Reactions of dicobalt octacarbonyl with dinucleating and mononucleating bis(imino)pyridine ligands

Article abstract of DOI:10.1039/C8DT03405B

This work focuses on the application of dicobalt octacarbonyl (Co₂(CO)₈) as a metal precursor in the chemistry of formally low-valent cobalt with redox-active bis(imino)pyridine [NNN] ligands. The reactions of both mononucleating mes

Full text of DOI:10.1039/C8DT03405B

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Reactions of dicobalt octacarbonyl with
dinucleating and mononucleating
bis(imino)pyridine ligands†
Cite this: Dalton Trans., 2018, 47,
15353
a
a
a
b
Ryan L. Hollingsworth, Jeffrey W. Beattie,
Amanda Grass, Philip D. Martin,
*
a
c
Stanislav Groysman * and Richard L. Lord
2 8
This work focuses on the application of dicobalt octacarbonyl (Co (CO) ) as a metal precursor in the
chemistry of formally low-valent cobalt with redox-active bis(imino)pyridine [NNN] ligands. The reactions
1
of both mononucleating mesityl-substituted bis(aldimino)pyridine (L ) and dinucleating macrocyclic
2
xanthene-bridged di(bis(aldimino)pyridine) (L ) with Co
2
(CO)
8
were investigated. Independent of the
2
metal-to-ligand ratio (1 : 1 or 1 : 2 ligand to Co
2
(CO)
8
), the reaction of the dinucleating ligand L with
2
Co (CO) produces a tetranuclear complex [Co
2
8
4
(L )(CO)10] featuring two discrete [Co
2 5
[NNN](CO) ] units.
1
In contrast, a related mononucleating bis(aldimino)pyridine ligand, L , produces different species at
1
1
different ligand to Co
2
(CO)
8
ratios, including dinuclear [Co
2 5 2 4
(CO) (L )] and zwitterionic [Co(L ) ][Co(CO) ].
2
Interestingly, [Co (L )(CO)10] features metal–metal bonds, and no bridging carbonyls, whereas
4
1
[Co
2
(CO)
5
(L )] contains cobalt centers bridged by one or two carbonyl ligands. In either case, treatment
Received 20th August 2018,
Accepted 27th September 2018
2 4
with excess acetonitrile leads to disproportionation to the zwitterionic [Co[NNN](NCMe) ][Co(CO) ] units.
The electronic structures of the complexes described above were studied with density functional theory.
All the obtained bis(imino)pyridine complexes serve as catalysts for cyclotrimerization of methyl propio-
DOI: 10.1039/c8dt03405b
rsc.li/dalton
2 8
late, albeit their reactivity is inferior compared with Co (CO) .
pyridinediimine alternatively termed bis(imino)pyridine) by
Introduction
1
0
the reduction of [NNN]FeX
2
precursors.
Alternatively,
The past decade has witnessed growing interest in the design formally reduced monometallic/bimetallic metal complexes
of bimetallic complexes due to the anticipated cooperativity with redox-active ligands can be obtained in a single step, if a
1
–3
between two nearby metal centers. One particularly fascinat- reduced metal precursor is available. While low-boiling
1
1
ing subfield of this research area involves redox-active dinu- Ni(CO)
4
is generally considered too toxic for contemporary
cleating ligands, which allow for the combination of metal– laboratory synthesis, formally Ni(0) complexes with dinucleat-
4
metal and metal–ligand cooperativity. Catalytically competent ing and mononucleating redox-active ligands can be con-
complexes of redox-active ligands generally require metals in veniently obtained by the direct reaction of Ni(COD)
formally low oxidation states.
2
with a
Such complexes can be variety of redox-active ligand precursors. Dicobalt octacarbo-
obtained by a “two-step” protocol, which involves first syn- nyl Co (CO) is one of the most decorated organometallic com-
thesis of the oxidized metal complexes (i.e. Ni(II) or Co(II)) then plexes, whose original synthesis by Mond and coworkers was
5
–8
12
2
8
9
13
a reduction. For example, Chirik described synthesis of [NNN] published in 1910. As Co (CO) is a solid, it does not consti-
2
8
Fe(N
2
)
2
and [NNN]Fe(CO)
2
complexes ([NNN] = redox-active tute as acute of an inhalation hazard as Ni(CO)
Co (CO)
metallic chemistry, in the synthesis of nanomaterials,
4
. As a result,
2
8
is frequently utilized as a pre-catalyst in organo-
1
4
15
a
Department of Chemistry, Wayne State University, 5101 Cass Ave, Detroit, MI,
1
6
metal clusters, and extended solids. Co
2
(CO)
8
also appears
4
8202, USA. E-mail: groysman@chem.wayne.edu
b
to be a viable candidate as a metal precursor for the synthesis
of the formally reduced cobalt complexes with redox-
active ligands, both mononucleating and dinucleating. Yet, the
reports of its use with such ligands are relatively scarce,
and generally require benzoquinone or closely related amido-
Lumigen Instrument Center, Wayne State University, 5101 Cass Ave, Detroit, MI,
4
8202, USA
c
Department of Chemistry, Grand Valley State University, Allendale, MI, 49401, USA.
E-mail: lordri@gvsu.edu
†
Electronic supplementary information (ESI) available: Synthetic procedures,
experimental data, and Cartesian coordinates. CCDC 1862093–1862097. For ESI
and crystallographic data in CIF or other electronic format see DOI: 10.1039/
c8dt03405b
1
7
phenolate as a co-reactant. Wender, Sternberg and Orchin
1
8
2 8
explored the reactivity of phenanthroline with Co (CO) ,
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 15353–15363 | 15353

View Article Online
Paper
Dalton Transactions
that resulted in the disproportionation of Co(0) into a Co(II) related ligand with methyl groups in the imine positions has
2
3
cation and two Co(1−) anions. More recently, Wieghardt and been previously synthesized.
coworkers explored the reactivity of Co (CO) with terpyri-
2
8
2
1
1
9
Synthesis and structures of the products of the reactions of L /L
with Co (CO)
dine. Chirik and coworkers prepared a bis(imino)pyridine
cobalt carbonyl complex, but this complex was obtained
2
8
2
0
indirectly using cobalt-dinitrogen precursor. We, and others, As mentioned earlier, this work focuses on the reactivity of
have previously shown that formally Ni(0) source, Ni(COD) , is Co (CO) as a precursor for the chemistry of formally low-
2
2
8
a highly useful precursor in the chemistry of low-valent nickel valent cobalt with redox-active bis(imino)pyridine ligands. To
2
with redox-active mononucleating and dinucleating ligands obtain a dinuclear complex of L , it was treated with one
1
2
featuring iminopyridine and bis(imino)pyridine chelates.
2 8
equivalent of Co (CO) at room temperature for 24 hours
Motivated by these findings, we became interested in the (Scheme 2). In contrast to our expectations, this reaction pro-
2
2 8 4
exploration of the chemistry of Co (CO) with bis(imino)pyri- duced tetranuclear complex [Co (L )(CO)10] (1), that is isolated
dine ligands. Herein we report synthetic, structural, and as red-violet crystals by recrystallization from diethyl ether in
density functional theory (DFT) study on the reactivity of 37% yield (with respect to the ligand). The yield of the tetra-
2
Co
2
(CO)
8
with dinucleating and mononucleating bis(imino) nuclear complex can be improved by using 1 : 2 ratio of L with
pyridine ligands.
