Site specificity in molecular hydrogen elimination from photodissociation of propane at 157 nm

Article abstract of DOI:10.1063/1.479507

Site effects on the molecular hydrogen elimination from propane at 157 nm excitation have been studied using the photofragment translational spectroscopic technique. Experimental results indicate that H₂ elimination from the internal carbon of propane (2,2-elimination) is predominant while eliminations from the terminal carbon (1,1- and 1,3-elimination) and the vicinal carbons (1,2-elimination) are minor. The translational energy distributions obtained for these processes also show that the dynamics of H₂ eliminations from different sites are significantly different. Relative branching ratios of the atomic hydrogen (H) and the molecular hydrogen (H₂) elimination processes were also determined.

Full text of DOI:10.1063/1.479507

Site specificity in molecular hydrogen elimination from photodissociation of propane at
157 nm
S. M. Wu, J. J. Lin, Y. T. Lee, and X. Yang
Citation: The Journal of Chemical Physics 111, 1793 (1999); doi: 10.1063/1.479507
View online: http://dx.doi.org/10.1063/1.479507
View Table of Contents: http://scitation.aip.org/content/aip/journal/jcp/111/5?ver=pdfcov
Published by the AIP Publishing
Articles you may be interested in
Distributions of angular anisotropy and kinetic energy of products from the photodissociation of methanol at 157
nm
J. Chem. Phys. 121, 11053 (2004); 10.1063/1.1814099
Site-specific dissociation dynamics of ethylene at 157 nm: Atomic and molecular hydrogen elimination
J. Chem. Phys. 113, 9668 (2000); 10.1063/1.1321044
Site specific dissociation dynamics of propane at 157 nm excitation
J. Chem. Phys. 112, 8027 (2000); 10.1063/1.481425
Photodissociation dynamics of trifluoroethylene at 157 nm excitation
J. Chem. Phys. 109, 10719 (1998); 10.1063/1.477770
Site and isotope effects on the molecular hydrogen elimination from ethylene at 157 nm excitation
J. Chem. Phys. 109, 2979 (1998); 10.1063/1.476888
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:
134.71.135.134 On: Fri, 21 Nov 2014 23:03:50

JOURNAL OF CHEMICAL PHYSICS
VOLUME 111, NUMBER 5
1 AUGUST 1999
COMMUNICATIONS
Site specificity in molecular hydrogen elimination from photodissociation
of propane at 157 nm
S. M. Wu
Department of Chemistry, National Taiwan University, Taiwan, Republic of China
J. J. Lin
Institute of Atomic and Molecular Science, Academia Sinica, Taipei, Taiwan, Republic of China
Y. T. Lee
Department of Chemistry, National Taiwan University, and Institute of Atomic and Molecular Science,
Academia Sinica, Taipei, Taiwan, Republic of China
X. Yanga)
Institute of Atomic and Molecular Science, Academia Sinica, Taipei, Taiwan, Republic of China
͑Received 23 April 1999; accepted 24 May 1999͒
Site effects on the molecular hydrogen elimination from propane at 157 nm excitation have been
studied using the photofragment translational spectroscopic technique. Experimental results indicate
that H2 elimination from the internal carbon of propane ͑2,2-elimination͒ is predominant while
eliminations from the terminal carbon ͑1,1- and 1,3-elimination͒ and the vicinal carbons
͑1,2-elimination͒ are minor. The translational energy distributions obtained for these processes also
show that the dynamics of H2 eliminations from different sites are significantly different. Relative
branching ratios of the atomic hydrogen ͑H͒ and the molecular hydrogen (H2) elimination processes
were also determined. © 1999 American Institute of Physics. ͓S0021-9606͑99͒01829-2͔
Photodissociation of saturated hydrocarbons ͑alkanes͒ is
a fundamentally important and interesting subject. It has
been established that molecular detachment processes play a
significant role in the photolysis of methane (CH4),1,2 ethane
(CH3CH3),3 propane (CH3CH2CH3),4 and n-butane
(CH3CH2CH2CH3)5 under vacuum ultraviolet ͑VUV͒ light
excitation. However, molecular hydrogen elimination pro-
cesses from saturated hydrocarbons have rarely been studied
systematically using modern experimental techniques. This
is partly due to the lack of access of high power pulsed lasers
in the VUV region where saturated hydrocarbon molecules
start to absorb.
