Controlling Metal-to-Oxygen Ratios via M=O Bond Cleavage in Polyoxovanadate Alkoxide Clusters

Article abstract of DOI:10.1021/acs.inorgchem.9b00389

In this manuscript, we further investigate the use of Lindqvist polyoxovanadate alkoxide (POV-alkoxide) clusters as homogeneous molecular models of reducible metal oxides (RMO), focusing on the structural and electronic consequences of forming one or two oxygen-deficient sites. We demonstrate the reactivity of a neutral POV-alkoxide cluster, [V₆O₇(OCH₃)₁₂]⁰, with a reductant, revealing routes for controlling metal-to-oxygen ratios in self-assembled polynuclear ensembles through post-synthetic modification. The outlook of this science is bolstered by the fact that, in both cases, O-atom removal reveals reduced V ions at the surface of the cluster. Extending our entry into small-molecule activation mediated by surface defect sites, we report the reactivity of mono- and divacant clusters with a model substrate, tert-butyl isocyanide, demonstrating the electronic consequences of small-molecule coordination to reduced ions in RMO materials.

Full text of DOI:10.1021/acs.inorgchem.9b00389

Forum Article
Cite This: Inorg. Chem. XXXX, XXX, XXX−XXX
pubs.acs.org/IC
Controlling Metal-to-Oxygen Ratios via MO Bond Cleavage in
Polyoxovanadate Alkoxide Clusters
Brittney E. Petel, Alex A. Fertig, Michela L. Maiola, William W. Brennessel, and Ellen M. Matson*
Department of Chemistry, University of Rochester, Rochester, New York 14627, United States
*
S Supporting Information
ABSTRACT: In this manuscript, we further investigate the
use of Lindqvist polyoxovanadate alkoxide (POV−alkoxide)
clusters as homogeneous molecular models of reducible metal
oxides (RMO), focusing on the structural and electronic
consequences of forming one or two oxygen-deficient sites. We
demonstrate the reactivity of a neutral POV−alkoxide cluster,
0
[
V O (OCH ) ] , with a reductant, revealing routes for
6
7
3 12
controlling metal-to-oxygen ratios in self-assembled polynu-
clear ensembles through post-synthetic modification. The
outlook of this science is bolstered by the fact that, in both
cases, O-atom removal reveals reduced V ions at the surface of
the cluster. Extending our entry into small-molecule activation
mediated by surface defect sites, we report the reactivity of
mono- and divacant clusters with a model substrate, tert-butyl
isocyanide, demonstrating the electronic consequences of small-molecule coordination to reduced ions in RMO materials.
INTRODUCTION
oxide clusters have been reported, access to structurally related
assemblies with controlled metal-to-oxygen ratios remains
challenging. This is principally due to well-established
synthetic procedures for the formation of metal oxide clusters,
■
Reducible metal oxides (RMOs) are prominent materials that
act as functional components in many industrially relevant
heterogeneous processes.
1
−3
Theoretical investigations have
14
which nominally invoke self-assembly protocols. Most
revealed that surface defects in these materials, such as O-atom
vacancies, act as active sites in catalysis for the activation of
commonly, synthetic chemists manipulate the structures of
plenary motifs (e.g., Keggin, Dawson, Anderson POM ions) via
basic degradation, affording the formation of mono-, di-, and
4
small molecules. Despite progress in the development of
RMOs as catalysts, an understanding of the factors that
influence the formation and reactivity of O-atom vacancies
remains a significant gap in knowledge. In particular, the
optimization of RMOs for the conversion of small, inert
substrates to energy-rich chemical fuels has been hampered by
a limited spectroscopic understanding of the role of single-
atom vacancies on extended surfaces during catalysis (vide
infra). Recent advances in microscopy probing single-atom
vacancies on the surface of bulk materials remain limited by
15
n−
trilacunary clusters. In this approach, entire [MO4]
subunits are removed from the surface of a POM structure.
While this strategy does lead to the controlled formation of
POMs with new metal-to-oxygen ratios, the nucleophilic,
oxygen-terminated surfaces of the lacunary clusters do not
appropriately model surface defects (i.e., reduced metal ions)
invoked in small-molecule activation involving heterogeneous
catalysts.
Interested in investigating the consequences of defect
formation at the surface of well-defined metal oxide materials,
our research group has been studying the O-atom-transfer
5
6,7
challenging sample preparation, poor resolution, thermal
8,9
10
drift, misassignment of signals, and artifact generation.
Furthermore, the lack of in situ analysis limits opportunities to
characterize reactive intermediates. Thus, to provide insight
into the formation of reactive defect sites and their role in
substrate activation, our research group has focused on probing
the chemistry of O-atom vacancies using homogeneous cluster
complexes as models for bulk RMOs.
(
OAT) reactivity of POMs, targeting removal of terminal oxo
5
16
moieties with retention of the polynuclear core. Specifically,
org ano-f unction aliz ed p olyoxovanadate ion s,
n
[
V O (OCH ) ] (n = 0, 1−), have captured our interest
6
7
3 12
because of their distinct physical and electrochemical proper-
1
4,17,18
ties (Figure 1).
Their composition features 6 terminal
One class of compounds well-suited for modeling the surface
chemistry and activity of bulk RMO materials is polyox-
11−13
Special Issue: Celebrating the Year of the Periodic Table: Emerging
Investigators in Inorganic Chemistry
ometalates (POMs).
These polynuclear clusters are three-
dimensional molecular fragments of metal oxides that consist
of multiple transition-metal oxyanions linked together by
bridging oxide units. Although many examples of these metal
Received: February 10, 2019
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.inorgchem.9b00389
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Forum Article
n
−
monoanionic form of the cluster [ Bu N][V O (OCH ) ]
4
6
7
3 12
−
(
1-V O ) resulted in OAT to the reductant and the isolation
6 7
o f
a
r e d u c e d p o l y o x o v a n a d a t e c o m p o u n d ,
n
−
−
[
Bu N][V O (OCH ) ] (2-V O ; Scheme 1). Herein,
4 6 6 3 12 6 6
we extend upon our initial report, targeting controlled
formation of multiple vacant sites within a POV−alkoxide
assembly. A correlation between the electronic structure of the
parent POV−alkoxide cluster and number of vacant sites
reveals predictable routes to structurally related clusters with
controlled metal-to-oxygen ratios. Finally, we summarize an
entry into modeling the surface chemistry of defect sites with
small-molecule substrates. Exploration of the coordination
chemistry of isonitrile ligands with reduced, site-differentiated
V ions embedded directly within the polyoxovandate clusters
provides a system for interrogating the initial steps in the
mechanism of small-molecule activation in RMOs (i.e.,
substrate binding). Collectively, these results provide a distinct
route to exploring the structural and electronic consequences
of selective vacancy generation and substrate binding on metal
oxide supports in heterogeneous catalysis.
Figure 1. POV−alkoxide clusters as models for vacancy formation in
RMOs.
vanadyl ions and 12 bridging oxygen-donating alkoxo ligands,
V
IV
constructing stable mixed-valent (V V
, where n = 0−4)
n
6−n
18−20
clusters that resemble redox-active metal oxide supports.
Additionally, the solubility of these clusters in an organic
solvent renders them amenable to in situ analysis via
spectroscopies reserved for homogeneous systems. Further-
more, extensive chemical and electrochemical studies on
polyoxovanadate alkoxide (POV−alkoxide) clusters have
demonstrated the cluster’s ability to redistribute electron
density following oxidative or reductive transformations,
making these homogeneous materials ideal systems for
modeling the diffuse electronic structure of semiconducting
RESULTS AND DISCUSSION
■
Correlating the Electronic Structure of POV−Alk-
oxides to OAT. The generation of a reduced polyoxovanadate
assembly in our preliminary manuscript marked the first
example of the creation of an oxygen-deficient site via MO
1
6,19−21
RMOs.
