anderson, B. B. Crystal and molecular structures of π-toluene
complexes of bis(pentafluorobenzene)cobalt and bis-
(pentafluorobenzene)nickel. Inorg. Chem. 1980, 19, 1221−1226.
(72) Weinberg, D. R.; Hazari, N.; Labinger, J. A.; Bercaw, J. E.
Iridium(I) and Iridium(III) Complexes Supported by a Diphenolate
Imidazolyl-Carbene Ligand. Organometallics 2010, 29, 89−100.
(73) Min, K. S.; Weyhermuller, T.; Bothe, E.; Wieghardt, K.
̈
Tetradentate Bis(o-iminobenzosemiquinonate(1−)) π Radical Ligands
and Their o-Aminophenolate(1−) Derivatives in Complexes of
Nickel(II), Palladium(II), and Copper(II). Inorg. Chem. 2004, 43,
2922−2931.
(74) Despagnet-Ayoub, E.; Henling, L. M.; Labinger, J. A.; Bercaw, J.
E. Addition of a phosphine ligand switches an N-heterocyclic carbene-
zirconium catalyst from oligomerization to polymerization of 1-hexene.
Dalton Trans. 2013, 42, 15544−15547.
(75) Chaudhuri, P.; Wieghardt, K. Phenoxyl Radical Complexes. In
Progress in Inorganic Chemistry. John Wiley & Sons, Inc 2002, 151−
216.
(76) Gurley, L.; Beloukhina, N.; Boudreau, K.; Klegeris, A.; McNeil,
W. S. The synthesis and characterization of a series of cobalt(II) beta-
N
DOI: 10.1021/acs.inorgchem.7b01906
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(95) Chaudhuri, P.; Verani, C. N.; Bill, E.; Bothe, E.; Weyhermuller,
Design and assessment of accuracy. Phys. Chem. Chem. Phys. 2005, 7,
3297−3305.
̈
T.; Wieghardt, K. Electronic Structure of Bis(o-
iminobenzosemiquinonato)metal Complexes (Cu, Ni, Pd). The Art
of Establishing Physical Oxidation States in Transition-Metal
Complexes Containing Radical Ligands. J. Am. Chem. Soc. 2001,
123, 2213−2223.
(96) Urbach, F. L.; Bereman, R. D.; Topich, J. A.; Harihara, M.;
Kalbache, Bj. Stereochemistry and Electronic-Structure of Low-Spin,
Square-Planar Cobalt(II) Chelates with Tetradentate Schiff-Base
Ligands. J. Am. Chem. Soc. 1974, 96, 5063−5069.
(97) Dong, W. K.; Duan, J. G.; Guan, Y. H.; Shi, J. Y.; Zhao, C. Y.
Synthesis, crystal structure and spectroscopic behaviors of Co(II) and
Cu(II) complexes with Salen-type bisoxime ligands. Inorg. Chim. Acta
2009, 362, 1129−1134.
(98) Cantalupo, S. A.; Fiedler, S. R.; Shores, M. P.; Rheingold, A. L.;
Doerrer, L. H. High-Spin Square-Planar CoII and FeII Complexes and
Reasons for Their Electronic Structure. Angew. Chem., Int. Ed. 2012,
51, 1000−1005.
(99) Chirik, P. J. Preface: Forum on Redox-Active Ligands. Inorg.
Chem. 2011, 50, 9737−9740.
(116) Weigend, F.; Haser, M.; Patzelt, H.; Ahlrichs, R. RI-MP2:
optimized auxiliary basis sets and demonstration of efficiency. Chem.
Phys. Lett. 1998, 294, 143−152.
(117) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. A consistent and
accurate ab initio parametrization of density functional dispersion
correction (DFT-D) for the 94 elements H-Pu. J. Chem. Phys. 2010,
132, 154104.
(118) Baerends, E. J.; Ellis, D. E.; Ros, P. Self-consistent molecular
Hartree-Fock-Slater calculations - I. The computational procedure.
Chem. Phys. 1973, 2, 41−51.
(119) te Velde, G.; Baerends, E. J. Numerical-Integration for
Polyatomic Systems. J. Comput. Phys. 1992, 99, 84−98.