ity of the atomic elimination process from the photodissocia-
tion of propane at 157 nm excitation has been investigated
using the Doppler spectroscopic technique by Tonokura
et al.13 Their experimental results showed that the hydrogen
atom products mostly come from the terminal carbons of
propane. The energy diagram of possible products from the
photodissociation of propane at 157 nm excitation is shown
in Fig. 1. At the excitation of this high photon energy ͑181.4
kcal/mol or 7.87 eV͒, methyl radical (CH3), methane mol-
ecule (CH4), atomic hydrogen ͑H͒, and molecular hydrogen
(H2) elimination processes, and many different triple disso-
ciation pathways are energetically accessible. The molecular
hydrogen could be eliminated via the following pathways:
Dynamics of the atomic hydrogen elimination processes
from the photodissociation of CH3CH3 and CH3CH2CH3 at
Lyman-␣ ͑121.6 nm͒ excitation have been recently studied
using the velocity map imaging technique,6 and CH4 has
been extensively investigated in detail with its isotopomers
in the past few years.2,7–9 Chandler and co-workers have also
measured the H2(␷,J) photofragment images following the
two-photon absorption which deposits 10.8–11.8 eV of en-
ergy in a methane molecule.2 They observed that there are
two molecular hydrogen elimination channels, which are the
direct two-body dissociation process (CH4˜CH2ϩH2) and
the three-body dissociation process (CH4˜CHϩH2ϩH).
Propane is the smallest saturated hydrocarbon that has suffi-
cient absorption at 157 nm excitation.10–12 The site specific-
CH CH CH ϩh␯ 157 nm ˜CH CH CHϩH
͒
͑1͒
͑2͒
͑3͒
͑4͒
͑5͒
͑6͒
͑7͒
͑
3
2
3
3
2
2
˜CH3CCH3ϩH2
˜CH3CHvCH2ϩH2
˜CH2CH2CH2ϩH2
˜CH3CwCHϩ2H2
˜CH2vCvCH2ϩ2H2
˜C3H5ϩH2ϩH.
The first channel is the molecular hydrogen elimination from
the terminal carbon of propane ͑1,1-elimination͒, the second
channel is from the internal carbon ͑2,2-elimination͒, the
third channel is from the vicinal carbons ͑1,2-elimination͒,
and the fourth channel is from the two terminal methyl
a͒
Author to whom all correspondence should be addressed. Electronic mail:
xmyang@po.iams.sinica.edu.tw
0021-9606/99/111(5)/1793/4/$15.00
1793
© 1999 American Institute of Physics
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:
134.71.135.134 On: Fri, 21 Nov 2014 23:03:50

1794
J. Chem. Phys., Vol. 111, No. 5, 1 August 1999
Wu et al.
this study is the ultraclean detector in the apparatus. The
ultrahigh vacuum (1ϫ10Ϫ12 Torr) in the electron impact
ionization region reduces the H2 background by almost 2
orders of magnitude in comparison to other similar appara-
tus. Thus, the detection of the molecular and atomic hydro-
gen channels becomes much easier. An unpolarized 157 nm
laser beam ͑Lambda Physik LPX 210i, F2 laser͒ with about 1
mJ/pulse laser power was focused to a spot size of
ϳ4 mmϫ4 mm in the interaction region. The laser beam was
crossed with the propane molecular beam perpendicularly in
at a distance of ϳ8 mm away from the nozzle tip. A 50 Hz
pulsed propane beam was produced by expanding the neat
propane sample from a solenoid valve ͑General Valve͒
through a 0.5-mm-diam orifice at the backing pressure of
about 80 Torr. The flight distance of the neutral products was
285 mm and the detection axis was orthogonal to both the
laser beam and the molecular beam. All experimental condi-
tions such as molecular number density, and laser intensity
in the interaction region were well controlled in order to
make meaningful comparisons. Multiphoton effect and mo-
lecular clustering effect were carefully checked and avoided.
