Electrochemical Investigation of Low-Valent Multiply MM Bonded Group VI Dimers: A Standard Chemical Reduction Leads to an Unexpected Product

Article abstract of DOI:10.1021/acs.organomet.0c00533

The synthesis and investigation of multimetallic complexes bearing metal-metal bonds have been significantly advanced over the last 30 years with exciting applications across many fields. While the outlook for synthetically innovative metal-metal species is promising, the reactivity of the classic M2X6 alkyl dimers (where M = W, Mo; X = CH2CMe3, CH2SiMe3) is still not fully understood. In particular, the redox characteristics of these complexes have not been explored by cyclic voltammetry (CV). Herein, W2(CH2CMe3)6 and Mo2(CH2CMe3)6 (1 and 2, respectively) were characterized using CV to show differing electrochemical behavior between the two species. Stoichiometric reactions, guided by the results of the CV experiments, led to the isolation of an alkylidyne-bridged dimer bearing a rare structural motif, W2(μ-CCMe3)2(CH2CMe3)4 (3). Single-crystal X-ray diffraction (SCXRD) and 1H and 13C nuclear magnetic resonance spectroscopy (NMR) were used to establish the connectivity of the structure in the solid and solution phases. Density functional theory (DFT) calculations on 1-3 and the previously reported W2(CSiMe3)2(CH2SiMe3)4 (4) were used to rationalize the reactivity of 1 to form 3 and the structural differences between 3 and 4. Complexes 3 and 4 constitute a case in which the stereoelectronic properties of the silyl neopentyl ligand can affect differences in structure and bonding.

Full text of DOI:10.1021/acs.organomet.0c00533

pubs.acs.org/Organometallics
Article
Electrochemical Investigation of Low-Valent Multiply MM Bonded
Group VI Dimers: A Standard Chemical Reduction Leads to an
Unexpected Product
Alon Chapovetsky, Prajay Patel, Cong Liu,* Alfred P. Sattelberger,* David M. Kaphan,*
and Massimiliano Delferro*
Cite This: https://dx.doi.org/10.1021/acs.organomet.0c00533
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: The synthesis and investigation of multimetallic
complexes bearing metal−metal bonds have been significantly
advanced over the last 30 years with exciting applications across
many fields. While the outlook for synthetically innovative metal−
metal species is promising, the reactivity of the classic M2X6 alkyl
dimers (where M = W, Mo; X = CH2CMe3, CH2SiMe3) is still not
fully understood. In particular, the redox characteristics of these
complexes have not been explored by cyclic voltammetry (CV).
Herein, W2(CH2CMe3)6 and Mo2(CH2CMe3)6 (1 and 2,
respectively) were characterized using CV to show differing
electrochemical behavior between the two species. Stoichiometric reactions, guided by the results of the CV experiments, led to the
isolation of an alkylidyne-bridged dimer bearing a rare structural motif, W2(μ-CCMe3)2(CH2CMe3)4 (3). Single-crystal X-ray
1
diffraction (SCXRD) and H and 13C nuclear magnetic resonance spectroscopy (NMR) were used to establish the connectivity of
the structure in the solid and solution phases. Density functional theory (DFT) calculations on 1−3 and the previously reported
W2(CSiMe3)2(CH2SiMe3)4 (4) were used to rationalize the reactivity of 1 to form 3 and the structural differences between 3 and 4.
Complexes 3 and 4 constitute a case in which the stereoelectronic properties of the silyl neopentyl ligand can affect differences in
structure and bonding.
microscopic, and computational techniques established that
the surface binding was the same for both materials (Figure
1A), the two exhibited different reaction profiles upon
exposure to propylene. The precatalyst 1-SiO2 underwent an
induction period followed by stable activity, consistent with the
formation of an active alkylidene species (Figure 1B, red). In
contrast, 2-SiO2 showcased a maximum conversion at the
onset of the experiment, followed by a rapid deactivation
(Figure 1B, blue).17
While the mechanism of activation of the supported
complexes remains unclear, we hypothesized that the
generation of a catalytically active M−C multiply bonded
species might involve a redox event. Therefore, we decided to
further characterize complexes 1 and 2 using cyclic
voltammetry (CV) to better understand their behavior under
reductive or oxidative conditions. The results from the CV
INTRODUCTION
■
Multimetallic complexes bearing singly and multiply bonded
metal ions play a major role in both biogenic and
anthropogenic chemistries.1,2 The study of these complexes
ushered in many breakthroughs in a broad range of fields such
as catalysis, photochemistry, and single-molecule magnets, with
significant strides made over the course of the last 30 years.3−9
Synthetically elusive targets, including but not limited to the
first quintuply bonded transition-metal complex Ar′Cr2Ar′ (Ar′
= C6H3-2,6(C6H3-2,6-iPr2)2), a series of complexes bearing a
uranium−M bond (where M = Ni, Pd, Pt), and collections of
Fe3 and Mn4 clusters with unexpected magnetic properties,
have all been prepared and investigated with respect to their
reactivity.6,10−13 While the outlook for synthetically novel
metal−metal-bonded complexes is promising, there are still
gaps in our understanding of the structure and reactivity of
some of the most prototypical examples of metal−metal
multiply bonded complexes, particularly those of the form
M2(CH2CMe3)6 where M = W (1), Mo (2).14−16
Received: August 10, 2020
In recent work, our group has found that, upon grafting the
catalytically inert complexes 1 and 2 onto silica, the new
materials 1-SiO2 and 2-SiO2 became active toward the
catalytic disproportionation of propylene to ethylene and 2-
butene (Figure 1B,C).17 While a combination of spectroscopic,
https://dx.doi.org/10.1021/acs.organomet.0c00533
© XXXX American Chemical Society
Organometallics XXXX, XXX, XXX−XXX
A

Organometallics
pubs.acs.org/Organometallics
Article
complex 1 exhibits a single irreversible oxidation feature at
−2.21 V. These observations can be rationalized via an EC type
process involving an electrochemical reduction (E) followed by
a chemical rearrangement (C) to yield a new species with
different electrochemical properties. Variable-scan-rate experi-
ments on complex 1 show the disappearance of the oxidative
feature at slow scan rates (100−10 mV/s) consistent with the
diffusion of a new species away from the electrode (Figure S5).
