Vibrational overtone activation of methylcyclopropene

Article abstract of DOI:10.1063/1.471315

Laser vibrational overtone activation has been used to investigate the reaction channel competition in the isomerization of 1-methylcyclopropene (MCPene). The vibrational overtone activation of three types of CH stretches (methyl, methylenic, and olefinic) in the 5v_CH and 6v_CH transitions initiated the isomerization and all three products (2-butyne, 1,3-butadiene, and 1,2-butadiene) were detected by gas chromatography. Stern-Volmer plots were constructed for the appearance of each individual product and the derived experimental specific rate coefficients were compared to those of the Rice-Ramsperger-Kassel-Marcus (RRKM) theory. The rate coefficients for the 6v_CH transitions were in good agreement with the predicted values but those for the 5v_CH transition were as much as a factor of 5 too large. Product ratios of 1,3-butadiene to 2-butyne and 1,2-butadiene to 2-butyne were independent of pressure. In general, these ratios were lower than the RRKM predicted ratios due to collisional deactivation. No evidence of mode specific behavior was observed in these product yield ratios.

Full text of DOI:10.1063/1.471315

Vibrational overtone activation of methylcyclopropene
D. L. Snavely, O. Grinevich, S. Hassoon, and G. Snavely
Citation: The Journal of Chemical Physics 104, 5845 (1996); doi: 10.1063/1.471315
View online: http://dx.doi.org/10.1063/1.471315
View Table of Contents: http://scitation.aip.org/content/aip/journal/jcp/104/15?ver=pdfcov
Published by the AIP Publishing
Articles you may be interested in
Photodissociation of methylazide: Observation of triplet methylnitrene radical
J. Chem. Phys. 105, 5798 (1996); 10.1063/1.472423
Breaking symmetry with hydrogen bonds: Vibrational predissociation and isomerization dynamics in HF–DF and
DF–HF isotopomers
J. Chem. Phys. 104, 9313 (1996); 10.1063/1.471677
Master equation simulations of the vibrational overtone activation of methylcyclopropene
J. Chem. Phys. 104, 5852 (1996); 10.1063/1.471316
Vibrational coherence effects on electronic curve crossing
J. Chem. Phys. 104, 5638 (1996); 10.1063/1.471803
Understanding the dynamic behavior of molecular vibrational states
AIP Conf. Proc. 172, 280 (1988); 10.1063/1.37373
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
142.150.190.39 On: Sat, 20 Dec 2014 18:02:51

Vibrational overtone activation of methylcyclopropene
D. L. Snavely, O. Grinevich, S. Hassoon, and G. Snavely
Department of Chemistry, Bowling Green State University, Bowling Green, Ohio 43403
͑Received 22 November 1995; accepted 16 January 1996͒
Laser vibrational overtone activation has been used to investigate the reaction channel competition
in the isomerization of 1-methylcyclopropene ͑MCPene͒. The vibrational overtone activation of
three types of CH stretches ͑methyl, methylenic, and olefinic͒ in the 5␯ and 6␯ transitions
CH
CH
initiated the isomerization and all three products ͑2-butyne, 1,3-butadiene, and 1,2-butadiene͒ were
detected by gas chromatography. Stern–Volmer plots were constructed for the appearance of each
individual product and the derived experimental specific rate coefficients were compared to those of
the Rice–Ramsperger–Kassel–Marcus ͑RRKM͒ theory. The rate coefficients for the 6␯
CH
transitions were in good agreement with the predicted values but those for the 5␯CH transition were
as much as a factor of 5 too large. Product ratios of 1,3-butadiene to 2-butyne and 1,2-butadiene to
2-butyne were independent of pressure. In general, these ratios were lower than the RRKM
predicted ratios due to collisional deactivation. No evidence of mode specific behavior was observed
in these product yield ratios. © 1996 American Institute of Physics. ͓S0021-9606͑96͒02515-0͔
INTRODUCTION
from nitromethane to D2O results from vibrational overtone
photoactivation of the CH stretch with a quantum yield of
3.1ϫ10Ϫ5 10
If overtone excitation produces species capable
Vibrational overtone activation of chemical reactions1
produces activated molecules with well defined internal
energies2 suitable for gaseous unimolecular reaction studies.