While our systems certainly do not constitute
16
absolute structural mimics of the surface of an extended
material, we believe that the similarities in the electronic
structures of the two systems justify further work into the study
of their chemistry.
bond cleavage. Evaluation of the electronic structure of the
−
parent POV−alkoxide cluster, complex 1-V O , and the
6
7
−
oxygen-deficient product, 2-V O , by electronic absorption
6
6
spectroscopy revealed a loss of intervalence charge-transfer
IV
We recently disclosed the formation of a single O-atom
vacancy at the surface of a POV−alkoxide cluster, marking the
first report of an unsupported (i.e., “ligand-free”) polynuclear
metal oxide assembly with a reduced ion formed as a result of
(IVCT) bands associated with electron transfer between V /
V
V ions. The removal of an O atom, coupled with the loss of
IVCT bands, indicates a two-electron reduction of the single
V
−
V
IV
III
V center in 1-V O (electronic distribution: V V 5) to V
6
7
16
−
MO bond cleavage. Our initial results demonstrated that
in the reduced cluster (2-V O ; electronic distribution:
6 6
IV III
the addition of a single equivalent of V(Mes) (THF) (Mes =
V
5V ). To experimentally evaluate the relationship between
3
2
,4,6-trimethylphenyl; THF = tetrahydrofuran) to the
the electronic structure of the parent POV−alkoxide cluster,
Scheme 1. Reactivity of POV−Alkoxide Clusters across Three Different Oxidation States, with the Vacancy Formation
V
Correlated to the Number of V Ions in the Starting Material
B
DOI: 10.1021/acs.inorgchem.9b00389
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Forum Article
V
specifically the number of V ions in the starting material, and
VO bond cleavage, we attempted the formation of multiple
vacant sites via the addition of excess reductant to complex 1-
sixth associated with solvent bound to the cluster (CH CN),
3
suggesting the formation of a cluster with reduced molecular
1
0
5
symmetry. The H NMR spectrum obtained for 5-V O is
6
−
V O . A subsequent VO cleavage event at a tetravalent
consistent with selective formation of the cis-vacancy product
6
7
II
vanadyl ion would necessitate transient formation of a V
upon removal of the second O atom.
0
center, likely unfavorable in the oxidizing environment of the
POM. Consistent with our hypothesis, despite the addition of
multiple equivalents of V(Mes) (THF) to 1-V O and
prolonged heating of the reaction mixture, the sole formation
of complex 2-V O was observed (Figure S1).
To confirm the formation of complex 5-V O (electronic
6
5
IV
III
distribution: V 4V 2), analyses via IR and electronic
absorption spectroscopies were conducted. IR spectroscopy
of POV−alkoxide clusters has been demonstrated to be a
useful tool for cluster charge-state analysis via examination of
the stretching frequencies of the characteristic metal−oxygen
−
3
6
7
−
6
6
V
To probe the influence of the number of V ions present in
the starting material to vacancy formation, a control experi-
ment was conducted with the fully reduced, dianionic POV−
bonds of the Lindqvist core [ν(VO
) and ν(O
−CH
); O
terminal oxo and O Upon removal
of two surface O atoms from 4-V O , two strong absorption
6 7
=
t
b
3
t
1
6,19,20,24
= bridging oxo].
b
n
IV
2−
2−
0
alkoxide cluster [ Bu N] [V O (OCH ) ] (3-V O ;
4
2
6
7
3
12
6
7
Scheme 1). The oxidation state distribution of V ions across
the Lindqvist core is VIV6, presenting an opportunity to
evaluate potential OAT from a tetravalent V ion imbedded
within an isovalent vanadate cluster. As expected, the addition
bands were observed in the IR spectrum, corresponding to the
−1
−1
ν(VO
) (964 cm ) and ν(O
−CH
) (1038 cm )
t
b
3
0
stretching frequencies of 5-V
O
6
(Figure 2a). The VO
5
t
2
−
of V(Mes) (THF) to 3-V O yielded no reaction, providing
3
6
7
V
further support of the hypothesis that the presence of V ions
within the POV−alkoxide cluster is a determinant in OAT
(
Figure S2).
Synthesis of a POV−Alkoxide Cluster with Multiple
O-Atom Vacancies. Theoretical investigations on the role of
surface defects have revealed the importance of O-atom
22
vacancies on the activity of RMO catalysts. Studies featuring
the synthesis and characterization of oxygen-deficient metal
oxides have demonstrated that multiple oxygen vacancies are
created on the surface of the heterogeneous catalyst via current
1
0,23
techniques such as surface annealing,
ment,
electron bombard-
1
0,23
23
or hydrothermal syntheses. Therefore, the
generation of a POV−alkoxide model system that allows for
the investigation of the effect of variable numbers of O-atom
vacancies on the electronic structure of RMOs is coveted
because this would more aptly describe the surface
composition of the bulk solid.
Given the apparent relationship between the vacancy
V
generation and number of available V O moieties in the
Lindqvist core, we shifted our focus to the neutral POV−
0
0
alkoxide, [V O (OCH ) ] (4-V O ). We hypothesized that
6
7
3 12
6
7
0
the electronic distribution of the neutral cluster 4-V O
V V 4) would support the generation of either a single or
multiple oxygen-deficient site(s), due to the two accessible V
centers in the hexavanadate core. Upon exposure of 4-V O to
6
7
V
IV
(
2
V
0
6
7
2
equiv of V(Mes) (THF), a color change to brown-red was
3
observed, consistent with the generation of an oxygen-deficient
cluster (Scheme 1). Analysis of the reaction mixture by
electrospray ionization mass spectrometry (ESI-MS, positive
mode) revealed a single peak corresponding to m/z 758. This
molecular weight is consistent with formation of the desired
0
0
product, [V O (OCH ) (MeCN) ] (5-V O ; Figure S3).
0
0
6
5
3 12
2
6
5
Figure 2. (a) IR spectra of complexes 4-V O , 5-V O , and 6-
6
7
6
5
While the product of the reaction is proposed to be neutral in
charge, we justify the observation of the parent ion in the
positive mode as a result of one-electron oxidation of the
cluster under the high voltages associated with electrospray
0
V O . (b) Electronic absorption spectra of complexes 4-V O , 5-
6
6
6
7
0
0
V O , and 6-V O collected in MeCN.
6
5
6
6
1
6,24,36−39
0
ionization.
Further support of OAT was obtained via
and O −CH stretching frequencies are shifted from 4-V O
−1 −1
b
3
6
7
1
analysis of the crude reaction mixture by H NMR spectros-
copy. Formation of the anticipated diamagnetic byproduct,
(VO , 966 cm ; O −CH , 1009 cm ), confirming the
t
b
3
formation of a reduced cluster core. Specifically, the changes in
the energy differences between ν(VO ) and ν(O −CH ) for
OV(Mes) , was observed, along with a series of new
3
t
b
3
0
−1
0
paramagnetic resonances, ranging from 85.31 to −22.46 ppm
complexes 4-V O (Δ = 43 cm ) and 5-V O (Δ = 74
6
7
6
5
−1
(
Figure S4). Unlike the three signals observed for the
cm ) are reminiscent of the separation between similar bands
16
−
monovacant cluster, 2-V O , the product of this reaction
in the IR spectra of reduced and oxidized POV−alkoxide
6
6
16,24
18−20
has six resonances, five assigned to the O−CH ligands and the
clusters previously reported by our group
and others.
3
C
DOI: 10.1021/acs.inorgchem.9b00389
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Forum Article
Furthermore, the reduction in the intensity of ν(VO ) in the
performed using CD CN, affording an opportunity for the
t
3
0
0
7
complex 5-V O compared to that of complex 4-V O is
more strongly coordinating MeCN ligand to bind to the
cluster. We postulate that MeCN makes for a better ligand to
6
5
6
indicative of the removal of two vanadyl O atoms.
0
Further evidence supporting the reduction of the hexavana-
the complex 5-V O , in comparison to THF, as a result of the
6
5
0
date core was obtained via analysis of 5-V O5 via electronic
reduced steric bulk of the solvent at the coordinating atom.