2 8
Co (CO) . Overnight recrystallization in diethyl ether produced
dark red crystals of 1 in 96% yield.
Complex 1 was characterized by NMR spectroscopy, IR spec-
1
troscopy, X-ray crystallography, and elemental analysis.
H
Results and discussion
NMR spectrum of 1 suggests a single species of approximate
2v symmetry in solution, based on one signal for all tert-butyl
Synthesis and characterization of the macrocyclic ligand
C
2
precursor L
13
groups and two signals for the xanthene methyl groups.
C
This study focuses on the reactivity of new dinucleating macro- spectrum is consistent with a single species of C2v symmetry
2
cyclic xanthene-bridged di(bis(imino)pyridine) L and pre- in solution, demonstrating the expected five signals in the ali-
1
12f
2
viously reported bis(imino)pyridine L . Macrocycle L is syn- phatic region and ten signals in the aromatic/imino region
thesized by the acid-catalyzed condensation of the corres- (see ESI†). However, two signals are observed for CO carbons
ponding dialdehyde with xanthene diamine linker (Scheme 1). (210.75 and 203.81 ppm), indicating that more than one type
2
,6-Pyridinedicarboxaldehyde is added to a methanol solution of CO ligand is present. The X-ray structure determination con-
2
of the xanthene linker (2,7-di-tert-butyl-9,9-dimethyl-9H- firms a tetranuclear complex of [Co (L )(CO)10] composition.
4
2
1
xanthene-4,5-diamine), with a catalytic amount of acetic Intriguingly, 1 crystallizes in two slightly different forms (1 as
acid, and refluxed for 24 hours. The product precipitates as a red plates and 1′ as red needles) that are two conformational
pale yellow solid, and is isolated in 85% yield. We note that no isomers (Fig. 1), suggesting some conformational flexibility of
template is required for the synthesis of this rigid xanthene- the macrocycle. Both structures demonstrate two discrete iden-
linked macrocycle, unlike our previously reported para-xylylene tical (crystallographic C
2
symmetry in both structures) bi-
di(bis(imino)pyridine) macrocycle. The ligand is character- metallic “[NNN]Co(CO)Co(CO) ” fragments. Each “[NNN]Co
” fragment is held together by a metal–metal
2
2
4
1
13
1
ized by H and C{ H} NMR spectroscopy and high-resolution (CO)Co(CO)
4
mass spectrometry (HRMS). NMR spectra demonstrate bond; no bridging carbonyls are observed. Co–Co bond dis-
effective C2v symmetry in solution, consistent with the rigid tances within these bimetallic units (2.80(1) Å) are slightly
“
bowl” structures observed for all complexes. Thus, four tert- longer than the Co–Co bond distance in the minor structural
butyl groups give rise to a single peak. In contrast, two isomer of Co (CO) that exhibits no bridging carbonyls
2
8
2
4
different signals are observed for the methyl groups, indicating (2.70 Å). No interaction is observed between two internal
the lack of exchange between the “inner” and the “outer” cobalts (Co2⋯Co2′ distances of 4.8 and 5.2 Å). The redox-
methyls (Mein and Meout, see Scheme 1). We note that a active nature of the [NNN] chelates is confirmed by the
1
0,12,20
changes in the imino CvN and C–C bonds (Table 1)
and by DFT calculations (see below). As the structure of 1′ is of
relatively low quality, only detailed structural parameters of 1
are presented in Table 1; however, metrics of 1′ are similar to
1
. The structure demonstrates electron delocalization along
the redox-active chelates, with elongated im–Nim bond
C
lengths of 1.330(4) and 1.314(4) Å and condensed Cim–Cpy
bond lengths of 1.426(4)/1.437(4) Å. Using the previously
2
5
described τ parameter, the geometry at both internal and
lateral cobalts in 1 is best described as distorted square
pyramidal.
The observed formation of complex 1 stands in sharp con-
trast to the anticipated di-cobalt complex, where each cobalt
2
Scheme 1 Synthesis of L .
15354 | Dalton Trans., 2018, 47, 15353–15363
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
1
2
2
2 8
Scheme 2 Reactivity of L and L with Co (CO) . For the structure of L , see Scheme 1.
2
liberates two/three CO molecules to form hypothetical distance between chelating bis(aldimino)pyridine units in L ,
2
“[Co
2
(L )(CO)2/3/4]” species (Scheme 2). We note that the reac- combined with its relative rigidity and the overall steric bulk.
tions of Co (CO) are strongly substrate-dependent, and may Because of these factors, each chelating unit reacts preferen-
2
8
lead to either partial or full substitution of carbonyl ligands. tially with one equivalent of Co
2 8
(CO) . To test this hypothesis,
In some cases, retention of the significant number of the car- we investigated the reactivity of a mononucleating analogue of
2
1
bonyl ligands and the overall dinuclear structure is observed L , L (Scheme 2), with Co (CO) . The synthesis and reactivity
2
8
1
12f
(
e.g. [(CHD)
2 2 4 2 2 6
Co (CO) ] or [(NHC) Co (CO) ] (CHD = 1,3-cyclo- of L with nickel were previously reported.
1
hexadiene, NHC = N-heterocyclic carbene) or (Fig. 2A and
B).
Mixing one equivalent of L with Co
Along the same lines, Agapie and coworkers have purple solution, from which complex 2 is isolated as a purple
recently reported that a dinucleating arene-bridged diphos- solid in 88% yield. X-ray quality crystals of 2 are obtained by
phine forms complex C in the reaction with Co (CO) recrystallization from pentane. Spectral and structural data for
Fig. 2C). In other cases, full ligand substitution at one of the 2 are consistent with [Co (L )(CO) ] composition, similar to
2 8
(CO) forms a red-
2
6,27
2
8
2
8
1
(
2
5
cobalts leads to the metal–metal bond splitting and formation the corresponding complexes of the dinucleating ligand. There
1
−
−
of ion pairs featuring [Co (CO)
4
]
as an anion (e.g. are, however, some notable differences between the systems.
I
1−
18,19,29
[
Co (terpy) ][Co (CO) ], Fig. 2D; terpy = terpyridine).
While NMR spectroscopy for 2 indicates a single species in
2
4
Interestingly, relatively small changes on the ligand periphery solution, IR spectroscopy suggests the presence of both term-
may have a significant effect on the outcome of the reaction inal and bridging carbonyls (see below for the discussion of IR
with Co (CO) . Thus, Wieghardt and coworkers showed that spectra). We also note the overall similarity between the UV-vis
2
8
while the reaction of unsubstituted terpy with Co
2
(CO)
8
spectra of 2 and 1 (see Fig. S12 and S13 in the ESI†), with the
I
1−
forms [Co (terpy)
2
][Co (CO)
4
], the reaction of 4,4′,4″-tri- difference being three peaks observed for 1 (λ = 719, 542, and
II
>
tert-butyl-2,2′:6′,2″-terpyridine
(t-terpy)
forms
[Co (t- 487 nm) and four peaks for 2 (767, 711, 556, 495 nm).
1
−
19
3 4 2
terpy) ][Co (CO) ] .