This technique has been used to measure the experimental
energy specific rate coefficient for many reactions in explo-
rations of the validity of RRKM theory. The vibrational over-
tone absorption spectrum for many hydrocarbons possesses
peaks which fit a simple vibrational progression for each
type of CH bond ͑methyl-in-plane, methyl-out-of-plane, ole-
finic, methylenic, etc.͒ in the molecule.3 This fact suggests
that the photon excitation energy could localize in one por-
tion of the molecule.
Unimolecular reactions such as ring openings, isomer-
izations, sigmatropic shifts, and dissociations have been in-
vestigated using vibrational overtone activation.4–18 Specific
rate coefficients derived from the Stern–Volmer plots have
been compared with theory. For the ring opening of cy-
clobutene, initiated by pumping the 5␯ and 6␯ methyl-
enic and olefinic stretches, the RRKM parameters which fit
the thermal activation data underestimated the experimental
Stern–Volmer specific rate coefficients. New ‘‘loose’’ transi-
tion state RRKM parameters with five vibrational frequen-
cies lowered from 660 to 450 cmϪ1 fit the overtone data;
however, the Arrhenius parameters resulting from the
‘‘loose’’ transition state calculation overestimated the ther-
mal rate constant. No difference was found in the specific
rate coefficient when specific oscillators were excited on ei-
ther one ring or the other in 1-cyclopropylcyclobutene.5 The
.
of fluorescence, laser-induced fluorescence ͑LIF͒ can moni-
tor state specific product yields. The photodissociation kinet-
ics of HONO2, ͑Ref. 11͒ HOH, ͑Ref. 12͒ HOOH ͑Ref. 13͒
were studied by OH fragment LIF. Combining infrared opti-
cal double resonance techniques with LIF ͑Refs. 14 and 15͒
provides spectroscopic access to vibrational combination
levels which are not accessible from the ground state.
For the isomerization of methyl isocyanide, wavelengths
within the CH stretch band contour were photolyzed to check
for rotational or vibrational hot band effects on the experi-
mental specific rate coefficient.16 A hot band based on the
lowest frequency bending mode was confirmed by spectros-
copy to contribute to the 5␯ stretch band contour.17 How-
CH
ever, the measured specific rate coefficient for 5␯
in-
CH
creased monotonically with increasing energy, showing no
hot band effect. The possible explanation of this result is that
the specific photolysis energy fell on part of the band contour
where CH stretch transitions occurred, not hot band transi-
tions. In contrast, measured specific rate coefficients for the
isomerization of allyl isocyanide, with overlapping absorp-
tion peaks for the methyl, methylenic and vinyl CH stretches,
did not increase monotonically with increasing energy.18 This
behavior was explained by hot bands, however, the appear-
ance of hot bands has not been confirmed by spectroscopy.
For the present work we chose the ring opening reaction
of methylcyclopropene ͑MCPene, 1͒, shown in Fig. 1, to
study the competition of three unimolecular reactions initi-
ated by vibrational overtone activation. This reaction pro-
duces three products: 2-butyne ͑2͒, 1,3-butadiene ͑3͒, and
1,2-butadiene ͑4͒.19 At total pressures of 20 torr, the thermal
kinetics of this ring opening reaction was investigated20 us-
ing MCPene diluted in n-butane to avoid dimerization or
polymerization. A common intermediate for all three reaction
products was postulated as the cyclopropene ring opens
forming a biradical ͑5͒ which undergoes one of three pos-
CH
CH
RRKM
behavior
of
tetramethyldioxetane6
and
quadricyclane7 was also confirmed using vibrational over-
tone excitation. Overtone excitation of the unimolecular dis-
sociation of t-butylhydroperoxide,8 simulated by master
equation calculations, fit experimental results in the low
pressure range,9 but did not explain the large curvature at
high pressures. Overtone initiated reactions have also been
observed in the liquid phase. For example, proton transfer
J. Chem. Phys. 104 (15), 15 April 1996
0021-9606/96/104(15)/5845/7/$10.00
© 1996 American Institute of Physics
5845
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
142.150.190.39 On: Sat, 20 Dec 2014 18:02:51

5846
Snavely et al.: Overtone activation of methylcyclopropene
FIG. 1. Reaction of methylcyclopropene to form 2-butyne, 1,3-butadiene,
and 1,2-butadiene.