Oxidative scans on complex 1 did not produce any well-
defined features.
Stoicheometric Reduction of 1. Stoichiometric reactions
were performed in order to isolate and characterize the
reduction product of 1. Potassium graphite (KC8) was chosen
as a stoichiometric reducing agent due to its potential (−3.10
V vs Fc/Fc+) and compatibility with THF.19 The addition of
1.1 equiv (per dimer) of KC8 to a 0.02 M solution of 1 in THF
at −36 °C resulted in an immediate color change from orange
to deep amber/brown. The solution was stirred at room
temperature for 30 min in order to ensure complete reduction.
Figure 1. Structures (A) and reaction profiles (B) for 1-SiO2 (red)
and 2-SiO2 (blue) during a 10 h plug flow run for the catalytic
disproportionation of propylene to ethylene and 2-butene. (C)
Reaction scheme and conditions for the plug flow experiment.
1
A H NMR spectrum of the crude reaction mixture in C6D6
exhibited two singlets attributed to the starting material (δ 2.54
and 1.24 ppm), a set of three new singlets (δ 1.53, 1.06, and
0.61 ppm, ratio of 9:18:4, respectively), and free neopentane
(δ 0.91 ppm). A broad baseline peak in the range of δ −5 to +3
ppm was attributed to a collection of ill-defined, paramagnetic
decomposition products. The persistence of the starting
material in the reaction NMR spectrum suggested that 2
equiv of KC8 is required per dimer. The reaction was repeated
under the previously described conditions, but with 2.2 equiv
(per dimer) of KC8 (Scheme 1) to produce the same visual
experiments were used to expand the chemical reactivity of
these species, leading to the isolation of an alkylidyne-bridged
dimer of the form W2(μ-CCMe3)2(CH2CMe3)4 (3). Finally,
the CV data, chemical reactivity results, and density functional
theory (DFT) were used to shed light on the fundamental
properties of these species. To our knowledge, this work
represents the first example where cyclic voltammetry was
employed to characterize and guide the reactivity of the
M2(CH2CMe3)6 dimeric complexes (M = W, Mo).18
Scheme 1. Reaction Scheme and Conditions for the
Reduction of Complex 1 to Produce Complex 3 and
Neopentane in Low Yields (5%)
RESULTS AND DISCUSSION
■
a
Electrochemical Characterization of 1. Complexes 1
and 2 were synthesized according to previously reported
procedures.17 Room-temperature cyclic voltammograms
(CVs) of 1 (1 mM) using a glassy-carbon working electrode
in a dry tetrahydrofuran (THF) solution of 0.1 M [nBu4N]-
[PF6] under a nitrogen atmosphere (Figure S1) yield a single,
irreversible feature at −2.91 V (vs Fc/Fc+) that is hypothesized
to correspond to the population of the π* LUMO orbitals of
the WW dimer (Figure 2). An oxidative return scan of
a
The remaining mass balance is attributed to W(0) decomposition
products.
observations. The 1H NMR (in C6D6) of the reaction mixture
taken after 30 min of stirring revealed only the three previously
observed singlets (δ 1.53, 1.06, and 0.61 ppm, same ratio), free
neopentane (δ 0.91 ppm), and the baseline features attributed
to decomposition. In situ experiments to quantify the
neopentane formed in the reduction are consistent with 0.4
equiv formed per 1 equiv of 1.
The THF reaction mixture was filtered through a plug of
activated basic alumina to remove soluble decomposition
products, and the solvent was removed under reduced pressure
1
to produce a bright yellow, air-sensitive solid. The H NMR
spectrum of the solid in C6D6 consisted solely of the three
singlets at δ 1.53, 1.06, and 0.61 ppm in the same integration
ratio (9:18:4) (Figure S3). Single crystals of complex 3 suitable
for X-ray diffraction (Figure 3 and Figure S2 and Table S1)
were grown out of a concentrated pentane solution (ca. 20 mg
of solid per 0.25 mL of pentane) at −36 °C. Complex 3 was
isolated in an overall yield of 5%. While the reduction is low
yielding, the synthesis of 3 was reproducible on larger scales.
Figure 2. Cyclic voltammograms of 1 mM solutions of 1 (black) and
3 (red) in THF solutions containing 0.1 M [nBu4N][PF6] under an
atmosphere of N2. The CV of 1 displays an irreversible electron
transfer process followed by the observation of a new species with a
more positive potential, while the CV of 3 exhibits a redox couple at
−2.29 V. The scan rates were 100 mV/s.
B
https://dx.doi.org/10.1021/acs.organomet.0c00533
Organometallics XXXX, XXX, XXX−XXX

Organometallics
pubs.acs.org/Organometallics
Article
Table 1. Selected Experimental and Computed Bond
a
Distances for 3 and Other Isostructural Complexes
complex
M
M−M (Å)
M−Npb (Å)
M−Npt (Å)
3 (EXP)
3 (DFT)
4 (EXP)
4 (DFT)
6 (EXP)
W
W
W
W
Nb
2.5974(10)
2.610
2.549(2)
2.582
1.967(7)
1.982
1.91(3)
1.990
2.121(13)
2.162
2.09(3)
2.128
2.897(2)
1.995(9)
2.160(9)
a
All reported values are averages. Npb = bridging neopentylidyne
ligand, Npt = terminal neopentyl ligand.