Time-of-flight ͑TOF͒ spectra of the photodissociation
products at m/eϭ1,2 from CH3CH2CH3 and at m/e
ϭ1,2,3,4 from CH3CD2CH3 at 157 nm excitation are shown
in Fig. 2. The experimental study on CH3CD2CH3 aims to
clarify the sources of molecular hydrogen elimination pro-
cesses. In order to know the exact relative contributions of
different products, detection efficiencies of H2, HD, and D2
were calibrated by measuring the signal intensities of the
continuous effusive beam consisting of an equimolar mixture
of H2, HD, and D2 at m/eϭ2, 3, and 4. Since total ionization
cross sections of H2 and D2 are known to be nearly
identical,19 HD is expected to have the same ionization cross
section as H2 and D2. From the relative detection of H2, HD,
D2, detection efficiency of the H atom at m/eϭ1 could be
estimated since the ratio of the ionization efficiency of
atomic and molecular hydrogen was known.19 Since disso-
ciative ionization of molecular hydrogen by the electron
bombardment at ϳ60 eV electron energy is negligible,19 the
TOF spectra at m/eϭ1 and 2 shown in Fig. 2͑a͒ should be
the photofragment signals of H and H2, respectively. The
TOF spectra at m/eϭ1, 2, 3, and 4 shown in Fig. 2͑b͒ are the
photofragment signals of H, D and H2, HD, and D2, respec-
tively. The D/H ratio from the photodissociation of
CH3CD2CH3 at 157 nm has been reported to be ϳ0.07:1 by
comparing the total areas under the Doppler profile.13 Since
the signals at m/eϭ1 come only from the H atom in this
experiment, the D atom contributions to m/eϭ2 should be
very small by comparing the signals and the D/H ratio. H
atom Rydberg tagging experiment in our laboratory also con-
firms this result. Thus, the signals at m/eϭ2 shown in Fig.
2͑b͒ should mostly come from the photodissociation prod-
ucts of H2. The solid curves shown in Fig. 2 are the simu-
lated TOF spectra using the translational energy distribu-
tions, shown in Fig. 3, for atomic and molecular hydrogen
elimination channels from the photodissociation of
CH3CH2CH3 and CH3CD2CH3. The simulations were carried
out using a recently developed software package, CMLAB3,
which is modified from the previous version, CMLAB2.20 All
FIG. 1. Energy diagram of some possible product channels from the photo-
dissociation of propane at 157 nm ͑7.87 eV͒ excitation. The relative energies
were calculated from the heat of formation of the species ͑see Ref. 21͒.
groups ͑1,3-elimination͒. The final three channels are the
triple dissociation processes. As we can see from the above,
H2 products may come from many possible combinations of
different sites.
The absorption spectra of the alkanes from propane to
n-octane in the VUV region are continuous with no clear
vibrational structure.10,14 There is weak absorption between
163.0 and 157.5 nm in all the spectra from ethane to
n-octane. However, no clear consensus has been reached yet
on the assignment of these absorption bands. Raymonda and
Simpson assigned those to an intramolecular charge transfer
excitation between adjacent C–C bonds,11 while Robin as-
signed as a Rydberg excitation to 3s,12 and Sandorfy and
co-workers suggested the excitation is localized mainly in
one ethyl group.10 Theoretical calculations15,16 shows that
the vertical transition energy to the 3s Rydberg orbital is
about 1 eV higher than the starting absorption energy of the
first absorption shoulder. This, however, is not necessarily
contradictory to the assignment of the lowest absorption of
propane to the 3s Rydberg absorption since vertical transi-
tion energy could be much higher than the adiabatic transi-
tion energy due to large geometric changes between the ini-
tial and final electronic states.17 In any case, the identity of
the first absorption shoulder needs to be further clarified.