sible hydrogen shift reactions to form product. Formation of
these products via homogeneous unimolecular reactions has
been documented20 with the following activation barriers:
37.8Ϯ0.37 kcal/mol, 2-butyne; 42.2Ϯ0.7 kcal/mol, 1,3-
butadiene; 43.8Ϯ1.5 kcal/mol, 1,2-butadiene. The stabilities
of these three products have a different order with 1,3-
butadiene being the most stable ͑Ϫ32.3 kcal/mol͒, 1,2-
butadiene being the least stable ͑Ϫ19.4 kcal/mol͒ and
2-butyne lying intermediate ͑Ϫ23.5 kcal/mol͒. Another pos-
sible biradical intermediate ͑6͒ has been suggested which
leads to a carbene ͑7͒ and subsequent insertion to yield
methylenecyclopropane ͑MCPane͒. MCPane has not been
observed as a product in this reaction; however, it is believed
that the carbene intermediate does form during thermolysis.
The ring opening of MCPene has been studied using
vibrational overtone activation21,22 using the 6-0 vinylic CH
excitation. Baggott and Law21 found the measured product
yield ratio of 1,3-butadiene to 2-butyne, y3/y2 , in the total
pressure range of 1 to 5 torr was independent of pressure but
smaller than that calculated by RRKM theory. The difference
in the experimental and calculated product ratio was attrib-
uted to inefficient collisional energy transfer. Unfortunately
the value of the calculated ratio was not strongly dependent
on the magnitude of ⌬E and could not be used to determine
the collisional energy transfer parameters. However, a ⌬E of
Ϫ730 cmϪ1 was determined from a master equation fitting of
the experimental quantum yield versus pressure. The overall
conclusion of this work was that the results at the vinyl ex-
citation behaved according to RRKM theory. Samarasinghe
and Snavely22 measured the disappearance rate of MCPene
FIG. 2. Energy ͑in wave numbers-cmϪ1͒ level diagram for the overtone
activation of methylcyclopropene. Numbers in parentheses are in kcal/mole.
17 093 cmϪ1 was assigned as the vinylic stretch. Figure 2
displays the energy diagram for the overtone activation of
1-methylcyclopropene.
Since 5 and 6 are implicated in this reaction profile per-
haps vibrational overtone activation can be used to create a
vibrationally excited species which resembles one or the
other of these two intermediates. Using selective excitation
of a particular CH stretch within the molecule, an influence
on the product yield ratios for the three products may be
observed. To evaluate this influence the experimental product
yield ratios can be compared to those predicted by calculated
RRKM specific rates. Six different excitation wavelengths
ranging over 3800 cmϪ1 in energy have been studied in this
work. In addition to the 6␯
vinyl stretch studied
CH
previously,21 we have photolyzed the 6␯ methyl and me-
CH
thylenic and 5␯
vinylic, methyl, and methylenic CH
CH
stretches. The product yields of all three products separately
were detected by gas chromatography.
with photolysis into the 6␯ vinylic CH stretch and calcu-
CH
lated the specific rate coefficient from the Stern–Volmer plot.
Their result was 1.66ϫ108 secϪ1 for the reaction. This value
appeared to be significantly greater than that obtained by
Baggott and Law.21
EXPERIMENT
1-methylcyclopropene was prepared according to the
procedure developed by Fisher and Applequist.23 Following
the synthesis the product was purified using several bulb-to-
bulb vacuum distillations at liquid nitrogen temperature. Fur-
ther distillation was required to enhance the purity of the
methylcyclopropene/methylenecyclopropane mixture using
water/ice temperatures and collecting the sample in a dry
ice/isopropanol bath. The distillate was collected over a
24–48 h period in these low temperature distillations. This
procedure was repeated a second time before storing the
product under vacuum in liquid nitrogen.
The overtone spectra of MCPene/MCPane in the 5␯
CH
and 6␯ transition regions have been previously recorded
CH
and assigned.22 The large peaks at 13 359 cmϪ1 in the 5␯
CH
transition and at 15 623 cmϪ1 in the 6␯ transition were
CH
assigned as the methylenic C–H stretch. The two smaller,
overlapping peaks to the blue of the methylenic peak, at
13 436 and 13 582 cmϪ1 ͑5␯ ͒, and at 15 736 and 15 893
CH
cmϪ1 ͑6␯ ͒, were assigned as the methyl out-of-plane and
CH
the methyl in-plane stretches, respectively. The single peak at
J. Chem. Phys., Vol. 104, No. 15, 15 April 1996
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
142.150.190.39 On: Sat, 20 Dec 2014 18:02:51

Snavely et al.: Overtone activation of methylcyclopropene
5847
FIG. 3. Stern–Volmer plot for the photolysis of the methylenic CH at
13 359 cmϪ1
FIG. 4. Stern–Volmer plot for the photolysis of the methyl in plane CH at
13 582 cmϪ1
.
