FTIR matrix-isolation study of the reaction products of vanadium atoms with propene: Observation of allylvanadium hydride as a precursor to sacrificial hydrogenation of propene

Article abstract of DOI:10.1021/ic200934r

Vanadium atoms have been reacted with different partial pressures of propene in Ar under matrix-isolation conditions, and the products have been observed using Fourier transform infrared (FTIR) spectroscopy. Under dilute propene in Ar conditions, new features are observed in the IR spectra corresponding to a C-H insertion product, identified here as H-V-(η³-allyl). Use of d₃-propene (CD ₃-CH=CH₂) demonstrates that the initial V-atom insertion occurs at the methyl group of the propene molecule, and DFT calculations have been used to support the identity of the initial product. Upon increasing the partial pressure of propene, additional features corresponding to propane (C₃H₈) are observed, with the hydrogen-atom source for the observed hydrogenation demonstrated to be additional propene units. Analysis of a systematic increase in the partial pressure of propene in the system demonstrates that the yield of propane correlates with the decrease of the allyl product, demonstrating the H-V(allyl) species as a reactive intermediate in the overall hydrogenation process. An overall mechanism is proposed to rationalize the formation of the insertion product and ultimately the products of hydrogenation, which agrees with previous gas-phase and matrix-isolation work involving propene and the related system, ethene.

Full text of DOI:10.1021/ic200934r

ARTICLE
pubs.acs.org/IC
FTIR Matrix-Isolation Study of the Reaction Products of Vanadium
Atoms with Propene: Observation of Allylvanadium Hydride As a
Precursor to Sacrificial Hydrogenation of Propene
†
‡
,†,‡
Matthew G. K. Thompson, Stephen W. C. Walker, and J. Mark Parnis*
†
Department of Chemistry, Chemical Sciences Building, 1600 West Bank Drive, Trent University, Peterborough, Ontario,
Canada, K9J 7B8
‡
Department of Chemistry, Chernoff Hall, 90 Bader Lane, Queen’s University, Kingston, Ontario, Canada, K7L 3N6
S Supporting Information
b
ABSTRACT: Vanadium atoms have been reacted with different
partial pressures of propene in Ar under matrix-isolation condi-
tions, and the products have been observed using Fourier
transform infrared (FTIR) spectroscopy. Under dilute propene
in Ar conditions, new features are observed in the IR spectra
corresponding to a CꢀH insertion product, identified here as
3
HꢀVꢀ(η -allyl). Use of d -propene (CD ꢀCHdCH ) de-
3
3
2
monstrates that the initial V-atom insertion occurs at the methyl
group of the propene molecule, and DFT calculations have been
used to support the identity of the initial product. Upon
increasing the partial pressure of propene, additional features
corresponding to propane (C H ) are observed, with the hydrogen-atom source for the observed hydrogenation demonstrated to
3
8
be additional propene units. Analysis of a systematic increase in the partial pressure of propene in the system demonstrates that the
yield of propane correlates with the decrease of the allyl product, demonstrating the HꢀV(allyl) species as a reactive intermediate in
the overall hydrogenation process. An overall mechanism is proposed to rationalize the formation of the insertion product and
ultimately the products of hydrogenation, which agrees with previous gas-phase and matrix-isolation work involving propene and
the related system, ethene.
’
INTRODUCTION
the H atoms are transferred to an additional ethene unit via the
VꢀH bond in vinylvanadium hydride, thereby transferring
hydrogen from one ethene to another (sacrificial hydrogena-
tion). The current investigation uses propene to confirm pre-
vious mechanistic aspects and to provide new spectroscopic
insights into the nature of the reaction products.
Metal-based reactions with hydrocarbons play a key role in
1
understanding many industrial and catalytic processes. Many of
these metal-based processes incorporate early transition metal
atoms, and while these can be very catalytically active, the specific
reaction steps of the catalytic processes are rarely known pre-
cisely. Given the paucity of detailed structural and spectroscopic
information about some of the simplest organometallic species,
overall reaction mechanisms in this field are difficult to confirm.
Thus, techniques that can directly explore intermediate steps of
an overall reaction pathway can provide invaluable insight into
the mechanism of metal-based catalytic processes.
4
The reactivity of ground- and excited-state V atoms with
5
,6
ethene and propene has been investigated in the gas phase.