Figure 3. Solid-state structure of 3 viewed (A) perpendicular to the
W−W bond and (B) down the W−W bond, highlighting the plane
formed by the C(1)−W(1)−C(1′)−W(1′) core. Thermal ellipsoids
are drawn at the 35% probability level. Hydrogen atoms are omitted
for clarity. Legend: tungsten (blue), carbon (gray). Selected structural
parameters: W(1)−W(1′) = 2.5974(10) Å, C(1)−W(1)−C(1′) =
97.4(3)°, W(1)−C(1)−W(1′) = 82.6(3)°, C(2)−W(1)−C(3) =
120.0(6)°.
be found in Figures S13−S18 and Tables S5−S7 in the
Supporting Information.
The CV of complex 3 under the previously described
conditions showed a redox couple at −2.29 V consistent with
the new feature observed in the CV of 1 (Figure 2 and Figures
S6−S8) in agreement with the hypothesis that 3 forms upon
the electrochemical reduction of 1. The large peak separation
between the reductive and oxidative features of 3 (170 mV at a
scan rate of 100 mV/s) suggests that the redox event is
accompanied by a chemical or structural rearrangement.
Anodic scans of 3 revealed two irreversible oxidative features
at 0.55 and 1.1 V (Figure S9); these are hypothesized to
involve the electrons in the W−W bond. The most unusual
aspect of 3 is the oxidation state of the tungsten atoms (+5)
being higher than that of the tungsten atoms of the starting
material 1 (+3) after the latter was treated with a strong
reductant. Given the low isolated yield (∼5%), the formation
of 3 is likely a result of a series of disproportionation reactions
that culminate in the deprotonation of two neopentyl ligands
by potassium neopentyl in situ to form 3.
The crystal structure of 3 consists of a dimer in which each
tungsten atom is formally in the +5 oxidation state. Each
tungsten atom is coordinated to two terminal neopentyl
ligands and two bridging alkylidyne groups in a pseudote-
trahedral fashion. The C(1)−W(1)−C(1′) and C(2)−W(1)−
C(3) planes are nearly perpendicular to each other, while the
W2C2 core is nearly coplanar (torsion angle of 11.34°). The
W−W bond length of 2.5974(10) Å is consistent with a single
bond.20 The average W−C bond distances for the bridging and
terminal neopentyl ligands are 1.967(7) and 2.121(13) Å,
respectively. The symmetry in the bond distances of bridging
W−C bonds suggests that they can be best described as
delocalized across the W2C2 core (vide infra). Only three other
dimers with similar coordination about the metal centers exist:
W2(μ-CSiMe3)2(CH2SiMe3)4 (4) and its tantalum (Ta2(μ-
CSiMe3)2(CH2SiMe3)4; 5) and niobium (Nb2(μ-CSi-
Me3)2(CH2SiMe3)4; 6) analogues.20−22 Only 4 and 6 have
been structurally characterized, and of those two, only 4
features a metal−metal bond. Finally, the closely related
rhenium analogue Re2(μ-CCMe3)2(OCMe3)4 (7), where the
terminal neopentyl ligands have been replaced with tert-
butoxides, has also been prepared.23 The W−W bond distance
in 3 is longer than that of 4 (2.549(2) Å), likely due to the
electronic effects exerted by the silyl group on the M−C bonds
involving the alkyl and alkylidyne carbons. It has been
previously reported that a silicon atom can influence the
electronic properties of α, β, and γ substituents through the
Electrochemical and Chemical Reduction of 2. The
Mo homologue of 1, complex 2, was similarly analyzed using
electrochemical methods. The CV of complex 2 in dry THF
yields a single, quasi-reversible feature at −1.39 V (Figure 4).
culmination of electronic, steric, and solvation effects.24−28
A
previous report attributes the contraction in the C−Si and C−
Cl bonds in (chloromethyl)trimethylsilane (Me3SiCH2Cl) to
the participation of unoccupied silicon 3d orbitals in
bonding.25 The contraction of W−C bonds on going from 3
to 4 in both the experimental and computational data provides
further support for this hypothesis. Additional bonding
parameters (experimental and computed) and their compar-
isons to other dimers with isostructural connectivity can be
found in Table 1. The 13C NMR spectrum of 3 (Figure S4)
features a resonance at δ 371.2 ppm corresponding to the
bridging alkylidyne carbons. This resonance is more deshielded
than that of 4 (δ 352.6 ppm) and less deshielded than that of
the Ta homologue 5 (δ 404.0 ppm). The experimental NMR
spectra for 3 are in good agreement with the computed data.
Additional experimental and calculated NMR parameters can
Figure 4. Cyclic voltammogram scanned at 100 mV/s of 2 (1 mM) in
a THF solution containing 0.1 M [nBu4N][PF6] under an atmosphere
of N2 displaying a reversible redox feature.
This reduction potential is significantly more positive than that
of 1, as expected from periodic trends.29,30 Variable-scan-rate
experiments on 2 produced a series of CVs with increasing
peak potentials and separations (Figure S10). The large peak
separation between the oxidative and reductive features (110
mV at 100 mV/s) suggests that upon reduction 2 undergoes
some type of electronic or structural rearrangement. A plot of
C
https://dx.doi.org/10.1021/acs.organomet.0c00533
Organometallics XXXX, XXX, XXX−XXX

Organometallics
pubs.acs.org/Organometallics
Article
the peak cathodic and anodic currents versus the square root of
the scan rate gives rise to plots with a linear fit, consistent with
a freely diffusing species in solution as opposed to a surface-
confined process (Figures S11 and S12). Reductive scans to
more negative potentials and oxidative scans on 2 did not yield
any well-defined features.