.
RESULTS AND DISCUSSION
The identity and relative purity of MCPene including the
major contaminant, MCPane, were confirmed by NMR and
gas chromatography ͑GC͒. Although the absolute purity of
the synthesis product was unknown, a close approximation
could be assumed by the comparison of peak areas from the
GC analysis for reactants and solvents from the synthesis.
The final purified mixture was about 95% MCPene, 5%
MCPane, and small impurities of tetrahydrofuran, 3-chloro-
2-methylpropene, 2-butyne, and 1,3-butadiene ͑listed in de-
creasing amounts͒.
Product analyses were conducted on a Hewlett-Packard
gas Chromatograph outfitted with a stainless steel, 30Јϫ1/8Љ
o.d., ␤,␤Ј-oxydipropiononitrile-bonded porasil column and
flame ionization detector. We found a superior separation
with this column which increased our sensitivity limits dra-
matically. For example, the retention times for MCPene:
MCPane: 1,3-butadiene: 1,2-butadiene: 2-butyne were 21
Ϯ0.5, 26Ϯ0.5, 29Ϯ0.3, 33Ϯ0.3, and 47Ϯ1 min. Also in our
separation MCPene and MCPane did not overlap. Samples
were collected prior to injection onto an evacuated gas sam-
pling loop cooled to liquid nitrogen temperature. Peak areas
were quantified using a Hewlett-Packard Integrator inter-
faced to the GC. The intensities of two of the products,
2-butyne and 1,3-butadiene, were calibrated against true
standards in order to calibrate the product yields.
Figures 3–8 display the Stern–Volmer plots for the
13 359, 13 582, 14 552, 15 623, 15 893, and 17 093 cmϪ1
photolysis, respectively, for the production of 2-butyne, 1,3-
butadiene, and 1,2-butadiene. The photolysis at 14 552 cmϪ1
corresponds to the 5␯ olefinic transition, the spectrum of
CH
which was observed for the first time in this work. These
Stern–Volmer plots were constructed using the kinetic
mechanism describing the three competitive reaction chan-
nels. The unimolecular reaction of methylcyclopropene pro-
duces 2-butyne, 1,3-butadiene, and 1,2-butadiene proceeding
through the activated complex according to the following
equations:
*
MCP→MCP
ka ,
͑1͒
͑2͒
͑3͒
͑4͒
͑5͒
*
MCP →2-butyne
k1 ⑀͒,
͑
*
MCP →1,3-butadiene
k2 ⑀͒,
͑
*
MCP →1,2-butadiene
k3 ⑀͒,
͑
*
MCP ϩM→MCP
kd ,
The photoisomerizations were performed in a glass
sample cell with quartz windows set at Brewster’s angle. The
cell was placed intracavity of a dye laser pumped by an
argon laser, and the contents photolyzed for a pre-determined
amount of time at constant power. At low pressures ͑between
0.1 and 2 Torr͒ photolysis duration was one hour, while at
higher pressures, photolyses lasted up to 6 h depending on
the power output of the dye laser at the photolysis frequency.
This procedure was repeated for pressures ranging from 0.1
to 6 Torr. The fourth and fifth vibrational overtones corre-
sponding to the methyl, methylenic and vinylic CH local
modes were used in the photolyses. Rhodamine 6G, DCM,
and pyridine 2 dyes were used for these photolyses.
FIG. 5. Stern–Volmer plot for the photolysis of the olefinic CH at 14 552
cmϪ1
.
J. Chem. Phys., Vol. 104, No. 15, 15 April 1996
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
142.150.190.39 On: Sat, 20 Dec 2014 18:02:51

5848
Snavely et al.: Overtone activation of methylcyclopropene
FIG. 8. Stern–Volmer plot for the photolysis of the olefinic CH at 17 093
cmϪ1
FIG. 6. Stern–Volmer plot for the photolysis of the methylenic CH at
15 623 cmϪ1
.