The initial products are identified as atomic metal and alkene π-
coordination products, but exact structures of the final products
are not confirmed. Song and co-workers also investigated
vanadium atom and cluster reactions with propene and halopro-
7
penes using mass spectrometry. Following alkene monomer
The technique of matrix isolation coupled with infrared
pickup, all clusters, save the ground-state metal atom, demon-
spectroscopy provides one method by which metal-atom reac-
tions with hydrocarbons can be straightforwardly investigated
while providing detailed information about the structure of
metalꢀorganic reactive intermediates and their reactivity. Early
transition metal atoms reacting with ethene have been previously
investigated by Thompson and Parnis, as well as by Cho and
Andrews, under matrix-isolation conditions. Such results de-
2,3
strated products consistent with CꢀH bond insertion and
7
elimination of H .
2
The present work is aimed toward further investigating the
CꢀH bond insertion chemistry of V atoms with propene. A
CꢀH bond insertion intermediate is observed, which is identi-
fied as allylvanadium hydride. At higher partial pressures of
2
3
monstrate that excited-state V atoms insert into CꢀH bonds to
3
yield a vinylvanadium hydride intermediate. Increasing ethene
Received: May 3, 2011
Published: June 28, 2011
partial pressure yields ethane (C H ), with the implication that
2
6
r 2011 American Chemical Society
7317
dx.doi.org/10.1021/ic200934r
|Inorg. Chem. 2011, 50, 7317–7323

Inorganic Chemistry
ARTICLE
propene, the allylvanadium hydride also undergoes further
reactivity to yield propane as the major product.
’
EXPERIMENTAL SECTION
The specific details of the apparatus used in these experiments have
2
been described previously. All gas samples were prepared using standard
manometric techniques and contained 0.1ꢀ2% propene (CH
9+% propene, Aldrich), perdeuteriopropene (CD
ꢀCDdCD
atom D, C/D/N Isotopes), or d -propene (CD ꢀCHdCH , 99.9% atom
3
ꢀCHdCH
2
,
9
3
2
, 99.7%
3
3
2
D, C/D/N isotopes) diluted in argon. Some experiments included 0.1ꢀ1%
H
2
O or D
2
O (99% atom D, Aldrich), purified by freezeꢀpumpꢀthaw
cycles. Experiments involving D O also involved exposure of the entire
2
interior of all components of the apparatus with D O vapor in three
2
repetitions to dissociatively chemisorb D
2
O, minimizing conversion of
, C.P. grade,
D
2
O to HOD, by H/D exchange. Additionally, propane (C H
3 8
Matheson) was used for verification as a spectroscopic reference material
where necessary. All gas samples were prepared by diluting the reagent gases
in ultra-high-purity argon (UHP grade, 99.9995%, Praxair).
Vanadium atoms were generated by resistively heating a thin strip of
the pure metal (0.25 mm thickness, 99.5% metal basis, Alfa Aesar) placed
between water-cooled copper electrodes using a 30 A alternating current
ꢀ
7
ꢀ8
variable-power supply under high-vacuum conditions (10 ꢀ10 Torr
system base pressure). The rate of deposition of metal atoms was
monitored by use of a quartz crystal microbalance. Typical metal
introduction rates were between 0.3 and 0.5 μmol/h. Metal atoms were
mixed in a flow of argon (up to ∼10 mmol/h), which contained the
above-listed reagent species, using an MKS 1100 series mass flow
controller. Note that the resistive heating of the filament unavoidably
irradiates the matrix during sample preparation with orange light.
Argon/propene mixtures containing the metal atoms were condensed
on a cold CsI spectroscopic window maintained near 17 K using an APD
Cryogenics (model HC-2) closed cycle helium refrigeration unit.
Additional irradiation of deposited matrices with light was performed
using a Kratos LH150 150W Xe arc lamp and various band-pass filters,
where indicated. Typical matrix deposition times were 75 min. In all
cases, IR absorption spectra of the condensed matrices were collected
Figure 1. Comparison of the infrared spectra obtained from replicate Ar
matrices formed containing (A) 1 to 1000 propene to Ar and (B) 1 to
1000 propene to Ar with V atoms. The uppermost trace is a subtraction
(
difference spectrum) of B ꢀ A. Negative features are consumed
precursor, while positive absorptions represent product features.
species. (Spectrally expanded portions of the infrared spectra,
highlighting the key absorption regions, are given in the Support-
ing Information.) Of the main product features observed, one
group (referred to as species A) is identified below as allylvana-
ꢀ1
using a 1 cm resolution Bomem MB 102 spectrometer, operating in
absorption mode.
3
dium hydride, HꢀVꢀ(η -C H ). In addition, several features
3
5
ꢀ1
(
2973.2, 1466.8, and 743.9 cm ) correspond well to absorp-
To confirm the identity of molecular species associated with observed
infrared absorptions, complementary density functional theory (DFT)
tions of propane isolated in argon (referred to as species B).
