Mechanistic Aspects of a Surface Organovanadium(III) Catalyst for Hydrocarbon Hydrogenation and Dehydrogenation

Article abstract of DOI:10.1021/acscatal.9b02800

Understanding the mechanisms of action for base metal catalysis of transformations typically associated with precious metals is essential for the design of technologies for a sustainable energy economy. Isolated transition-metal and post-transition-metal catalysts on oxides such as silica are generally proposed to effect hydrogenation and dehydrogenation by a mechanism featuring either σ-bond metathesis or heterolytic bond cleavage as the key bond activation step. In this work, an organovanadium(III) complex on silica, which is a precatalyst for both olefin hydrogenation and alkane dehydrogenation, is interrogated by a series of reaction kinetics and isotopic labeling studies in order to shed light on the operant mechanism for hydrogenation. The kinetic dependencies of the reaction components are potentially consistent with both the σ-bond metathesis and the heterolytic bond activation mechanisms; however, a key deuterium incorporation experiment definitively excludes the simple σ-bond metathesis mechanism. Alternatively, a two-electron redox cycle, rarely invoked for homologous catalyst systems, is also consistent with experimental observations. Evidence supporting the formation of a persistent vanadium(III) hydride upon hydrogen treatment of the as-prepared material is also presented.

Full text of DOI:10.1021/acscatal.9b02800

Research Article
pubs.acs.org/acscatalysis
Cite This: ACS Catal. 2019, 9, 11055−11066
Mechanistic Aspects of a Surface Organovanadium(III) Catalyst for
Hydrocarbon Hydrogenation and Dehydrogenation
David M. Kaphan,* Magali S. Ferrandon, Ryan R. Langeslay, Gokhan Celik, Evan C. Wegener,
Cong Liu, Jens Niklas, Oleg G. Poluektov, and Massimiliano Delferro*
Chemical Sciences and Engineering Division, Argonne National Laboratory, 9700 S. Cass Ave., Lemont, Illinois 60439, United
States
S
* Supporting Information
ABSTRACT: Understanding the mechanisms of action for base metal catalysis
of transformations typically associated with precious metals is essential for the
design of technologies for a sustainable energy economy. Isolated transition-
metal and post-transition-metal catalysts on oxides such as silica are generally
proposed to effect hydrogenation and dehydrogenation by a mechanism
featuring either σ-bond metathesis or heterolytic bond cleavage as the key bond
activation step. In this work, an organovanadium(III) complex on silica, which is
a precatalyst for both olefin hydrogenation and alkane dehydrogenation, is
interrogated by a series of reaction kinetics and isotopic labeling studies in order
to shed light on the operant mechanism for hydrogenation. The kinetic dependencies of the reaction components are
potentially consistent with both the σ-bond metathesis and the heterolytic bond activation mechanisms; however, a key
deuterium incorporation experiment definitively excludes the simple σ-bond metathesis mechanism. Alternatively, a two-
electron redox cycle, rarely invoked for homologous catalyst systems, is also consistent with experimental observations. Evidence
supporting the formation of a persistent vanadium(III) hydride upon hydrogen treatment of the as-prepared material is also
presented.
KEYWORDS: heterogeneous catalysis, supported organometallic catalysis, mechanistic study, hydrogenation, organovanadium
own right.7 Isolated surface transition-metal and post-
transition-metal catalysts for hydrogenation and/or dehydro-
INTRODUCTION
■
The development of nonprecious-metal-based catalysts for
hydrocarbon hydrogenation and dehydrogenation increases in
genation are extensively studied and have been synthesized
from scandium,8 yttrium,8 titanium,9 zirconium,9,10 hafnium,9
importance, in conjunction with the shift of the energy
vanadium,4c,11 chromium,12 iron,13 cobalt,14 zinc,15 gallium,16
economy toward shale gas feedstocks.1 Platinum nanoparticles,
and indium17 inorganic and organometallic molecular
usually employed as a bimetallic alloy with tin, zinc, gallium, or
precursors.
other promoting metals, are the most widely used catalysts for
Mechanistic study of supported organometallic catalysts is
light hydrocarbon dehydrogenation.2 Chromium oxide sup-
complicated by heterogeneity, mass-transfer effects, as well as
ported on silica or alumina is a viable nonprecious-metal
an increase in the kinetics of oxidative and hydrolytic
decomposition, relative to analogous solvated or solid-state
alternative, and it has been adopted for industrial applica-
tions.2a,3
organometallics, as a result of their dispersion on high-surface-
Vanadium oxides have also shown promise as catalysts for
dehydrogenation, primarily in the context of oxidative
area support materials.7b Nevertheless, significant effort has
been devoted to understanding these mechanisms.14 Broadly,
dehydrogenation, but also under anaerobic conditions.4
hydrogenation and dehydrogenation over isolated metal
These catalysts, typically synthesized by impregnation or
catalysts, both those synthesized from organometallic
sol−gel processes, are characterized as a mixture of one-
dimensional and two-dimensional polymers and oligomers, as
precursors, as well as those synthesized by other methods,
well as some isolated vanadium sites.5 Recent work in this area
generally proceeds through one of two mechanistic manifolds:
heterolytic bond activation (Scheme 1A) and metal hydride σ-
suggests that dehydrogenation occurs at the isolated sites;
however, the low abundance of these sites in catalytically
bond metathesis (Scheme 1B).
Isolated Cr, Co, Fe, Zn, Ga, and In are catalytically active in
the hydrogenation and dehydrogenation of light hydrocarbons
and likely proceed via the heterolytic bond activation
relevant material complicates the confirmation of this
hypothesis.4b,6
The application of supported transition-metal complexes
synthesized via the chemisorption of well-defined molecular
precursors has emerged as a compelling tool for the study of
both isolated and polyatomic metal oxide active sites, and they
have shown promise as active and selective catalysts in their
Received: July 3, 2019
Revised: October 21, 2019
© XXXX American Chemical Society
11055
DOI: 10.1021/acscatal.9b02800
ACS Catal. 2019, 9, 11055−11066