6
absorption spectroscopy (Figure 2b). The absorption spectra
Analysis of the intracluster V −O bond distances reveals that,
v
0
V
IV
of the parent cluster 4-V O (electronic distribution: V V 4)
upon removal of two surface O atoms, the reduced V ions
pucker toward the center of the cluster, resulting in shortened
V −O bonds [O = central u -O atom, 2.0666(17) and
6
7
2
contain two IVCT bands assigned to charge transfer between
IV
V
−1
−1
V
and V metal centers (386 nm, ε = 6975 M cm ; 998
v
c
c
6
−
1
−1
nm, ε = 1375 M cm ). A loss of these IVCT bands
associated with electron transfer between V (d ) and V (d )
centers in the mixed-valent starting material is observed upon
2.0760(17) Å]. These bond lengths compare well with similar
V
0
IV
1
metrics in the previously reported O-atom-vacancy-containing
product, [V O (OCH ) OTf] (OTf = trifluoromethylsulfo-
6
6
3 12
0
16
the formation of 5-V O , consistent with a reduction of all d
6
5
nate), with a V −O distance of 2.08(4) Å. Likewise, to
accommodate the movement of the reduced ions toward the
v
c
III
vanadyl moieties to V ions upon the removal of two surface
O atoms. An additional weak transition is observed in the
electronic absorption spectrum, centered at 524 nm (ε = 468
center of the cluster, the V −O −V (V = terminal vanadyl
v
b
t
t
ion) bond angles are contracted, averaging ∼105°.
−
1
−1
M
cm ). A similar band was observed in the spectrum of 2-
0
Each V ion resolved in the crystal structure of 5-V O
6
5
−
−1
−1
V O (526 nm, ε = 473 M cm ) and is assigned to a spin-
6
6
occupies a distinct position in the unit cell. This allows for
bond-valence-sum (BVS) calculations to be performed for this
POV−alkoxide cluster, shedding insight into the oxidation
state distributions of V centers across the Lindqvist core
IV
16
forbidden, d−d excitation of a V ion.
Crystals suitable for structural analysis were grown from the
slow diffusion of diethyl ether into a concentrated THF
0
solution of 5-V O . Refinement of the data revealed the
6
5
(
Table S2). Unsurprisingly, our calculations suggest that the
expected Lindqvist structure with two O-atom vacancies
formed on cis-V ions (Figure 3 and Table 1; for full
III
two reduced metal centers can be assigned as V , while the
remaining vanadyl sites occupy a tetravalent oxidation state. A
charge distribution of V 4V
IV III
0
for complex 5-V O5 is
2
6
consistent with its electronic absorption spectrum, which
bears no IVCT bands (vide supra). As in the case of
[
V O (OCH ) OTf], the BVS values calculated for individual
6 6 3 12
V ions deviate from the integer values of the anticipated
valence states, consistent with the established electronic
16
communication between V ions across the Lindqvist core.
Sole formation of the cis-vacancy product is surprising.
While formation of a second vacant site cis to the original O-
atom vacancy is statistically favored, we note that we never
1
observe additional resonances in the H NMR spectrum that
would be consistent with the formation of a trans-vacancy
product. Because the parent POV−alkoxide cluster is a well-
25
established Robin and Day class II delocalized system, IVCT
processes prevent localization of the oxidation states
throughout the cluster (supported by BVS calculations and
Figure 3. Molecular structures of 5-V O60 (left) and 6-V O (right)
shown with 50% probability ellipsoids. H atoms and solvent
6
6
6
molecules coordinated to the cluster have been removed for clarity.
0
0
electronic characterization of 4-V O ). Thus, the selective
6
7
In the case of complex 6-V O , the asymmetric unit cell contains half
6
6
generation of the cis product can be justified by the observed
of the POV−alkoxide cluster located at a crystallographic inversion
center.
0
5
structural perturbations of 5-V O , which require truncation
6
of the V −O bond distance. Positioning the vacant sites cis to
v
c
0
18
Table 1. Structural Parameters of Complexes 4-V O , 5-
one another allows for the formation of a “square” V O site
2 2
6
7
0
16
V O , and [V O (OCH ) (OTf)]
with enhanced stability, as observed in the “V −O −V ” (V1−
0
v
b
v
6
5
6
6
3 12
O−V2) bond angles of 5-V O (Figure 4 and Table 1).
0
6
5
5
-V O (E =
[V O (OCH ) (OTf)]
6 6 3 12
6
5
0
However, we cannot rule out an electronic influence, in which
bond
4-V O
N)
(E = O)
6
7
Vv−E (Å)
2.112(2),
.104(3)
2.052(8)
2
V −O (Å)
2.0666(17),
.0760(17)
105
2.079(4)
105
v
c
2
V −O −V (avg.)
v
b
c
(
deg)
V −O −V (deg)
96.78
2.3686
1.605
v
b
v
V−O (avg.) (Å)
2.25
1.60
2.317
1.585
c
V O (avg.) (Å)
t
t
crystallographic parameters, see Table S1). Each reduced V
0
center (V ) has a bound acetonitrile (MeCN), occupying the
v
Figure 4. Portion of 5-V O5 with other atoms in the POV−alkoxide
6
vacant site. While crystallization was not conducted in the
presence of MeCN, characterization of the cluster was
cluster removed to highlight distortion of the V−O −V bond angle
compared to the starting material 4-V O .
b
0
6
7
D
DOI: 10.1021/acs.inorgchem.9b00389
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Forum Article
case the VV center remaining after reduction of the first
vanadyl moiety to yield a monovacant cluster is localized to the
equatorial plane of the remaining vanadyl ions. This indeed is
apparent in the BVS calculations previously reported for
(Figure 2a). These features are significantly shifted from both
0
the parent cluster, 4-V O , and the monovacant anionic
6
7
−
16,20
cluster, 2-V O6 (Figure S7).
Previous reports have shown
6
a correlation between the vibrational energies of both the
[
V O (OCH ) OTf], where oxidation of the monovacancy-
ν(VO ) and ν(O -CH ) stretching frequencies and the
6
6
3 12
t
b
3
−
containing complex 2-V O6 is exclusively localized to vanadyl
ions positioned cis to the reduced V ion.
oxidation state distribution of vanadyl ions that compose the
6
16
20
POV−alkoxide cluster. For example, the monoanionic
−
Accessing a Neutral POV−Alkoxide Cluster with a
Single O-Atom Vacancy. The ability to remove two O atoms
from the POV−alkoxide cluster confirms our hypothesis that
the number of possible oxygen vacant sites is dependent on the
Lindqvist POV−alkoxide cluster 1-V O has absorbance
6
7
−
1
−1
features at 953 cm [ν(VO )] and 1047 cm (ν(Ob-
t
CH )). Upon oxidation of the hexavanadate cluster to 4-
3
0
−1
V O , the stretching frequencies shift to 964 and 1009 cm ,
6
7
V
16
0
6
available V moieties in the cluster core. Intrigued by whether
respectively. In the case of complex 6-V O , strong ν(V
6
we could selectively form a cluster containing a single vacancy
O ) and ν(O −CH ) frequencies are observed at 968 and 1030
t
b
3
−
1
−
from the neutral POV−alkoxide precursor, we shifted our
cm , whereas the reduced species 2-V O bears analogous
6
6
−1
16
focus to probing the generation of a neutral, monovacant
features at 951 and 1047 cm , respectively. These distinct
0
cluster via VO bond cleavage of 4-V O . The addition of 1
6
7
shifts in ν(VO ) and ν(O −CH ) are consistent with a one-
t
b
3
0
equiv of V(Mes) (THF) to 4-V O resulted in an
−
3
6
7
electron oxidation of 2-V O upon formation of the neutral
6 6
instantaneous color change from green to red-brown,
suggesting vacancy formation (Scheme 1). Analysis of the
0
0
species 6-V O . Evidence for the reduction of 4-V O
7
6
6
6
through the removal of a single O atom is also obtained
from electronic absorption spectroscopy (Figure 2b). The
absorption spectrum of the monovacant complex consists of
two IVCT bands, albeit with lower-intensity molar absorptivity
1
crude product by H NMR spectroscopy revealed a mixture of
the anticipated diamagnetic byproduct OVMes , small
3
0
5
amounts of complex 5-V O , and three new paramagnetically
6
shifted and broadened resonances (δ 25.44, 18.27, and −12.69;
0
values than those of the parent cluster 6-V O (394 nm, ε =
2
6
6
Figure S5). A similar three-peak pattern was observed in the
−1
−1
−1
−1
766 M cm ; 998 nm, 450 M cm ). The decrease in the
1
−
H NMR spectra of 2-V O , suggesting that the newly shifted
6
6
intensity of the IVCT bands is a result of the lower probability
resonances correspond to the formation of a product with a
V
IV
of electronic exchange between V /V ions as a result of the
similar reduction in the molecular symmetry (O → C ), albeit
V
III
h
4v
reduction of one V center to V , further confirming the
16
in a different charge state. These qualifications fit well with
V IV III
oxidation state assignment of V V V . The electronic
absorption spectra of 6-V O also contain a small absorption
at 530 nm (ε = 891 M cm ), corresponding to a spin-
4
formation of the desired product, the neutral monovacant
0
0
0
6
6
cluster [V O (OCH ) (MeCN)] (6-V O ). However, ad-
−1
−1
6
6
3 12
6
6
justing the reaction conditions by altering the time, temper-
IV
IV
2
2
forbidden d (V )−d (V ) excitation.