.
tional to the ratio of the product yields and, hence, to the
ratio of the integrated gas chromatography peaks for each
product, Aj . Since
where ka is the photoactivation rate constant, and kd is a
collisional deactivation constant ͑torrϪ1 sϪ1͒. The product
formation is described by first-order kinetic equations which
are independent of the other channels:
k ⑀͒/k ⑀͒ϭY /Y ϭA /A ,
͑9͒
͑
͑
i
j
i
j
i
j
the following equation can be derived:
1/kappϭ1/k ϩk MA / k k ⑀͒ ⌺A ͒ .
*
d P /dtϭk ⑀͒ MCP ,
͔
͑6͒
͑7͒
͓
͔
͑
͓
j
j
͑10͒
͓
͑
͑
͔
i
a
d
j
a
j
Y ϭ d P /dt͒/I,
͑ ͓
͔
j
j
This equation describes the jth channel of the unimolecular
reaction. Plotting 1/kapp vs MAj/⌺Ai yields the specific rate
coefficient for the jth channel. MAj/⌺Ai can be interpreted
as the partial pressure of the jth product in the mixture.
The experimental Stern–Volmer plots ͑Figs. 3–8͒ appear
linear. For plots of this type three different pressure regions
have been identified9 according to the properties of the reac-
tive distribution. The reactive distribution is the energy dis-
tribution of all activated molecules which eventually form
product. The experimental value for the slope results from
the average over the reactive distribution and the specific rate
coefficient. Since the reactive distribution changes with pres-
sure the Stern–Volmer plots curve. This curved region is
called region 2. If kdP is approximately equal to k͑⑀͒ for the
pressure range under study, then as the total pressure changes
the curvature will be observed. For most overtone activation
experiments the k͑⑀͒ is low enough that at typical pressure
ranges, no curvature is observed. Vibrational overtone ex-
periments are usually performed in region 3 where the shape
of the reactive distribution does not change. The curvature
͑region 2͒ has been observed for the overtone activation of
methyl isocyanide16,24 since k͑⑀͒ for this reaction is large
where Pj is the jth product concentration, I is the number of
absorbed photons, and Y j is the yield of the jth product. The
derivative in Eq. ͑6͒ is constant because the concentration of
*
MCP is constant in terms of the steady-state approximation
treatment and the low conversions ͑around 2%͒.
From the aforementioned considerations the reaction ki-
netics can be described using the general form of the Stern–
Volmer equation where kapp shows the rate of appearance of
all three products:
1/kappϭ1/kaϩkdM/ka
k
⑀͒ϩk ⑀͒ϩk ⑀͒ ,
͑8͒
͓
͑
͑
͑
͔
1
2
3
where M is overall pressure in the reaction vessel. One can
see from Eqs. ͑6͒ and ͑7͒ that the ratio of kj͑⑀͒’s is propor-
͑4ϫ108 secϪ1 for the 5␯ transition͒. The current Stern–
CH
Volmer plots lie in region 3 where, due to frequent collisions,
only the high energy wing of the initial distribution reacts.
For this work kd is 8.28ϫ106 torrϪ1 sϪ1 using a hard sphere
collisional deactivation model and k͑⑀͒ ranges from 1.2ϫ106
to 5ϫ106 sϪ1. With these values the curvature is expected
around 0.5 torr, too low to be observed in our experimental
pressure range. In region 3 the ratio of the Stern–Volmer
slope and intercept yields kd/k͑⑀͒. The experimental slopes
and intercepts are presented in Table I; the errors in these
FIG. 7. Stern–Volmer plot for the photolysis of the methyl-in-plane CH at
15 893 cmϪ1
.
J. Chem. Phys., Vol. 104, No. 15, 15 April 1996
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
142.150.190.39 On: Sat, 20 Dec 2014 18:02:51

Snavely et al.: Overtone activation of methylcyclopropene
5849
TABLE I. Experimental Stern–Volmer slopes ͑in s/cm3 torr͒ and intercepts ͑in s/cm3͒.