Others showed that the full release of Consistent with the spectral data, two different structural isomers
all CO ligands and splitting of Co–Co interaction is possible in (2a and 2b, Fig. 3) are observed for the solid-state structure
1
7
both mononucleating and dinucleating systems. One could of 2, co-crystallizing in the same unit cell. Isomer 2a contains
hypothesize that the observed formation of a tetranuclear square-pyramidal Co1 center ligated by the [NNN] chelate and
(
instead of dinuclear) complex in the present system indepen- two carbonyls, and distorted tetrahedral Co2 center ligated by
2
dent of the L -to-Co CO ratio results from the relatively long four carbonyls. 2b contains a square-pyramidal Co3 center,
2
8
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 15353–15363 | 15355

View Article Online
Paper
Dalton Transactions
Fig. 1 X-ray crystal structures of 1’ (left – top and side view) and 1 (right – top and side view), 50% probability ellipsoids. H atoms and co-crystal-
lized solvents are omitted for clarity. Selected bond distances (Å) and angles (°) for 1: Co2–Co1 2.798(1), Co2⋯Co2’ 4.815(1), Co1–N1 1.920(3), Co1–
N2 1.924(2), Co2–N3 1.833(3), Co1–C2 1.754(4), Co1–C1 1.773(4), Co1–C3 1.778(4), Co1–C4 1.784(4), Co2–C5 1.774(4), Co2–Co1–C2 150.2(1), C4–
Co1–C1 137.1(2).
a
Table 1 Selected bond lengths for 1, 3, and 4
CimvNim
Cim–Cpy
b
1
3
1.330(4)/1.314(4)c
1.437(4)/1.426(4)c
1.414(4)/1.415(4)
1.456(4)/1.475(4)
1.420(4)/1.424(4)
1.420(4)/1.423(4)
1.312(4)/1.309(4)
d
d
1
.288(4)/1.262(4)
1.305(4)/1.326(4)
.319(4)/1.321(4)
4
1
a
Due to relatively low C–C bond precision of structure 2, its metrics
b
2
are not discussed. Due to crystallographic C symmetry of 1, two
different bond distances were observed. Bond distances in fully
bound bis(imino)pyridine in 3. Bond distances in bis(imino)pyridine
c
d
bound through a single iminopyridine.
2 8
Fig. 2 Examples of products of the reactions of Co (CO) with various
19,26–28
ligands.
ligated by bis(aldimino)pyridine and two carbonyls, and trigo-
nal bipyramidal Co4 center, displaying five carbonyl ligands.
The major difference between the isomers is in the number of
bridging carbonyls: one for 2a and two for 2b. Accordingly, 2b lated in 57% yield. In a sharp contrast to 1 or 2, complex 3 is
manifests a significantly shortened Co⋯Co distance of 2.594(1) paramagnetic, displaying four broad proton resonances in
Å, vs. 2.703(1) Å in 2a. The electronic structure of 2b was approximately 20 ppm range. Purple crystals of 3 were
also investigated computationally (vide infra). We note that obtained from cold saturated THF solution; the structure is
Co (CO) is also known to be structurally fluxional, displaying given in Fig. 4 below. Structure determination of 3 reveals a
2
8
2
4,30,31
1
structures with and without bridging carbonyls.
Treatment of Co (CO)
the formation of purple solutions from which complex 3 is iso- similarly to terpy. Interestingly, in [Co(terpy) ][Co(CO) ] the
zwitterion pair [Co(L )
2
][Co(CO)
4
]. Thus, at 2 : 1 ratio, mono-
(CO)
8
1
2
8
with two equivalents of L leads to nucleating bis(aldimino)pyridine ligand reacts with Co
2
1
9
2
4
15356 | Dalton Trans., 2018, 47, 15353–15363
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
Fig. 4 X-ray structure of 3, 50% probability ellipsoids.
Co(CO) anion, and co-crystallized THF are omitted for clarity.
Selected bond distances (Å) and angles (°): Co1–N1 2.012, Co1–N2
.030(2), Co1–N3 1.945(2), Co1–N4 1.978(2), Co1–N5 1.810(2),
H atoms,
[
4
]
2
N3–Co1–N4 161.0(1), N1–Co1–N5 140.8(1), N2–Co1–N5 138.2(1).
symmetric species of [Co (L)(CO) ] type. At 1 : 1 ratio, however,
2
5
1
the behavior of the mononucleating ligand L diverges signifi-
cantly from the behavior of the dinucleating systems, forming
1
[
Co(L ) ][Co(CO) ], similar to the previously reported bipy/
2
4
terpy chemistry. Thus, we conclude that it is the dinucleating
2
nature of L in the present case that is the responsible for the
stability of the tetranuclear complex 1.
As demonstrated in this work (complex 4) and elsewhere in
2 n
the literature, the Co–Co bond in Co (CO) (L)8−n is capable of
undergoing heterolytic bond cleavage upon reaction with
various L-type ligands including pyridine, isocyanide, and
bond distances (Å): Co1–Co2 2.703(1), Co1–C1 2.087(6), Co2–C1 1.815(7), phosphine. Inspired by the formation of complex 3, we
Fig. 3 X-ray structure of 2a and 2b, 50% probability ellipsoids. Selected
3
2
Co1–C2 1.783(6), Co2⋯C2 2.663(6), Co3–Co4 2.594(1), Co3–C6 1.960(6),
Co4–C6 1.894(6), Co3–C7 1.814(6), Co4–C7 2.220(6).
attempted to split Co–Co bonds in 1 by treating it first with
PPh . Addition of PPh to 1 (4 equivalents) leads to the for-
mation of a green-brown solution. P NMR spectrum con-
3
3
31
tained a new signal around 40 ppm, indicating formation of a
1
cationic cobalt center exhibited pseudooctahedral coordi- new product. H NMR, however, suggested formation of a
nation, with nearly symmetric binding of terpy ligands. In con- mixture of products, which we were not able to separate. The
1
trast, [Co(L ) ][Co(CO) ] demonstrates penta-coordinate (τ = reactivity of 1 with excess acetonitrile was explored next.
2
4
0
.34) geometry. One of the [NNN] chelates is bound through Dissolution of dark red-purple 1 in acetonitrile leads to the for-
all three nitrogens, while the other is bound only through one mation of green solution, from which complex 4 was isolated
of the iminopyridines, with the second imine unbound. As by recrystallization from acetonitrile/ether mixture. Proton
3
expected, while the κ -bound [NNN] exhibits full redox deloca- NMR spectrum of complex 4 (taken in CD
3
CN) demonstrates
lization, typical CvN double and C–C single bonds consistent the expected five resonances in the aromatic region, two dis-
with a neutral chelate are observed for the unbound imine tinct methyl group resonances, and a single resonance for the
arms. The electronic structure calculations on the cation in 3 tert-butyl groups. Most significantly, the spectrum contains a
are discussed below.
