Photoinduced ethane formation from reaction of ethene with matrix-isolated Ti, V, or Nb atoms

Article abstract of DOI:10.1021/jp0447542

The reactions of matrix-isolated Ti, V, or Nb atoms with ethene (C ₂H₄) have been studied by FTIR absorption spectroscopy. Under conditions where the ethene dimer forms, metal atoms react with the ethene dimer to yield matrix-isolated ethane (C₂H₆) and methane. Under lower ethene concentration conditions (～1:70 ethene/ Ar), hydridic intermediates of the types HMC₂H₃ and H₂MC ₂H₂ are also observed, and the relative yield of hydrocarbons is diminished. Reactions of these metals with perdeuterioethene, and equimolar mixtures of C₂H₄ and C₂D ₂, yield products that are consistent with the production of ethane via a metal atom reaction involving at least two C₂H₄ molecules. The absence of any other observed products suggests the mechanism also involves production of small, highly symmetric species such as molecular hydrogen and metal carbides. Evidence is presented suggesting that ethane production from the ethene dimer is a general photochemical process for the reaction of excited-state transition-metal atoms with ethene at high concentrations of ethene.

Full text of DOI:10.1021/jp0447542

J. Phys. Chem. A 2005, 109, 9465-9470
9465
Photoinduced Ethane Formation from Reaction of Ethene with Matrix-Isolated Ti, V, or Nb
Atoms
Matthew G. K. Thompson and J. Mark Parnis*
Departments of Chemistry, Queen’s UniVersity, Kingston, Ontario K7L 3N6, and Trent UniVersity,
Peterborough, Ontario K9J 7B8, Canada
ReceiVed: NoVember 17, 2004; In Final Form: August 24, 2005
The reactions of matrix-isolated Ti, V, or Nb atoms with ethene (C
spectroscopy. Under conditions where the ethene dimer forms, metal atoms react with the ethene dimer to
yield matrix-isolated ethane (C
) and methane. Under lower ethene concentration conditions (∼1:70 ethene/
Ar), hydridic intermediates of the types HMC and H MC are also observed, and the relative yield of
hydrocarbons is diminished. Reactions of these metals with perdeuterioethene, and equimolar mixtures of
and C , yield products that are consistent with the production of ethane via a metal atom reaction
involving at least two C molecules. The absence of any other observed products suggests the mechanism
2 4
H ) have been studied by FTIR absorption
2 6
H
2
H
3
2
2 2
H
C
H
2 4
2 4
D
2 4
H
also involves production of small, highly symmetric species such as molecular hydrogen and metal carbides.
Evidence is presented suggesting that ethane production from the ethene dimer is a general photochemical
process for the reaction of excited-state transition-metal atoms with ethene at high concentrations of ethene.
Introduction
particular with ethene. Specifically, the reactions of Zr and Y
with ethene investigated by Porembski and Weisshaar2
1,22
The reactions of hydrocarbons with metal atoms are of interest
due to their value as prototypes for metal cluster and metal
surface reactions. Such studies often provide opportunities to
spectroscopically characterize chemical species which may exist
as intermediates in overall processes. These characterizations,
often achieved using the technique of matrix isolation, provide
insight into reaction processes that are important in both
industrial and catalytic contexts.
Of the hydrocarbons, ethene (C2H4) is one of the most
important molecules for industrial synthesis and is the prototypi-
cal molecule for the study of alkene reactions. Many matrix
isolation studies have been carried out on the reactions of ethene
with transition-metal atoms.1 Most of this work has focused
on determining the structure of the initial ethene-metal reaction
product, often leading to the conclusion that formation of a
π-bonded complex is the primary reaction process. In some
suggest that the thermodynamic barriers associated with a
stepwise conversion from a coordination complex of the form
MC2H4 through an intermediate, HMC2H3, to a final complex,
H2MC2H2, are either nonexistent or reasonably small. Products
predicted in this theoretical study have been observed in the
gas-phase kinetics work of Carroll et al. with Zr atoms and
2
3
ethene. In contrast, the theory work of Siegbahn concerning
the reaction of metal hydrides (MHx, x ) 1-3) with ethene
suggests exoergicity for the insertion of ethene into the M-H
1
7
bond of the metal hydride to form Hx-1MC2H5. Thus, gas-
phase transition-metal atoms have been shown to be capable of
dehydrogenation or hydrogenation of unsaturated organic
molecules.
-12
Gas-phase work from our laboratory on the reaction of C2H4
2
4
with the niobium atom and small niobium clusters shows that
these reactions lead to formal dehydrogenation of ethene.