The stoichiometric reduction of 2 was performed in
accordance with the procedure that was established for 1.
Addition of 2.2 equiv of KC8 to a THF solution of 2 (0.02 M)
at −36 °C resulted in a color change from yellow to green to
1
deep amber/brown. The H NMR spectrum of the reaction
mixture after it was stirred at room temperature for 30 min
showed only the starting material 2 (δ 2.76 and 1.24 ppm)
along with broad baseline features attributed to decomposition
products. Attempts to modify the reaction using different
solvents (THF, diethyl ether, pentane) and conditions
(number of equivalents of KC8, added 18-crown-6, temper-
ature) were unsuccessful in affording an isolable reduction
product of 2.
Electronic Structures of 1 and 2. Density functional
theory (DFT) calculations were performed at the M06L/def2-
TZVP level using Gaussian16 in order to rationalize the
reactivity of 1 and 2 and the structural discrepancies between 3
and 4.31,32 Natural bond orbital (NBO) analyses of 1 and 2
produced a pair of molecular orbital (MO) diagrams (Figures
S19 and S20 and Tables S8 and S9). The frontier orbitals on
tungsten and molybdenum were attributed to constructive and
destructive overlap between eg-symmetric (under pseudo-D3d
symmetry with the z axis down the MM bond) dyz and dxz
orbitals resulting in π and π* interactions for the HOMO and
LUMO, respectively. The MO diagrams for 1 and 2 are
consistent with a triple bond (Tables S8−S9 and Figures S19−
S20) made up of a σ bond and two π bonds. The bond indices
for 1 and 2 were calculated to be 2.691 and 2.677, respectively.
Finally, the HOMO−LUMO gaps for 1 (68.75 kcal/mol, 2.98
eV) and 2 (69.21 kcal/mol, 3.00 eV) were only 0.46 kcal/mol
(0.02 eV) apart, consistent with previously reported UV−vis
measurements.17 Addition of two electrons to 1 and 2
generates 12− and 22−, respectively, and leads to changes in
their geometries and MO diagrams (Figures S21 and 22 and
Tables S10 and S11). Complexes 12− and 22− both exhibit a
lowering of the HOMO−LUMO gap (25.41 kcal/mol, 1.10 eV
for 12− and 22.40 kcal/mol, 0.97 eV for 22−). Upon reduction,
1 undergoes a rotation with respect to the W−W bond
(C(1)−W(1)−W(1′)−C(1′) angle of 27.8° in comparison to
60° for 1 (Figure 5A) and a migration of one of the neopentyl
ligands toward the W−W core (Figure 5B). The C(3′)−W(1)
distance (2.676 Å) is longer than a that of a W−C single bond,
and the low bond index of 0.253 between C(3′) and W(1) is
suggestive of a nonbonding interaction between the two. While
the population of the LUMO π* orbitals with two electrons is
expected to decrease the W−W bond order from a triple bond
to a double bond, the opposite is observed.30,33,34
Figure 5. Geometry-optimized structures of 12− and 22− viewed (A,
C) down the M−M bond, highlighting the eclipsed conformation of
the molecule, and (B, D) perpendicular to the M−M bond,
highlighting the distortion around W(1′) and C(3′) in comparison
to Mo(1′) and C(3′). The C(1)−W(1)−W(1′)−C(1′) angle is
reduced to 27.8°, while the C(1)−Mo(1)−Mo(1′)−C(1′) is reduced
to 39.2° in comparison to 60° for the parent complexes. tert-Butyl
groups and hydrogen atoms are omitted for clarity. Legend: tungsten
(blue), molybdenum (teal), carbon (gray).
Figure 6. DFT calculated (A) LUMO of 1 and (B) HOMO of 12−
viewed perpendicular to the W−W bond and (C) a simplified
schematic depiction of the orbital rotation, alignment, and bonding
between the Eg orbitals of the tungsten atoms, also viewed
perpendicularly. The contour isovalue is set at 0.062.
The electronic structure of 1 features a σ-bonding
2
interaction between tungsten A1g orbitals (dz ) and two π-
bonding interactions between the tungsten Eg orbitals (dyz and
dxz), formally resulting in a triple bond. The LUMO orbitals of
2
2
1 consist of a pair of orbitals with Eg symmetry (dx −y and dxy).
Upon reduction, population of the LUMO of 1 results in a
rotation around the W−W axis to align the remaining Eg
symmetric tungsten orbitals, resulting in a δ-bonding
interaction (Figure 6). The overall bond index is increased
from 2.69 to 2.971. Consequently, 12− exhibits a shortening of
the metal−metal bond from 2.270 to 2.254 Å and a weakening
of the W−C bonds (the average W−C bond distance increases
from 2.16 to 2.25 Å). Interestingly, it has been previously
postulated that the staggered geometry around the bimetallic
D
https://dx.doi.org/10.1021/acs.organomet.0c00533
Organometallics XXXX, XXX, XXX−XXX

Organometallics
pubs.acs.org/Organometallics
Article
core is governed by steric ligand repulsions while the
electronics of the M−M bond favor an eclipsed conforma-
tion.35 The shortened W−W bond distance, coupled with the
steric bulk afforded by the neopentyl ligands, and the
weakening of the W−C bonds are hypothesized to lead to a
loss of a neopentyl ligand and subsequent reactivity.