8
Additional features are observed corresponding to matrix-iso-
calculations were carried out using the Gaussian 03 software package
9
9
lated CO2 and matrix-isolated CO, which are unavoidably
present in a trace amount during metal experiments. Similarly,
using the B3LYP density functional and the 6-311G(3df,3pd) basis set
for all of the atoms used. Geometries were fully relaxed during the
optimization cycles, and the optimized geometries were confirmed by
analysis of the calculated vibrational wavenumber values.
ꢀ1
an absorption at 1305 cm is also observed, corresponding to
CH4. When a similar set of experiments was performed using 1
9
to 1000 d -propene in Ar, related absorptions were observed with
6
absorption positions shifting in accordance with deuterium
’
RESULTS
substitution (see Table 1). Additionally, a very weak absorption
ꢀ1
9
When V atoms are codeposited with dilute propene in Ar (1 to
is observed at 993 cm , an indication of CD isolated in Ar.
4
1
000, C H to Ar) a number of new features can be observed in
When a similar set of experiments was performed using 1 to
3
6
the corresponding infrared spectra. Figure 1 provides the full
infrared spectra obtained from replicate matrices containing
1000 d -propene in Ar, related absorptions were observed with
6
absorption positions shifting in accordance with deuterium
substitution. Additionally, a very weak absorption is observed
0
.1% propene in Ar with and without inclusion of V atoms.
ꢀ
1
9
The use of difference spectra obtained by subtracting the metal-
free spectrum from the metal-containing spectrum serves to
illuminate product-related absorptions. All product absorptions
are observed in the difference spectrum as positive features, while
negative features are associated with consumed precursor.
Analysis of the difference spectrum for 0.1% propene with and
without V atoms shows a series of new features. The main
wavenumber values associated with product absorptions are
summarized in Table 1. All of these product absorptions are
consistent with new hydrocarbon or organometallic intermediate
at 993 cm , an indication of CD isolated in Ar.
4
The wavenumber ratio of the most intense absorptions in the
ꢀ
1
h -propene and d -propene at 1549 and 1117 cm , respectively,
6
6
suggests a VꢀH-containing product arising from metal-atom
insertion into the CꢀH bond of propene. To further investigate
this and determine which of the propene CꢀH bonds is the site
of insertion, additional experiments were performed using 3,3,3-
d -propene, CD ꢀCHdCH . When the difference spectrum for
3
3
2
matrices containing 1 to 400 3,3,3-d -propene to Ar is analyzed, a
3
series of related features is observed, along with absorptions at
7
318
dx.doi.org/10.1021/ic200934r |Inorg. Chem. 2011, 50, 7317–7323

Inorganic Chemistry
ARTICLE
Table 1. Comparison of Product Absorption Features for V atoms Reacting with Varied Concentration of Propene Isotopomers
a
under Matrix-Isolation Conditions
1
to 1000
1 to 100
propene
to Ar
1 to 1000
propane
in Ar
1 to 1000
1 to 100
1 to 400
(methyl)-propene
to Ar
propene
to Ar
d
6
-propene
to Ar
d
6
-propene
to Ar
gas-phase
-propane
d
3
d
8
wavenumber/
wavenumber/
wavenumber/ wavenumber/
wavenumber/
wavenumber/ wavenumber/
ꢀ1
ꢀ1
ꢀ1
ꢀ1
ꢀ1
ꢀ1
ꢀ1
cm
cm
identity
cm
identity
A
cm
cm
identity
cm
identity
cm
identity
1
1
1
9
8
7
6
2
549.2
480.9
207.5
74.5
10.1
28.7
A
A
A
A
A
A
A
B
1549.2 (very weak)
1117.2
1092.5
A
A
1435.1
1117.2
1081.1
772.5
A
A
A
A
A
780.8
638.4
561.7
A
A
A
B
576.4
24.1
973.2
2973.8
B
B
B
B
B
B
B
2973.1
2941.6
2902.8
2882.7
1473.5
1467
2226.6
2226.6
2150.4
2119.9
2077.8
1080.6
1063.5
B
B
B
B
B
B
2225
2148
2120
2080
1064
712
2
2
2
941.2
902.8
882.1
1473.2
1466.7
387.1
obscured by precursor
050.4
obscured by precursor
743.1
1
1
473.5
466.8
B
B
1080.4
B
1
1387.2
1372.6
1050.4
918.5
1
B
B
7
1
43.9
305.1
B
744.1
573.9
993.7
B
CD
4
572
CH
4
a
Gas-phase d
8
-propane values obtained from Gough et al., ref 10.
new positions. Major new product features are observed
these results suggest that net formation of species B involves
more than one propene molecule, while species A requires a
single propene unit.
(
see Table 1), along with weak CꢀH and CꢀD stretching
ꢀ
1
region absorptions. No feature is observed at 1549 cm
.