However, these results have been characterized only by mass
spectrometry, thereby providing no insight into the detailed
structure(s) of the product species involved in the reaction
mechanism. Our motivation for the present study was to identify
spectroscopically the intermediates involved in these gas-phase
reactions. To our surprise, the major product observed in the
present study at high ethene concentrations is the hydrogenated
product ethane (C2H6). This serendipitous result has given us
unexpected insight into the reactions of metal atoms with ethene,
as discussed in detail below.
1
0
works, such as that of Lee et al. with Ti atoms and ethene,
C-H bond insertion products are observed. C-H insertion is
believed to follow π-coordination of ethene, although Lee et
al. did not find evidence for a π-coordinated complex. Their
work demonstrates the photoproduction of vinyltitanium hydride
(HTiC2H3), as well as a dihydridoorganotitanium complex (H2-
TiC2H2). As well, the laser ablation matrix isolation work of
1
1
Cho and Andrews for Zr atoms reacting with ethene shows
evidence for C-H insertion intermediates that have been
implicated in H2 elimination chemistry. These examples show
C-H insertion for early transition metals; however, similar
insertion products have also been observed recently by Cho and
Andrews for the reaction of Pt atoms with ethene.
Transition-metal atom insertions into C-H bonds have
received a fair amount of attention in theoretical studies.
Experimental Details
12
Metal atoms were generated by resistively heating a thin
filament of the pure metal (Nb, V, or Ti (A.D. McKay, New
York, 99.999%)) in a water-cooled furnace. The rate of metal
atom deposition was continuously monitored with the use of a
quartz crystal microbalance built into the resistive furnace
1
3-22
These studies together predict reaction trends for metal atoms
on insertion into the C-H bonds of small hydrocarbons, in
25
assembly. Note that the generation of metal atoms using metal
filaments results in the unavoidable generation of visible light,
*
To whom correspondence should be addressed. E-mail: mparnis@
trentu.ca.
1
0.1021/jp0447542 CCC: $30.25 © 2005 American Chemical Society
Published on Web 10/01/2005

9466 J. Phys. Chem. A, Vol. 109, No. 42, 2005
Thompson and Parnis
Figure 2. Portions of the infrared spectra of argon matrices containing
ethene (1:70 ethene/Ar), with (dotted line) and without (solid line) V
atoms. Features associated with ethane, the ethene monomer, and ethene
dimer are labeled E, M, and D, respectively.
Figure 1. Portions of the FTIR absorption spectra of argon matrices
containing (from bottom to top) 10% ethene, 10% ethene with Ti atoms,
1
0% ethene with V atoms, 10% ethene with Nb atoms, and 0.1% ethane.
deposited using an ethane/argon mixture. Methane is also formed
-
1
as a product, giving rise to an absorption at 1304 cm . No
other products from reaction of ethene with metal atoms were
observed under 1:10 ethene/Ar conditions.
which irradiates the sample during matrix deposition. Ethene
C.P. grade, 99.5%, Matheson of Canada) or ultra-high-purity
(
ethene (99.99%, Sigma-Aldrich) was used without further
purification. Perdeuterioethene (99.99 atom % D) was obtained
from C/D/N Isotopes (Quebec, Canada). For verification of the
spectrum of ethane, chemical-purity ethane (99.5%, Matheson)
and d6-ethane (C/D/N Isotopes, 99.99 atom % D) were also
employed. For all of the matrices, argon (99.995%, Matheson)
was employed as the host gas. Unless stated otherwise, matrices
were of the mole ratio metal:C2H4:Ar ) 0.5:10:100. Matrices
were deposited for 1.5 h (unless otherwise noted) with a gas
flow rate of 0.5 sccm onto a KBr optical window maintained
at 17 K using an APD Displex refrigeration apparatus. Fourier
transform infrared absorption spectroscopic analyses were
performed using a Bomem Michelson MB-102 FTIR spectrom-
When production of ethane was observed, it was in all cases
accompanied by measurable depletion of absorptions due to
26
matrix-isolated ethene dimers, which give rise to absorptions
-1
at 3096 and 2980 cm in the C-H stretching mode region.
Note that ethene exists only in the dimer form at, and above,
concentrations of 1:10 ethene/Ar. Under conditions where
features associated with the ethene dimer are not present, no
ethane production is observed. Where the ethene monomer and
dimer coexist, depletion of both the monomer and dimer is
observed, with concomitant production of ethane and hydridic
species (see below). This effect is evident in Figure 2, where
spectra are shown from the C-H stretching region of matrices
generated with a 1:70 ethene/argon mixture, with and without
the inclusion of V metal atoms. Similar experiments were done
with Ti, V, or Nb atoms in pure ethene matrices. In these cases,
the yield of ethane was substantially greater than in 1:10 ethene/
Ar matrices, but otherwise no other significant changes were
noted beyond some changes in peak shape. Methane was also
produced in amounts that appear to be correlated with the
increase in ethane production.