A photofragment translational spectroscopic technique
was used to study the photodissociation of propane in this
work. The main thrust of this study is to understand the
molecular hydrogen elimination processes from propane.
The apparatus used in this work has been previously
described.18 Briefly, it contains two source chambers, a main
chamber and a rotatable detector. The most crucial part for
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:
134.71.135.134 On: Fri, 21 Nov 2014 23:03:50

J. Chem. Phys., Vol. 111, No. 5, 1 August 1999
Site specificity in molecular hydrogen elimination
1795
FIG. 3. Translational energy distributions used to fit the TOF spectra of the
products of the atomic and molecular hydrogen elimination channels from
the photodissociation of CH3CH2CH3 and CH3CD2CH3. Curves a1 and a2
are, respectively, the translational energy distributions of the H and H2
elimination channels from CH3CH2CH3. Curves b1, b2, b3, and b4 are,
respectively, the translational energy distributions of the H, H2, HD, and D2
elimination channels from CH3CD2CH3. The relative heights of the distri-
butions are arbitrary.
energy release of the H atom elimination is also very small,
indicating that the H atom process likely proceeds via a sec-
ondary dissociation process without an exit barrier. There are
some small, but obvious, differences between the two trans-
lational energy distributions for the H atom elimination pro-
cesses from CH3CD2CH3 and CH3CH2CH3. At higher ener-
gies, the translational energy distribution for the
H
FIG. 2. TOF spectra of the photodissociation products from ͑a͒
CH3CH2CH3 and ͑b͒ CH3CD2CH3. The spectra were accumulated over 100
k laser shots, and calibrated by the detection efficiencies of different species.
The open shapes corresponding to curve 1, 2, 3, and 4 are the experimental
results at m/eϭ1, 2, 3, and 4, respectively. Curve 1, 2 in ͑a͒, and curve 1, 2,
3, 4 in ͑b͒ are the simulated TOF calculated from the translational energy
distributions which are curves a1, a2, b1, b2, b3, and b4 shown in Fig. 3,
respectively.
elimination from CH3CH2CH3 is quite larger than from
CH3CD2CH3. This may mainly be due to the H detachment
from the internal carbon. The argument is also supported
from the preliminary results of the photodissociation of
CD3CH2CD3. Curve b2 is the translational energy distribu-
tion obtained for the H2 elimination ͑1,1- and 1,3-
elimination͒ from the photodissociation of CH3CD2CH3 and
peaks roughly around 20–30 kcal/mol. It is immediately
clear that the translational energy distribution ͑b2͒ from
CH3CD2CH3 is significantly different from that ͑a2͒ for the
H2 elimination from CH3CH2CH3. Curve b3 in Fig. 3 is the
translational energy distribution of the HD elimination ͑1,2-
elimination͒ and peaks at ϳ61 kcal/mol, and curve b4 in Fig.
3 is the translational energy distribution of the D2 elimination
͑2,2-elimination͒ and peaks at ϳ31 kcal/mol. The transla-
tional energy distribution ͑b3͒ for the HD elimination pro-
cess from CH3CD2CH3 is also significantly different from
that for the H2 elimination from CH3CH2CH3, while the
translational energy distribution ͑b4͒ for the D2 elimination
from CH3CD2CH3 is quite similar to that for the H2 elimina-
tion from CH3CH2CH3. The relative branching ratios of the
molecular hydrogen elimination processes from CH3CD2CH3
are listed in Table I. The ratio for the overall molecular hy-
drogen (H2, HD, and D2) to the H elimination channel from
CH3CD2CH3 is determined to be 2.2, which is quite close to
the H2 /H branching ratio for the normal propane ͑2.1͒. From
the relative branching ratios ͑Table I͒ of the three molecular
elimination channels, it is clear that the 2,2-elimination from
the photodissociation of CH3CD2CH3 is predominant ͑75%͒,
i.e., the molecular hydrogen products mostly come from the
central carbon. Other types of H2 elimination are minor. In
the translational energy distributions were obtained assuming
that the dissociation products were from a binary dissocia-
tion process.