xy
x −y
ature, solvent, and equivalents of V(Mes) (THF) added
3
Electronic Properties of Mono- and Divacant POV−
Alkoxide Clusters. Interested in further understanding the
electronic consequences of the removal of a second O atom
from the surface of the POV−alkoxide cluster, the electro-
0
resulted in nominal changes in the purity of 6-V O .
6
6
Given the challenges associated with the selective generation
0
of 6-V O via vacancy formation, attempts to access this
6
6
−
molecule via the oxidation of 2-V O6 were performed with 1
equiv of FcBF (Fc = ferrocenium; Scheme 1; confirmed by H
0
5
0
6
chemical profiles of complexes 5-V O and 6-V O were
1
6
6
6
4
interrogated by cyclic voltammetry (CV; Figure 5). The open-
NMR; Figure S6). While this independent synthesis resulted in
0
+
0
circuit potential of 5-V
6
O
5
(−0.39 V vs Fc/Fc ) suggests that
generation of the neutral species 6-V O6 as the sole
6
the divacant molecule is in its most reduced state, consistent
with the oxidation state distribution assignment predicted via
paramagnetic product, challenges in isolating this reactive
n
complex from the byproduct [ Bu N][BF ] persisted. The
4
4
BVS calculations and the electronic absorption spectrum
enhanced purity associated with the oxidative preparation of 6-
IV III
0
(V
4
V
). Two oxidation events are observed, centered at
2
V O afforded opportunities to rigorously confirm the vacancy
6
6
formation via X-ray crystallography, however, by selecting
deep-brown crystals from co-crystallized translucent salts.
Indeed, crystals suitable for analysis were grown from the
slow diffusion of pentane into a concentrated THF solution of
0
6
−
complex 6-V O synthesized via the oxidation of 2-V O6
6
6
with 1 equiv of FcBF . Structural refinement revealed the
4
expected Lindqvist structure with a single O atom removed
and replaced with MeCN (Figure 3; for full crystallographic
parameters, see Table S3). The neutral monovacant cluster
crystallized with a crystallographic inversion center, causing
VO/V−MeCN disorder. While this disorder within the unit
0
6
cell resulted in unusable bond metrics for complex 6-V O ,
6
structural analysis unambiguously confirms MO bond
cleavage.
To further investigate the perturbations in the electronic
0
V IV III
structure of 6-V O (electronic distribution: V V V ), IR
6
6
4
and electronic absorption spectroscopies were employed. In
0
the IR spectrum of the highest purity sample of 6-V O , two
0
6
6
Figure 5. Cyclic voltammograms of complexes 4-V O (green). 5-
6
7
0
0
strong absorptions were observed, corresponding to the ν(V
V O (red), and 6-V O6 (blue) collected in DCM with 0.1 M
6 5 6
−
1
−1
n
O ) (968 cm ) and ν(O −CH ) (1030 cm ) frequencies
[ Bu N][PF ] as the supporting electrolyte.
4 6
t
b
3
E
DOI: 10.1021/acs.inorgchem.9b00389
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Forum Article
+
E1/2 values of −0.305 and 0.256 V (vs Fc/Fc ). In comparison
The activation of CO by metal oxide supports is a well-
established heterogeneous transformation used to mediate the
0
to the electrochemical profile of 6-V O , which contains three
reversible events (E1 = −0.704, −0.155, and +0.519 V vs Fc/
Fc ), the removal of a second O atom further destabilizes the
redox chemistry of the Lindqvist core. This result follows the
trend seen in CV of the monovacant cluster, where reduction
of the polyoxovanadate assembly results in the loss of a single
redox event from the electrochemical profile of the initial
polyoxovanadate assembly (Figure 5).
6
6
32,33
/2
emission of this toxic environmental contaminant.
We
+
therefore began modeling substrate binding on bulk materials
with the addition of an electron-rich carbon monoxide
t
analogue, tert-butyl isocyanide (CN Bu), to the divacant
0
t
cluster. Treatment of 5-V O5 with 2 equiv CN Bu results in
6
t
0
isolation of the desired product [V O (OCH ) (CN Bu) ]
6
5
3 12
2
Previous work from our laboratory has established that
POV−alkoxide clusters exhibit fast electron-transfer kinetics
t
0
Scheme 2. Synthesis of 7-V O (CN Bu) and 8-
6
6
2
2
1
t
−
V O (CN Bu)
and quasi-reversible redox couples.
To evaluate the
6
6
consequences of O-atom removal from the periphery of the
hexavanadate cluster on the electrochemical properties of the
Lindqvist core, we examined the reversibility, diffusion
coefficients (D ), and electron-transfer rate constants (k ) for
0
0
0
0
all redox processes of 5-V O5 and 6-V O6 (Figures S8−S11
6
6
and Tables S4 and S5). Cyclic voltammograms of each event at
scan rates varying from 100 to 2000 mV s− were obtained to
1
determine the relationship between the peak current (i ) or
p
peak separation (ΔE ) and scan rate. The increase of ΔE with
p
p
0
increasing scan rate indicates that each event for 5-V O and
-V O is quasi-reversible (Figures S9a and S11a).
6
5
0
21,26
6
To
6
6
determine the diffusion coefficient (D ) associated with each
0
redox couple of the vacant clusters, the Randles−Sevcik
2
1,27−29
equation was used.
The comparable values of D for
0
0
0
0
−6
both 4-V O and complexes 5-V O5 and 6-V O6 (∼10 )
6
7
6
6
suggest that diffusion of the complexes through a solution is
not influenced by the formation of up to two vacant sites. The
Nicholson method was then used to determine the electron-
21,30,31
transfer rate constant.
of the resulting oxygen-deficient clusters 5-V O5 and 6-V O6
maintain k values comparable to that of 4-V O (Figures S12
These calculations show that both
0
0
6
6
0
t
0
0
6
7
[7-V O (CN Bu) ] in excellent yield (Scheme 2, 91%). The
6 5 2
and S13 and Table S6). The retention of fast electron-transfer
rates following the removal of one or two surface O atom(s)
suggests preservation of the delocalized electronic structure of
the hexanuclear core upon vacancy formation. This is further
supported by the IR spectra, electrochemical characterization,
1
t
0
H NMR spectrum of complex 7-V O (CN Bu) is composed
6
5
2
of five broad paramagnetic resonances that are slightly shifted
0
from the parent cluster 5-V O5 (Figure S14). Notably, we
6
were not able to observe a sixth resonance assigned to the
isonitrile tert-butyl moiety, likely because of fluxional
coordination of this ligand in solution.
0
0
0
and BVS calculations of 4-V O , 5-V O , and 6-V O , which
6
7
6
5
6
6
are consistent with extensive electron delocalization within
both the parent cluster and reduced scaffolds.
t
0
Analysis of the IR spectrum of 7-V O (CN Bu) reveals a
6
5
2
−1
new feature at 2195 cm , corresponding to ν(CN) of the
Modeling Substrate Interaction with Vacant Sites:
III
isonitrile ligands bound to the site-differentiated V ions
0
Reactivity of [V O (OCH ) (MeCN)2 ] and
6
5
3
1 2
(Figures 6 and S15). This value is substantially shifted from
−
[
V O (OCH ) ] with Isocyanide. With increasing concerns
t
−1
6
6
3 12
that of free CN Bu (2133 cm ), consistent with coordination
of the substrate to the coordinatively unsaturated, electron-rich
V
on the environmental impact of chemical pollutants, the
activation of energy-poor, gaseous molecules using metal oxide
surfaces has become one of the most important industrial
III
34
ion. Indeed, the observed shift in ν(CN) resembles
1
,3
processes for the production of energy-rich chemical fuels.