ACS Catalysis
Research Article
hydride transfer to the surface under conditions where
appreciable thermal dehydrogenation is thermodynamically
favorable (>450 °C).8,21 In order to rationalize the kinetic
incompetence of the heterolytic activation mechanism for
these metals, Hock and co-workers draw analogy to gas-phase
diatomic M−O bond dissociation energies, which are more
than 50 kcal/mol higher for Sc and Y than for metals known to
undergo heterolytic bond activation, such as Cr, Fe, and Co.11
Notably, while many of the isolated transition-metal catalysts
for hydrogenation and dehydrogenation have the filled d-
orbitals required for a mechanism involving activation of C−H
or H−H bonds by oxidative addition, no experimental or
computation evidence for this mechanism has been reported
for these systems.
Scheme 1. Mechanisms of Hydrogenation and
Dehydrogenation for Isolated Transition-Metal and Post-
Transition-Metal Catalysts on Silica: (A) Heterolytic
Cleavage Mecahnism and (B) σ-Bond Metathesis
Mechanism
It is not clear a priori if isolated vanadium on silica should
behave more similar to the group III and IV transition metals
and favor the σ-bond metathesis mechanism, or if it is more
likely to follow the heterolytic bond activation mechanism by
analogy to the later first-row transition-metal catalysts. In 2013,
Taoufik and co-workers described the synthesis of a well-
defined V(V) oxo mesityl complex on silica ((≡SiO)2VV(O)-
Mes), and its application as a catalyst for selective non-
oxidative propane dehydrogenation (Scheme 2A).4c Our group
Scheme 2. (A) (≡SiO)2VV(O)Mes Catalysis of Propane
Dehydrogenation Reported by Taoufik and Coworkers,4c
and (B) (≡SiO)2VIII(THF)Mes Catalysis of Propylene
Hydrogenation Reported by Delferro and Co-workers11a
mechanism. The commercial CrOx/Al2O3 catalyst for propane
dehydrogenation (the CATOFIN process)2a is thought to
initiate the catalytic cycle by heterolytic C−H bond activation,
based on labeling and operando studies; however, it is not clear
if catalytic turnover occurs via the heterolytic bond activation
or σ-bond metathesis mechanism once the Cr−H is
generated.3,12 Coperet and co-workers provide strong evidence
́
for the heterolytic bond activation mechanism for isolated Co
on silica, synthesized by the thermolytic molecular precursor
approach,18 including a para-hydrogen-induced polarization
experiment consistent with pairwise hydrogen incorporation,
as well as a detailed study of the kinetics of H/D exchange and
the hydrogenation process.14 This mechanism has also been
validated computationally for Cr, Co, and Zn, among
others.12,15a,17,19
Supported group III (Sc and Y) and IV (Ti, Zr, and Hf)
organometallic catalysts are proposed to hydrogenate olefins
by metal hydride insertion into the CC double bond,
followed by direct hydrogenolysis of the resulting metal alkyl
by σ-bond metathesis with dihydrogen to regenerate the metal
hydride. These d0 metal centers are highly electrophilic and
feature unusually low electron counts, because of the low
coordination number enforced by the structure of the oxide
surface. The mechanism of bond activation for these catalysts
is generally inferred by analogy to the homogeneous
literature.20 In support of this inference, it was demonstrated
that, while bipodal Sc and Y alkyls are competent hydro-
genation catalysts, authentically prepared isolated tripodal Sc
and Y sites on silica were incapable of effecting olefin
hydrogenation, presumably because of their inability to
heterolytically activate dihydrogen across the M−O bond.
These hydride catalysts are only metastable and undergo
later found that the V(III) homologue on silica
((≡SiO)2VIII(THF)Mes, 1) was active at room temperature
for olefin and alkyne hydrogenation (Scheme 2B).11a However,
little mechanistic information pertaining to these systems has
been reported. In 2018, Taoufik and co-workers reported that,
when the vanadium(V) catalyst system was employed for
hydrogenation above 200 °C, some pairwise hydrogen
incorporation occurred based on a para-hydrogen-induced
polarization study; however, it was inconclusive if this was a
dominant reaction pathway, because of potential relaxation
from the paramagnetic vanadium metal center.22 In this work,
a detailed investigation of the mechanism of olefin hydro-
genation catalyzed by V(III) catalyst 1 is reported and
interpreted in the context of its activity as a catalyst for both
hydrogenation and dehydrogenation.
11056
DOI: 10.1021/acscatal.9b02800
ACS Catal. 2019, 9, 11055−11066

ACS Catalysis
Research Article
1B). When the reaction profile for dehydrogenation with
catalyst 1 is compared to that of the vanadium(V) oxo species
reported by Taoufik and co-workers, 1 displays a significantly
shorter induction period and a comparable turnover
frequency.4c Based on these observations, it is plausible that
the vanadium(V) precatalyst is reduced in its induction period
and converges on the same active species that is generated
from the vanadium(III) precatalyst. This is also consistent with
the suggestion by Taoufik and co-workers that their catalyst is
reduced to V(III) under high-temperature hydrogenation
conditions (>250 °C).11b
RESULTS AND DISCUSSION
■
Bipodal organovanadium(III) sites on silica ((≡SiO)2V-
(THF)Mes, 1) were shown in our previous report to display
activity as a precatalyst for olefin hydrogenation;11a however,
the activity of this material as a precatalyst for alkane
dehydrogenation, and its performance relative to the
vanadium(V) analogue, was unknown.4c When the hydro-
genation of propylene was performed in a plug flow reactor at
22 °C with 100 mg of 1 (2.8% V (w/w)), and a 20 mL/min
flow rate of 1.2% propene and 42% hydrogen, an initial
conversion of 92% was observed, with gradual deactivation
over a 20 h period, in which time ∼120 turnovers had occurred
(see Figure 1A).23 When the hydrogenation was run under
Alternating between hydrogenation and dehydrogenation in
the plug flow reactor without recovering the catalyst revealed
that subjection to dehydrogenation conditions deactivates the
catalyst for hydrogenation, suggesting that the low-temperature
hydrogenation is effected by a thermally metastable catalyst
structure, not persistent at temperatures required for
dehydrogenation. Thus, insights from the hydrogenation
reaction cannot be applied to infer mechanistic details about
the dehydrogenation by the concept of microscopic reversi-
bility. Nonetheless, the catalytic mechanism of the hydro-
genation was intriguing, as there are very few well-defined
molecular vanadium complexes capable of promoting the
hydrogenation of unsaturated hydrocarbons. Gambarotta and
co-workers reported a V(II) hydride dimer that affected olefin
hydrogenation by a bimetallic V(II)/V(III) redox cycle.24
Toste and co-workers later reported a V(V) bisimido complex,
which was a catalyst for selective semihydrogenation of
alkynes, and activated dihydrogen by heterolytic cleavage
across the VN double bond.25 For this reason, a mechanistic
study on the low-temperature hydrogenation was pursued in
order to elucidate the nature of this process in the context of
the existing literature regarding isolated metal ion catalysis.
The nature of supported organometallic catalysts leads to
increased oxygen and moisture sensitivity, relative to molecular
analogues, mass-transfer effects, and surface heterogeneity,
each of which represent intrinsic complications for detailed
mechanistic study, and as a result, relatively few such studies
have appeared to date. In order to understand the mechanism
of action of our recently reported supported organovanadium-
(III) catalyst (1), conditions were required for reproducible
batch olefin hydrogenation that could be validated as free from
mass-transfer effects, and therefore kinetically pertinent to the
mechanism of hydrogenation. Hydrogen uptake was the
targeted as the method of kinetic reaction monitoring in
order to minimize the analytical load of the kinetic study,
compared to overhead sampling and analysis by gas
chromatography. The hydrogen uptake kinetics were validated
relative to the reaction sampling method in the hydrogenation
of 4-tBu-styrene (2) run in triplicate in an eight-well high-
throughput parallel pressure reactor, with two reactions
monitored by overhead sampling and the third by gas uptake
(see Figure 2). The reactions were run with 0.25 M 4-tBu-
styrene and 0.35 mol % of 1 at 50 °C under 200 psi of H2 with
1,3,5-tri-tert-butylbenzene as an internal standard.
Figure 1. (A) Hydrogenation of propylene in a plug flow reactor
catalyzed by 1. (B) Dehydrogenation of propane in a plug flow reactor
catalyzed by 1.
differential conversion at 22 °C, a rate of 6.1 × 10−4 s−1 was
obtained (see Figure S1 in the Supporting Information).
Dehydrogenation of propane was then performed at 550 °C
with 200 mg of 1 (1% V (w/w)), and a 5 mL/min flow rate of
2.3% propane, resulting in an initial turnover frequency 6.5 ×
10−5 s−1. The dehydrogenation, while modest in activity, was
very stable, decreasing only from 8.1% conversion to 4.6%
conversion over 200 h on stream, with high selectivity (94% to
92% propylene selectivity and >99% mass balance) (see Figure
The reaction profile from hydrogen uptake obtained by
monitoring pressure changes in the reactor headspace are in
good agreement with the kinetic profile from overhead
sampling. Periodic discontinuities in this hydrogen uptake
profile correspond to roughly 60 s settling intervals during
which the common overhead stirring for the three reactors was
halted to allow the catalyst to settle and prevent obstruction of
the sampling needle. During this period, hydrogen dissolution
11057
DOI: 10.1021/acscatal.9b02800
ACS Catal. 2019, 9, 11055−11066