2-butyne
Intercept
1,3-butadiene
1,2-butadiene
Wave number ͑cmϪ1
͒
Slope
Intercept
Slope
Intercept
Slope
13359
13582
14552
15623
15895
17093
1.57ϫ1013
8.27ϫ1013
3.78ϫ1013
2.72ϫ1014
6.86ϫ1014
1.07ϫ1014
1.78ϫ1015
2.53ϫ1015
8.64ϫ1014
1.43ϫ1015
2.15ϫ1015
1.80ϫ1014
2.35ϫ1014
9.02ϫ1014
1.04ϫ1014
1.44ϫ1016
1.89ϫ1016
8.46ϫ1014
2.66ϫ1014
8.40ϫ1014
1.18ϫ1014
4.65ϫ1016
4.66ϫ1016
2.05ϫ1015
found for the lower energy photolyses at 13 359 and 13 582
cmϪ1, respectively. At the low energies only 2-butyne is pro-
duced because the other products possess higher activation
barriers. This difference may indicate a problem with the
RRKM calculation near the barrier or it may arise from dif-
ferences in the collisional deactivation with total energy. A
complete master equation simulation of the experimental
data is required to answer these questions.26
slopes are all below 3%. Experimental k͑⑀͒ values, tabulated
as exp k͑⑀͒ in Table II, were derived from the slopes and
intercepts assuming the hard-sphere collision frequency. The
errors are included within the parentheses in Table II.
An RRKM calculation was undertaken using the RRKM
program in Gilbert’s UNIMOL suite ͑Version 1992͒. This pro-
gram calculates the specific rate coefficients using the fol-
lowing expression:25
The measured specific rate coefficient for the appearance
of 2-butyne reported earlier by Snavely and Samarsinghe22
for the 17 093 cmϪ1 is much larger than the value reported in
this work. Furthermore the value for the total disappearance
of MCPene reported in Ref. 25 is larger than the one in this
work calculated by totalling the product yields in our study.
A possible explanation is that our MCPene reactant was 95%
pure compared to purities of 61% and 75% in the other stud-
ies. The published spectra in both Refs. 25 and 26 indicate
large methylenecyclopropane impurities. In the GC separa-
tions for the current work we could separate completely the
MCPene from the MCPane which was not possible in the
previous work. Due to this improved analysis, we have a
high level of confidence in the current results.
Table III and Fig. 10 compare the experimental product
yield ratios to those calculated from the RRKM rate coeffi-
cients according to Eq. ͑9͒. In the case where several prod-
ucts are produced from a common intermediate the product
ratios are given by the ratios of the specific rate coefficients
for the particular products. These RRKM rate coefficient ra-
tios should be considered as the zero pressure limits. The
experimental product yield ratio for 1,3-butadiene to
2-butyne is designated as y13/y2 while that for 1,2-butadiene
to 2-butyne is designated as y12/y2 . Product yield ratios are
independent of the laser power but should vary with energy.
These experimental ratios are independent of pressure. The
values for the experimental y13/y2 are 0.22, 0.094, and 0.10
for the 17 093, 15 893, and 15 623 cmϪ1 photolysis, respec-
EϪE
0
ϩ
k ⑀͒ϭ
͑
␳
E ͒dE
ϩ
h␳ E͒ ,
͑11͒
͑
͓
͑
͔
͵
ͩ
ͪ
ϩ
Ͳ
0
ϩ
where ␳(E) and ␳ ͑Eϩ͒ are the densities of states of the
reactant and transition state. The input requires the critical
energy for each channel, vibrational frequencies and mo-
ments of inertia for the reactant and transition state. The
vibrational frequencies were taken from Ref. 21. Since we
assumed that there was no change in the moment of inertia
from reactant to transition state geometry the values of the
moments of inertia did not influence the results. The internal
methyl rotors can be treated as hindered or unhindered
rotors.25 In this work the internal methyl rotors were de-
scribed as hindered rotors and included as low-energy oscil-
lators. The angular momentum was chosen to be conserved
in this calculation.25 The Gilbert program uses a grain size of
100 cmϪ1 for the density of states calculation. The exact
count method was used for the density of states. Figure 9
compares these calculated specific rate coefficients to those
derived from our Stern–Volmer slopes. The values of calc
k͑⑀͒ for the specific photolysis energies are also tabulated in
Table II.
The comparison of experimental to theoretical specific
rate coefficients is good overall for all three products. For
2-butyne the rate for the high energy photolyses differs by
only 10%. However, larger differences of 60% and 82% are
TABLE II. Experimental specific rate coefficients compared to RRKM specific rate coefficients.