-1
eter in the range of 4000-400 cm . Where performed, sample
irradiations were done using a 150 W Hg arc lamp (Kratos
model LH-150), using either the full focused output of the lamp
(
UV-vis irradiation) or the focused lamp output filtered to
remove wavelengths below 400 nm (visible irradiation).
Results
Portions of the infrared spectra obtained prior to, and
following, the co-condensation of gas-phase Ti, V, or Nb atoms
with 10% ethene in argon are given in Figure 1. For comparison,
the spectrum of 0.1% ethane in argon is also shown, which
demonstrates clearly that the major product obtained in the
metal-containing matrices is unperturbed matrix-isolated ethane.
The major unobscured ethane absorptions obtained by metal
atom reactions with ethene are observed at 2944.9, 2919.4,
Co-deposition of Ti or V atoms with lower concentration
mixtures of ethene in Ar (1:70 ethene/Ar) gave rise to
diminished yields of ethane, accompanied by new features in
-
1
the region of 1650-1450 cm . In the case of Ti, these
absorptions have been previously identified in 1:200 ethene/Ar
10
matrices by Lee et al. as being associated with HTiC2H3 (1500
-
1
-1
cm ) and H2TiC2H2 (1585 and 1611 cm ). These features
show photochemical behavior identical to that described previ-
ously by Lee et al.; e.g., irradiation of deposition samples
containing both species with 466 nm light gave rise to depletion
-
1
2885.9, 1465.0, and 1373.1 cm , wavenumber positions that
are identical to those observed for matrix-isolated ethane

Ethane Formation from Ethene with Ti, V, or Nb
J. Phys. Chem. A, Vol. 109, No. 42, 2005 9467
Figure 3. Portions of the infrared spectra of argon matrices containing
ethene (1:70 ethene/Ar) with (from bottom to top) 5 h of deposition
with no metal, with 4.5 h of deposition with V atoms, following 10
min of irradiation at >455 nm, following an additional 10 min of
irradiation at >400 nm, following an additional 10 min of irradiation
with full UV-vis output of the lamp, and following an additional 1260
min of irradiation with full UV-vis output of the lamp. Features
associated with matrix-isolated ethane and water and those assigned
Figure 4. Portions of the FTIR absorption spectra of argon matrices
containing (from bottom to top) 10% perdeuterioethene, 10% perdeu-
terioethene with V atoms, and 0.1% perdeuterioethane. The asterisk
denotes a feature due to matrix-isolated CO, generated by thermal
desorption of CO in the apparatus during sample preparation.
2 2 2
in the text to H VC H are denoted E, W, and A, respectively.
(C2D6). The results of these experiments confirm that perdeu-
of HTiC2H3 and significant growth of H2TiC2H2. Prolonged
irradiation of such matrices resulted in the complete destruction
of both titanium hydride species, in accord with the observations
of Lee et al. However, in the present work, such prolonged
irradiation also resulted in concomitant production of both ethane
and methane, an observation not previously reported.
terioethane is the only major hydrocarbon product in these
reactions. Absorptions due to CO2, CO, and traces of CH4
associated with the metal vaporization approach were also
observed, as noted below for the C2H4 reactions. A feature at
993 cm is also observed which is assigned to CD4 by
comparison with spectra of CD4 in an Ar matrix.
-
1
27
Features analogous to those of the H2TiC2H2 species were
The reaction of an equimolar mixture of 5% each of C2H4
and C2D4 with vanadium atoms was also studied. Infrared
spectra obtained for this reaction show numerous weak and
broad features in the C-H and C-D stretching regions. Many
of these absorptions correspond to features associated with C2H6
and C2D6. We speculate that the additional observed absorptions
are associated with one or more isotopomers of ethane, due to
the close proximity of these absorptions to the known modes
of the isotopically pure ethanes. Spectral congestion did not
allow for the identification of mixed-isotope product(s).