ACS Catalysis
Research Article
The initial product of hydrogenolysis of the bipodal
vanadium(III) mesityl precatalyst is likely to be the analogous
vanadium(III) monohydridide complex; however, further
reactivity could plausibly furnish either hydride transfer to
the silica surface, generating a tripodal vanadium(III) species,
or, alternatively, oxidative addition of a further equivalent of
dihydrogen could generate a vanadium(V) trihydride species.
Hydride transfers to open surface siloxane rings are commonly
observed for early transition-metal hydrides, and silica
supported tantalum(III) monohydrides reversibly form small
amounts of tantalum(V) tridhydride under an atmosphere of
dihydrogen.26
Several spectroscopic methods were employed to further
characterize the species generated upon exposure of 1 to a
hydrogen atmosphere. A hydrogen-treated sample of 1
(hereafter referenced as 4) was first interrogated by CW X-
band EPR spectroscopy (see Figure 4). The resulting spectrum
Figure 2. Comparison of overhead sampling and H2 uptake as a
method for tracking the kinetics of hydrogenation by catalyst 1.
is slow and the catalyst is not suspended in the reaction
mixture, arresting the progress of the reaction. The reaction
proceeded with an approximate half-life of 45 min, with no sign
of styrene polymerization or side-product formation on the
time scale of the hydrogenation process.
Having validated the hydrogen uptake analytical method, the
reaction was repeated without overhead sampling in order to
remove the periodic discontinuities in the rate. The reaction
profile obtained from this experiment (Figure 3, blue trace)
Figure 4. CW X-band EPR spectrum of 4 (1 after hydrogen
treatment) and 4 after brief exposure to atmospheric oxygen. (See
spectrum of 1 in Figures S5 and S7 in the Supporting Information.)
showed only weak EPR signals, a broad peak at g ≈ 4, and a
relatively minor contribution of more narrow signals at g ≈ 2,
which we attribute to a small amount of V4+ (d1 configuration).
This small amount of V4+ is likely due to adventitious oxidation
of the more-reduced vanadium, while V3+ (d2 configuration) is
EPR silent, because of a large zero-field splitting.27 Upon
exposure to air, sample 4 showed an intense EPR signal typical
of V4+ in vanadyl complexes.27,28 Since dioxygen is strongly
oxidizing and generated V4+, the majority of the vanadium in
hydrogen-treated 4 is likely composed of a more-reduced state.
Since V2+ is expected to give substantial EPR signals which
were not observed, the majority of vanadium in sample 4 is
inferred to be in the V3+ oxidation state.
Vanadium K-edge X-ray absorption measurements were
performed on 1 and 4 to gain further information on effect of
hydrogen treatment on the local coordination environment
and oxidation state of the precatalyst. Changes in the pre-edge
and main absorption feature in the X-ray Absorption Near
Edge Structure (XANES) spectra (Figure 5, top) indicate a
change to the local coordination environment of V after
hydrogen treatment. However, the edge energy of 4 and 1
overlap to within the error of the measurement (5472.8 0.1
eV), providing further evidence that the vanadium center
remains in the 3+ oxidation state.29 The decreased intensity of
the feature at ∼1.5 Å observed in the Extended X-ray
Figure 3. Comparison of the hydrogen uptake reaction profile for 1
as-prepared (blue) and with a 30 min hydrogen pretreatment
(orange).
deviated significantly from simple first-order kinetic behavior,
with a slight increase in rate over the first 15% conversion,
suggestive of an induction period. When the reaction was
immediately preceded by a 30 min hydrogen treatment prior to
the addition of the substrate, the initial rate of the reaction
increased and the induction period was not observed (Figure 3,
orange trace). The initial rate the reaction was (1.22 0.07) ×
10−4 M/s under standard conditions (0.25 M 4-tBu-styrene,
200 psi of H2, 25 mg of 1 [1.4 mmol, 0.35 mol %], 50 °C; for a
discussion of error, see page S7 in the Supporting
Information). Elimination of the induction period upon
hydrogen treatment is consistent with the hydrogenolysis of
the vanadium mesityl occurring with a higher kinetic barrier
than the subsequent catalytic turnover of the reaction.
11058
DOI: 10.1021/acscatal.9b02800
ACS Catal. 2019, 9, 11055−11066

ACS Catalysis
Research Article
Brønsted acidic residual silanol sites). Furthermore, it was
reasoned that a surface vanadium hydride species 4 would be
active for insertion and oligomerization of ethylene, while a
tripodal vanadium species, lacking a hydride or alkyl for
insertion, would be relatively sluggish or inactive in this
transformation. Indeed, when 1 was subjected to 1 atm of
hydrogen for 1 h in a J. Young NMR tube, followed by
degassing and replacing the atmosphere with ethylene, rapid
oligomerization was observed with no apparent induction
period (see Figure S9 in the Supporting Information).
Interestingly, when vanadium mesityl complex 1 was subjected
to an atmosphere of ethylene without any hydrogen pretreat-
ment, an induction period was observed, followed by the
formation of a similar profile of oligomerization products,
concomitant with the formation of 2,4,6-trimethylstyrene (5),
presumably from the insertion of ethylene into the vanadium-
mesityl bond, followed by rapid β-hydride elimination (see
Scheme 3, as well as Figure S9). This oligomerization behavior
Scheme 3. Generation of Vanadium Hydride 2 by
Hydrogenolysis of 1 or Insertion of Ethylene Followed by β-
H Elimination
Figure 5. Vanadium K-edge XANES (top) and EXAFS (bottom)
spectra of 1 and 4 showing a decrease in the magnitude of the major
feature at ∼1.5 Å potentially consistent with a decrease in
nonhydrogen coordination number.
Absorption Figue Structure (EXAFS) spectrum of 4, compared
to 1 suggests a decrease in nonhydrogen coordination number,
consistent with the hydrogenolysis of the mesityl group to
form a vanadium hydride, although changes in symmetry at the
vanadium center may also account for this decrease in
magnitude (Figure 5, bottom).
Diffuse reflectance infrared Fourier transform spectroscopy
(DRIFTS) was employed in search of vanadium hydride
stretching resonances for hydrogen-treated 1 both in situ and
ex situ; however, these studies were inconclusive. This is not
surprising as early transition-metal hydrides are difficult to
characterize via infrared (IR) analysis, because of their
sensitivity and low molar absorptivity (see Figure S17 in the
Supporting Information).21 It is also possible that partial
activation to generate a low concentration of the isolated
metal hydride species explains the difficulty in spectroscopic
characterization.
More information about the nature of the species generated
by hydrogen treatment of 1 was obtained by indirect chemical
means. A hydrogen-treated sample of 1 was decomposed with
methanol-d4, which resulted in a complex mixture of species by
1H NMR spectroscopy, including both HD and H2, likely from
proteolytic decomposition of a vanadium hydride, although it
is conceivable that hydrides transferred to the surface could
also lead to the generation of HD/H2 (Figure S8 in the
Supporting Information; note that the formation of H2 may be
explained by rapid H/D exchange between methanol and
is consistent with the characterization of the hydrogen-treated
material as a vanadium(III) hydride (4), although it does not
specifically rule out the possibility of a tripodal species, as
silica-supported tripodal Ta(III) complexes have also been
shown to polymerize ethylene under certain conditions.30 Each
of these pieces of data based on spectroscopic and chemical
reactivity experiments are individually inconclusive but, taken
together, suggest the formation of vanadium hydride 4 upon
hydrogenolysis of mesityl complex 1.
In order to gain more information about the mechanism of
the reaction by probing the kinetic dependence on various
reaction components, it was first necessary to demonstrate that
the rate of the reaction was intrinsic to the chemistry occurring
at the organovanadium sites, and was not dominated by mass-
transfer effects. Mass-transfer limitations at the gas/liquid
interface could be ruled out by comparing the rate of hydrogen
dissolution to that of the reaction. By omitting the substrate
and catalyst from the reactor, the kinetics of dissolution could
be directly observed by monitoring the pressure of the reactor
headspace immediately after dosing hydrogen (see Figure S10
in the Supporting Information). The resulting hydrogen
pressure profile displayed first-order kinetics with an
approximate half-life of 18 s (k = 3.78 × 10−2 s−1, or an
11059
DOI: 10.1021/acscatal.9b02800
ACS Catal. 2019, 9, 11055−11066