To better identify the carrier of the observed absorptions,
irradiation of the metal-containing matrices was performed. A
series of 5 min irradiations, with wavelength ranges of light >495,
Results from samples of V co-condensed with 1 to 100 d6-
propene to Ar show two sets of absorptions also corresponding
to species A and B. The former decrease with increasing propene
partial pressure, while those of species B increase. The absorptions
455, 400 nm and finally with a band-pass range with maximum
transmittance at 320 nm showed no changes in the matrix spectra,
indicating no photochemical reactivity of the species within the
matrix environment. In addition, to investigate any influence of
deposition temperature on the reaction chemistry during matrix
formation, a set of comparison experiments was performed where
the matrix-deposition temperature was raised to 25 K. The compar-
ison with experiments done at lower temperature showed no
significant effect on hydrogenation product yield.
due to species B are consistent with the spectrum of gas-phase d -
propane, providing further support for the species B identification
8
10
as propane. Matrices formed containing 1 to 100 3,3,3-d -propene
3
(CD ꢀCHdCH ) to Ar with and without metal atoms showed
3
2
many CꢀH and CꢀD modes corresponding to isotopic propane
species. However, the relative complexity of the isotopic products
makes it difficult to assign the specific isotopomer of propane
conclusively. The complexity of the pattern suggests that more than
one propane isotopomer may be generated.
Experiments employing increased concentration from 1 to
1
5
000 propene to Ar through to a maximum concentration of 1 to
0 propene to Ar were performed to investigate the dependence
Using CD CHdCH at various concentrations, a similar set
of features demonstrating the behavior of species A can also be
3
2
ꢀ
1
of the products on the availability of the precursor. With 1 to 100
propene to Ar, the features associated with species A showed
decreased intensity in the higher concentration experiments while
those of species B increased in intensity, with new features observed
at 2941.2, 2902.1, 2882.1, 1387.1, and 1051.1 cm . On the basis of
this behavior, we can identify species A as a primary product and
species B (propane) as a secondary product. A comparison of the
major product absorptions with an infrared spectrum of 0.1%
propane isolated in Ar is given as Figure 2. The complete set of
absorptions and their tentative identity is given in Table 1.
identified. Again, no HꢀV stretch is observed near 1550 cm
.
Furthermore, at higher concentration, these species A features do
not appear. Instead, spectra yield features with characteristics
similar to species B, described above. The pattern of behavior is
consistent with species B, but in the isotopic case, it is possible
that there is more than one isotopomeric form of the species B
product, depending on the H-atom source for hydrogenation
(i.e., d -propene or impurity H O).
ꢀ
1
6
2
Finally, that species B demonstrates an apparent hydrogena-
tion of propene when V atoms are present along with an ob-
Figure 3 shows the comparative yield of each species as a
function of propene concentration. With increasing propene
concentration, the measured area of the main features attributed
to species A decrease while the main features for species B
increase. In addition, it was observed that increasing the gas-flow
rate for matrix deposition showed marked enhancements of
species B, with only modest increases in species A. Together,
served consumption of unavoidable impurity H O in the corre-
sponding gas implies that hydrogenation could occur by reaction
2
of V atoms with H O and propene. Experiments were performed
2
where 1% D O was added to a sample containing 1% propene in
2
Ar. No evidence for new CꢀD modes was observed. Similarly,
samples containing d -propene with H O showed no evidence
6
2
for new CꢀH features following codeposition with V atoms.
7
319
dx.doi.org/10.1021/ic200934r |Inorg. Chem. 2011, 50, 7317–7323

Inorganic Chemistry
ARTICLE
Figure 4. Optimized B3LYP/6-311(3df,3pd) structure of doublet
3
HꢀV(η -CH
2 2
CHCH ). The molecule is predicted to belong to the
s
C point group and is pictured from two sides for clarity.
and quartet are lower in energy than the starting products. As a
result, the observed features corresponding to species A were
compared with calculated frequencies for both the doublet and
the quartet species. Only the doublet insertion species provides a
good match for the calculated wavenumber values. The structure
of the optimized geometry for the doublet species is given as
Figure 4.
In Table 2, the comparison of the calculated wavenumber
3
values for the doublet HꢀVꢀ(η -allyl) with those observed in
the experiment are given. A comparison of the same data for the
quartet spin state of the same species is given in the Supporting
Information.
Figure 2. Comparison of the infrared spectrum of 0.1% propane in Ar
with the difference spectrum obtained when V atoms react with 1%
propene in Ar. Unless obscured by the propene precursor, all propane
modes are observed in the product matrix at the same position with
similar relative intensities as the reference propane sample matrix.