Other minor features sometimes appear in the spectra of the
metal-containing matrices associated with CO2 (2349 and 667
-
1
seen at 1552 and 1644 cm when V atoms were employed
with 1:70 ethene/Ar gas mixtures (see Figure 3). The sample
was subjected to a sequence of irradiations similar to those
employed by Lee et al. in the Ti work noted above. Irradiation
with light of wavelength >455 nm for 10 min resulted initially
in the growth of these features; however, further irradiation at
these wavelengths resulted in their modest depletion. Subsequent
irradiation with broad-band UV-vis light for 10 min resulted
in further depletion of these features, with the complete
destruction of these features following prolonged irradiation
(
∼15 h). Given that the modes we observe with V under these
-
1
-1 28
conditions have a relative intensity profile and photochemical
behavior similar to those of the dihydridotitanium species of
Lee et al., we tentatively assign these new features at 1552 and
cm ) and CO (2137 cm ). Both species are well-known
matrix isolation impurities, in this case generated as byproducts
of metal vaporization by desorption from the apparatus walls
during sample formation. In accord with this supposition, the
intensities of these features varied from experiment to experi-
ment in a manner that did not correlate with production of ethane
or methane. For each metal used, a new absorption was
1
644 to H2VC2H2. Following irradiations of the sample contain-
ing V atoms with visible light, there is a very small increase in
the features associated with ethane. Subsequent broad-band
UV-vis irradiation of the sample gave a more pronounced
growth of the ethane features, particularly following prolonged
irradiation (see Figure 3, top spectrum).
For vanadium atoms, the analogous reactions were carried
out with argon containing 10% perdeuterioethene (C2D4).
Portions of the resultant spectra are given in Figure 4, along
with the corresponding spectrum for 0.1% perdeuterioethane
-1
sometimes observed at 1193 (Ti), 1205 (V), or 1263 (Nb) cm .
In each case, the absorption could be photobleached during
irradiation of the matrix with visible light. There was no other
apparent change in the recorded spectra upon elimination of
this species. The yield of this species varied significantly
between similarly prepared samples, despite the fact that

9
468 J. Phys. Chem. A, Vol. 109, No. 42, 2005
Thompson and Parnis
production of ethane was essentially constant. Therefore, we
conclude that this set of absorptions is due to the production of
a metal-containing species, via a competitive metal atom reaction
which we are unable to identify.
reaction of nickel atoms with ethene under relatively concen-
trated conditions of ethene in argon, and in pure ethene. They
2
have assigned several observed vibrational modes to Ni(C2H4)2,
which appear at exactly the same positions as major ethane
-1
modes (2945, 2886, and 1465 cm , for 1:10 C2H4/Ar) observed
here in argon. The published spectra also show absorptions at
Discussion
-
1
2
918 and ∼1375 cm . The first of these modes has been
Ethane Production in Low-Temperature Matrices. The
results presented above demonstrate that ethane is the major
infrared-observable, matrix-isolated product for the reactions
of Ti, V, or Nb atoms with ethene in matrices containing the
ethene dimer. Organometallic species that are believed to be
precursors to dehydrogenation products are observed in our work
when relatively low concentrations of ethene are employed,
where the ethene monomer is present. In the case of Ti, these
species have been identified previously as titaniocyclopropene
dihydride and vinyltitanium hydride.10 At intermediate concen-
trations, where a relatively low abundance of the ethene dimer
coexists with the matrix-isolated ethene monomer, production
of ethane and metal hydride species can be observed simulta-
neously.
2
assigned to ν(C-H) of Ni(C2H4)3, but given both the close
wavenumber matching and observation of the other three
fundamental modes of ethane, it is highly likely that these
absorptions are also due to ethane. A similar group of bands
-
1
(
2943, 2888, and 1463 cm , average values quoted from
various ethene, Ar, or Xe matrix experiments) assigned by Huber
4
et al. to Pd(C2H4)2 are almost certainly due to ethane features
as well. The remaining modes of the group of bands assigned
by Huber et al. to Ni(C2H4)2 have been reassigned to Ni(C2H4)
8
by Merle-Mejean et al., and confirmed in the work of Lee et
9
al. Interestingly, Merle-Mejean et al. did not observe the feature
-
1
at 1465 cm under their experimental conditions, though they
do predict a vibrational mode of Ni(C2H4) to occur at about
-
1
1
460 cm , resulting from the coupling of CdC stretching and
At higher ethene concentrations (1:10 ethene/Ar and higher),
metal hydride species are not observed, which suggests that, if
they are formed, they exist only as transient species under these
conditions due to their consumption in the formation of ethane.