signal at 1.96 ppm, whose integration signifies the presence of
The experiments above suggest significant differences four acetonitrile molecules. The X-ray structure of 4 (Fig. 5)
between the mononucleating and the dinucleating systems. At demonstrates a di-cationic, formally di-Co(I), complex balanced
1
−
−
2
1
: 2 ratio of chelating [NNN] units to Co
2
(CO)
8
, both dinucleat- by two [Co (CO)
4
]
anions. As in the other L structures
ing and mononucleating bis(aldimino)pyridines form non- reported in this manuscript (1a/1b), the two [NNN] chelates
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 15353–15363 | 15357

View Article Online
Paper
Dalton Transactions
spectra leads to several noteworthy conclusions. First, as
anticipated, only compound 2 demonstrates the peak associ-
−1
ated with a bridging carbonyl around 1800 cm (bridging car-
bonyls generally occur below 1850 cm− ). The second note-
1
worthy feature is the presence of a strong peak in the
860–1890 cm− area in the spectra of all the compounds. The
1
1
presence of a strong peak in this region is often associated
with a [Co (CO)4]− ion. In contrast, any of the known
isomers of Co (CO) (C2v/D2d/D3d), which contain Co(0), do not
feature signals in this area. Compounds 3–5 contain genuine
1
−
33
2
8
3
1
Co (CO)4]− counter-ion, and thus the presence of a single
very strong carbonyl resonance around 1870 cm is well justi-
1
−
[
−1
fied. What explains the presence of the peak around 1880/
890 cm− in the spectra of compounds 1 and 2, that contain
1
1
4
formally Co(0) oxidation states in the “[Co(CO) ]” units? One
possible explanation for this phenomenon is that the redox
non-innocence within the “[NNN]Co(CO)” fragment of dinuc-
lear structures extends to the [Co(CO)
Co(CO)
4
] anions are omitted for clarity. Selected bond distances (Å) and making it effectively [Co (CO)
4
4
] to which it is linked,
] , to probe this and other elec-
angles (°): Co1–N1 1.934(3), Co1–N2 1.936(3), Co1–N3 1.814, Co1–N4 tronic structure questions, we turned to DFT calculations.
.030(3), Co1–N5 1.928(3), N1–Co1–N2 160.0(1), N3–Co1–N5 155.0(1).
Fig. 5 X-ray structure of 4, 50% probability ellipsoids. H atoms and
[
1
−
−
2
DFT calculations
Calculations at the B3LYP-D3/6-31G(d) level of theory were per-
formed on model complexes of the di- and mono-nucleating
are nearly co-parallel, displaying interplanar angle of 6
degrees. Both cobalt centers are distorted square pyramidal,
exhibiting τ values of 0.07 and 0.18. Bis(imino)pyridine
metrics (see Table 1) and DFT calculations (below) indicate
that [NNN] chelates are reduced, and that Co centers are likely
Co(II). Similar chemistry was observed with the mononuclear
analogue: dissolution of complex 2 in acetonitrile forms
complex 5. While we were not able to obtain the structure of
complex 5, its NMR and IR characterization suggests [Co[NNN]
3
4
2
t
ligands. Because L has a Bu group meta to the iminopyri-
dine connection, we reasoned that replacing these aryl groups
with Ph would provide a similar steric environment at the Co
3
1
center (L ). The ortho Me groups in L were kept due to their
steric impact, but the para Me was removed for computational
efficiency; i.e. 2,6-dimethylphenyl was modeled instead of
4
mesityl (L ). Only half of each dinucleating complex was
3
modeled: 1 as a neutral singlet dicobalt species with L
2 4
(NCMe) ][Co(CO) ] formulation. In addition, the UV-vis spec-
denoted i, 2b as a neutral singlet dicobalt species with L
denoted iib, 3 as a cationic triplet monocobalt species with L
denoted iii, 4 as a cationic singlet monocobalt species with L
4
4
trum of 5 closely resembles that of 4. Both compounds display
three peaks in the 400–900 nm region featuring similar wave-
length and absorptivity values (Fig. S15 and S16†).
3
denoted iv, and 5 as a cationic singlet monocobalt species
4
with L denoted v (see Fig. S50† for the schematic description
IR spectroscopy
of all models). We analyze these compounds in reverse order
The IR spectra for all compounds in the carbonyl region are below due to their increasingly complicated electronic
summarized in Fig. 6. The comparison between different IR structures.
iv and v optimized to similar structures with Co–N bond
lengths that all agreed within 0.01 Å, and deviated from the
X-ray structure of 4 by at most 0.04 Å for the equitorial aceto-
nitrile ligand (Table 2). Both iv and v show intraligand bond
lengths of 1.32, 1.43, and 1.37 Å for the Cim–Nim, Cim–Cpyr, and
Cpyr–Npyr bond lengths, respectively, suggesting that the ligand
Table 2 Summary of selected bond lengths in 4, iv, and v where pyr
and im stand for pyridine and imine
Bond
4
iv
v
Co–NCMeax
Co–NCMeeq
2.030
1.928
1.814
1.934
1.936
2.000
1.889
1.833
1.945
1.945
2.003
1.893
1.833
1.942
1.943
Fig. 6 IR spectra of compounds 1–5 in 1700–2300 cm− range. CC oo –– NN pi my r
1
ATR-FTIR was used to record all IR spectra of powdered samples.
Co–Nim′
15358 | Dalton Trans., 2018, 47, 15353–15363
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
1
+
is a radical anion. Visualization of the spin density for iv con- whereas our modeling of [Co(L )] by itself, which should
firms this (see ESI† for the spin density of v), which implies more accurately represent the form in solution, predicts the
the bis(acetonitrile) complex is best described as having a low- triplet to be favored consistent with the experimentally
LS
II
spin cobalt(II) ion ( Co ) antiferromagnetically coupled to an observed paramagnetism. Both iiiS=1 and iiiS=0 are predicted to
anionic bis(imino)pyridine ligand radical (Fig. 7). This assign- feature a monoanionic terdentate bis(imino)pyridine radical
LS
II
HS
II
LS
II
ment of Co is consistent with the short equitorial Co–N antiferromagnetically coupled to
Co and
Co ions,
bonds and elongated Co–NCMeax bond length. respectively, as shown by the spin density plots in Fig. 8. The
iii was optimized as a cationic triplet due to the paramag- bidentate bis(imino)pyridine is predicted to be neutral in
1
netic broadening observed in the H NMR spectra of 3. As each structure, as evidenced by a lack of spin density on that
Table 3 shows, however, the optimized Co–N bond lengths for ligand.