ACS Catalysis
Research Article
Figure 6. Kinetic dependencies of 4-tBu-styrene hydrogenation catalyzed by 1. (A) Rate dependence on loading of 1. (B) Double log plot showing
first-order kinetics, with respect to 1. (C) Rate dependence on the initial concentration of 4-tBu-styrene. (D) Double reciprocal plot showing
saturation kinetics, with respect to styrene concentration. (E) Rate dependence on H2 pressure. (F) Double reciprocal plot showing saturation
kinetics with H2 pressure. (G) Rate dependence on the concentration of THF added to the reaction mixture. (H) Reciprocal plot consistent with
THF dissociation kinetics at high THF concentrations. (I) Rate dependence on H2 pressure in the saturation regime, with respect to substrate
concentration.
approximate initial rate of 1.43 × 10−3 M/s), which is more
than an order of magnitude faster than the rate of a
representative hydrogenation (1.22 × 10−4 M/s). Furthermore,
the observation of first-order kinetic dependence, with respect
to catalyst loading, is inconsistent with gas−liquid mass-
transfer limitations (vide infra). If mass-transfer limitations at
the bulk liquid/particle interface controlled the kinetics of the
reaction, then decreasing the catalyst particle size would
increase exterior particle surface area and, by extension,
reaction rate. To evaluate this, the catalyst was prepared with
silica sieved to three different ranges of particle size125−177
μm, 177−250 μm, and 250−500 μmand no effect on
reaction rate was observed (rates of 2.03 × 10−4 M/s, 1.96 ×
10−4 M/s, and 2.00 × 10−4 M/s, respectively, were
observed31). These experiments suggest that the kinetics of
the reaction are not governed by mass-transport limitations at
the gas/liquid or bulk liquid/particle interface, and the kinetic
dependencies of the reaction components should faithfully
reflect the reaction occurring at the vanadium active site.32
Having validated that the hydrogenation of 4-tBu-styrene
(2) with 1 was not mass-transfer-limited, the kinetic depend-
ence of the reaction was evaluated with respect to H2, 4-tBu-
styrene, 1, and THF (see Figure 6).33 A first-order rate
dependence on the loading of vanadium precatalyst 1 was
observed over a loading range of 0.35 mol % to 1.38 mol %
(see eq 1 and Figure 6A). This is confirmed by a double
logarithmic plot with a slope of 1.07 (ln[rate] vs ln[mol % 1];
see Figure 6B). The rate dependence, with respect to the
concentration of 4-tBu-styrene, was nonlinear, with close to
first-order behavior at low substrate loading, and only marginal
increases in rate observed after ∼1 M substrate concentration
(Figure 6C). This kinetic dependence could be linearized by
11060
DOI: 10.1021/acscatal.9b02800
ACS Catal. 2019, 9, 11055−11066

ACS Catalysis
Research Article
plotting the reciprocal of rate versus the reciprocal of the
styrene concentration, which is consistent with saturation
kinetics (see eq 2 and Figure 6D). Varying the pressure of
hydrogen in the reactor between 50 psi and 400 psi also led to
a nonlinear kinetic response; however, this deviation from first-
order behavior was much more subtle than that observed in the
study of substrate dependence (Figure 6E). Nonetheless, the
rate dependence on hydrogen pressure could also be linearized
in a double reciprocal plot, suggesting a similar saturation
mechanism for hydrogen (eq 3 and Figure 6F). The effect of
the addition of THF to the reaction mixture was then studied
in order to determine if THF dissociates from the vanadium
metal center between the resting state and rate-determining
transition state of the reaction. If THF dissociation is relevant
to the catalytic mechanism of the reaction, then added THF
should inhibit the reaction, and the reciprocal of the rate of the
reaction should be proportional to the concentration of THF.
Indeed, added THF was found to inhibit the reaction (Figure
6G); however, the reciprocal plot deviated from linearity at low
concentrations of THFthat is, THF inhibition is more
extreme than expected at low concentrations of THF, based on
a simple dissociation mechanism (see eq 4 and Figure 6F).
Two scenarios could account for this deviation from simple
dissociative kinetic behavior (see Figure S13 in the Supporting
Information): the first equivalents of THF could result in an
increase in coordination number, forming a bis(THF)
vanadium adduct, which then must dissociate one THF
molecule to perform the chemistry; alternatively, there may
exist a minority of highly active lower-coordinate vanadium
sites lacking in THF, which are transformed back to the
[(≡SiO)2V(THF)Mes] structure upon the addition of small
amounts of THF. The latter hypothesis is consistent with the
observation of small amounts (<5 mol %, relative to grafted
vanadium) of THF that are liberated upon chemisorption of
(THF)VMes3 in the synthesis of 1.
residual starting material during the reaction by reversible
hydride insertion (Figure 7, top). Deuterium incorporation at
Order in catalyst:
d(3)
dt
∝ [1]
(1)
Figure 7. (Top) Profile of 1H NMR signals for residual styrene during
hydrogenation catalyzed by 1 under an atmosphere of D2. (Bottom)
2H NMR spectrum at partial conversion showing deuterium
incorporation into the ethyl group of the product, as well as the β-
position of the residual styrene.
Order in substrate:
d(3)
dt
[2]
(C + [2])
∝
(2)
(3)
(4)
Order in hydrogen:
the β-position of the styrene was verified by the observation of
[H2]
(C + [H2])
d(3)
dt
2
new resonances at 5.2 and 5.7 ppm in the H NMR spectrum
∝
at partial conversion (260 min, Figure 7, bottom). Upon
completion of the reaction, ethyl benzene was afforded with
50% deuterium incorporation at the benzylic methylene, and
82% deuterium incorporation at the terminal methyl position
based on integration of the resulting 1H NMR spectrum. If the
insertion were irreversible, the expected deuterium incorpo-
ration would be 50% at the methylene and 33% at the methyl
position. The observation of exactly 50% deuterium incorpo-
ration at the methylene position and excess deuterium
incorporation at the methyl group implies that vanadium
hydride insertion occurs with complete regiospecificity,
affording the vanadium alkyl only at the more-substituted
benzylic position. This regiochemical outcome is in contrast
with several recently reported first-row transition-metal-based
catalysts for olefin hydrogenation, which add the deuteride at
Order in THF:
d(3)
dt
1
∝
at high [THF]
(C + [THF])
In order to evaluate the reversibility of the migratory
insertion step of the reaction, the hydrogenation of styrene was
conducted in a J. Young NMR tube under an atmosphere of
1
2
D2. The reaction was followed by both H and H NMR
spectroscopy. As the reaction progressed, the signals for the
protons in the β-position of the styrene starting material
diminished more rapidly than the signals associated with the
proton at the α-position or those associated with the aromatic
ring, suggesting that deuterium was incorporated into the
11061
DOI: 10.1021/acscatal.9b02800
ACS Catal. 2019, 9, 11055−11066