the flat variant of 3 being only 0.7 kcal/mol (0.03 eV) higher in
energy than its puckered homologue. In both cases, the
HOMO orbitals consist of a constructive interaction between
two tungsten ag orbitals (in D2h symmetry), leading to a σ
bond between the two metal centers (Figure 8A).38−40 This
Complex 22− is isoelectronic but not isostructural with 12−
(Figure 5B,C). While 22− exhibits a shorter Mo−Mo distance
(2.213 Å to 2.191 Å with a bond index of 3.000) and longer
Mo−C distance (2.1740 Å to 2.268 Å), the C(1)−Mo(1)−
Mo(1′)−C(1) angle of 39.2° is larger than that of 12− at
27.81° (Figure 5C). Finally, the neopentyl ligands on 22−
remain aligned away from the Mo−Mo core, with no
propensity for a bridging ligand (Figure 5D). The discrepancy
between the structures of 12− and 22− is hypothesized to
explain the fact that a molybdenum homologue of 3 could not
be isolated.
Electronic Structures of 3 and 4. Calculated geometries
of 3 and its silyl analogue 4 (Figures S23 and S24 and Tables
S12 and S13) show that a shift from C to Si leads to a decrease
in the W−W and W−C bond distances, consistent with the
crystallographic results. A structural overlay of 3 and 4 (Figure
7) reveals a reasonably strong alignment between the W2C2
Figure 8. Visual representation of the binding interactions within the
W2C2 core depicting the (A) W−W σ bond from the overlap of two
tungsten ag orbitals, (B) W−C σ bonds from the overlap of tungsten
b2g and ligand pz orbitals, and (C) the W−C π-bonds from the
overlap of tungsten b1g and ligand py orbitals.
interpretation is also consistent with the absence of a Nb−Nb
bond in 6, where the orbitals responsible for the metal−metal
bonds are unpopulated. The W−C σ bonds within the W2C2
core are due to overlap between the tungsten b2g and ligand pz
orbitals (Figure 8B). While similar bonding interactions have
been identified in both WC3H4 and C4H4, it is the π-bonding
interactions within the four-membered core that set 3
apart.36,41 Overlap between tungsten orbitals possessing b1g
symmetry and py orbitals on the core carbon atoms give rise to
a fully conjugated π interaction within the W2C2 core (Figure
8C), making 3 an example of a metalloaromatic complex.42−44
The metalloaromaticity observed in 3 is contingent on the
presence of two opposite atoms that possess orbitals with b1g
symmetry, which are absent for WC3H3 and C4H4. The
absence of these orbitals and, subsequently, aromatic
stabilization in WC3H3 and C4H4 explains their stability in
comparison to 3.36,41 Finally, the LUMO orbital of 3 consists
of a single π* antibonding oribital. The population of this
orbital is hypothesized to disrupt the aromatic bonding around
the W2C2 core. This is consistent with the behavior exhibited
in the CV of 3 (Figure 2 and Figures S6−S8).
On the Formation of 3 from 1. Given the above
calculations, it is not unlikely that the reduction of 1 with KC8
gives rise to the sterically hindered and reactive 12−. A possible
decomposition route for 12− involves the dissociation of a
neopentyl ligand, which can act as a base in situ, resulting in a
series of deprotonation and disproportionation reactions to
generate 3 in low yields. The dependence of the reaction on
the stoichiometry of KC8 and its reproducibility suggests that a
more complex equilibrium might exist in solution. Finally,
complex 2 is hypothesized to undergo a reduction to produce
22−. However, the less pronounced structural rearrangement
calculated for 22− is hypothesized to inhibit further reactivity.
Figure 7. Overlay of the geometry-optimized structures of 3 (red)
and 4 (blue) highlighting the loss of planarity in the C(1)−W(1)−
C(1′)−W(1′) core. 3 and 4 were aligned along the plane formed by
two tungsten atomsand a carbon atom. The displacement between the
C(1′) atoms of 3 and 4 is 0.11 Å. The terminal ligands and CMe3/
SiMe3 fragments on the bridging ligands are removed for clarity.
cores. The W2C2 torsion angles of 3 and 4 are only 3.36° apart
(11.34° and 14.80° for 3 and 4, respectively). When the
electronic structures of 3 and 4 are compared, the low-energy
orbitals of complex 4 showcase a greater electronic
contribution from the silicon atoms to the M−C carbons in
comparison to the contribution of the quaternary carbons to
the M−C carbon atoms in 3. The electronic influence of the
silicon atoms on the W−C atoms is consistent with the
hypothesis that the silicon atom contributes to the bonding
nature of the ligand and facilitates the strengthening of the
metal−metal bond.
The stable nature of 3 in comparison to its monometallo-
cyclobutadiene (WC3) and cycobutadiene (C4) analogues
suggests that electronic contributions from the second
tungsten metal play a key role in the stability of this structural
motif.36 Previous DFT analyses on variously substituted W2C2
complexes have shown that puckering around the rings can
strongly affect their electronic structures and bonding.37
Subsequently, NBO analysis was performed on both the fully
relaxed (puckered, Table S12) and D2h constrained (flat, Table
S13) structures of 3. Both structures were isoelectronic, with
E
https://dx.doi.org/10.1021/acs.organomet.0c00533
Organometallics XXXX, XXX, XXX−XXX

Organometallics
pubs.acs.org/Organometallics
Article
Massimiliano Delferro − Chemical Sciences and Engineering
Division, Argonne National Laboratory, Lemont, Illinois
60439, United States; orcid.org/0000-0002-4443-165X;
Email: delferro@anl.gov
CONCLUSION
■
This work represents the first report in which a triply bonded
species has been characterized by cyclic voltammetry. Complex
1 exhibited electrochemical behavior consistent with an EC-
type pathway, where an electrochemical reduction leads to a
chemical rearrangement, while complex 2 showcased a
positively shifted reversible wave. The chemical reduction of
1 and 2 was also examined and produced divergent results.