2-butyne
Expt. k͑⑀͒ Calc. k͑⑀͒
1,3-butadiene
Expt. k͑⑀͒ Calc. k͑⑀͒
1,2-butadiene
Expt. k͑⑀͒ Calc. k͑⑀͒
Wave number ͑cmϪ1
͒
13 359
13 582
14 552
15 623
15 895
17 093
͑7.41Ϯ0.12͒104 3.16ϫ104
͑2.82Ϯ0.04͒105 5.13ϫ104
͑3.63Ϯ0.08͒105 2.75ϫ105
͑1.58Ϯ0.01͒106 1.17ϫ106 ͑1.34Ϯ0.04͒105 1.26ϫ105 ͑4.78Ϯ0.04͒104 5.25ϫ104
͑2.63Ϯ0.02͒106 1.58ϫ106 ͑3.98Ϯ0.04͒105 3.02ϫ105 ͑1.48Ϯ0.02͒105 5.62ϫ104
͑4.89Ϯ0.05͒106 5.37ϫ106 ͑1.02Ϯ0.01͒106 1.78ϫ106 ͑4.68Ϯ0.04͒105 6.03ϫ105
J. Chem. Phys., Vol. 104, No. 15, 15 April 1996
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
142.150.190.39 On: Sat, 20 Dec 2014 18:02:51

5850
Snavely et al.: Overtone activation of methylcyclopropene
FIG. 9. Comparison of RRKM calculated rate coefficients with experiment.
FIG. 10. Comparison of RRKM calculated and experimental product ratios.
tively. At a lower photolysis energy, less 1,3-butadiene was
produced relative to the 2-butyne. This is as expected since
the barrier to 1,3-butadiene is higher. As the photolysis en-
ergy increases so does the y12/y2 experimental ratio.
Our RRKM y13/y2 ratios at 17 093, 15 893, and 15 623
cmϪ1 are 0.33, 0.19, and 0.11, respectively, indicating a
smooth increase in the amount of 1,3-butadiene relative to
2-butyne with increasing energy. The y12/y2 RRKM ratio
does not increase monotonically with energy due to the prox-
imity of the activation barrier where calculated specific rates
fluctuate with the sum of states of the transition state. This
nonmonotonic behavior is not seen in the experimental
y12/y2 ratios.
In general, the experimental product yield ratios are
lower than those of the RRKM calculations. Part of the dif-
ference between the experimental and calculated values is
collisional deactivation which decreases the yield of 1,3-
butadiene since this product has a higher barrier than
2-butyne and is deactivated faster by collisions. As the pres-
sure decreases and less deactivation occurs more 1,3-
butadiene would be produced. The RRKM ratios are ‘‘zero
pressure’’ ratios and should be larger than the experimental
values. The product ratios calculated by solving the master
equation problem are needed to evaluate the difference due
to collisional energy transfer.
cients and experimental values is usually very good, these
product yield ratios are in good agreement and show no evi-
dence of mode specific reaction.
CONCLUSIONS
Vibrational overtone activation has been used to study
the ring opening reaction of methylcyclopropene to form
2-butyne, 1,3-butadiene, and 1,2-butadiene from 0.25 to 6
torr pressure. Six wavelengths corresponding to the olefinic,
methylenic and methyl-in-plane CH stretches were chosen
for photolysis. The experimental specific rate coefficients
were determined directly from the Stern–Volmer slopes and
intercepts using a hard sphere collision model. At the sixth
quantum level the experimental values compared well with
the RRKM calculated values being on average a factor of 1.2
different. But at the lower energies near the activation barrier
the experimental values were high by as large as a factor of
5. A reexamination of the RRKM calculation parameters in
conjunction with a full master equation simulation is needed
to understand these differences. Experimental product ratios
of 1,3-butadiene to 2-butyne and 1,2-butadiene to 2-butyne
were independent of pressure in the range from 0.25 to 6 torr.
The experimental ratios are generally lower than those pre-
dicted by RRKM theory for all photolysis energies. When
comparing the experimental product yield ratios to those
from theory no differences due to the local mode excitation
were found.