ACS Catalysis
Research Article
both the α- and β-position of styrenyl substrates.34 One
hypothesis to explain this observation is that the surface may
enforce a lower coordination number21 than would typically be
obtained for a molecular organometallic intermediate,
increasing the stabilization from developing η2 or η3 binding
modes when the vanadium is entering the benzylic position.
Based on these observations, consistent mechanisms can be
drawn for both the σ-bond metathesis mechanism (Scheme 4)
vanadium hydride (4, A) represents the resting state at low
concentrations of the styrene, transitioning to vanadium alkyl
species B at high concentrations of styrene, which would
explain the saturation kinetics in styrene concentration.
Dihydrogen would then cleave the vanadium alkyl through a
σ-bond metathesis mechanism, possibly through a lower-
coordinate intermediate, such as C. The nonlinearity of the
kinetic dependence on hydrogen may be taken into account by
a partial change in the rate-determining step from hydro-
genolysis by σ-bond metathesis at low hydrogen concentration,
to olefin insertion at higher pressures of hydrogen.
Scheme 4. Plausible σ-Bond Metathesis Mechanism of
Hydrogenation Consistent with Kinetic Data, But
Inconsistent with Deuterium Incorporation Pressure
Dependence
The heterolytic cleavage mechanism starts with bipodal or
tripodal species D (Ln = hydride, alkyl, silanol, or other X-type
ligand), which reversibly binds the styrenyl substrate as an
olefin complex (intermediate E), followed by heterolytic
activation of hydrogen to afford intermediate G. Reversible
migratory insertion generates vanadium alkyl H, which then
undergoes 1,2-elimination, releasing the hydrogenated product
and regenerating D. In this mechanism the resting state of the
cycle is species D at low concentration of styrene, shifting to π-
complex E at elevated styrene concentration, resulting in
saturation kinetics. Vanadium alkyl complex H can be ruled
out as the resting state at high styrene concentration for this
mechanism by the observation that the reaction is roughly first
order in hydrogen under saturation conditions, with respect to
the styrene (2.5 M 4-tBu-styrene; see Figure 6I). If H were the
resting state at high styrene concentration, then, under those
conditions, the reaction would be zero order in hydrogen, as
incorporation of H2 into the active site is prerequisite for the
formation of H. In contrast, the observation of a positive rate
dependence in hydrogen is consistent with the assignment of E
as the resting state, as hydrogen has yet to be incorporated at
that point in the catalytic cycle. The nonlinearity in hydrogen
pressure dependence is rationalized in this mechanism by a
buildup of intermediates F and/or G in the speciation of the
resting state, in which hydrogen has already been activated.
While direct evidence for each of the elementary steps in this
mechanism and those below is not obtained for this system,
these processes are precedented and commonly proposed for
well-defined homogeneous organometallic complexes. For
general rate laws and derivations for the σ-bond metathesis
and heterolytic bond activation mechanisms, see pages S19 and
S20, respectively, in the Supporting Information. In order to
evaluate if THF displacement by various ligands in the system
is energetically plausible, the thermodynamics of displacement
of THF by styrene (to form a π-complex) and by dihydrogen
(to form a σ-complex) was examined. The displacement of
THF from the vanadium complex following heterolytic
activation of dihydrogen was found to be 10.7 kcal/mol uphill
(see Figure S19 in the Supporting Information). This is a very
reasonable value for a reversible pre-equilibrium to generate a
high-energy intermediate in this system. Note that the lower-
valent vanadium(III) metal center is expected to be
significantly less oxophilic than its vanadium(V) analogue,
which contributes to the relatively low energetic cost for THF
dissociation. Dissociative ligand displacement is well-prece-
dented for this class of organometallic complex, including for
the molecular precursor to 1, (THF)VMes3.35 Displacing THF
from the bipodal vanadium monohydride with dihydrogen was
found to be uphill by 27.0 kcal/mol. This is value would be too
large for a high-energy intermediate in a reaction pathway that
proceeds at a reasonable rate at room temperature; however,
we do not discount the plausible intermediacy of a dihydrogen
and the heterolytic bond activation mechanism (Scheme 5) of
hydrogenation over isolated transition-metal sites on oxide
supports. In the σ-bond metathesis mechanism, a bipodal
Scheme 5. Plausible Heterolytic Bond Activation
Mechanism of Hydrogenation Consistent with Kinetic
Data and with Pressure Independent Deuterium
Incorporation
11062
DOI: 10.1021/acscatal.9b02800
ACS Catal. 2019, 9, 11055−11066

ACS Catalysis
Research Article
sigma complex, since the variable coordination-site geometries
present on the amorphous surface may decrease the
thermodynamic penalty of this equilibrium sufficiently to
bring it into the accessible range for a reaction between room
temperature and 50 °C. For further discussion of the viability
of THF displacement in these mechanisms, including
computational thermodynamic data, see page S17 in the
Supporting Information.
be consistent with a mechanism in which both hydrogenolysis
and β-hydride elimination were first-order, with regard to D2.
If, instead of σ-bond metathesis, the mechanism of vanadium
alkyl cleavage in Scheme 4 proceeded via oxidative addition of
dihydrogen to form vanadium(V) intermediate J, followed by
C−H reductive elimination (Scheme 6), it is possible that β-
Scheme 6. Plausible Redox Mechanism of Hydrogenation
Consistent with Kinetic Data
While mechanisms for the hydrogenation reaction by both
the σ-bond metathesis and heterolytic bond activation
pathways are both plausible and consistent with the kinetic
data, inspection of these reaction networks reveals a key
difference in the reaction sequences that can be leveraged for
their differentiation: the extent of excess deuterium incorpo-
ration is predicted to have a different pressure dependence for
the two mechanisms. When the hydrogenation is performed
with D2, the fraction of deuterium labeling at the β-position in
the product is governed by the ratio of product formation to β-
hydride elimination from the vanadium alkyl intermediate (C/
H). For the σ-bond metathesis mechanism, hydrogenolysis of
the alkyl group in intermediate C is first order in hydrogen,
whereas hydrogen is not involved in the β-hydride elimination,
making that elementary step zero-order, with regard to
hydrogen. This mechanism predicts that increasing the D2
pressure should decrease the amount of excess deuterium
incorporated in the product. In contrast, for the heterolytic
bond activation mechanism, the hydrogen molecule has
already been incorporated in vanadium alkyl intermediate H.
Both the 1,2-elimination step to form the hydrogenated
product, and the β-hydride elimination to reform the styrene
are zero-order in D2, which suggests that the extent of excess
deuterium incorporation should be independent of D2
pressure. When the reaction was performed under one, two,
and four atmospheres of D2, no discernible changes in the
extent of excess deuterium incorporation were observed, with a
measured deuterium fraction at the β-methyl group of 82%,
83%, and 81%, respectively (see Table 1). The independence
hydride elimination from the oxidized species, J, could be
faster than β-hydride elimination from vanadium(III) alkyl B.
If this were the case, then both the forward hydrogenolysis step
and backward β-hydride elimination step would be first-order,
with regard to hydrogen. Since these two steps have the same
order in hydrogen, no effect on the extent of deuterium
incorporation upon varying D2 pressure would be predicted,
which is consistent with the experimental outcome. In this
mechanism, vanadium alkyl intermediate B sits off-cycle and
represents the resting state at high styrene concentration. The
deviation from linearity in the rate dependence on hydrogen
pressure may be taken into account by a buildup of
vanadium(V) hydride species I and/or J. Such a redox-based
mechanism for olefin hydrogenation has not been proposed for
other isolated transition-metal ions on silica that catalyze
hydrogenation. However, many of these catalysts have a d0
electron configuration and are incapable of oxidative addition,
particularly those based on group III and IV transition metals.
While this redox cycling mechanism would require a delicate
balance of elementary rate constants, it is consistent with
experimental observations. While no direct precedent for the
intermediacy of vanadium(V) trihydride complexes has been
reported in the literature, some support for the kinetic
accessibility of this complex may be inferred from the
observation of the homologous tantalum(V) trihydride on
silica by Bassett and co-workers.30 While the paramagnetic
vanadium(III) hydride would be difficult to detect by solid-
state 1H NMR spectroscopy, the vanadium(V) trihydride
would be diamagnetic and, therefore, would be observable if it
were persistent and low enough in energy to be significantly
populated in the resting state. When the hydrogen-treated
Table 1. Effect of D2 Pressure on Excess Deuterium
Incorporation at the β-Position
D2 pressure (atm)
% deuterium at β-position
1
2
4
82
83
81
of the amount of excess deuterium incorporation on the
pressure of deuterium requires that the elementary steps of the
catalytic cycle forward and backward from the vanadium alkyl
intermediate must have the same order, with regard to
hydrogen. This is inconsistent with the σ-bond metathesis
mechanism, as drawn in Scheme 4, and this mechanism can be
ruled out for the hydrogenation of styrenyl substrate by
vanadium catalyst 1.
1
material was inspected by H NMR spectroscopy, no hydride
This deuterium pressure dependence study on the extent of
deuterium incorporation is consistent with the heterolytic
bond activation mechanism, as both the forward and backward
steps from the vanadium alkyl intermediate (H) are zero-order,
with regard to hydrogen; however, this observation would also
resonances were detectable; however, this does not discount
their mechanistic relevance as high-energy intermediates (see
Figure S21 in the Supporting Information). For a general rate
law and rate law derivation for the redox mechanism of
hydrogenation, see page S22 in the Supporting Information.
11063
DOI: 10.1021/acscatal.9b02800
ACS Catal. 2019, 9, 11055−11066