this structure are inconsistent with the X-ray structure,
i and iib were optimized as neutral species differing in the
especially Co–N (one of the Co–Nim bonds in the terdentate aryl substituents on the bis(imino)pyridine ligand. We were
4
ligand) that is predicted to be too long by ∼0.4 Å. This value unable to reproduce the geometry of 2a in these bimetallic
far exceeds the typical structural error in DFT and, along with model calculations. The crystallographic bond lengths of 1 are
the asymmetric distortion of the Co–Nim bonds in the triden- reproduced well by i (±0.02 Å) with the exception of the Co–Co
tate ligand, suggests a high-spin electron configuration at the distance that is ∼0.1 Å too short (see Table S2†). Intraligand
II
Co center. Thus, we also optimized iii as a singlet and much bond lengths in i are 1.32, 1.43, and 1.37 Å for Cim–Nim
,
better structural agreement is observed with a maximum devi-
Cim–Cpyr, and Cpyr–Npyr. Slightly larger discrepancies are seen
ation of 0.03 Å in the Co–Npyr bond. This structure is predicted between 2b and iib, with deviations up to 0.04 Å except: (i) two
−
1
to be higher in free energy than the triplet by 6.0 kcal mol
of the bridging Co–carbonyl bonds have errors of 0.08 and
−
1
(
ΔE(SCF) = 1.5 kcal mol ). It is important to note that our 0.14 Å, and (ii) better agreement in the Co⋯Co interaction that
2
DFT modeling only considers the cation of the [Co(L )
2
]
is only 0.03 Å too short (see Table S3†). Intraligand bond
[
Co(CO) ] ion pair. It is possible that intermolecular interactions lengths in iib are 1.32, 1.42, and 1.37 Å for Cim–Nim, Cim–Cpyr
,
4
cause the singlet to be lower in energy in the solid-state, and Cpyr–Npyr. These intraligand bond lengths, which are
similar to those in iv and v, suggest that the bis(imino)pyridine
ligand is a radical anion. However, closed-shell wavefunctions
were found for both i and iib (after starting with open-shell,
broken symmetry guess wavefunctions). To address the
charges of the metal centers, the Natural Bonding Orbitals
3
5
were generated to analyze the Natural Atomic Charges. Co1
3
and Co2 (Co1 bound to L , see Fig. 1) have charges of 0.42 and
4
−
0.50 in i, and Co3 and Co4 (Co3 bound to L , see Fig. 3) have
charges of 0.47 and −0.52 in iib; these values that differ by ∼1
suggest that the cobalt bound to the iminopyridine is in a
higher oxidation state than the one that is only bound to car-
bonyls. Given the neutral overall charge, and the anionic
ligand charge inferred from the intraligand bond lengths, this
I
would lead to assignment of the oxidation states as Co and
0
I−
Co , respectively. However, a Co was inferred from the experi-
mental carbonyl stretches. i is calculated to have a frequency at
913 cm− that corresponds to stretching of the carbonyls on
1
1
3
6
4
the Co(CO) fragment. iib is calculated to have frequencies at
1
806 and 1845 cm− that correspond to the asymmetric and
1
Fig. 7 Spin density isosurface plot (iso = 0.002 au) for iv. Blue and
white represent excess of α and β density.
symmetric stretches of the bridging carbonyls, respectively,
and the Co(CO) fragment has stretches between
978–2034 cm− . Thus, it seems reasonable that the observed
3
1
1
−
1
stretch at 1802 cm in 2 can be attributed to bridging carbo-
nyls. However, the low frequency mode in i suggests formu-
1
Table 3 Summary of selected bond lengths in 3, iiiS=1, and iiiS=0. N and
II
I−
N
2
are in the bidentate ligand and N
3
–N
5
are in the tridentate ligand, as lation as Co /Co instead, even though no evidence for spin
polarization at the Co bound to the bis(imino)pyridine is
shown in Fig. 4
observed. The Natural Atomic Charges are not capable of dis-
Bond
3
iiiS=1
iiiS=0
I
0
II
1−
tinguishing between these valence limits (Co /Co ↔ Co /Co
)
Co–N
Co–N
Co–N
Co–N
Co–N
1
2
3
4
5
2.012
2.030
1.945
1.978
1.810
2.050
2.069
2.084
2.365
1.928
1.992 as they do not correspond to oxidation states.
2.047
2 8
Bridged and unbridged Co (CO) have been studied exten-
sively to determine what kind of theoretical analysis can deter-
1.944
1.998
1.838 mine the bonding (or lack thereof) between the metal
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 15353–15363 | 15359

View Article Online
Paper
Dalton Transactions
Fig. 8 Spin density isosurface plots (iso = 0.002 au) for iiiS=1 (left) and iiiS=0 (right). Blue and white represent excess of α and β density.
Table 4 Topological properties at the bond critical point in bridged Co
kinetic energy density, and V(r) is the potential energy density
2
(CO)
8
, unbridged Co
2
(CO)
8
, i, and iib. ρ is the electron density, G(r) is the
2
2
2
2
2
Species
bcp
ρ (10 au)
∇ ρ (10 au)
G(r) (10 au)
V(r) (10 au)
G(r)/ρ
|V(r)|/G(r)
Unbridged Co
Bridged Co (CO)
2 8
i
iib
2
(CO)
8
(3,−1)
(3,+1)
(3,−1)
(3,+1)
3.84
4.69
3.23
4.30
1.69
15.3
5.25
12.9
1.25
4.56
1.77
3.68
8.28
7.26
4.63
4.60
0.33
0.97
0.55
0.86
1.7
1.2
1.3
1.1
3
7
centers. Topological analyses were performed for the pre- and the value for i of 1.3 is only slightly larger and much
viously characterized Co (CO) species, i, and iib for a compari- smaller than the value for unbridged Co (CO) of 1.7.
2
8
2
8
3
8
son at a common level of theory. Table 4 summarizes these Collectively, we interpret these values to indicate that a cobalt–
results. For Co (CO) , important differences of the bond criti- cobalt bond exists in i but not iib, however the difference in
2
8
cal point (bcp) between the bonded (unbridged) and bonding is smaller than in Co
2 8
(CO) . A source function ana-
unbonded (bridged) species include: (i) the nature of the bcp, lysis of these bond critical points also supports this interpret-
(
ii) lower density in the bonded species, (iii) much smaller ation and the difference in oxidation state between the Co
3
7d,39
Laplacian values for the bonded species, (iv) lower kinetic centers (see ESI†).
energy per electron (G(r)/ρ), and (v) a larger |V(r)|/G(r) ratio.
The bcp and electron density are similar for i and iib, but
differences are observed in the other metrics. Whereas nearly
3
7d
Reactivity of tetranuclear, dinuclear, and mononuclear cobalt
complexes in alkyne cyclotrimerization
an order of magnitude difference was observed for the Having synthesized cobalt complexes 1–5, we have conducted
Laplacian in Co (CO) , a difference of only 2–3 times (0.0525 initial screening of their catalytic reactivity. We specifically
2
8
vs. 0.129) is seen in i and iib. The kinetic energy per electron focused on cyclotrimerization of ethyl propiolate, as (1)
4
0
in both i and iib is larger than unbonded Co
2 8
2 8
(CO) but smaller Co (CO) is known to catalyze alkyne cyclotrimerization; (2)
than bonded Co (CO) , with an appreciable difference of ∼0.3 it was previously shown that bimetallic complexes exhibit
2
8
4
1
for i and iib (vs. 0.6 for dicobalt octacarbonyl). Finally, the cooperative reactivity in cyclotrimerizations; (3) it is often
V(r)|/G(r) ratio that Gatti and Lasi recommended as a useful considered a superior cyclotrimerization substrate compared
metric for distinguishing M–M bonded and non-bonded com- with more electron-rich or bulkier substrates. The results are
|
4
1
pounds does not approach the value of 2, even for unbridged summarized in Table 5. Reactions were carried out under
3
7d
Co
2
(CO)
8
, that they observed.
However, the value of this three different sets of conditions: 40 °C in CD Cl , 80 °C in
2 2
ratio for iib is quite similar to bridged Co (CO) (1.2 vs. 1.1), C D (toluene-d ), and 70 °C in CD CN. The reactivity studies
2
8
7
8
8
3
15360 | Dalton Trans., 2018, 47, 15353–15363
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
2 8
Table 5 Reactivity of complexes 1–5 and Co (CO) in cyclotrimeriza-
tion of ethyl propiolate under three different sets of conditions
Summary and conclusions
2 8
In summary, we have investigated the capability of Co (CO) to
serve as a direct metal precursor for the formally low oxidation
state complexes of cobalt with redox-active mononucleating
and dinucleating bis(imino)pyridine ligands. Formation of
tetranuclear, dinuclear and mononuclear complexes was
observed. Spectroscopic, structural, and computational studies
Cat.