Overall the experimental product yield ratios increase
monotonically with energy with no unusual behavior at a
particular photolysis energy. The largest difference occurs at
15 893 cmϪ1 for the y13/y2 ratio where the experimental
value is a factor of 2 lower than the RRKM value. Given that
a factor of 2 comparison between calculated rates coeffi-
ACKNOWLEDGMENTS
Thanks to Professor I. Oref for help with the RRKM
calculations. This work was supported in part by the Office
of Naval Research.
TABLE III. Experimental and calculated product yield ratios.
y
13/y2
y12/y2
Wave number ͑cmϪ1
͒
Expt.
Calc.
Expt.
Calc.
1 K. V. Reddy and M. J. Berry, Chem. Phys. Lett. 52, 111 ͑1977͒.
2 F. F. Crim, Annu. Rev. Phys. Chem. 35, 657 ͑1984͒.
3 B. Henry, Vib. Spectra Struct. 10, 269 ͑1981͒.
4 J. M. Jasinski, J. K. Frisoli, and C. B. Moore, J. Chem. Phys. 79, 1312
͑1983͒.
15 623
15 895
17 093
0.10
0.094
0.22
0.11
0.19
0.33
0.026
0.041
0.088
0.045
0.036
0.11
J. Chem. Phys., Vol. 104, No. 15, 15 April 1996
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
142.150.190.39 On: Sat, 20 Dec 2014 18:02:51

Snavely et al.: Overtone activation of methylcyclopropene
5851
5 J. M. Jasinski, J. K. Frisoli, and C. B. Moore, J. Chem. Phys. 87, 3826
͑1983͒.
16 S. Hassoon, N. Rajapakse, and D. L. Snavely, J. Phys. Chem. 96, 2576
͑1991͒.
6 B. D. Cannon and F. F. Crim, J. Chem. Phys. 75, 1752 ͑1981͒.
7 D. G. Lishan, K. V. Reddy, G. S. Hammond, and J. E. Leonard, J. Phys.
Chem. 92, 656 ͑1988͒.
17 S. Hassoon and D. L. Snavely, J. Phys. Chem. 99, 2511 ͑1993͒.
18 J. Segall and R. N. Zare, J. Chem. Phys. 89, 5704 ͑1988͒.
19 R. Srinivasan, J. Amer. Chem. Soc. 91, 6250 ͑1969͒.
20 H. Hopf, G. Wachholz, and R. Walsh, Chem. Ber. 118, 3570 ͑1985͒.
21 J. E. Baggott and D. W. Law, J. Chem. Phys. 88, 900 ͑1988͒.
22 S. Samarasinghe and D. L. Snavely, Chem. Phys. Letters 150, 386 ͑1988͒.
23 F. Fisher and D. E. Applequist, J. Org. Chem. 30, 2089 ͑1965͒.
24 D. W. Chandler and J. A. Miller, J. Chem. Phys. 85, 4502 ͑1986͒.
25 R. G. Gilbert and S. C. Smith, Theory of Unimolecular and Recombina-
tion Reactions ͑Blackwell Scientific, Oxford, 1990͒, p. 71.
26 I. Ouporov, O. Grinevich, and D. L. Snavely, following paper, J. Chem.
Phys. 104, 5852 ͑1996͒.
8 D. W. Chandler, W. E. Farneth, and R. N. Zare, J. Chem. Phys. 77, 4447
͑1982͒.
9 D. W. Chandler and J. A. Miller, J. Chem. Phys. 81, 455 ͑1984͒.
10 D. M. Goodall and I. Chetwood, Chem. Phys. Lett. 129, 291 ͑1986͒.
11 P. R. Fleming, M. Li, and T. R. Rizzo, J. Chem. Phys. 94, 2425 ͑1991͒.
12 R. L. Vander Wal and F. F. Crim, J. Phys. Chem. 93, 5331 ͑1989͒.
13 P. R. Fleming, M. Li, and T. R. Rizzo, J. Chem. Phys. 95, 865 ͑1991͒.
14 X. Luo, P. R. Fleming, and T. R. Rizzo, J. Chem. Phys. 96, 5659 ͑1992͒.
15 M. Li and T. R. Rizzo, J. Chem. Phys. 94, 889 ͑1991͒.
J. Chem. Phys., Vol. 104, No. 15, 15 April 1996
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
142.150.190.39 On: Sat, 20 Dec 2014 18:02:51