’ DISCUSSION
Identification of Species A and B. Analysis of the infrared
spectra of matrices formed by co-condensation of V atoms with
dilute propene in Ar shows that a clear series of product features
appears in the spectra. The lack of change in any of the spectra
following irradiation suggests that either the product species are
quite stable or these species have no chromophore for UVꢀvis
absorption in the employed range. By directly comparing the
relative amounts of these species as the availability of propene is
increased, species B maximizes at the expense of species A (see
Figure 3). These facts imply that species A is a primary product,
which upon increase of availability of propene is transformed into
product B.
The major features associated with species A are maximized
when propene concentration is lowest. In the most dilute
ꢀ
1
experiments, a major absorption is observed near 1549 cm
ꢀ
1
for V/C H , and this absorption shifts to 1117 cm when V/
3
6
C D experiments are performed. The ratio of the absorptions
3
6
for the observed VꢀH/VꢀD is 1.404, and this is precisely the
value that would be anticipated for the ratio of a VꢀH/VꢀD
stretching absorption. This is strong evidence in support of
assignment of this mode as an uncoupled single VꢀH stretching
mode. The absorption position of organometallic species having
Figure 3. Comparison of the yield of the metal CꢀH insertion product
“
A” and the yield of propane with propene availability. For each product,
ꢀ
1 2,3
the measured peak areas has been normalized relative to the largest area
in the corresponding data set.
a vanadium monohydride unit is in the vicinity of 1560 cm
.
Thus, the V atom has clearly inserted into a CꢀH bond of the
propene molecule, and species A is consistent with a metal
monohydride, with likely chemical formula HV(C H ).
Thus, when hydrogenation is observed, it involves hydrogen
transfer from propene precursor molecules only.
3
5
There are four unique CꢀH bonds in propene, CH ꢀ
3
Computational Results. To confirm the proposed structure
CHdCH : one allylic position and three vinylic positions. The
2
3
of species A as HꢀVꢀ(η -allyl), B3LYP/6-311G(3df,3pd)
bond strength of the CꢀH bond at the allylic position is
ꢀ
1
calculations were performed for the main spin states expected
for this structure. The doublet, quartet, and sextet spin states of
the molecule are possibilities; however, of these, only the doublet
361 kJ mol , much weaker than the other vinylic CꢀH bonds,
3
ꢀ
1 11
which are closer to 444 kJ mol . Therefore, it is energetically
favorable for a metal atom to preferentially insert into the methyl
3
7
320
dx.doi.org/10.1021/ic200934r |Inorg. Chem. 2011, 50, 7317–7323

Inorganic Chemistry
ARTICLE
Table 2. Comparison of the Predicted Wavenumber Values
Table 2. Continued
6-311G(3df,3pd)
3
(Unscaled) for Isotopomers of Doublet (η -Allyl) Vanadium
rel int
expt
Hydride
1145.00
1.00
0.03
0.10
0.03
0.14
0.02
0.00
0.03
0.02
0.01
1117.6
1
1
219.35
250.59
6
-311G(3df,3pd)
rel int
expt
3
1440.78
doublet, HV(η ꢀCH
2 2
CHCH )
1
2
2
3
3
3
506.32
226.04
371.24
054.75
126.42
197.61
1475.9
2
2
3
3
3
4
5
6
7
8
9
9
1
1
1
1
1
1
1
3
3
3
3
3
32.8
73.8
04.2
48.4
64.4
12.7
69.5
97.2
96.7
33.7
29.6
47.9
011.3
230.5
240.7
415.4
480.9
520.1
601.3
051.4
058.1
125.3
196.2
199.0
0.21
0.05
0.03
0.05
0.15
0.06
0.00
0.01
0.12
0.00
0.00
0.01
0.07
0.02
0.05
0.00
0.01
0.08
1.00
0.00
0.03
0.01
0.00
0.00
624.1
728.7
CꢀH bond. When experiments were performed using dilute
ꢀ
3 3 3
1
3,3,3-d -propene, CD ꢀCHdCH in Ar, the 1549 cm absorp-
tion was completely absent and no new features were observed in
ꢀ1
this area. However, the 1117 cm absorption was present and as
974.5
intense as in related experiments where perdeuteriopropene was
employed. Using this observation, it seems clear that insertion of the
metal atom occurs selectively at the allylic position of the hydro-
carbon. Additionally, the complete shifting to almost the exact
wavenumber position observed in pure C D suggests that this
1207.5
1480.9
1549.2
3
6
observed feature is a pure uncoupled VꢀH (VꢀD) stretch.