The reduction of 2 with KC8, independent of reaction
conditions, did not produce any isolable products. In contrast,
the reduction of 1 with 2.2 equiv of KC8 (per dimer) gave rise
to the alkylidyne-bridged, metalloaromatic dimer W2(μ-
CSiMe3)2(CH2SiMe3)4 (3). Complex 3 represents one of
t h r e e r e p o r t e d s t r u c t u r e s w i t h t h e M 2 ( μ -
CXMe3)2(CH2XMe3)4 (X = C, Si) structural motif and the
only one where X = C. NBO analyses of 1 and 2 and their
doubly reduced congeners suggest that, upon reduction, an
electronic and structural rearrangement leads to a contraction
in the M−M bond distance, which, given the steric profile of
the neopentyl ligands, could explain the reactivity of the
complexes. The solid-state and calculated structures of 3 were
compared to those of 4, showing an increase in the W−W
bond distance on going from the Me3SiC− ligand to the
Me3CC− ligand. These structural trends are rationalized via
long-distance effects of the electronically unsaturated Si atom,
which influences the structure of the W2C2 core. The results
presented in this work confirm that cyclic voltammetry can
serve as a powerful tool for studying metal−metal-bonded
species. Future work will involve the elucidation of
mechanistically relevant intermediates and reactivity studies
of the new species toward small-molecule activation and
catalysis.
Authors
Alon Chapovetsky − Chemical Sciences and Engineering
Division, Argonne National Laboratory, Lemont, Illinois
60439, United States
Prajay Patel − Chemical Sciences and Engineering Division,
Argonne National Laboratory, Lemont, Illinois 60439,
United States
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.organomet.0c00533
Funding
This work was supported by the U.S. Department of Energy
(DOE), Office of Basic Energy Sciences, Division of Chemical
Sciences, Geosciences, and Biosciences, Catalysis Science
Program, under contract no. DE-AC-02-06CH11357. DFT
calculations were performed using the computing resources
provided by the Laboratory Computing Resource Center
(LCRC) at Argonne National Laboratory.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Charlotte Stern (Northwestern University) for
crystallographic support, Karen Mulfort (Argonne National
Laboratory) for allowing us access to electrochemical equip-
ment, and Ryan Langeslay (Shepherd Chemical) for relevant
preliminary results. We also thank Professor Bruce Bursten
(WPI) for enlightening discussions on metalloaromaticity.
This work is dedicated to Professor Malcolm L. H. Green
(Oxford), who passed away on July 24th, 2020.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.organomet.0c00533.
■
sı
REFERENCES
■
General discussion of synthetic and experimental
procedures, instrumentation, characterization of materi-
als, and rate laws and their derivations (PDF)
(1) Lindahl, P. A. Metal-metal bonds in biology. J. Inorg. Biochem.
2012, 106, 172−178.
(2) Buchwalter, P.; Rose, J.; Braunstein, P. Multimetallic Catalysis
́
Based on Heterometallic Complexes and Clusters. Chem. Rev. 2015,
115, 28−126.
Cartesian coordinates of the calculated structures (XYZ)
Accession Codes
(3) Gray, H. B.; Maverick, A. W. Solar Chemistry of Metal
Complexes. Science 1981, 214, 1201−1205.
CCDC 1999807 contains the supplementary crystallographic
data for this paper. These data can be obtained free of charge
via www.ccdc.cam.ac.uk/data_request/cif, or by emailing
data_request@ccdc.cam.ac.uk, or by contacting The Cam-
bridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
(4) Heyduk, A. F.; Nocera, D. G. Hydrogen Produced from
Hydrohalic Acid Solutions by a Two-Electron Mixed-Valence
Photocatalyst. Science 2001, 293, 1639−1641.
(5) Corcos, A. R.; Pap, J. S.; Yang, T.; Berry, J. F. A Synthetic
Oxygen Atom Transfer Photocycle from a Diruthenium Oxyanion
Complex. J. Am. Chem. Soc. 2016, 138, 10032−10040.
(6) Vaddypally, S.; Jovinelli, D. J.; McKendry, I. G.; Zdilla, M. J.
Covalent Metal−Metal-Bonded Mn4 Tetrahedron Inscribed within a
Four-Coordinate Manganese Cubane Cluster, As Evidenced by
Unexpected Temperature-Independent Diamagnetism. Inorg. Chem.
2017, 56, 3733−3737.
AUTHOR INFORMATION
Corresponding Authors
■
Cong Liu − Chemical Sciences and Engineering Division,
Argonne National Laboratory, Lemont, Illinois 60439,
United States; orcid.org/0000-0002-2145-5034;
Email: congliu@anl.gov
Alfred P. Sattelberger − Chemical Sciences and Engineering
Division, Argonne National Laboratory, Lemont, Illinois
60439, United States; Email: asattelberger@anl.gov
David M. Kaphan − Chemical Sciences and Engineering
Division, Argonne National Laboratory, Lemont, Illinois
60439, United States; orcid.org/0000-0001-5293-7784;
Email: kaphand@anl.gov
(7) Krogman, J. P.; Thomas, C. M. Metal−metal multiple bonding in
C3-symmetric bimetallic complexes of the first row transition metals.
Chem. Commun. 2014, 50, 5115−5127.