ACS Catalysis
Research Article
CONCLUSIONS
General discussion of synthetic and experimental
procedures, instrumentation, characterization of materi-
als, and rate laws and their derivations (PDF)
■
Of the two classes of mechanisms commonly proposed for
hydrogenation and dehydrogenation catalyzed by isolated
transition-metal and post-transition-metal complexes on silica,
for organovanadium precatalyst 1, the heterolytic bond
activation mechanism is consistent with the kinetic and
isotope labeling data reported, while the σ-bond metathesis
mechanism is excluded. The key experiment to differentiate
these mechanisms was the observation that the amount of
excess deuterium incorporation in the product was independ-
ent of the D2 pressure under which the reaction was run. An
alternative mechanism based on a vanadium(III)/vanadium-
(V) redox cycle is also consistent with the experimental data.
This mechanism, typically associated with late-transition-metal
molecular catalysts, is generally not favored for homologous
isolated transition-metal or post-transition-metal catalysts on
oxides such as silica.
AUTHOR INFORMATION
Corresponding Authors
*E-mail: kaphand@anl.gov (D. M. Kaphan).
*E-mail: delferro@anl.gov (M. Delferro).
ORCID
David M. Kaphan: 0000-0001-5293-7784
Ryan R. Langeslay: 0000-0003-2915-9309
Gokhan Celik: 0000-0001-8070-5219
Cong Liu: 0000-0002-2145-5034
Jens Niklas: 0000-0002-6462-2680
Oleg G. Poluektov: 0000-0003-3067-9272
Massimiliano Delferro: 0000-0002-4443-165X
■
Propane dehydrogenation catalyzed by 1, although not
studied in depth in this report, displays activity and selectivity
reminiscent of other isolated first-row transition-metal sites on
silica (such as Cr, Fe, and Co), plausibly suggesting a common
mechanism, for which heterolytic cleavage is typically
proposed. The active catalyst for low-temperature hydro-
genation must be a metastable structure that follows a different
mechanism than the dehydrogenation (vide supra). Although
the data for the hydrogenation reaction presented above are
consistent with both the heterolytic cleavage mechanism and
the redox cycling mechanism, one possibility is that the
vanadium hydride redox cycling mechanism occurs at low
temperatures, while under conditions required to drive the
thermodynamics of dehydrogenation, the more active catalytic
species is not persistent and the heterolytic cleavage
mechanism dominates. The interpretation of data in the
context of multiple possible mechanisms and then selection of
those that most simply explain the experimental observations is
consistent with recent direction from the editorial office of this
journal.36
Differentiation of the heterolytic cleavage mechanism and
the redox cycling mechanism may be possible by in situ
spectroscopic techniques in the presence of elevated pressures
of dihydrogen; however, the sensitivity of 1 has thus far
complicated these experiments. Nonetheless, efforts to this end
are ongoing, in addition to efforts to describe these energetic
landscapes, from a computational perspective. Understanding
the mechanisms by which base transition metals catalyze
reactions typically associated with precious metals such as
hydrogenation and dehydrogenation is of great importance to
the development of a new generation of sustainable catalyst
materials. Well-defined vanadium catalysts are less well-
studied, compared to some of their first-row transition-metal
neighbors; however, this element has a range of properties that
make its chemistry unique, and increased attention to its
catalytic applications is warranted.
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. Use
of the Advanced Photon Source is supported by the U.S.
Department of Energy, Office of Science, and Office of the
Basic Energy Sciences, under Contract No. DE-AC02−
06CH11357. MRCAT operations are supported by the
Department of Energy and the MRCAT member institutions.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors thank Prof. A. Fout and her research group for
assistance in attempting a para-hydrogen induced polarization
experiment on this system, J. Moher for experimental
assistance, Dr. J. Kropf for consultation on X-ray absorption
spectroscopy, and Dr. A. Sattelberger for helpful discussion.
We thank Dr. F. Dogan Key for assistance with solid-state
NMR spectroscopy.
REFERENCES
■
(1) Marcilly, C. Present Status and Future Trends in Catalysis for
Refining and Petrochemicals. J. Catal. 2003, 216, 47−62.
(2) (a) Sattler, J. J.; Ruiz-Martinez, J.; Santillan-Jimenez, E.;
Weckhuysen, B. M. Catalytic Dehydrogenation of Light Alkanes on
Metals and Metal Oxides. Chem. Rev. 2014, 114, 10613−10653.
(b) Bhasin, M. M.; McCain, J. H.; Vora, B. V.; Imai, T.; Pujado, P. R.
Dehydrogenation and Oxydehydrogenation of Paraffins to Olefins.
Appl. Catal., A 2001, 221, 397−419. (c) Sun, P. P.; Siddiqi, G.;
Vining, W. C.; Chi, M. F.; Bell, A. T. Novel Pt/Mg(In)(Al)O
Catalysts for Ethane and Propane Dehydrogenation. J. Catal. 2011,
282, 165−174.
(3) Lillehaug, S.; Børve, K. J.; Sierka, M.; Sauer, J. Catalytic
Dehydrogenation of Ethane over Mononuclear Cr(III) Surface Sites
on Silica. Part I. C-H Activation by σ-Bond Metathesis. J. Phys. Org.
Chem. 2004, 17, 990−1006.
(4) (a) Rossetti, I.; Fabbrini, L.; Ballarini, N.; Oliva, C.; Cavani, F.;
Cericola, A.; Bonelli, B.; Piumetti, M.; Garrone, E.; Dyrbeck, H.;
Blekkan, E. A.; Forni, L. V-Al-O Catalysts Prepared by Flame
Pyrolysis for the Oxidative Dehydrogenation of Propane to Propylene.
Catal. Today 2009, 141, 271−281. (b) Barman, S.; Maity, N.; Bhatte,
K.; Ould-Chikh, S.; Dachwald, O.; Haessner, C.; Saih, Y.; Abou-
Hamad, E.; Llorens, I.; Hazemann, J. L.; Kohler, K.; D’Elia, V.; Basset,
J. M. Single-Site VOx Moieties Generated on Silica by Surface
Organometallic Chemistry: A Way to Enhance the Catalytic Activity
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acscatal.9b02800.
XYZ coordinate files for computational assessment of
ligand displacement from vanadium to support mecha-
nistic discussion. Jmol files (ZIP)
11064
DOI: 10.1021/acscatal.9b02800
ACS Catal. 2019, 9, 11055−11066