This is supported by the work of Siegbahn which suggests that
second-row transition-metal hydride species can incorporate an
CH2 deformation motions. This mode has subsequently been
9
-1
reported by Lee et al. at 1468 cm , following co-condensation
of Ni vapor with ethene in argon matrices. As the spectra of
Lee et al. show no evidence for any of the other modes of
ethane, and the experiments were carried out at relatively low
ethene concentrations where ethene dimers are not present, we
1
7
ethene molecule by a barrierless process involving direct
insertion of the ethene molecule into the M-H bond. For Ti
and V, this process may involve photoexcitation of the metal
hydride species. Such a process would result in formation of
HM(C2H2)C2H5, which would be expected to be the direct
precursor to ethane in our experiments by reductive elimination
of the hydride and alkyl fragments. Note that, since insertion
of bare transition-metal atoms into C-H bonds is generally
exothermic, such a reductive elimination process would be
expected to exhibit a significant potential energy barrier. To
overcome such a barrier, the process would need to be photon-
assisted or exhibit thermodynamics significantly different from
that of the bare metal atom, the latter possibly resulting from
the presence of one or more additional ligands on the metal
atom.
2
-1
believe that the mode observed by Huber et al. at 1465 cm
is due to matrix-isolated ethane, but the mode reported at 1468
-
1
cm by Lee et al. is due to NiC2H4. The latter assignment is
confirmed by the observation of a C shift in the 1468 cm
1
3
-1
-
1
mode of -15.5 cm , as anticipated for a metal-ethene
9
,11
-1
π-complex,
and far greater than the shift of -4 cm
observed in our work for ethane. The observation of features
attributable to ethane in the work of Huber et al. is most likely
due to metal atom reactions with ethene dimers present in their
samples at the high ethene concentrations employed. We suggest
that, in cases where there is doubt about the origin of the 1465
-
1
13
cm mode, C-isotopic substitution could be used profitably
to distinguish between formation of ethane and the formation
of a metal-ethene π-complex.
Mechanistic Considerations. It is obvious by stoichiometric
considerations alone that the production of ethane from ethene
must involve the reaction of at least two ethene molecules,
provided there is no other source of hydrogen available.
However, the ubiquitous presence of water in all matrix samples
raises the concern that water may be the source of hydrogen
incorporated into ethane. When metal atoms are co-deposited
with samples containing ethene and argon, there is a significant
loss of features associated with water. To investigate water as
a hydrogen donor in our work, experiments which incorporated
D2O in the sample gas were performed to determine if any D
atoms would be included in the observed ethane. No evidence
for inclusion of deuterium was observed, thereby eliminating
water as the formal source of hydrogen in our experiments. Since
the depletion of water is evident in all metal-containing
experiments, it is clear that some reaction of metal atoms with
water is occurring on deposition. Indeed, we see a feature near
The production of ethane in the matrix isolation environment
following the reaction of iron atoms with ethene has been
previously reported by Kafafi et al. at 13.5-15 K. Cardenas
7
and Shevlin also observed production of ethane following warm-
up of matrices containing iron atoms in ethene deposited at 77
2
9
K. In the work of Kafafi et al., absorptions assigned to ethane
-
1
(
C2H6) at 1465.9 and 1374.5 cm and methane (CH4) at 1305.1
-1
cm were observed on co-condensation of iron atoms with high
concentrations of ethene in argon (Fe:C2H4:Ar ) 0.6:12.4:100).
7
Prolonged UV irradiation of such matrices resulted in an increase
in ethane and methane absorptions, with a concomitant decrease
in the diethene-iron complex (Fe(C2H4)2). They propose that
methane and ethane are produced following the disproportion-
ation of the diethene-iron π-complex, in which production of
an unobserved iron carbide species is also speculated. Note that
we also observe an increase in the amount of ethane upon
prolonged irradiation of our matrices. Gentle annealing of the
matrix is always a feature of prolonged irradiations with broad-
band light, and we propose that such annealing results in
additional metal atom-ethene dimer reactions occurring.
Vibrational modes of ethane are also evident in earlier
published matrix isolation work on the reaction of transition-
metal atoms with ethene in argon. Huber et al. have studied the
-
1
1584 cm which may be attributed to a metal hydride stretch
associated with a metal-water reaction product (see Figure 3).
Note that such a feature near this position has been attributed
3
0
to HVOH by Kauffman et al., and to a vanadium cluster-
3
1
water reaction product by Zhou et al. Thus, it would appear
that we have observed vanadium water chemistry occurring,

Ethane Formation from Ethene with Ti, V, or Nb
J. Phys. Chem. A, Vol. 109, No. 42, 2005 9469
and that such chemistry is unrelated to the formation of ethane
in this work.
irradiation from the filament, since we observe no other
intermediates on deposition that show concentration relationships
that link them to ethane production.