(8) Berry, J. F. The role of three-center/four-electron bonds in
superelectrophilic dirhodium carbene and nitrene catalytic inter-
mediates. Dalton Trans. 2012, 41, 700−713.
(9) Berry, J. F.; Lu, C. C. Metal−Metal Bonds: From Fundamentals
to Applications. Inorg. Chem. 2017, 56, 7577−7581.
(10) Nguyen, T.; Sutton, A. D.; Brynda, M.; Fettinger, J. C.; Long,
G. J.; Power, P. P. Synthesis of a Stable Compound with Fivefold
F
https://dx.doi.org/10.1021/acs.organomet.0c00533
Organometallics XXXX, XXX, XXX−XXX

Organometallics
pubs.acs.org/Organometallics
Article
Bonding Between Two Chromium(I) Centers. Science 2005, 310,
844−847.
(29) Kadish, K. M.; Lacombe, D. A.; Anderson, J. E. Electro-
chemistry of molybdenum and tungsten cyclopentadienyl carbonyl
complexes, [M(CO)3Cp]2, [M(CO)3Cp]+, [M(CO)3Cp]−, and
M(CO)3Cp where M = Mo and W. Inorg. Chem. 1986, 25, 2246−
2250.
(30) Chisholm, M. H.; Clark, D. L.; Huffman, J. C.; Van der Sluys,
W. G.; Kober, E. M.; Lichtenberger, D. L.; Bursten, B. E. The
tungsten-tungsten triple bond. 13. Bisalkyl tetracarboxylates of
dimolybdenum and ditungsten. Triple bonds between metal atoms
with the valence molecular orbital description π4δ2. J. Am. Chem. Soc.
1987, 109, 6796−6816.
(31) Weigend, F.; Ahlrichs, R. Balanced basis sets of split valence,
triple zeta valence and quadruple zeta valence quality for H to Rn:
Design and assessment of accuracy. Phys. Chem. Chem. Phys. 2005, 7,
3297−3305.
(32) Zhao, Y.; Truhlar, D. G. A new local density functional for
main-group thermochemistry, transition metal bonding, thermochem-
ical kinetics, and noncovalent interactions. J. Chem. Phys. 2006, 125
(19), 194101−194108.
(33) Beshouri, S. M.; Rothwell, I. P. Electrochemical behavior of
tungsten(VI) aryloxides and catecholates. Inorg. Chem. 1986, 25,
1962−1964.
(34) Chisholm, M. H.; Clark, D. L.; Huffman, J. C.; Van der Sluys,
W. G. The tungsten-tungsten triple bond. 14. Factors influencing
triple bonds between tungsten atoms having valence molecular orbital
configurations of σ2π4 and π4δ2 and preparation and characterization
of bis(neopentyl)bis(acetato)bis(diethyldithiocarbamato)ditungsten.
J. Am. Chem. Soc. 1987, 109, 6817−6825.
(35) Bursten, B. E.; Cotton, F. A.; Green, J. C.; Seddon, E. A.;
Stanley, G. G. Molecular orbital and spectroscopic studies of triple
bonds between transition-metal atoms. 1. The d3-d3 Mo2L6
compounds (L = OR, NR2, CH2R). J. Am. Chem. Soc. 1980, 102,
4579−4588.
(36) Cui, M.; Lin, R.; Jia, G. Chemistry of Metallacyclobutadienes.
Chem. - Asian J. 2018, 13 (8), 895−912.
́
́
(11) Hernandez Sanchez, R.; Bartholomew, A. K.; Powers, T. M.;
́
Menard, G.; Betley, T. A. Maximizing Electron Exchange in a [Fe3]
Cluster. J. Am. Chem. Soc. 2016, 138, 2235−2243.
(12) Hlina, J. A.; Pankhurst, J. R.; Kaltsoyannis, N.; Arnold, P. L.
Metal−Metal Bonding in Uranium−Group 10 Complexes. J. Am.
Chem. Soc. 2016, 138, 3333−3345.
(13) Queen, J. D.; Phung, A. C.; Caputo, C. A.; Fettinger, J. C.;
Power, P. P. Metathetical Exchange between Metal−Metal Triple
Bonds. J. Am. Chem. Soc. 2020, 142, 2233−2237.
(14) Chisholm, M. H.; Eichhorn, B. W.; Folting, K.; Huffman, J. C.;
Ontiveros, C. D.; Streib, W. E.; Van der Sluys, W. G. Preparation and
characterization of sodium heptachloropentakis(THF)ditungstate(1‑).
A synthetically useful precursor for X3WWX3 compounds where X
= CH2-tert-Bu, NMe2 and O-tert-Bu. Inorg. Chem. 1987, 26, 3182−
3186.
(15) Gilbert, T. M.; Landes, A. M.; Rogers, R. D. Synthesis and
electronic properties of triply bonded hexakis(fluoroalkoxy)-
dimolybdenum complexes. Structure of Mo2[OCMe(CF3)2]6 and
investigation of the nature of the frontier orbitals in triply bonded
M2X6 compounds. Inorg. Chem. 1992, 31, 3438−3444.
(16) Huq, F.; Mowat, W.; Shortland, A.; Skapski, A. C.; Wilkinson,
G. Crystal structure of hexakis(trimethylsilylmethyl)dimolybdenum. J.
Chem. Soc. D 1971, 18, 1079−1080.