In the reaction of V atoms with ethene, photoexcitation was
3ꢀ7
required,
provided by the metal source during sample deposi-
4
tion. In the case of propene, ground-state V atoms have been
5
shown to react with propene, though reaction of excited-state V
atoms is faster. Thus, when considering reactivity factors, the initial
6
3
2 2
doublet, DV(η -CD CDCD )
1
2
2
2
3
3
4
5
6
6
6
7
8
9
1
1
1
1
1
2
2
2
2
2
85.92
19.06
71.77
83.59
00.74
33.40
25.77
42.53
17.99
38.30
87.62
62.69
39.99
30.62
015.37
060.24
144.61
230.82
432.41
223.80
227.08
306.13
370.04
374.37
0.30
0.12
0.02
0.01
0.08
0.03
0.00
0.02
0.13
0.01
0.00
0.01
0.11
0.02
0.01
0.00
1.00
0.02
0.25
0.01
0.04
0.00
0.00
0.01
V atom may be a ground state V or a photoexcited V* atom, and in
this particular case, it is likely that both reactions proceed via a
parallel set of reaction steps.
Following insertion of the metal into the CꢀH bond of the
methyl group, two isomeric forms are possible, specifically, an open
structure where the π-bond is not coordinated to the V atom and a
closed structure where π-coordination to V has occurred. The open
1
561.7
638.4
HꢀVꢀCH ꢀCHdCH structure, bearing an η -propenyl substi-
2
2
ꢀ1
tuent, should have an absorption in the region of 1650 cm , since
there is an asymmetric CdC in the structure. However, no evidence
for such an absorption is apparent in any of the observed infrared
spectra. Instead, it is likely that π-coordination of the alkene at the V
atom is occurring. Such donation of the π-bond to an oxidized metal
atom would reasonably be expected to be a near-barrierless
association and would not be inhibited under the cryogenic condi-
tions of matrix isolation. Therefore, it is likely that this association
1117.2
1406.6
3
occurs spontaneously, forming a ring-like η -C H structure on the
3
5
3
metal center. Such η -propenyl species are well known in organo-
metallic and complexation chemistry and can be considered as
fluxionally interconvertible with isomeric resonance forms best
1
considered as a coordinated allylic group.
3
To confirm the structure, B3LYP/6-311G(3df,3pd) calcula-
2 2
doublet - DV(η -CD CHCH )
1
2
2
2
3
3
4
6
6
7
8
9
9
1
95.38
22.51
86.34
89.93
21.30
69.51
66.30
07.89
83.01
83.43
14.46
27.96
42.62
071.63
0.33
0.12
0.03
0.02
0.07
0.04
0.02
0.07
0.03
0.03
0.10
0.10
0.00
0.03
tions were performed on the doublet, quartet, and sextet
3
structures of the HꢀV(η -allyl) structure. Of these, only the
doublet and quartet were lower in energy than the starting
materials, and on the basis of the comparisons given in Table 2,
the doublet structure shows a strong match between the observed
and calculated wavenumber positions and relative intensities for
absorptions. The quartet species (see the Supporting Information
for summarized calculation data), by comparison, shows a very poor
correlation. Therefore, on the basis of all of this evidence, the most
likely structure of the species A product is the doublet state of
576.7
772.9
3
HꢀV(η -allyl). The structure of the optimized doublet species is
given in Figure 4.
7
321
dx.doi.org/10.1021/ic200934r |Inorg. Chem. 2011, 50, 7317–7323

Inorganic Chemistry
ARTICLE
Scheme 1. Proposed Reaction Scheme for V Atoms Reacting with Propene under Matrix-Isolation Conditions
V + hν f Vꢁ
2
1
ꢁ
ꢁ
V + CH ꢀ CHdCH f ðVðη -propeneÞÞ f H ꢀ Vðη -propenylÞ
3
2
Under low partial pressure of C H :
3
6
1
3
H ꢀ Vðη -propenylÞ f H ꢀ Vðη -allylÞ
Under high partial pressure of C3H6 :
1
1
H ꢀ Vðη -propenylÞ + CH ꢀ CHdCH f CH CH CH ꢀ Vðη -propenylÞ
3
2
3
2
2
1
CH3CH2CH2 ꢀ Vðη -propenylÞð + hνÞ f CH3CH2CH3 + VðC3H4Þ
VðC3H4Þð + hνÞ f f VðC2Þ + CH4
Further literature observations of these types of coordination
complexes can be used to support this identification. Allyl
π-coordination complexes have been observed and well char-
acterized by infrared and Raman techniques for some transition
species proposed is therefore an analog to intermediates pro-
posed in the case of hydrogenation of ethene by HꢀVꢀOH and
4
by HꢀVꢀCHdCH . Thus, it would not be unexpected for H
2
transfer from the propenyl species to the propyl unit yielding
propane and a V(C H ) species. These proposals are the direct
1
2ꢀ16
metals,
although no complexes containing only allyl groups
3
4
and V atoms have been reported in the literature. The observed
infrared features associated with species A are consistent with a
coordinated allyl group. Thus, the experimental and computa-
tionally available data listed in Table 2 are consistent with the
assignment of the features corresponding to species A as doublet
analogue to the hydrogen transfer observed for the case of
2,4
ethylꢀVꢀvinyl species in the V/ethene system.