Computational investigations on RMO-catalyzed small-mole-
cule activation have determined that substrate binding to
defect sites at the surface of these materials is a critical step in
2
2
catalysis. However, while a mechanistic understanding is
crucial for the design of improved catalysts, the electronic
consequences of substrate coordination to oxygen-deficient
sites during catalysis remains poorly understood. In previous
work, we have demonstrated that the vacant POV−alkoxide
−
cluster 2-V O is able to activate molecular oxygen, resulting
6
6
−
in the quantitative formation of 1-V O via the restoration of
6
7
16
an O atom into the vacant site. As such, we hypothesized
that the mono- and divacant clusters could be used to model
substrate binding at defect sites on heterogeneous surfaces,
providing a means to explore the consequences of altering the
t
t
0
Figure 6. IR spectra of CN Bu, 7-V O (CN Bu) , and 8-
V O (CN Bu) .
6 6
6
5
2
III
t
−
metal-to-oxygen ratios on the reactivity of reduced V centers.
F
DOI: 10.1021/acs.inorgchem.9b00389
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Forum Article
values reported for isonitrile ligands bound to other low-valent
CV of the isonitrile-coordinated complexes (7-
−
1
35
t
0
t
−
V ions, ranging from 2150 to 2200 cm . Further analysis of
V O (CN Bu) and 8-V O (CN Bu) ) was used to explore
6 5 2 6 6
t
0
2
t
the IR spectrum of 7-V O (CN Bu) reveals the characteristic
the electronic consequences of binding CN Bu to the reduced
6
5
III
ν(VO ) and ν(O −CH ) vibrational modes observed in
V
center (Figure 7). In the case of the monovacant cluster,
t
b
3
−
1
Lindqvist POV−alkoxide clusters (968 and 1049 cm ,
respectively; Figures 6 and S15). These values closely resemble
those of complex 5-V O5 (964 and 1038 cm ), indicating
0
−1
6
retention of the structural integrity of the POV−alkoxide
ensemble upon coordination of the isonitrile ligands. Likewise,
the negligible shifts in both bands indicate limited changes in
the electronic structure of the vanadyl core upon substrate
coordination. Further support for this observation can be made
through analysis of the electronic absorption spectrum of 7-
t
0
V O (CN Bu) , which reveals no significant changes in the
6
5
2
t
absorption features, following the addition of CN Bu (Figure
S16).
t
Likewise, we investigated the reactivity of CN Bu with a
monovacant cluster. In comparing the electrochemical profile
of the mono- and divacant molecules, we observed that the
0
electronic structure of 5-V O resembles that of the anionic
6
5
1
−
monovacant cluster, 2-V O6 , in that both are the most
6
−
IV III
5
0
5
reduced state of the cluster (2-V O , V V ; 5-V O ,
4V 2). Furthermore, challenges in isolating 6-V O6 in
appreciable amounts prompted the use of complex 2-V O6 as
6
6
6
IV III 16
0
−
V
Figure 7. Cyclic voltammograms of complexes 2-V
6
O
6
(dark blue,
6
0
t
0
−
solid), 5-V O (red, solid), 7-V O (CN Bu) (orange, dashed), and
-V O (CN Bu) (light blue, dashed) collected in DCM with 0.1 M
6
5
6
5
2
6
t
−
8
[ Bu
the monovacant starting material. The addition of 1 equiv of
6
6
n
t
−
N][PF ] as the supporting electrolyte.
4
6
CN Bu to 2-V O resulted in isolation of the desired product
6
6
t
−
t
−
[
V O (OCH ) (CN Bu)] (8-V O (CN Bu) ) in quantita-
6 6 3 12 6 6
1
the CV curve of the hexavanadate assembly following substrate
tive yield (Scheme 2, 98%). Analysis of the product by H
NMR spectroscopy reveals slight shifts of the three para-
magnetic resonances observed in complex 2-V O , indicating
t
−
coordination [8-V O (CN Bu) ] in dichloromethane (DCM)
6
6
−
contains three quasi-reversible redox events, shifted anodically
6
6
by ∼0.10 V relative to the respective parent cluster. However,
retention of the charge state of the Lindqvist core (Figure
t
0
in the case of complex 7-V O (CN Bu) , in addition to a
6
5
2
S17). These shifts in the paramagnetic signals are consistent
slight shift in the redox potential (∼0.05 V), a third reversible
electrochemical event was observed for the cluster upon
coordination of isonitrile (+0.92 V). We justify the appearance
of this more-oxidizing redox event as a result of stabilization of
the cluster core in higher oxidation states due to the strong σ-
1
with changes in the H NMR spectrum observed upon the
t
formation of 7-V O (CN Bu), suggesting analogous isocyanide
6
5
coordination to the anionic monovacant cluster. Further
support for formation of the desired product was garnered
via ESI-MS (positive mode), which revealed the presence of a
t
donating properties of the two CN Bu ligands.
t
−
new parent ion peak that corresponds to 8-V O (CN Bu)
t
0
6
6
Next, we examined the redox profile of 7-V O (CN Bu) in
t
−
6
5
2
(
m/z 921; 8-V O (CN Bu) + Na + CH CN; Figure S18).
6 6 3
DCM to study substrate coordination to the multivacant
The electronic consequences of substrate coordination to
the surface defect site were evaluated using IR and electronic
0
model. The CV spectrum of the parent cluster 5-V O5
6
contains two quasi-reversible redox events centered at
absorption spectroscopies. The IR spectrum of 8-
+
−
0.305 and 0.256 V (vs Fc/Fc ). Similar redox events are
t
−
V O (CN Bu) revealed, once again, no changes in the
t
0
6
6
observed in the electrochemical profile of 7-V O (CN Bu) ,
6 5 2
energies of the ν(VO ) and ν(O −CH ) vibrational modes
t
b
3
albeit shifted anodically by ∼0.1 V. These slight shifts are
consistent with changes in the electrochemical profile observed
−
1
(
951 and 1051 cm , respectively) from that of the starting
material 2-V O (951 and 1047 cm , respectively; Figure
S15). The lack of shifts observed in these transitions suggests
a similar distribution of oxidation states of vanadyl ions
following coordination of the isonitrile moiety at the site-
−
−1
t
−
6
6
in 8-V O (CN Bu) (vide supra). However, unlike the cyclic
6 6
1
6
t
−
voltammogram of complex 8-V O (CN Bu) , the appearance
6
6
of an additional oxidizing event, located at E = 0.9164 V (vs
1
/2
+
Fc/Fc ), marks a notable change in the electrochemical profile
of 7-V O (CN Bu) . This surprising result suggests that the
addition of two electron-rich CN Bu ligands to the cis-V sites
of the molecular model results in the cooperative stabilization
of higher oxidation state distributions of the polynuclear core.
III
t
0
differentiated V ion. This is analogous to the observed lack of
6 5 2
t
III
change in the electronic structure of the Lindqvist core upon
coordination of two isonitrile entities in the case of 7-
t
0
V O (CN Bu) . In addition to the vanadyl-based stretching
6
5
2
t
−
frequencies, the IR spectrum of 8-V O (CN Bu) possesses a
6
6
−
1
new transition located at 2189 cm , which was assigned to
ν(CN) of the new isonitrile ligand (Figure 6). Electronic
absorption spectroscopy was used to further confirm the
CONCLUSION
■
In this work, we have demonstrated the selective formation of
neutral mono- and divacant POV−alkoxide clusters. The
formation of O-atom vacancies was found to be dependent on
t
−
oxidation state distribution of complex 8-V O (CN Bu) . A
6
6
V
comparison of the spectrum of the isonitrile-bound POV−
the number of available V ions in the hexavandate starting
−
−
0
alkoxide cluster to that of 2-V O revealed no noticeable
materials (e.g., 1-V O7 and 4-V O ). This work presents an
6
6
6 6 7
changes, indicating retention of the oxidation state distribution
unprecedented route to controlling metal-to-oxygen ratios on
Lindqvist POM clusters. The IR and electronic absorption
IV III
of the V ions (V V ; Figure S16).