Solv., T
CD Cl , 40 °C
mol%
% Conv.
%1,2,4/%1,3,5 revealed a variety of products including metal–metal bonded
penta-coordinate complex, mono- and dicarbonyl bridged
1
2
4
5
3
2
2
1
2
1
2
2
2
20
28
21
24
12
61
16/4
16/12
14/9
14/10
6/6
dimers, and mononuclear/dinuclear complexes in which
cobalt centers are isolated and feature labile ligands (NCMe)
coordinated to the metal. The species with interacting Co
centers reported here show a similar trend to Co (CO) with
Co
2
(CO)
8
35/26
2
8
less Co–Co bonding as the number of bridging carbonyls
increases, despite shorter distances between the metals, even
though different oxidation states are involved (i.e. no longer
1
2
4
5
3
Co
C
7
D
8
, 80 °C
1
2
1
2
2
2
66
71
65
49
51
99
38/28
41/30
38/27
29/20
28/23
63/36
9
two d centers). The complexes with a well-defined mono-
metallic environment, even if it was embedded in a dinuclear
2
(CO)
8
8
II
complex, were best characterized as Co with a bis(iminopyri-
4
5
Co
CD
3
CN, 70 °C
1
2
2
12
16
17
7/5
7/9
12/5
dine) ligand radical. Species 1 and 2 exhibit intraligand bond
lengths suggesting a ligand radical as well, however, neither
the spectroscopy nor the calculations clearly point to a particu-
2
(CO)
I
0
II
I−
lar set of oxidation states (Co /Co ↔ Co /Co ) for the metals.
Overall, this work demonstrates promise of Co (CO) in its
2
8
reveal that while complexes 1–5 are competent cyclotrimeriza- capacity to serve as a precursor for multimetallic complexes
tion catalysts, their reactivity is inferior to the reactivity of with redox-active ligands.
Co
plexes 1, 2, 4, and 5 exhibit low, fairly similar conversion rates
ranging between 20–28%. In contrast, Co (CO) exhibits 60%
conversion. It is also worth noting that complex 3 demon-
2 8
(CO) . At relatively low reaction temperature (40 °C), com-
2
8
Conflicts of interest
strates the lowest reactivity at room temperature, consistent There are no conflicts to report.
with the structure lacking vacant positions available for cataly-
sis. At higher temperature (80 °C in toluene-d ) all complexes
8
exhibit similar reactivity (50–70% conversion), including
complex 3, likely indicating loss of one of the bis(imino)pyri-
Acknowledgements
2 8
dines. Again, Co (CO) is more reactive and exhibits nearly SG is grateful to National Science Foundation (NSF) for current
quantitative conversion at this temperature. Finally, we also support under grant number CHE-1349048. RLL thanks the
tested the reactivity of acetonitrile adducts in this solvent. The NSF for computational resources (CHE-1039925 to the
reactivity of the Co
2
(CO)
8
pre-catalyst was also tested in aceto- Midwest
Undergraduate
Computational
Chemistry
nitrile, although it is likely that Co (CO) is also converted into Consortium). Compound characterization was carried out at
2
8
3
2,33
a zwitterion in this solvent.
exhibited low reactivity, including Co
reaction inhibition by coordinating solvent.
While it is clear that complexes 3–5 demonstrate inferior
reactivity compared with Co (CO) due to the lack of available
In acetonitrile, all complexes Lumigen Instrument Center at Wayne State University. We
2
(CO) , likely due to the thank Thilini Hollingsworth for experimental assistance.
8
2
8
References
positions, it is less clear why complexes 1 and 2 are less reac-
tive. One possible reason is the significantly increased steric
1 A. L. Gavrilova and B. Bosnich, Chem. Rev., 2004, 104, 349–
383.
bulk around the metal(s), compared with Co
detrimental to cyclotrimerization that requires several vacant
positions. It is also possible, however, that while Co (CO)
exhibits a cooperative bimetallic effect in cyclotrimeriza-
2 8
(CO) , which is
2 (a) N. Mankad, Chem. Commun., 2018, 54, 1291–1302;
(b) I. G. Powers and C. Uyeda, ACS Catal., 2017, 7, 936–958;
(c) P. Buchwalter, J. Rosé and P. Braunstein, Chem. Rev.,
2015, 115, 28–126; (d) J. F. Berry, J. Chem. Sci., 2015, 127,
209–214; (e) C. M. Thomas, Comments Inorg. Chem., 2011,
32, 14–38; (f) R. M. Haak, S. J. Wezenberg and A. W. Kleij,
Chem. Commun., 2010, 46, 2713–2723.
2
8
4
1b,c
tion,
complexes 1 and 2 disproportionate in solution into
zwitterions and therefore catalyze this reaction in a monome-
tallic fashion. Our future endeavors will focus on other cata-
lytic applications of these systems.
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 15353–15363 | 15361

View Article Online
Paper
Dalton Transactions
3
(a) F. Gloaguen and T. B. Rauchfuss, Chem. Soc. Rev., 2009,
8, 100–108; (b) S. C. Lee, W. Lo and R. H. Holm, Chem.
Rev., 2014, 114, 3579–3600; (c) S. Friedle, E. Reisner and
S. J. Lippard, Chem. Soc. Rev., 2010, 39, 2768–2779.
For selected recent examples, see: (a) D.-H. Kwon,
M. Proctor, S. Mendoza, C. Uyeda and D. H. Ess, ACS
Catal., 2017, 7, 4796–4804; (b) M. J. Behlen, Y.-Y. Zhou,
T. J. Steiman, S. Pal, D. R. Hartline, M. Zeller and
C. Uyeda, Dalton Trans., 2017, 46, 5493–5497;
2007, 46, 7880–7889; (b) C. C. Lu, E. Bill, T. Weyhermüller,
E. Bothe and K. Wieghardt, J. Am. Chem. Soc., 2008, 130,
3181–3197; (c) A. Bheemaraju, J. W. Beattie, R. L. Lord,
P. D. Martin and S. Groysman, Chem. Commun., 2012, 48,
9595–9597; (d) A. Bheemaraju, J. W. Beattie, E. G. Tabasan,
P. D. Martin, R. L. Lord and S. Groysman, Organometallics,
2013, 32, 2952–2962; (e) A. Bheemaraju, R. L. Lord,
P. Müller and S. Groysman, Organometallics, 2012, 31,
2120–2123; (f) B. R. Reed, M. Yousif, R. L. Lord,
M. McKinnon, J. Rochford and S. Groysman,
Organometallics, 2017, 36, 582–593.
3
4
(
c) K. T. Horak, A. Velian, M. W. Day and T. Agapie, Chem.