Although no new features are observed in the infrared
spectrum that could be consistent with a V(C H ) species, it is
3
4
worth noting that under high concentration of propene there
3
HꢀV(η -allyl).
appears to be formation of CH . There is also a corresponding
4
3
ꢀ1
Identification of the HꢀV(η -allyl) species makes it possible
observation of CD at 993 cm when d -propene is used. Since
4
6
to consider the behavior of this complex with increasing propene
concentration to yield species B. The set of product absorptions
no absorptions are observed consistent with a V(C H ) species,
3 4
it is believed that this species is able to continue to react spon-
listed as species B correlate very well with propane (and d -
taneously to yield a final product, in parallel with CH . In mass
spectrometric results on the V/propene system, Song et al. did
8
4
1
0
propane, see Figure 2). The origin of the hydrogen for
hydrogenation is demonstrated as propene, implicating a sacri-
ficial hydrogenation mechanism. We previously observed sacri-
ficial hydrogenation in the case of ethene reacting with V
see loss of a C fragment from propene units during dehydro-
1
7
genation on clusters. Therefore, it is conceivable that such a
process also occurs here, and this process may be photoassisted
during matrix formation by the prolonged and unavoidable
irradiation from the metal filament prior to the initial infrared
acquisition.
2
,4
atoms, in which ethane was similarly formed under high ethene
concentration by a sacrificial hydrogenation mechanism.
As noted in the Results section, no enhancement of the
hydrogenated product was observed when matrices were formed
at the elevated temperature of 25 K. However, increasing
reagent-gas flow rates during deposition resulted in increased
hydrogenation product yield. These results suggest that the
enhancement is due to increased availability of reactant in the
reaction zone during matrix formation. The increase in reagent
availability to metal atoms and metallic species provides en-
hanced opportunity for intermediate species to react with a
Assuming V(C H ) yields CH as a final product, the remain-
3
4
4
ing fragment would have the molecular formula VC . This same
2
final product is expected in the analogous reactions where
2
sacrificial hydrogenation occurs with ethene, and in those
experiments, no conclusive spectral evidence for VC was
2
17
observed. Formation of stronger CꢀH bonds in CH would
4
support thermochemical favorability in the proposed steps, as
would the final product VC , which is known to have an overall
2
3
ꢀ1 18
second precursor molecule. Since HꢀV(η -allyl) is believed to
binding energy between 550 and 590 kJ mol . However, the
absence of VC (or other organometallic species) in any of the
3
be the direct precursor to propane formation, its dependence on
conditions which provide only a single propene molecule for
reaction supports such claims.
2
related experiments is perplexing and suggests that, if present, the
molecule must be below the threshold for detection using
infrared spectroscopy. To further examine this, we conducted
Previous work with increased ethene concentration suggested
that following the CꢀH bond insertion an additional alkene unit
could directly insert into the VꢀH bond, yielding a coordinated
DFT calculations on several isomers of VC (see Supporting
2
Information) for both doublet and quartet spin states. These
2,4
alkyl group on the metal as well as a coordinated vinyl group. In
the current work, the same conditions apply, with the available
hydrogen coming from the coordinated allyl unit. Initially, the V
atom is inserted into the allylic position of propene, generating
computations indicate that if VC were formed as a major final
2
product, we would expect to observe at least one of its absorp-
tions. The absence of such an observation remains a challenging
point for further study (see Supporting Information).