The hydrogenation of ethene involving a second ethene
molecule requires one of the ethene molecules to act as a
sacrificial hydrogen atom donor, while the other ethene molecule
acts as a hydrogen atom acceptor. Mechanistic steps required
for such a process are supported by the calculations of
SCHEME 1: Reaction Mechanism for Production of
Ethane from Ethene Disproportionation with
Transition-Metal Atoms
2
1,22
Porembski and Weisshaar for early transition metals,
experimental and theoretical investigations by Carroll et al.,
hν
1
9
M + 2C
2
H
4
9
8 M(C
2
H
4
)
2
*
and the theoretical work on metal hydrides and ethene by
Siegbahn.17 In the first of these, early second-row transition-
metal atoms are predicted to coordinate ethene through the
π-bond to form a metallocyclopropane species. Zr atom reac-
tions with C2D432 show no kinetic isotope effect in the rate of
atom removal, thus supporting the supposition that π-complex-
ation of ethene is the initial reaction step. For Ti and V,
π-complexation must involve photoexcitation of the metal atom,
since experimental gas-phase measurements of the rate of
M(C H ) * f HMC H (C H )
2
4 2
2
3
2
4
HMC H (C H ) f H MC H (C H )
2
3
2
4
2
2
2
2
4
H MC H (C H ) f HM(C H )C H
5
2
2
2
2
4
2
2
2
hν
HM(C H )C H
98 MC H + C H
2 2 2 6
2
2
2
5
MC H ff MC + H
2
2
2
2
depletion of both of these metals by ethene show that the ground
states of both elements are not reactive.33 Further, the first
One can understand why the yield of ethane increases
nonlinearly with ethene concentration, since the ethene dimers
have a very low concentration in 1:70 ethene/Ar samples, but
are essentially the only form ethene takes in 1:10 ethene/Ar
matrices. As well, the production of additional ethane following
prolonged irradiation in 1:70 samples is in accord with the idea
that photoannealing of the matrix over very long periods of time
will cause an increase in the number of metal atoms in close
proximity to ethene dimers, yielding additional production of
ethane. Subsequent destruction of the MC2H2 species is implicit
in the lack of observation of any other products save methane,
which it cannot generate directly.
allowed excited states of Ti and V are accessible with visible
3
4
light corresponding to wavelengths (629 and 611 nm) present
in emission from the heated Ti or V filaments in our experi-
ments. In both cases, such excitation would lead to an electronic
configuration similar to that of ground-state niobium atoms
(having partial occupancy of the valence s-orbital), and therefore,
it is reasonable to suggest that this excitation will result in similar
chemistry between the three metallic elements, as observed.
This coordination step is believed to be followed by insertion
of the coordinated metal into a C-H bond to form hydrido-,
and subsequently dihydrido-, organometallic intermediates. The
barrier for the first of these insertions for early transition metals
has been calculated to be fewer than 2 kcal/mol for Y and
The overall reaction process would be M* + 2C2H4 f MC2
C2H6 + H2, which would give rise to no new major
+
spectroscopically observable hydrocarbon products, other than
ethane. While we have no evidence for a metal carbide product,
we speculate that it must be present for the validity of these
reactions and that its infrared spectrum is not observed under
our conditions. Note that Kafafi et al. propose the formation of
a similar metal carbide with iron atoms following prolonged
irradiation of their matrix samples, which was also not observed
19,21,22
19
17
Zr,
and nonexistant for Nb. In the work of Siegbahn,
metal hydrides are predicted to be able to insert ethene directly
into an M-H bond without prior formation of a π-bonded ethene
1
7
complex. Such a direct insertion is predicted to be barrierless
for early second-row transition metals up to Nb, for each of the
mono-, di-, or trihydrides, and increasingly thermodynamically
favored for the di- and trihydrides compared with the mono-
hydride.
7
in their work. The fundamental of NbC has been experimentally
-
1
35
determined to occur at 993 cm (ωe). TiC and VC are
These studies together provide the basis for development of
a formal scheme for alkene hydrogenation through sacrificial
hydrogen donation by another alkene. The observation of
perdeuterioethane in our work for the analogous reaction of V
atoms with perdeuterioethene conclusively demonstrates that the
formal hydrogen atom source must be ethene. The observation
of ethane production only under conditions where the ethene
dimer exists implicates one ethene molecule in the ethene dimer
as the actual source of hydrogen in our work. We believe that
the production of ethane by disproportionation of ethene at
transition-metal atom centers is a general reaction of ethene
under high-concentration conditions in low-temperature matri-
ces. The generality of this process is demonstrated by the
production of ethane both in our work with early-transition-
metal atoms and in the work of others using late transition
metals. The general observation of ethane production under
conditions of higher ethene concentration demonstrates that there
exists a richer chemistry for metal atom reactions with ethene
than has been previously recognized, specifically under condi-
tions where a second ethene molecule is available to react
immediately with primary products.