Commun., 2014, 50, 4427–4429; (d) D. L. J. Broere,
D. K. Modder, E. Blokker, M. A. Siegler and J. I. van der 13 (a) E. W. Abel and F. G. A. Stone, Q. Rev., Chem. Soc., 1970,
Vlugt, Angew. Chem., Int. Ed., 2016, 55, 2406–2410;
e) R. L. Hollingsworth, A. Bheemaraju, N. Lenca,
24, 498–552; (b) L. Mond, H. Hirtz and M. D. Cowa,
J. Chem. Soc. Trans., 1910, 97, 798.
(
R. L. Lord and S. Groysman, Dalton Trans., 2017, 46, 5605– 14 For selected recent examples, see: (a) G. G. Melikyan and
5
616; (f) C. Camp, D. Toniolo, J. Andrez, J. Pécauta and
M. Mazzanti, Dalton Trans., 2017, 46, 11145–11148;
g) S. Suseno, K. T. Horak, M. W. Day and T. Agapie,
B. Anker, Organometallics, 2015, 34, 4194–4197;
(b) A. C. Coleman, C. Long, A. Meetsma, B. L. Feringa,
W. R. Browne and M. T. Pryce, Dalton Trans., 2009, 7885–
7887; (c) T. León and E. Fernández, Chem. Commun., 2016,
52, 9363–9366; (d) R. Imayoshi, H. Tanaka, Y. Matsuo,
M. Yuki, K. Nakajima, K. Yoshizawa and Y. Nishibayashi,
Chem. – Eur. J., 2015, 21, 8905–8909; (e) M. A. Henderson,
J. Luo, A. Oliver and J. S. McIndoe, Organometallics, 2011,
30, 5471–5479; (f) S. Karabiyikoglu, B. A. Boon and
C. A. Merlic, J. Org. Chem., 2017, 82, 7732–7744;
(g) X. Creary, J. Org. Chem., 2018, 83, 136–144;
(h) F. Hoffmann, J. Wagler and G. Roewer, Organometallics,
2014, 33, 5622–5625; (i) A. Dahan and M. Portnoy, Chem.
Commun., 2002, 2700–2701; ( j) Y. Li, J. A. Krause and
H. Guan, Organometallics, 2018, 37, 2147–2158.
(
Organometallics, 2013, 32, 6883–6886; (h) S. Fortier, J. J. Le
Roy, C.-H. Chen, V. Vieru, M. Murugesu, L. F. Chibotaru,
D. J. Mindiola and K. G. Caulton, J. Am. Chem. Soc., 2013,
1
35, 14670–14678.
V. Lyaskovskyy and B. de Bruin, ACS Catal., 2012, 2, 270–
79.
S. Enthaler, K. Junge and M. Beller, Angew. Chem., Int. Ed.,
008, 47, 3317–3321.
5
6
2
2
7
8
K. G. Caulton, Eur. J. Inorg. Chem., 2012, 435–443.
Q. Knijnenburg, S. Gambarotta and P. H. M. Budzelaar,
Dalton Trans., 2006, 5442–5448.
9
For selected references, see: (a) S. Fortier, J. J. Le Roy,
C.-H. Chen, V. Vieru, M. Murugesu, L. F. Chibotaru, 15 For selected recent examples, see: (a) L. Wang, Z. Zeng,
D. J. Mindiola and K. G. Caulton, J. Am. Chem. Soc., 2013,
35, 14670–14678; (b) S. Fortier, O. González-del Moral,
C.-H. Chen, M. Pink, J. J. Le Roy, M. Murugesu,
C. Ma, Y. Liu, M. Giroux, M. Chi, J. Jin, J. Greeley and
C. Wang, Nano Lett., 2017, 17, 3391–3395; (b) C. Vollmer,
R. Thomann and C. Janiak, Dalton Trans., 2012, 41, 9722–
9727; (c) C.-Y. Kao, B. Li, Y. Lu, J.-W. Yoo and A. J. Epstein,
J. Mater. Chem. C, 2014, 2, 6171–6176; (d) Z. Wang, X. Hou,
J. Shen and T. Li, RSC Adv., 2016, 6, 89503–89509;
(e) J. A. Delgado, C. Claver, S. Castillón, D. Curulla-Ferré
and C. Godard, ACS Catal., 2015, 5, 4568–4578.
1
D. J. Mindiola and K. G. Caulton, Chem. Commun., 2012,
4
8, 11082–11084; (c) T. D. Manuel and J.-U. Rohde, J. Am.
Chem. Soc., 2009, 131, 15582–15583; (d) B. R. Reed,
S. A. Stoian, R. L. Lord and S. Groysman, Chem. Commun.,
2
015, 51, 6496–6499; (e) D. Zhu, I. Thapa, I. Korobkov,
S. Gambarotta and P. H. M. Budzelaar, Inorg. Chem., 2011, 16 (a) A. Franken, T. D. McGrath and F. G. A. Stone, Dalton
5
0, 9879–9887; (f) T. W. Meyers and L. A. Berben, Chem.
Trans., 2009, 7353–7355; (b) S. Wolf and C. Feldmann,
Dalton Trans., 2012, 41, 8455–8459; (c) A. Franken,
T. D. McGrath and F. G. A. Stone, Organometallics, 2010, 29,
2234–2247; (d) H. D. Park, M. Dincă and Y. Román-
Leshkov, ACS Cent. Sci., 2017, 3, 444–448.
Commun., 2013, 49, 4175–4177; (g) S. C. Bart, K. Chlopek,
E. Bill, M. W. Bouwkamp, E. Lobkovsky, F. Neese,
K. Wieghardt and P. J. Chirik, J. Am. Chem. Soc., 2006, 128,
1
3901–13912; (h) M. W. Bouwkamp, S. Bart, E. J. Hawrelak,
R. J. Trovitch, E. Lobkovsky and P. J. Chirik, Chem. 17 (a) A. A. Zolotukhin, M. P. Bubnov, A. S. Bogomyakov,
Commun., 2005, 3406–3408.
N. A. Protasenko, G. K. Fukin, I. D. Grishin and
V. K. Cherkasov, Inorg. Chim. Acta, 2016, 440, 16–20;
(b) A. L. Smith, L. A. Clapp, K. I. Hardcastle and J. D. Soper,
Polyhedron, 2010, 29, 164–169; (c) N. A. Protasenko,
A. I. Poddel’sky, A. S. Bogomyakov, G. K. Fukin and
V. K. Cherkasov, Inorg. Chem., 2015, 54, 6078–6080;
(d) S. Bin-Salamon, S. H. Brewer, E. C. Depperman,
S. Franzen, J. W. Kampf, M. L. Kirk, R. K. Kumar, S. Lappi,
K. Peariso, K. E. Preuss and D. A. Shultz, Inorg. Chem.,
2006, 45, 4461–4467.
1
0 (a) S. C. Bart, E. Lobkovsky and P. J. Chirik, J. Am. Chem.
Soc., 2004, 126, 13794; (b) S. C. Bart, E. Lobkovsky, E. Bill,
K. Wieghardt and P. J. Chirik, Inorg. Chem., 2007, 46, 7055–
7
063; (c) A. M. Tondreau, C. Milsmann, E. Lobkovsky and
P. J. Chirik, Inorg. Chem., 2011, 50, 9888–9895.
1 L. Mond, C. Langer and F. Quincke, J. Chem. Soc. Trans.,
1
1
890, 57, 749–753.
1
2 For selected examples, see: (a) C. C. Lu, E. Bill,
T. Weyhermüller, E. Bothe and K. Wieghardt, Inorg. Chem.,
15362 | Dalton Trans., 2018, 47, 15353–15363
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
1
1
2
8 I. Wender, H. W. Sternberg and M. Orchin, J. Am. Chem.
Soc., 1952, 74, 1216–1219.