As shown in Fig. 3, curve a1 is the translational energy
distribution of the H atom elimination and curve a2 is that of
the H2 elimination from the photodissociation of
CH3CH2CH3. From Fig. 2͑a͒, H2 elimination from propane
is clearly the dominant process, while H elimination is mi-
nor. The H2 /H branching ratio was determined to be 2.1 by
calculating the integral ratio of the two channels in the
center-of-mass frame. The translational energy distributions
for the H and H2 elimination processes are significantly dif-
ferent. The translational energy distribution for the H2 elimi-
nation process ͑curve a2͒ has two peaks. The lower energy
peak is at ϳ30 kcal/mol and the higher one is at ϳ61
kcal/mol, indicating there are at least two different dynami-
cal processes involved in the H2 elimination from normal
propane. The translational energy distribution of the H elimi-
nation process ͑curve a1͒ peaks at 0 kcal/mol.
Curve b1 in Fig. 3 is the translational energy distribution
for the H elimination from the photodissociation of
CH3CD2CH3. Similar to the case of CH3CH2CH3, the trans-
lational energy distribution also peaks at 0 kcal/mol. The
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:
134.71.135.134 On: Fri, 21 Nov 2014 23:03:50

1796
J. Chem. Phys., Vol. 111, No. 5, 1 August 1999
Wu et al.
TABLE I. Relative yields of molecular hydrogen elimination channels from
the photodissociation of CH3CH2CH3 and CH3CD2CH3.
picture provided above for the molecular hydrogen elimina-
tion should be correct. Therefore it is clear that the isotope
effects are not significant. From Fig. 4, all translational en-
ergy distributions for molecular hydrogen elimination pro-
cesses extend to higher energy limit, indicating that molecu-
lar hydrogen elimination processes are likely primary
dissociation processes. And from the relative branching ra-
tios, it is apparent that molecular H2 elimination processes
from propane are very nonstatistical, while the H atom elimi-
nation is likely a statistical process.
Molecule
1,1-E and 1,3-E
1,2-E
2,2-E
CH3CH2CH3
CH3CD2CH3
9a
6
14a
19
77a
75
aThe relative yields come from the fitting ratios for curves 1, 2, and 3 shown
in Fig. 4.
the H2 elimination from the terminal CH3 groups, there are
two possible types of elimination processes: 1,1-elimination
and 1,3-elimination. In this experiment, we are not able to
differentiate the two pathways. A closer look at these disso-
ciation pathways is under way in our laboratory. The site
specificity observed for the molecular elimination here is
very interesting since it is exactly the opposite of the atomic
hydrogen elimination in which the dominant process is the H
elimination from the terminal CH3 groups. It is also interest-
ing to notice that the dynamics for all these processes are
significantly different from each other, indicating that no iso-
tope scrambling occurs in the dissociation process of pro-
pane.
Since total H2 elimination from CH3CH2CH3 is the com-
bination of different site H2 elimination processes: 1,1- and
1,3-elimination, 1,2-elimination, and 2,2-elimination, the ex-
perimental TOF spectrum at m/eϭ2(H2) from CH3CH2CH3
can be fitted using the translational energy distributions of
these different processes obtained from the photodissociation
of CH3CD2CH3. Figure 4 shows the experimental and simu-
lated TOF spectra of the H2 product. The agreement between
experiment and simulation is quite good. The relative
branching ratios obtained for different H2 elimination pro-
cesses from the photodissociation of CH3CH2CH3 are listed
in Table I. The results are very similar to those obtained for
the photodissociation of CH3CD2CH3, indicating that the
From the above experimental investigations, a very clear
picture of the atomic and molecular hydrogen elimination
processes is presented. Molecular hydrogen elimination from
the photodissociation of propane at 157 nm excitation is
found to be significantly more important than the atomic
hydrogen elimination process. Molecular hydrogen elimina-
tion mostly proceeds via direct two-body dissociation path-
ways, and is primarily from the internal carbon ͑2,2-
elimination͒. The dynamics of different molecular hydrogen
elimination processes from different sites are also signifi-
cantly different.