ACS Catalysis
Research Article
in the Oxidative Dehydrogenation of Propane. ACS Catal. 2016, 6,
5908−5921. (c) Szeto, K. C.; Loges, B.; Merle, N.; Popoff, N.;
Quadrelli, E. A.; Jia, H. P.; Berrier, E.; De Mallmann, A.; Delevoye, L.;
Gauvin, R. M.; Taoufik, M. Vanadium Oxo Organometallic Species
Supported on Silica for the Selective Non-Oxidative Dehydrogenation
of Propane. Organometallics 2013, 32, 6452−6460.
(5) (a) Wegener, S. L.; Marks, T. J.; Stair, P. C. Design Strategies for
the Molecular Level Synthesis of Supported Catalysts. Acc. Chem. Res.
2012, 45, 206−14. (b) Carrero, C. A.; Schloegl, R.; Wachs, I. E.;
Schomaecker, R. Critical Literature Review of the Kinetics for the
Oxidative Dehydrogenation of Propane over Well-Defined Supported
Vanadium Oxide Catalysts. ACS Catal. 2014, 4, 3357−3380.
(6) Fukudome, K.; Ikenaga, N.; Miyake, T.; Suzuki, T. Oxidative
Dehydrogenation of Propane Using Lattice Oxygen of Vanadium
Oxides on Silica. Catal. Sci. Technol. 2011, 1, 987−998.
(7) (a) Pelletier, J. D.; Basset, J. M. Catalysis by Design: Well-
Defined Single-Site Heterogeneous Catalysts. Acc. Chem. Res. 2016,
49, 664−77. (b) Coperet, C.; Comas-Vives, A.; Conley, M. P.; Estes,
D. P.; Fedorov, A.; Mougel, V.; Nagae, H.; Nunez-Zarur, F.; Zhizhko,
P. A. Surface Organometallic and Coordination Chemistry toward
Single-Site Heterogeneous Catalysts: Strategies, Methods, Structures,
and Activities. Chem. Rev. 2016, 116, 323−421.
(8) Getsoian, A. G. B.; Hu, B.; Miller, J. T.; Hock, A. S. Silica-
Supported, Single-Site Sc and Y Alkyls for Catalytic Hydrogenation of
Propylene. Organometallics 2017, 36, 3677−3685.
(9) Zakharov, V. A.; Ryndin, Yu. A. Surface Hydride Complexes of
Group-IV Transition-Metals as Active-Sites for Polymerization and
Hydrogenation Reactions. J. Mol. Catal. 1989, 56, 183−193.
(10) (a) Schwartz, J.; Ward, M. D. Silica-Supported Zirconium
Hydrides as Isomerization or Hydrogenation Catalysts for Long-
Chain Olefins. J. Mol. Catal. 1980, 8, 465−469. (b) Hamzaoui, B.;
Eter, M. E.; Abou-Hamad, E.; Chen, Y.; Pelletier, J. D. A.; Basset, J.-
M. Well-Defined Single-Site Monohydride Silica-Supported Zirco-
nium from Azazirconacyclopropane. Chem. - Eur. J. 2015, 21, 4294−
4299.
(11) (a) Sohn, H.; Camacho-Bunquin, J.; Langeslay, R. R.; Ignacio-
De Leon, P. A.; Niklas, J.; Poluektov, O. G.; Liu, C.; Connell, J. G.;
Yang, D.; Kropf, J.; Kim, H.; Stair, P. C.; Ferrandon, M.; Delferro, M.
Isolated, Well-Defined Organovanadium(III) on Silica: Single-Site
Catalyst for Hydrogenation of Alkenes and Alkynes. Chem. Commun.
2017, 53, 7325−7328. (b) Zhivonitko, V. V.; Skovpin, I. V.; Szeto, K.
C.; Taoufik, M.; Koptyug, I. V. Parahydrogen-Induced Polarization
Study of the Silica-Supported Vanadium Oxo Organometallic
Catalyst. J. Phys. Chem. C 2018, 122, 4891−4900.
(12) Conley, M. P.; Delley, M. F.; Nunez-Zarur, F.; Comas-Vives,
A.; Coperet, C. Heterolytic Activation of C-H Bonds on Cr(III)-O
Surface Sites Is a Key Step in Catalytic Polymerization of Ethylene
and Dehydrogenation of Propane. Inorg. Chem. 2015, 54, 5065−5078.
(13) Hu, B.; Schweitzer, N. M.; Zhang, G.; Kraft, S. J.; Childers, D.
J.; Lanci, M. P.; Miller, J. T.; Hock, A. S. Isolated FeII on Silica as a
Selective Propane Dehydrogenation Catalyst. ACS Catal. 2015, 5,
3494−3503.
(14) Estes, D. P.; Siddiqi, G.; Allouche, F.; Kovtunov, K. V.;
Safonova, O. V.; Trigub, A. L.; Koptyug, I. V.; Coperet, C. C-H
Activation on Co,O Sites: Isolated Surface Sites Versus Molecular
Analogs. J. Am. Chem. Soc. 2016, 138, 14987−14997.
Catalyzed by Ga(i-Bu)3 Grafted onto γ-Alumina and Silica. ACS
Catal. 2018, 8, 7566−7577. (b) Getsoian, A.; Das, U.; Camacho-
Bunquin, J.; Zhang, G. H.; Gallagher, J. R.; Hu, B.; Cheah, S.;
Schaidle, J. A.; Ruddy, D. A.; Hensley, J. E.; Krause, T. R.; Curtiss, L.
A.; Miller, J. T.; Hock, A. S. Organometallic Model Complexes
Elucidate the Active Gallium Species in Alkane Dehydrogenation
Catalysts Based on Ligand Effects in Ga K-Edge Xanes. Catal. Sci.
Technol. 2016, 6, 6339−6353.
(17) Liu, C.; Camacho-Bunquin, J.; Ferrandon, M.; Savara, A.; Sohn,
H.; Yang, D.; Kaphan, D. M.; Langeslay, R. R.; Ignacio-De Leon, P.
A.; Liu, S.; Das, U.; Yang, B.; Hock, A. S.; Stair, P. C.; Curtiss, L. A.;
Delferro, M. Development of Activity-Descriptor Relationships for
Supported Metal Ion Hydrogenation Catalysts on Silica. Polyhedron
2018, 152, 73−83.
(18) (a) Male, J. L.; Niessen, H. G.; Bell, A. T.; Don Tilley, T.
Thermolytic Molecular Precursor Route to Active and Selective
Vanadia-Zirconia Catalysts for the Oxidative Dehydrogenation of
Propane. J. Catal. 2000, 194, 431−444. (b) Terry, K. W.; Tilley, T. D.
Trialkoxysiloxy Complexes as Precursors to Mo2·4SiO2 (M = Ti, Zr,
Hf) Materials. Chem. Mater. 1991, 3, 1001−1003. (c) Rulkens, R.;
Tilley, T. D. A Molecular Precursor Route to Active and Selective
Vanadia-Silica-Zirconia Heterogeneous Catalysts for the Oxidative
Dehydrogenation of Propane. J. Am. Chem. Soc. 1998, 120, 9959−
9960.
(19) Hu, B.; Getsoian, A.; Schweitzer, N. M.; Das, U.; Kim, H.;
Niklas, J.; Poluektov, O.; Curtiss, L. A.; Stair, P. C.; Miller, J. T.;
Hock, A. S. Selective Propane Dehydrogenation with Single-Site CoII
on SiO2 by a Non-Redox Mechanism. J. Catal. 2015, 322, 24−37.
(20) (a) Watson, P. L. Methane Exchange-Reactions of Lanthanide
and Early-Transition-Metal Methyl Complexes. J. Am. Chem. Soc.
1983, 105, 6491−6493. (b) Rabaa, H.; Saillard, J. Y.; Hoffmann, R.
Hydrogen-Hydrogen and Carbon-Hydrogen Activation Reactions at
d0 Metal Centers. J. Am. Chem. Soc. 1986, 108, 4327−4333.
(c) Thompson, M. E.; Baxter, S. M.; Bulls, A. R.; Burger, B. J.;
Nolan, M. C.; Santarsiero, B. D.; Schaefer, W. P.; Bercaw, J. E. σ-Bond
Metathesis for Carbon-Hydrogen Bonds of Hydrocarbons and Sc-R
(R = H, Alkyl, Aryl) Bonds of Permethylscandocene Derivatives.
Evidence for Noninvolvement of The π System in Electrophilic
Activation of Aromatic and Vinylic C-H Bonds. J. Am. Chem. Soc.
1987, 109, 203−219.
(21) Coperet, C.; Estes, D. P.; Larmier, K.; Searles, K. Isolated
Surface Hydrides: Formation, Structure, and Reactivity. Chem. Rev.
2016, 116, 8463−8505.
(22) Zhivonitko, V. V.; Skovpin, I. V.; Szeto, K. C.; Taoufik, M.;
Koptyug, I. V. Parahydrogen-Induced Polarization Study of the Silica-
Supported Vanadium Oxo Organometallic Catalyst. J. Phys. Chem. C
2018, 122, 4891−4900.
(23) This number of turnovers assumes that all vanadium sites are
equally active, which is unlikely, and, thus, represents an under-
estimate.
(24) Ghazar, A.; Gambarotta, S.; Yap, G. P. A. Oxidative Addition of
a Dinuclear and Divalent Vanadium Hydride to an Olefin C−H Bond,
Leading to Catalytic Hydrogenation. Organometallics 2001, 20, 5008−
5010.
(25) La Pierre, H. S.; Arnold, J.; Toste, F. D. Z-Selective
Semihydrogenation of Alkynes Catalyzed by a Cationic Vanadium
Bisimido Complex. Angew. Chem., Int. Ed. 2011, 50, 3900−3903.
(26) (a) Rataboul, F.; Baudouin, A.; Thieuleux, C.; Veyre, L.;
Coperet, C.; Thivolle-Cazat, J.; Basset, J. M.; Lesage, A.; Emsley, L.
Molecular Understanding of the Formation of Surface Zirconium
Hydrides Upon Thermal Treatment under Hydrogen of [(≡SiO)Zr-
(15) (a) Schweitzer, N. M.; Hu, B.; Das, U.; Kim, H.; Greeley, J.;
Curtiss, L. A.; Stair, P. C.; Miller, J. T.; Hock, A. S. Propylene
Hydrogenation and Propane Dehydrogenation by a Single-Site Zn2+
on Silica Catalyst. ACS Catal. 2014, 4, 1091−1098. (b) Camacho-
Bunquin, J.; Aich, P.; Ferrandon, M.; Getsoian, A.; Das, U.; Dogan, F.;
Curtiss, L. A.; Miller, J. T.; Marshall, C. L.; Hock, A. S.; Stair, P. C.
Single-Site Zinc on Silica Catalysts for Propylene Hydrogenation and
Propane Dehydrogenation: Synthesis and Reactivity Evaluation Using
an Integrated Atomic Layer Deposition-Catalysis Instrument. J. Catal.
2017, 345, 170−182.
t
(CH2 Bu)3] by Using Advanced Solid-State NMR Techniques. J. Am.
Chem. Soc. 2004, 126, 12541−12550. (b) Rosier, C.; Niccolai, G. P.;
Basset, J.-M. Catalytic Hydrogenolysis and Isomerization of Light
Alkanes over the Silica-Supported Titanium Hydride Complex
(≡SiO)3TiH. J. Am. Chem. Soc. 1997, 119, 12408−12409.
(c) Soignier, S.; Taoufik, M.; Le Roux, E.; Saggio, G.; Dablemont,
C.; Baudouin, A.; Lefebvre, F.; de Mallmann, A.; Thivolle-Cazat, J.;
Basset, J.-M.; Sunley, G.; Maunders, B. M. Tantalum Hydrides
(16) (a) Szeto, K. C.; Jones, Z. R.; Merle, N.; Rios, C.; Gallo, A.; Le
Quemener, F.; Delevoye, L.; Gauvin, R. M.; Scott, S. L.; Taoufik, M.
A Strong Support Effect in Selective Propane Dehydrogenation
11065
DOI: 10.1021/acscatal.9b02800
ACS Catal. 2019, 9, 11055−11066