-1
predicted to have ωe in the range 869-1006 cm . The MCC
dicarbides of Ti, V, and Nb are known to have an M-C
-1 36-38
fundamental mode in the range of 500-600 cm .
As
ethene in argon matrices (1:10 ethane/Ar) has strong IR
-1
absorption in the range 920-985 cm , it is possible that spectral
features due to metal monocarbides are obscured by the strong
ethene absorptions in this spectral range. However, given the
low concentration of metal in our samples, it is more likely
that features due to metal carbide species are simply too weak
to observe.
The reaction of niobium atoms with ethene has been studied
19
24
in the gas phase, by both Weisshaar and ourselves, using
fast-flow metal atom reactors. In both cases, Nb atoms are
removed by ethene at rates very close to the gas-kinetic collision
frequency, supporting a barrierless association reaction to form
a π-complex as a first, rate-limiting step. In our flow reactor
work, product mass data indicate that production of NbC2H2 is
the major active process in the gas phase. As well, the kinetic
behavior of the reaction of this product with a second ethene
molecule exhibits sequential addition behavior. The observation
of NbC2H2 as the major product in the gas phase contrasts the
current matrix isolation work, in which no product except ethane
is observed. This difference is undoubtedly due to the fact that
We propose Scheme 1 as a framework for understanding the
present work. All steps in this scheme must be fast under

9470 J. Phys. Chem. A, Vol. 109, No. 42, 2005
Thompson and Parnis
the gas-phase processes involve only the ethene monomer,
whereas our matrix work involves both the ethene monomer
and the ethene dimer. Intermediates such as those identified by
Lee et al. are likely the precursors to the gas-phase reactivity,
leading to partial or full dehydrogenation of ethene. We predict
that ethane production would be observed in the gas phase, if
the gas-phase reaction process were to be carried out at high
enough partial pressures of ethene for reaction of two ethene
molecules at the metal center before any metal-hydride bond
cleavage were to occur.
(2) Huber, H.; Ozin, G. A.; Power, W. J. J. Am. Chem. Soc. 1976, 98,
6
9
508.
3) Ozin, G. A.; Power, W. J. Inorg. Chem. 1977, 16, 212.
4) Huber, H.; Ozin, G. A.; Power, W. J. Inorg. Chem. 1977, 16, 6,
(
(
1
0
79.
(5) Ozin, G. A.; Power, W. J.; Upton, T. H.; Goddard, W. A., III. J.
Am. Chem. Soc. 1978, 100, 4750.
6) Parker, S. F.; Peden, C. H. F.; Barrett, P. H.; Pearson, R. G. Inorg.
Chem. 1983, 22, 2813.
7) Kafafi, Z. H.; Hauge, R. H.; Margrave, J. L. J. Am. Chem. Soc.
1985, 107, 7550.
(
(
(8) Merle-Mejean, T.; Cosse-Mertens, C.; Bouchareb, S.; Galan, F.;
Mascetti, J.; Tranquille, M. J. Phys. Chem. 1992, 96, 9148.
In both our work and the work of Kafafi et al., absorptions
(9) Lee, Y. K.; Hannachi, Y.; Xu, C.; Andrews, L.; Manceron, L. J.
7
attributed to CH4 are observed, the latter involving photoex-
Phys. Chem. 1996, 100, 11228.
citation of a metal atom-ethene complex. In the analogous
(10) Lee, Y. K.; Manceron, L.; Papai, I. J. Phys. Chem. A 1997, 101,
9650.
-1
reactions with C2D4, the formation of CD4 (993 cm ) demon-
strates that the presence of methane in the resultant spectra must
be due to a competitive metal atom-ethene reaction. In the
absence of any further spectroscopic information related to
methane production, we can only speculate on the mechanism
for methane formation by early-transition-metal atoms. It is clear
that the mechanism must involve a C-C bond cleavage as a
critical step. Such a process likely involves further reaction of
hydridic metal complexes from reactions of metal atoms with
ethene, likely involving intermediates such as HMC2H3 and H2-
MC2H2 identified by Lee et al. for Ti and proposed for V in
this work.
(11) Cho, H.-G.; Andrews, L. J. Phys. Chem. A 2004, 108, 3965.
(12) Cho, H.-G.; Andrews, L. J. Phys. Chem. A 2004, 108, 6272.
(13) McIntosh, D. F.; Ozin, G. A.; Messmer, R. P. Inorg. Chem. 1980,
19, 3321.