(120) Fonseca Guerra, C.; Snijders, J. G.; te Velde, G.; Baerends, E. J.
Towards an order-N DFT method. Theor. Chem. Acc. 1998, 99, 391−
403.
(100) Morris, W. D.; Wolczanski, P. T.; Sutter, J.; Meyer, K.;
Cundari, T. R.; Lobkovsky, E. B. Iron and Chromium Complexes
Containing Tridentate Chelates Based on Nacnac and Imino- and
Methyl-Pyridine Components: Triggering C-X Bond Formation. Inorg.
Chem. 2014, 53, 7467−7484.
(101) Williams, V. A.; Wolczanski, P. T.; Sutter, J.; Meyer, K.;
Lobkovsky, E. B.; Cundari, T. R. Iron Complexes Derived from
{nacnac-(CH(2)py)(2)}(−) and {nacnac-(CH(2)py)(CHpy)}(n)
Ligands: Stabilization of Iron(II) via Redox Noninnocence. Inorg.
Chem. 2014, 53, 4459−4474.
(102) Boehme, C.; Frenking, G. Electronic structure of stable
carbenes, silylenes, and germylenes. J. Am. Chem. Soc. 1996, 118,
2039−2046.
(103) Herrmann, W. A. N-heterocyclic carbenes: A new concept in
organometallic catalysis. Angew. Chem., Int. Ed. 2002, 41, 1290−1309.
(104) Fulmer, G. R.; Miller, A. J. M.; Sherden, N. H.; Gottlieb, H. E.;
Nudelman, A.; Stoltz, B. M.; Bercaw, J. E.; Goldberg, K. I. NMR
Chemical Shifts of Trace Impurities: Common Laboratory Solvents,
Organics, and Gases in Deuterated Solvents Relevant to the
Organometallic Chemist. Organometallics 2010, 29, 2176−2179.
(105) Evans, D. F. The Determination of the Paramagnetic
Susceptibility of Substances in Solution by Nuclear Magnetic
Resonance. J. Chem. Soc. 1959, 2003−2005.
(106) Piguet, C. Paramagnetic susceptibility by NMR: The ’solvent
correction’ removed for large paramagnetic molecules. J. Chem. Educ.
1997, 74, 815−816.
(107) Rhile, I. J.; Markle, T. F.; Nagao, H.; DiPasquale, A. G.; Lam,
O. P.; Lockwood, M. A.; Rotter, K.; Mayer, J. M. Concerted Proton−
Electron Transfer in the Oxidation of Hydrogen-Bonded Phenols. J.
Am. Chem. Soc. 2006, 128, 6075−6088.
(108) Bain, G. A.; Berry, J. F. Diamagnetic corrections and Pascal’s
constants. J. Chem. Educ. 2008, 85, 532−536.
(109) TURBOMOLE Version 7.1; TURBOMOLE Gmbh, Karlsruhe,
Germany, 2013.
(110) PQS Version 2.4; Parallel Quantum Solutions, Fayetteville, AR,
USA, 2001.
(111) Baker, J. An Algorithm for the Location of Transition-States. J.
Comput. Chem. 1986, 7, 385−395.
(112) Budzelaar, P. H. M. Geometry optimization using generalized,
chemically meaningful constraints. J. Comput. Chem. 2007, 28, 2226−
2236.
(113) Becke, A. D. Density-Functional Exchange-Energy Approx-
imation with Correct Asymptotic-Behavior. Phys. Rev. A: At., Mol., Opt.
Phys. 1988, 38, 3098−3100.
(114) Perdew, J. P. Density-Functional Approximation for the
Correlation-Energy of the Inhomogeneous Electron-Gas. Phys. Rev. B:
Condens. Matter Mater. Phys. 1986, 33, 8822−8824.
(115) Weigend, F.; Ahlrichs, R. Balanced basis sets of split valence,
triple zeta valence and quadruple zeta valence quality for H to Rn:
O
DOI: 10.1021/acs.inorgchem.7b01906
Inorg. Chem. XXXX, XXX, XXX−XXX