5
G
DOI: 10.1021/acs.inorgchem.9b00389
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Forum Article
spectroscopies, as well as CV, established the retention of a
high degree of electron delocalization following the formation
of reduced V centers. This is a significant result in light of the
mmol) and 5 mL of THF. V(Mes)
3
(THF) (0.067 g, 0.139 mmol, 2.2
equiv) in 4 mL of THF was added dropwise to the solution with
stirring. The color of the mixture immediately changed from green to
red-brown. The reaction was stirred for an additional 1 h to ensure
completion, after which the solvent was removed under reduced
pressure to yield a red solid. The residue was stirred in diethyl ether
III
semiconducting properties of many RMO materials. Addition-
ally, coordination of a carbon monoxide analogue, tert-butyl
isocyanide, provides insight into the electronic consequences
of substrate binding at these surface vacant sites in materials.
Future work investigating POV−alkoxide clusters as molecular
models for heterogeneous catalysts will expand upon probing
the role of surface defects in small-molecule activation
processes and is currently underway in our laboratory.
(10 mL) for 30 min and then filtered over a bed of Celite (2 cm) to
give a red/black solid. The red/black solid was continuously washed
with diethyl ether until the filtrate ran clear, ensuring complete
removal of the byproduct OV(Mes) . The solid was then extracted
3
with THF (5 mL × 3), and any volatiles were removed under vacuum
0
to yield the product 5-V O5 as a red solid (90%). Crystals suitable
6
for X-ray analysis were grown from the slow diffusion of diethyl ether
into a concentrated solution of the product in THF. H NMR (500
1
EXPERIMENTAL SECTION
■
MHz, CD CN): δ 85.31, 27.10, 25.86, −1.80, −6.76, −22.46. FT-IR
3
−
1
General Considerations. All manipulations were carried out in
the absence of water and dioxygen using standard Schlenk techniques
or in a UniLab MBraun inert-atmosphere drybox under a dinitrogen
atmosphere except where specified otherwise. All glassware was oven-
dried for a minimum of 3 h and cooled in an evacuated antechamber
prior to use in the drybox. Unless otherwise noted, solvents were
dried and deoxygenated on a Glass Contour System (Pure Process
(ATR, cm ): 1038 (O−CH ), 964 (VO). UV−vis (CH CN): 524
3
3
−
1
− 1
nm (ε = 468 M cm ). Elem anal. Calcd for
V O C H N · / C H O (MW = 858.17 g mol ) : C, 23.79; H,
1
−1
6
17 16 42
2
4
4
8
5.17; N, 3.26. Found: C, 23.63; H, 4.87; N, 3.11. Note: Crystals used
for X-ray analysis were crushed up and dried under vacuum and
subsequently submitted for elemental analysis.
0
Synthesis of [V O (OCH ) (MeCN)] (6-V O ). Method A. In a
6
6
3 12
6
6
0
Technology, LLC) and stored over activated 3 Å molecular sieves
glovebox, a 20 mL scintillation vial was charged with 4-V O (0.0602
g, 0.0762 mmol) and 6 mL of THF. V(Mes )(THF) (0.0370 g,
6
7
n
purchased from Fisher Scientific prior to use. [ Bu N]-
4
3
−
−
20
n
2−
[
V O (OCH ) ] (1-V O ), [ Bu N] [V O (OCH ) ] (3-
6 7 3 12 6 7 4 2 6 7 3 12
18 18
0.0765 mmol, 1 equiv) in 4 mL of THF was added dropwise to the
solution with stirring. The color of the mixture immediately changed
from green to red-brown. The reaction was stirred for 2 h, after which
the solvent was removed under reduced pressure to yield a brown
solid. The residue was stirred in diethyl ether (10 mL) overnight and
then filtered over a bed of Celite (2 cm) to give a brown solid. The
solid was continuously washed with diethyl ether until the filtrate ran
2
−
0
0
n
V O7 ),
[
[V O (OCH )
]
(4-V O7 ),
[ Bu N]-
4
16
6
6
7
3
1 2
6
−
16
V O (OCH ) ] (2-V O ), [V O (OCH ) (OTf)], and V-
6 6 3 12 6 6 6 6 3 12
40
(
Mes) (THF) were prepared according to published procedures.
3
Ferrocenium tetrafluoroborate was purchased from Sigma-Aldrich and
used as received. Celite 545 (J. T. Baker) was dried in a Schlenk flask
for 48 h under vacuum while heating to at least 150 °C prior to use in
the glovebox. tert-Butyl isocyanide was purchased from Sigma-Aldrich
and used as received.
clear, ensuring complete removal of the byproduct OV(Mes) . The
3
solid was then extracted with THF (5 mL × 3), and any volatiles were
1H NMR spectra were recorded at 500 and 400 MHz on Bruker
DPX-500 and DPX-400 MHz spectrometers locked on the signal of
deuterated solvents. All chemical shifts were reported relative to the
0
6
0
5
removed under vacuum to yield a mixture of 6-V O and 5-V O .
6
6
1
H NMR (400 MHz, CD CN): δ 25.44, 18.27, −12.69. FT-IR (ATR,
3
−1
cm ): 1030 (O−CH ), 968 (VO). UV−vis (CH CN): 394 nm (ε
3
3
peak of the residual H signal in deuterated solvents. CD CN was
−1
−1
−1
−1
3
=
M
8
2766 M cm ), 530 nm (ε = 891 M cm ), 998 nm (ε = 450
purchased from Cambridge Isotope Laboratories, degassed by three
freeze−pump−thaw cycles, and stored over fully activated 3 Å
molecular sieves. IR [Fourier transform infrared (FT-IR) and
attenuated total reflectance (ATR)] spectra of complexes were
recorded on a Shimadzu IRAffinity-1 Fourier transform infrared
−
1
−1
1
cm ). Elem anal. Calcd for V O C H N· / C H O (MW =
6 18 14 39 4 4 8
1
−
33.12 g mol ): C, 21.62; H, 4.96; N, 1.68. Found: C, 21.81; H,
4.62; N, 1.38. Note: Crystals used for X-ray analysis were crushed up and
dried under vacuum and subsequently submitted for elemental analysis.
−
1
Method B. In a glovebox, a 20 mL scintillation vial was charged
spectrophotometer and are reported in wavenumbers (cm ).
Electronic absorption measurements were recorded at room temper-
ature in anhydrous MeCN in a sealed 1 cm quartz cuvette with an
Agilent Cary 60 UV−vis spectrophotometer. Mass spectrometry
−
with 2-V O6 (0.0501 g, 0.0493 mmol), FcBF (0.0139 g, 0.0509
6
4
mmol, 1 equiv), and 6 mL of MeCN. The reaction was stirred for 3 h,
after which the solvent was removed under reduced pressure to yield a
brown solid. The solid was stirred in pentane (10 mL) for 30 min.
The solid was then filtered and continuously washed with pentane
until the filtrate ran clear. The solid was then extracted with DCM,
and any volatiles were removed under vacuum to yield a mixture of 6-
L
analyses were performed on an Advion Expression compact mass
spectrometer equipped with an electrospray probe and an ion-trap
mass analyzer. Direct injection analysis was employed in all cases with
a sample solution in MeCN. CV experiments were recorded with a
Bio-Logic SP200 potentiostat/galvanostat and the EC-Lab software
suite. All measurements were performed in a three-electrode system
cell configuration that consisted of glassy carbon (⌀ = 3.0 mm) as the
working electrode (CH Instruments, USA), a Pt wire as the counter
0
n
V O and [ Bu N][BF ]. Crystals suitable for X-ray analysis were
6
6
4
4
grown from the slow diffusion of pentane into a concentrated solution
of the product in THF. Note: Elemental analysis on this sample was not
performed because of our inability to separate the tetrabutylammonium
0
+
tetrafluoroborate salt from compound 6-V O .
electrode (CH Instruments, USA), and an Ag/Ag nonaqueous
6
6
t
0
n
Synthesis of [V O (OCH ) (CNC(CH ) ) ] [7-V O (CN Bu) ].
reference electrode with 0.01 M AgNO in 0.05 M [ Bu N][PF ] in
6 5 3 12 3 3 2 6 5 2
3
4
6
0
In a glovebox, a 15 mL pressure vessel was charged with 5-V O
5
MeCN (BASi, USA). All electrochemical measurements were
6
(0.050 g, 0.060 mmol) and 5 mL of DCM. tert-Butyl isocyanide (15
performed at room temperature in a N -filled glovebox. Dry DCM
2
n
μL, 0.133 mmol, 2.2 equiv) was added to the solution with stirring.