(14) Blomberg, M. R. A.; Siegbahn, P. E. M.; Svensson, M. J. Phys.
Chem. 1992, 96, 9794.
15) Siegbahn, P. E. M.; Blomberg, M. R. A.; Svensson, M. J. Am.
Chem. Soc. 1993, 115, 1952.
16) Papai, I.; Mink, J.; Fournier, R.; Salahub, D. R. J. Phys. Chem.
1993, 97, 9986.
(
(
(
(
(
17) Siegbahn, P. E. M. J. Am. Chem. Soc. 1993, 115, 5803.
18) Fournier, R. J. Chem. Phys. 1995, 102, 5396.
19) Carroll, J. J.; Haug, K. L.; Weisshaar, J. C.; Blomberg, M. R. A.;
Siegbahn, P. E. M.; Svensson, M. J. Phys. Chem. 1995, 99, 13955.
(20) Lawson, D. B.; DeKock, R. L. J. Phys. Chem. A 1999, 103, 1627.
(21) Porembski, M.; Weisshaar, J. C. J. Phys. Chem. A 2001, 105, 4851.
In summary, the production of ethane is a general process in
the chemistry of matrix-isolated ethene with metal atoms, when
the concentration of ethene is sufficiently high for the ethene
dimer to be present. We have proposed a general mechanistic
scheme for ethene disproportionation to form ethane, molecular
hydrogen, and metal carbides which is supported by this work
and by observations from previous work with ethene. We have
shown that gas-phase and matrix isolation reaction chemistries
are distinct, and that differences in the observed chemistry are
due to the availability of a second ethene molecule to react
promptly with primary organometallic products at high ethene
concentrations.
(
(
22) Porembski, M.; Weisshaar, J. C. J. Phys. Chem. A 2001, 105, 6655.
23) Carroll, J. J.; Haug, K. L.; Weisshaar, J. C. J. Am. Chem. Soc.
1
993, 115, 6962.
(24) Parnis, J. M.; Escobar-Cabrera, E.; Thompson, M. G. K.; Jacula,
J. P.; Lafleur, R. D.; Guevara-Garc ´ı a, A.; Mart ´ı nez, A. M.; Rayner, D. M.
J. Phys. Chem. A 2005, 109, 7046.
(25) Moskovits, M.; Ozin, G. A. Appl. Spectrosc. 1972, 26, 487.
(26) Rytter, E.; Gruen, D. M. Spectrochim. Acta, A 1979, 35A, 199.
(27) Nielsen, A. H.; Nielsen, H. H. Phys. ReV. 1938, 54, 118.
(28) Nakamoto, K. Infrared and Raman Spectra of Inorganic and
Coordination Compounds, 4th ed.; Wiley: New York, 1978.
(
(
29) Cardenas, G.; Shevlin, P. B. J. Org. Chem. 1984, 49, 4726.
30) Kauffman, J. W.; Hauge, R. H.; Margrave, J. L. J. Phys. Chem.
1
985, 89, 3547.
31) Zhou, M.; Dong, J.; Zhang, L.; Qin, Q. J. Am. Chem. Soc. 2001,
123, 135.
(32) Porembski, M.; Weisshaar, J. C. J. Phys. Chem. A 2000, 104, 1524.
33) Ritter, D.; Carroll, J. J.; Weisshaar, J. C. J. Phys. Chem. 1992, 96,
0636.
Acknowledgment. We gratefully acknowledge the support
(
of the Natural Sciences and Engineering Research Council
(NSERC) for operating funds via the Discovery Grant Program.
(
M.G.K.T. thanks NSERC for scholarship funding through the
PGS program. We acknowledge the many stimulating conversa-
tions with A. Martinez and A. Guevara (UNAM) that were most
helpful in preparing this work.
1
(
(
34) Moore, C. E. Natl. Bur. Stand. Circ. (U. S.) 1947, 467, 1.
35) Simard, B.; Presunka, P. I.; Loock, H.-P.; B e´ rces, A.; Launila, O.
J. Chem. Phys. 1997, 107, 307.
(
(
(
36) Wang, X.; Ding, C.; Wang, L. S. J. Phys. Chem. 1997, 101, 7699.
37) Li, X.; Wang, L.-S. J. Chem. Phys. 1999, 111, 8389.
38) Zhai, H.-J.; Liu, S.-R.; Li, X.; Wang, L.-S. J. Chem. Phys. 2001,
References and Notes
(1) McIntosh, D.; Ozin, G. A. J. Organomet. Chem. 1976, 121, 127.
115, 5170.