ACS Catalysis
Research Article
Supported on MCM-41 Mesoporous Silica: Activation of Methane
and Thermal Evolution of the Tantalum-Methyl Species. Organo-
metallics 2006, 25, 1569−1577.
(27) Krzystek, J.; Ozarowski, A.; Telser, J.; Crans, D. C. High-
Frequency and -Field Electron Paramagnetic Resonance of Vanadium-
(IV, III, and II) Complexes. Coord. Chem. Rev. 2015, 301, 123−133.
(28) Smith, T. S.; LoBrutto, R.; Pecoraro, V. L. Paramagnetic
Spectroscopy of Vanadyl Complexes and Its Applications to Biological
Systems. Coord. Chem. Rev. 2002, 228, 1−18.
(29) Wong, J.; Lytle, F. W.; Messmer, R. P.; Maylotte, D. H. K-Edge
Absorption Spectra of Selected Vanadium Compounds. Phys. Rev. B:
Condens. Matter Mater. Phys. 1984, 30, 5596−5610.
(30) Chen, Y.; Ould-Chikh, S.; Abou-Hamad, E.; Callens, E.;
Mohandas, J. C.; Khalid, S.; Basset, J.-M. Facile and Efficient
Synthesis of the Surface Tantalum Hydride (≡SiO)2TaIIIH and Tris-
Siloxy Tantalum (≡SiO)3TaIII Starting from Novel Tantalum Surface
Species (≡SiO)TaMe4 and (≡SiO)2TaMe3. Organometallics 2014, 33,
1205−1211.
(31) For the study of particle size effects, 50 mg of catalyst was used,
instead of the standard conditions (25 mg).
(32) Rate-limiting diffusion of physisorbed substrates is not excluded
by these experiments, but is generally inconsistent with the kinetic
experiments described below.
(33) In lieu of performing each reaction in triplicate, the error
associated with the reaction was evaluated under standard conditions
both in parallel on the same day, and when performed on different
days with separate batches of reagents (see Figure S4 in the
Supporting Information). Five reactions run in parallel under the
identical conditions were found to have an average initial rate of 1.22
× 10−4 M/s with a standard deviation of 0.07 × 10−4 M/s. The same
conditions performed on five different days was found to have an
average initial rate of 1.26 × 10−4 M/s with a standard deviation of
0.16 × 10−4 M/s. When possible, directly compared rates were
collected in parallel, to minimize error. By studying the error of the
standard conditions, a more complete understanding of this error is
obtained, and a wider set of conditions can be evaluated, with respect
to each reaction variable.
́
(34) (a) Leonard, N. G.; Chirik, P. J. Air-Stable α-Diimine Nickel
Precatalysts for the Hydrogenation of Hindered, Unactivated Alkenes.
ACS Catal. 2018, 8, 342−348. (b) Tokmic, K.; Markus, C. R.; Zhu,
L.; Fout, A. R. Well-Defined Cobalt(I) Dihydrogen Catalyst:
Experimental Evidence for a Co(I)/Co(III) Redox Process in Olefin
Hydrogenation. J. Am. Chem. Soc. 2016, 138, 11907−11913.
(c) Friedfeld, M. R.; Shevlin, M.; Margulieux, G. W.; Campeau, L.
C.; Chirik, P. J. Cobalt-Catalyzed Enantioselective Hydrogenation of
Minimally Functionalized Alkenes: Isotopic Labeling Provides Insight
into the Origin of Stereoselectivity and Alkene Insertion Preferences.
J. Am. Chem. Soc. 2016, 138, 3314−3324.
(35) Seidel, V. W.; Kreisel, G. Darstellung und Eigenschaften von
Triarylvanadin-Verbindungen. Z. Anorg. Allg. Chem. 1977, 435, 146−
152.
(36) Scott, S. L. The Burden of Disproof. ACS Catal. 2019, 9, 4706−
4708.
11066
DOI: 10.1021/acscatal.9b02800
ACS Catal. 2019, 9, 11055−11066