The reaction was stirred for 2 h at 70 °C, after which the solvent was
removed under reduced pressure to yield a red solid. The residue was
stirred in pentane (10 mL) for 30 min and then filtered over a bed of
Celite (2 cm) to give a red solid. The solid was washed with 10 mL of
pentane to remove any unreacted tert-butyl isocyanide. The solid was
then extracted with DCM (5 mL × 3), and any volatiles were
that contained 0.1 M [ Bu N][PF ] was used as the electrolyte
4
6
solution.
Single crystals of 5-V O50 and 6-V O60 were mounted on the tip of
6
6
a thin glass optical fiber (goniometer head) and mounted on a XtaLab
Synergy-S dual-flex diffractometer equipped with a HyPix-6000HE
HPC area detector for data collection at 100.00(10) K. The structure
41
was solved using SHELXT-2018/2 and refined using SHELXL-
42
t
0
2
018/3. Elemental analyses were performed on a PerkinElmer 2400
removed under vacuum to yield the product 7-V
red solid (0.060 g, 0.066 mmol, 91%). H NMR (500 MHz,
6
O
5
(CN Bu)
2
as a
1
series II analyzer, at the CENTC Elemental Analysis Facility,
University of Rochester, Rochester. NY.
−
1
CD CN): δ 27.20, 25.96, −1.11, −6.75, −19.69. FT-IR (ATR, cm ):
3
0
Synthesis of [V O (OCH ) (MeCN) ] (5-V O ). In a glovebox,
2195 (CN), 1049 (O−CH ), 968 (VO). UV−vis (CH CN):
6
5
3 12
2
6
5
3
3
0
−1
−1
a 20 mL scintillation vial was charged with 4-V O7 (0.050 g, 0.063
532 nm (ε = 1871 M cm ). Elem anal. Calcd for
6
H
DOI: 10.1021/acs.inorgchem.9b00389
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Forum Article
1
−1
V O C H N · / C H O (MW = 942.33 g mol ): C, 29.32; H,
(3) Ved
2017, 7, 341−366.
́
rine, J. C. Heterogeneous catalysis on metal oxides. Catalysts
6
17 22 54
2
4
4
8
5
.99; N, 2.97. Found: C, 29.40; H, 5.72; N, 2.66.
Synthesis of [ Bu N][V O (OCH ) (CNC(CH ) )] (8-
V O (CN Bu) ). In a glovebox, a 15 mL pressure vessel was charged
n
4
6
6
3
12
3 3
(4) Ruiz Puigdollers, A.; Schlexer, P.; Tosoni, S.; Pacchioni, G.
Increasing oxide reducibility: the role of metal/oxide interfaces in the
formation of oxygen vacancies. ACS Catal. 2017, 7, 6493−6513.
t
−
6
6
−
with 2-V O6 (0.055 g, 0.0539 mmol) and 6 mL of DCM. tert-Butyl
6
isocyanide (7.25 μL, 0.0641 mmol, 1.2 equiv) was added to the
solution with stirring. The reaction was stirred for 2 h, after which the
solvent was removed under reduced pressure to yield 8-
V O (CN Bu) as a red solid (0.0579 g, 0.0527 mmol, 98%). H
NMR (500 MHz, CD CN): δ 25.41, 23.86, 3.07, 1.59, 1.35, 0.97,
(
5) Ek, M.; Ramasse, Q. M.; Arnarson, L.; Georg Moses, P.; Helveg,
S. Visualizing atomic-scale redox dynamics in vanadium oxide-based
catalysts. Nat. Commun. 2017, 8, 1−9.
6) Lauhon, L. J.; Ho, W. Effects of temperature and other
experimental variables on single molecule vibrational spectroscopy
with the scanning tunneling microscope. Rev. Sci. Instrum. 2001, 72,
t
−
1
6
6
(
3
−
1
−
15.87. FT-IR (ATR, cm ): 2189 (CN), 1051 (O−CH ), 951
3
−
1
−1
(
VO). UV−vis (CH CN): 430 nm (ε = 763 M cm ), 522 nm (ε
3
2
16−223.
−
1
−1
=
(
512 M cm ). Elem anal. Calcd for V O C H N ·CH Cl
6 18 33 81 2 2 2
1
−
(7) Bai, C. Scanning Tunneling Microscopy and Its Application, 2nd
revised ed.; Shenghai Scientific & Technical Publishers: Shenghai,
China, 2000; Vol. 2.
MW = 1184.58 g mol ): C, 34.47; H, 7.06; N, 2.36. Found: C,
3
4.73; H, 7.11; N, 2.19.
(
8) Lapshin, R. V. Automatic drift elimination in probe microscope
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.inorg-
chem.9b00389.
■
images based on techniques of counter-scanning and topography
feature recognition. Meas. Sci. Technol. 2007, 18, 907−927.
(9) Marinello, F.; Balcon, M.; Schiavuta, P.; Carmignato, S.; Savio, E.
Thermal drift study on different commercial scanning probe
microscopes during the initial warming-up phase. Meas. Sci. Technol.
2011, 22, 094016.
S
1
H NMR, ESI-MS (positive mode), UV−vis, and IR
(
10) Diebold, U. The surface science of titanium dioxide. Surf. Sci.
Rep. 2003, 48, 53−229.
11) Hayashi, Y. Hetero and lacunary polyoxovanadate chemistry:
Synthesis, reactivity and structural aspects. Coord. Chem. Rev. 2011,
spectroscopic data for all complexes, crystallographic
0
0
parameters of complexes 5-V O5 and 6-V O , BVS
calculations of complex 5-V O , and the diffusion
6
6
6
(
0
6
5
parameters and electron-transfer rate constants for
2
55, 2270−2280.
0
0
5
0
complexes 4-V O , 5-V O , and 6-V O (PDF)
6
7
6
6
6
(12) Wang, S. S.; Yang, G. Y. Recent advances in polyoxometalate-
catalyzed reactions. Chem. Rev. 2015, 115 (11), 4893−4962.
Accession Codes
(
13) Long, D. L.; Burkholder, E.; Cronin, L. Polyoxometalate
CCDC 1883474 and 1896509 contain 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
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
clusters, nanostructures and materials: from self assembly to designer
materials and devices. Chem. Soc. Rev. 2007, 36, 105−121.
(14) Yin, P.; Li, D.; Liu, T. Solution behaviors and self-assembly of
polyoxometalates as models of macroions and amphiphilic poly-
oxometalate-organic hybrids as novel surfactants. Chem. Soc. Rev.
2012, 41, 7368−7383.
(15) Hutin, R.; Long, D.-L.; Cronin, L. Comprehensive Inorganic
AUTHOR INFORMATION
Corresponding Author
E-mail: matson@chem.rochester.edu.
ORCID
William W. Brennessel: 0000-0001-5461-1825
Ellen M. Matson: 0000-0003-3753-8288
Author Contributions
B.E.P., A.A.F., and M.L.M. contributed to the experimental
findings. All authors participated in data interpretation and
analysis. The manuscript was written through contributions of
B.E.P. and E.M.M. All authors have given approval to the final
version of the manuscript.
Chemistry II; Elsevier, 2013; Vol. 2.
16) Petel, B. E.; Brennessel, W. W.; Matson, E. M. Oxygen-atom
vacancy formation at polyoxovanadate clusters: homogeneous models
■
(
*
for reducible metal oxides. J. Am. Chem. Soc. 2018, 140, 8424−8428.