An overall mechanism accounting for the observed products,
over the range of concentrations studied in this work, is given in
Scheme 1. According to Scheme 1, replacement of each of the H
atoms with deuterium atoms would yield perdeuterio analogs of
all products, consistent with the observed data. However, for the
1
the HꢀVꢀ(η -propenyl) complex. In the absence of collision
with a second propene molecule, the observations suggest that
3
the initially formed complex isomerizes to the HꢀV(η -
3
propenyl), i.e., HꢀV(η -allyl) complex. However, at higher
propene concentration, this isomerization to form the HꢀV-
3
(
η -allyl) complex is in direct competition with propene inser-
3,3,3-d -propene case, hydrogenation of the propene molecule
3
tion into the HꢀV bond. The alkene insertion intermediate
could lead to two possible hydrogenation products, depending
7
322
dx.doi.org/10.1021/ic200934r |Inorg. Chem. 2011, 50, 7317–7323

Inorganic Chemistry
ARTICLE
on which of the two C atoms on propene receives the H atom
formerly on the metal during alkene insertion. While it should be
possible to identify CD H on the basis of its infrared spectrum,
V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick,
D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui,
Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith,
T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.;
Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.;
Pople, J. A. Gaussian 03, revision D.01; Gaussian, Inc.: Wallingford, CT,
2
2
the complexity of the potential propane isotopomers and their
very similar spectra in the CꢀH and CꢀD absorption regions
makes the current analysis difficult at best when considered with
the wealth of propene precursor features in the same areas. Thus,
while a full mechanistic investigation involving careful isotopic
substitution should be possible, it would require a substantial
investigation in order to confirm the geometric details of the
latter part of this mechanism.
2004.
(9) We isolated CO, CO , CH , and CD in Ar as reference
2 4 4
standards in our own laboratory in this experimental system and
observed these wavenumber values using the FTIR apparatus
described here.
(
10) Gough, K. M.; Baudais, F. L.; Casal, H. L. J. Chem. Phys. 1986,
4, 549.
11) Data obtained from CRC Handbook of Chemistry and Physics,
70th ed.; Weast, R. C., Ed. in Chief; CRC Press: Boca Raton, FL,
8
’
SUMMARY AND CONCLUSIONS
(
The nature of V-atom reactivity with propene evolves as the
1
1
1
5
989ꢀ1990.
12) Andrews, D. C.; Davidson, G. J. Chem. Soc. Dalton Trans. 1972,
3, 1381.
concentration of propene is increased. Under low partial-pres-
sure conditions, a CꢀH bond insertion intermediate is observed,
which is consumed with increasing partial pressures of propene
to yield a hydrogenation product, propane. Additionally, under
high propene concentration, products consistent with CꢀC
bond cleavage appear to occur, as evidenced by the formation
(
(13) Davidson, G.; Andrews, D. C. J. Chem. Soc., Dalton Trans. 1972,
, 126.
(14) Andrews, D. C.; Davidson, G. J. Organomet. Chem. 1973,
5, 383.
of CH at high propene partial pressures. Given the current
4
(15) Andrews, D. C.; Davidson, G. J. Organomet. Chem. 1977,
results, the only viable mechanism is one in which the final
V-containing products are small, highly symmetric species that
are unobserved under the present conditions. Otherwise, the
mechanism and product distribution are very similar to pre-
viously reported observations wherein ethene is analogously
hydrogenated sacrificially.
124, 181.
(16) Chensdaya, T. B.; Leites, L. A.; Aleksanyan, V. T. J. Organomet.
Chem. 1978, 148, 85.
(
(
17) Li, X.; Wang, L.-S. J. Chem. Phys. 1999, 111, 8389.
18) Gupta, S. K.; Gingerich, K. A. J. Chem. Phys. 1981, 74, 3584.
’
ASSOCIATED CONTENT
S
Supporting Information. This material is available free
b
of charge via the Internet at http://pubs.acs.org.
’
AUTHOR INFORMATION
Corresponding Author
*E-mail: mparnis@trentu.ca.
’
REFERENCES
(
1) See, for example: (a) Crabtree, R. The Organometallic Chemistry
of the Transition Metals, 4th ed.; John Wiley & Sons, Inc.: Hoboken, NJ,
005. (b) Cotton, F. A., Wilkinson, G. Advanced Inorganic Chemistry, 5th
2
ed.; Wiley: New York, 1988. (c) Gates, B. C. Catalytic Chemistry; John
Wiley & Sons, Inc.: Hoboken, NJ, 1992.
(
2) Thompson, M. G. K.; Parnis, J. M. J. Phys. Chem. A. 2005,
109, 9465.
(
(
(
3) Cho, H.-G.; Andrews, L. J. Phys. Chem. A. 2007, 111, 5201.
4) Thompson, M. G. K.; Parnis, J. M. Inorg. Chem. 2008, 47, 4045.
5) Ritter, D.; Carroll, J. J.; Weisshaar, J. C. J. Phys. Chem. 1992,
9
9
6, 10636.
(
(
6) Senba, K.; Matsui, R.; Honma, K. J. Phys. Chem. 1995, 99, 13992.
7) Song, L.; Freitas, J. E.; El-Sayed, M. A. J. Phys. Chem. 1990,
4, 1604.
8) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
(
Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.;
Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.;
Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson,
G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda,
R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai,
H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken,
V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;
Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.;
Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski,
7
323
dx.doi.org/10.1021/ic200934r |Inorg. Chem. 2011, 50, 7317–7323