(17) Chapovetsky, A.; Langeslay, R. R.; Celik, G.; Perras, F. A.;
Pruski, M.; Ferrandon, M. S.; Wegener, E. C.; Kim, H.; Dogan, F.;
Wen, J.; Khetrapal, N.; Sharma, P.; White, J.; Kropf, A. J.; Sattelberger,
A. P.; Kaphan, D. M.; Delferro, M. Activation of Low-Valent, Multiply
M−M Bonded Group VI Dimers toward Catalytic Olefin Metathesis
via Surface Organometallic Chemistry. Organometallics 2020, 39,
1035−1045.
(18) Cotton, F.A.; Murillo, C.A.; Walton, R. A. Multiple Bonds
between Metal Atoms, 3rd ed.; Springer Science: New York, 2005.
(19) Connelly, N. G.; Geiger, W. E. Chemical Redox Agents for
Organometallic Chemistry. Chem. Rev. 1996, 96, 877−910.
(20) Chisholm, M. H.; Cotton, F. A.; Extine, M.; Murillo, C. A.
Detailed structure of bis(μ-trimethylsilylmethylidyne)-tetrakis-
(trimethylsilylmethyl)ditungsten(W-W). Inorg. Chem. 1978, 17,
696−698.
(21) Mowat, W.; Wilkinson, G. Elimination stabilized alkyls. Part III.
Trimethylsilylmethyl and neopentyl alkyls of transition metals. J.
Chem. Soc., Dalton Trans. 1973, 10, 1120−1124.
(22) Huq, F.; Mowat, W.; Skapski, A. C.; Wilkinson, G. Crystal
structure of bis-μ-(trimethylsilylmethylidyne)tetrakis-
(trimethylsilylmethyl)diniobium(V). A new type of carbon bridging
group. J. Chem. Soc. D 1971, 0, 1477−1478.
(23) Toreki, R.; Schrock, R. R.; Davis, W. M. Conversion of rhenium
alkylidyne complexes that contain unsupported metal-metal double
bonds into relatives that contain μ-alkylidyne ligands. J. Organomet.
Chem. 1996, 520, 69−78.
(24) Sommer, L. H.; Dorfman, E.; Goldberg, G. M.; Whitmore, F. C.
The Reactivity with Alkali of ChlorineCarbon Bonds Alpha, Beta
and Gamma to Silicon. J. Am. Chem. Soc. 1946, 68, 488−489.
(25) Whitmore, F. C.; Sommer, L. H. Organo-silicon Compounds.
II. Silicon Analogs of Neopentyl Chloride and Neopentyl Iodide. The
Alpha Silicon Effect. J. Am. Chem. Soc. 1946, 68, 481−484.
(26) Sommer, L. H.; Whitmore, F. C. Organo-silicon Compounds.
III. α- and β-Chloroalkyl Silanes and the Unusual Reactivity of the
Latter. J. Am. Chem. Soc. 1946, 68, 485−487.
(27) Berkefeld, A.; Guerra, C. F.; Bertermann, R.; Troegel, D.; Daiß,
J. O.; Stohrer, J.; Bickelhaupt, F. M.; Tacke, R. Silicon α-Effect: A
Systematic Experimental and Computational Study of the Hydrolysis
of Cα- and Cγ-Functionalized Alkoxytriorganylsilanes of the Formula
Type ROSiMe2(CH)nX (R = Me, Et; n = 1, 3; X = Functional
Group). Organometallics 2014, 33, 2721−2737.
(28) Goodman, L.; Konstam, A. H.; Sommer, L. H. d-Orbital
Participation in Silane Conjugation. J. Am. Chem. Soc. 1965, 87,
1012−1016.
(37) Jemmis, E. D.; Giju, K. T. Organometallic Analogs of the
Cyclobutadiene Dication: An Ab Initio MO and Density Functional
Study of the Symmetrical Planar and Puckered [WL2(μ-CR)]2
Complexes (L = H, Me, F, OH; R = H, F, Me). Organometallics
1997, 16, 1425−1429.
(38) Chisholm, M. H.; Heppert, J. A.; Kober, E. M.; Lichtenberger,
D. L. 1,3-Ditungstacyclobutadienes. 2. Synthesis of alkoxide
derivatives of W2(μ-CSiMe3)2(CH2SiMe3)4 and investigations of the
electronic structures of the M2(μ-CSiMe3)2 core as a function of dn-dn
interactions (n = 0, M = Ta; n = 1, M = W; n = 2, M = Re).
Organometallics 1987, 6, 1065−1073.
(39) Jemmis, E. D.; Prasad, B. V. Theoretical studies of (μ-CR)
bridging complexes. Polyhedron 1988, 7, 871−879.
(40) Jemmis, E. D.; Prasad, B. V. Reaction of tungsten complexes
L2W(μ-CR)2WL2 with acetylenes C2R2. A theoretical study. J. Am.
Chem. Soc. 1987, 109, 2560−2563.
(41) Frogley, B. J.; Wright, L. J. Recent Advances in Metallaaromatic
Chemistry. Chem. - Eur. J. 2018, 24, 2025−2038.
(42) Bursten, B. E. On the stability of early-transition-metal
metallacyclobutadiene complexes. J. Am. Chem. Soc. 1983, 105,
121−122.
(43) Suresh, C. H.; Frenking, G. 1,3-Metal−Carbon Bonding and
Alkyne Metathesis: DFT Investigations on Model Complexes of
Group 4, 5, and 6 Transition Metals. Organometallics 2012, 31, 7171−
7180.
(44) Suresh, C. H.; Frenking, G. Direct 1−3 Metal−Carbon
Bonding and Planar Tetracoordinated Carbon in Group 6 Metal-
lacyclobutadienes. Organometallics 2010, 29, 4766−4769.
G
https://dx.doi.org/10.1021/acs.organomet.0c00533
Organometallics XXXX, XXX, XXX−XXX