(b) C. Lee, W. Yang and R. G. Parr, Phys. Rev. B: Condens.
Matter Mater. Phys., 1988, 37, 785–789; (c) A. D. Becke, Phys.
Rev. A, 1988, 38, 3098–3100; (d) A. D. Becke, J. Chem. Phys.,
1993, 98, 1372–1377; (e) P. J. Stephens, F. J. Devlin,
C. F. Chabalowski and M. J. Frisch, J. Phys. Chem., 1994, 98,
11623–11627; (f) S. Grimme, J. Antony, S. Ehrlich and
H. Krieg, J. Chem. Phys., 2010, 132, 154104; (g) S. Grimme,
S. Ehrlich and L. Goerigk, J. Comput. Chem., 2011, 32,
1456–1465.
9 J. England, E. Bill, T. Weyhermüller, F. Neese, M. Atanasov
and K. Wieghardt, Inorg. Chem., 2015, 54, 12002–12018.
0 A. C. Bowman, C. Milsmann, C. C. H. Atienza,
E. Lobkovsky, K. Wieghardt and P. J. Chirik, J. Am. Chem.
Soc., 2010, 132, 1676–1684.
2
1 (a) T. S. Hollingsworth, R. L. Hollingsworth, T. Rosen and
S. Groysman, RSC Adv., 2017, 7, 41819–41829;
(
b) T. S. Hollingsworth, R. L. Hollingsworth, R. L. Lord and 35 E. D. Glendening, A. E. Reed, J. E. Carpenter and
S. Groysman, Dalton Trans., 2018, 47, 10017–10024. F. Weinhold, NBO Version 3.1.
2 J. W. Beattie, D. J. SantaLucia, D. W. White and 36 The value for B3LYP/6-31G(d) of 0.952 was used for the
2
2
2
2
2
S. Groysman, Inorg. Chim. Acta, 2017, 460, 8–16.
3 S. Takano, Y. Takeuchi, D. Takeuchi and K. Osakada,
Chem. Lett., 2014, 43, 465–467.
4 T. Y. Garcia, J. C. Fettinger, M. M. Olmstead and
A. L. Balch, Chem. Commun., 2009, 7143–7145.
5 A. W. Addison, N. T. Rao, J. Reedijk, J. van Rijn and
G. C. Verschoor, J. Chem. Soc., Dalton Trans., 1984, 1349–1356.
6 (a) G. Winkhaus and G. Wilkinson, J. Chem. Soc., 1961, 602;
scaling factor. I. M. Alecu, J. Zheng, Y. Zhao and
D. G. Truhlar, J. Chem. Theory Comput., 2010, 6, 2872–2887.
37 (a) C. Bo, J.-P. Sarasa and J.-M. Poblet, J. Phys. Chem., 1993,
97, 6362–6366; (b) P. Macchi and A. Sironi, Coord. Chem.
Rev., 2003, 238–239, 383–412; (c) O. Kluge, M. Finger and
J. Reinhold, Inorg. Chem., 2005, 44, 6494–6496; (d) C. Gatti
and D. Lasi, Faraday Discuss., 2007, 135, 55–78;
(e) R. Ponec, G. Lendvay and J. Chaves, J. Comput. Chem.,
2008, 29, 1387–1398; (f) J. C. Green, M. L. H. Green and
G. Parkin, Chem. Commun., 2012, 48, 11481–11503;
(g) D. Tiana, E. Francisco, P. Macchi, A. Sironi and
A. M. Pendás, J. Phys. Chem. A, 2015, 119, 2153–2160.
38 (a) R. F. W. Bader, Chem. Rev., 1991, 91, 893–928; (b) T. Lu
and F. Chen, J. Comput. Chem., 2011, 33, 580–592.
(b) E. O. Fischer, P. Kuzel and H. P. Fritz, Z. Naturforsch., B,
1
961, 16b, 138; (c) J. Newman and A. R. Manning, J. Chem.
Soc., Dalton Trans., 1973, 1593–1601.
2
2
2
3
3
7 H. Van Rensburg, R. P. Tooze, D. F. Foster and
A. M. Z. Slawin, Inorg. Chem., 2004, 43, 2468–2470.
8 K. T. Horak, A. Velian, M. W. Day and T. Agapie, Chem.
Commun., 2014, 50, 4427–4429.
39 (a) R. F. W. Bader and C. Gatti, Chem. Phys. Lett., 1998, 287,
233–238; (b) C. Gatti, F. Cargnoni and L. Bertini, J. Comput.
Chem., 2003, 24, 422–436.
9 H. Behrens and W. Z. Aquila, Anorg. Allg. Chem., 1967, 356,
8–21.
0 G. G. Sumner, H. P. Klug and L. E. Alexander, Acta 40 (a) W. Hübel and C. Hoogzand, Chem. Ber., 1960, 93, 103–
Crystallogr., 1964, 17, 732–742.
115; (b) U. Krüerke and W. Hübel, Chem. Ber., 1961, 94,
2829–2856; (c) C. Hoogzand and W. Hübel, Tetrahedron
Lett., 1961, 637–643; (d) M. A. Bennett and P. B. Donaldson,
Inorg. Chem., 1978, 17, 1995–2000.
1 (a) R. L. Sweany and T. L. Brown, Inorg. Chem., 1977, 16,
4
2
15–421; (b) G. Aullón and S. Alvarez, Eur. J. Inorg. Chem.,
001, 3031–3038.
3
2 For selected references, see: (a) F. Calderazzo, F. Masi and 41 (a) K. Yamamoto, H. Nagae, H. Tsurug and K. Mashima,
G. Pampaloni, Inorg. Chim. Acta, 2009, 362, 4291–4297;
b) W. Hieber and R. Wiesboeck, Chem. Ber., 1958, 91,
230–1234; (c) A. Sacco, Gazz. Chim. Ital., 1953, 83, 632;
d) A. Sisak and L. Markó, J. Organomet. Chem., 1987, 330,
01–206; (e) E. Fuhrmann, Chem.-Ztg., 1988, 112, 295.
3 L. T. Mika, R. Tuba, I. Tóth, S. Pitter and I. T. Horváth,
Organometallics, 2011, 30, 4751–4764.
4 See ESI† for full computational details: (a) S. H. Vosko,
L. Wilk and M. Nusair, Can. J. Phys., 1980, 58, 1200–1211;
Dalton Trans., 2016, 45, 17072–17081; (b) R. J. Baxter,
G. R. Knox, P. L. Pauson and M. D. Spicer, Organometallics,
1999, 18, 197; (c) R. J. Baxter, G. R. Knox, J. H. Moir,
P. L. Pauson and M. D. Spicer, Organometallics, 1999, 18,
206–214; (d) K. Yamamoto, H. Tsurugi and K. Mashima,
Organometallics, 2016, 35, 1573–1581; (e) H.-Z. Chen,
S.-C. Liu, C.-H. Yen, J.-S. K. Yu, Y.-J. Shieh, T.-S. Kuo and
Y.-C. Tsai, Angew. Chem., Int. Ed., 2012, 51, 10342; (f) S. Pal
and C. Uyeda, J. Am. Chem. Soc., 2015, 137, 8042–8045.
(
1
(
2
3
3
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 15353–15363 | 15363