(
17) Monakhov, K. Y.; Bensch, W.; Kogerler, P. Semimetal-
̈
functionalised polyoxovanadates. Chem. Soc. Rev. 2015, 44, 8443−
8483.
(
18) Spandl, J.; Daniel, C.; Brudgam, I.; Hartl, H. Synthesis and
̈
structural characterization of redox active dodecamethoxoheptaox-
ohexavanadium clusters. Angew. Chem., Int. Ed. 2003, 42, 1163−1166.
IV
V
(
19) Daniel, C.; Hartl, H. A mixed-valence V /V alkoxo-
n±
polyoxovanadium cluster series [V O (OCH ) ] exploring the
6
8
3 11
influence of a μ-Oxo ligand in a spin frustrated structure. J. Am. Chem.
Soc. 2009, 131, 5101−5114.
IV
V
Notes
(20) Daniel, C.; Hartl, H. Neutral and cationic V /V mixed-
n+
The authors declare no competing financial interest.
valence alkoxo-polyoxovanadium clusters [V
O
6
7
(OR)12
]
(R = −
CH −C H ): structural, cyclovoltammetric and IR-spectroscopic
3
2
5
ACKNOWLEDGMENTS
The authors acknowledge the National Science Foundation for
financial support of this research through Grants CHE-
653195 and CHE-1725028 (MRI). The authors also
acknowledge generous financial support from the University
of Rochester.
investigations on mixed valency in a hexanuclear core. J. Am. Chem.
Soc. 2005, 127, 13978−13987.
■
(
21) VanGelder, L. E.; Kosswattaarachchi, A. M.; Forrestel, P. L.;
Cook, T. R.; Matson, E. M. Polyoxovanadate-alkoxide clusters as
multi-electron charge carriers for symmetric non-aqueous redox flow
batteries. Chem. Sci. 2018, 9, 1692−1699.
1
(22) Mars, P.; van Krevelen, D. W. Oxidations carried out by means
of vanadium oxide catalysts. Chem. Eng. Sci. 1954, 3, 41−59.
REFERENCES
(
23) Pan, X.; Yang, M.-Q.; Fu, X.; Zhang, N.; Xu, Y.-J. Defective
■
(
1) McFarland, E. W.; Metiu, H. Catalysis by doped oxides. Chem.
TiO2 with oxygen vacancies: synthesis, properties and photocatalytic
applications. Nanoscale 2013, 5, 3601−3614.
(24) Li, F.; Carpenter, S. H.; Higgins, R. F.; Hitt, M. G.; Brennessel,
W. W.; Ferrier, M. G.; Cary, S. K.; Lezama-Pacheco, J. S.; Wright, J.
T.; Stein, B. W.; Shores, M. P.; Kozimor, S. A.; Neidig, M. L.; Matson,
Rev. 2013, 113, 4391−427.
(
2) Lee, D. W.; Yoo, B. R. Advanced metal oxide (supported)
catalysts: Synthesis and applications. J. Ind. Eng. Chem. 2014, 20,
947−3959.
3
I
DOI: 10.1021/acs.inorgchem.9b00389
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Forum Article
E. M. Polyoxovanadate−alkoxide clusters as a redox reservoir for iron.
Inorg. Chem. 2017, 56, 7065−7080.
(
25) Robin, M. B.; Day, P. Mixed Valence ChemistryA Survey and
Classification; World Scientific: New York, 1968; Vol. 10.
(
26) Ortiz, M. E.; Nunez-Vergara, L. J.; Squella, J. A. Voltammetric
́
̃
determination of the heterogeneous charge transfer rate constant for
superoxide formation at a glassy carbon electrode in aprotic medium.
J. Electroanal. Chem. 2003, 549, 157−160.
(
27) Liu, Q.; Shinkle, A. A.; Li, Y.; Monroe, C. W.; Thompson, L.
T.; Sleightholme, A. E. S. Non-aqueous chromium acetylacetonate
electrolyte for redox flow batteries. Electrochem. Commun. 2010, 12,
1
(
634−1637.
28) Sleightholme, A. E. S.; Shinkle, A. A.; Liu, Q.; Li, Y.; Monroe,
C. W.; Thompson, L. T. Non-aqueous manganese acetylacetonate
electrolyte for redox flow batteries. J. Power Sources 2011, 196, 5742−
5
(
745.
29) Liu, Q.; Sleightholme, A. E. S.; Shinkle, A. A.; Li, Y.;
Thompson, L. T. Non-aqueous vanadium acetylacetonate electrolyte
for redox flow batteries. Electrochem. Commun. 2009, 11, 2312−2315.
(
30) Nicholson, R. S.; Shain, I. Theory of stationary electrode
polarography. single scan and cyclic methods applied to reversible,
irreversible, and kinetic Systems. Anal. Chem. 1964, 36, 706−723.
(
31) Muhammad, H.; Tahiri, I. A.; Muhammad, M.; Masood, Z.;
Versiani, M. A.; Khaliq, O.; Latif, M.; Hanif, M. A comprehensive
heterogeneous electron transfer rate constant evaluation of dissolved
oxygen in DMSO at glassy carbon electrode measured by different
electrochemical methods. J. Electroanal. Chem. 2016, 775, 157−162.
(
32) Nie, L.; Mei, D.; Xiong, H.; Peng, B.; Ren, Z.; Hernandez, X. I.
P.; De La Riva, A.; Wang, M.; Engelhard, M. H.; Kovarik, L.; Datye,
A. K.; Wang, Y. Activation of surface lattice oxygen in single-atom Pt/
CeO for low-temperature CO oxidation. Science 2017, 358, 1419−
2
1
(
423.
33) Cheng, T.; Xiao, H.; Goddard, W. A. Nature of the active sites
for CO reduction on copper nanoparticles; suggestions for optimizing
performance. J. Am. Chem. Soc. 2017, 139, 11642−11645.
(
34) Crabtree, R. H. The Organometallic Chemistry of the Transition
Metals, 4th ed.; Wiley-Interscience: Hoboken, NJ, 2005.
35) Silverman, L. D.; Dewan, J. C.; Giandomenico, C. M.; Lippard,
(
S. J. Molecular structure and ligand-exchange reactions of
trichlorotris(tert-butyl isocyanide)vanadium(III). Synthesis of the
hexakis(tert-butyl isocyanide)vanadium(II) cation. Inorg. Chem. 1980,
1
(
9, 3379−3383.
36) Meyer, R. L.; Brennessel, W. W.; Matson, E. M. Synthesis of a
gallium-functionalized polyoxovanadate-alkoxide cluster: Toward a
general route for heterometal installation. Polyhedron 2018, 156, 303−
3
(
11.
37) VanGelder, L. E.; Brennessel, W. W.; Matson, E. M. Tuning the
redox profiles of polyoxovanadate-alkoxide clusters via heterometal
installation: toward designer redox reagents. Dalton Trans. 2018, 47,
3
(
698−3704.
38) Li, F.; VanGelder, L. E.; Brennessel, W. W.; Matson, E. M. Self-
assembled, iron-functionalized polyoxovanadate alkoxide clusters.
Inorg. Chem. 2016, 55, 7332−7334.
(
39) Zhang, T.; Sole-Daura, A.; Hostachy, S.; Blanchard, S.; Paris,
C.; Li, Y.; Carbo, J. J.; Poblet, J. M.; Proust, A.; Guillemot, G.
Modeling the oxygen vacancy at a molecular vanadium(III) silica-
supported catalyst. J. Am. Chem. Soc. 2018, 140, 14903−14914.
(
40) Hamaed, A.; Mai, H. V.; Hoang, T. K. A.; Trudeau, M.;
Antonelli, D. Functionalized porous silicas with unsaturated early
transition metal moieties as hydrogen storage materials: comparison
of metal and oxidation state. J. Phys. Chem. C 2010, 114, 8651−8660.
(
41) Sheldrick, G. M. SHELXT, version 2018/2. Acta Crystallogr.,
Sect. A: Found. Adv. 2015, 71, 3−8.
42) Sheldrick, G. M. SHELXL, version 2018/3. Acta Crystallogr.
015, C71, 3−8.
(
2
J
DOI: 10.1021/acs.inorgchem.9b00389
Inorg. Chem. XXXX, XXX, XXX−XXX