This work was supported by the National Research
Council, the Academia Sinica of R.O.C., and the China Pe-
troleum Corporation. The authors would like to acknowledge
S. Harich for the development of the convenient software
package, CMLAB3. We appreciate the helpful discussions
with Professor A. M. Mebel.
1 B. H. Mahan and R. Mandal, J. Chem. Phys. 37, 207 ͑1962͒.
2 A. J. R. Heck, R. N. Zare, and D. W. Chandler, J. Chem. Phys. 104, 4019
͑1996͒.
3 H. Okabe and J. R. McNesby, J. Chem. Phys. 34, 668 ͑1961͒.
4 H. Okabe and J. R. McNesby, J. Chem. Phys. 37, 1340 ͑1962͒.
5 H. Okabe and D. A. Becker, J. Chem. Phys. 39, 2549 ͑1963͒.
6 W. M. Jackson, R. J. Price II, D. D. Xu, J. D. Wrobel, M. Ahmed, D. S.
Peterka, and A. G. Suits, J. Chem. Phys. 109, 4703 ͑1998͒.
7 D. H. Mordaunt, I. R. Lambert, G. P. Morley, M. N. R. Ashfold, R. N.
Dixon, C. M. Western, L. Schnieder, and K. H. Welge, J. Chem. Phys. 98,
2054 ͑1993͒.
8 A. J. R. Heck, R. N. Zare, and D. W. Chandler, J. Chem. Phys. 104, 3399
͑1996͒.
9 J.-H. Wang and K. Liu, J. Chem. Phys. 109, 7105 ͑1998͒.
10 B. A. Lombos, P. Sauvageau, and C. Sandorfy, J. Mol. Spectrosc. 24, 253
͑1967͒.
11 J. W. Raymonda and W. T. Simpson, J. Chem. Phys. 47, 430 ͑1967͒.
12 M. B. Robin, Higher Excited States of Polyatomic Molecules ͑Academic,
New York, 1974͒, Vol. 1, p. 105, and references therein.
13 K. Tonokura, Y. Matsumi, M. Kawasaki, and K. Kasatani, J. Chem. Phys.
95, 5065 ͑1991͒.
14 S. Katagiri and C. Sandorfy, Theor. Chim. Acta 4, 203 ͑1966͒.
15 A. Richartz, R. J. Buenker, and A. D. Peyerimhoff, Chem. Phys. 31, 187
͑1978͒.
16 V. Galasso, Chem. Phys. 161, 189 ͑1992͒.
17 A. M. Mebel, S. H. Lin, and C.-H. Chang, J. Chem. Phys. 106, 2612
͑1997͒.
18 J. J. Lin, D. W. Hwang, S. Harich, Y. T. Lee, and X. Yang, Rev. Sci.
Instrum. 69, 1642 ͑1998͒.
FIG. 4. Site-specific TOF spectrum of the H2 elimination channel from the
photodissociation of CH3CH2CH3. The open circles are the experimental
data, while the solid lines are the fits to the spectrum using the translational
energy distributions of the molecular hydrogen elimination processes from
CH3CD2CH3 shown in Fig. 3. Curves 1, 2, and 3 are fitted from the H2, HD,
and D2 elimination processes, respectively; curve 4 is the overall fit.
19 H. Tawara and T. Kato, At. Data Nucl. Data Tables 36, 167 ͑1987͒.
20 X. Zhao, Ph.D. thesis, University of California, Berkeley, 1988; J. D.
Myers, Ph.D. thesis, University of California, Berkeley, 1992.
21 Handbook of Chemistry and Physics, edited by D. R. Lide ͑CRC Press,
Boca Raton, FL, 1995͒.
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:
134.71.135.134 On: Fri, 21 Nov 2014 23:03:50