Photodissociation dynamics of 1,1-difluoroethylene at 157 nm excitation

Article abstract of DOI:10.1063/1.477780

Photodissociation of 1,1-difluoroethylene (F₂CCH₂) at 157 nm has been investigated using photofragment translational spectroscopy. Five dissociation channels have been experimentally observed; molecular HF elimination, H atom elimination, molecular hydrogen (H₂) elimination, F atom elimination, and double bond breaking. Molecular HF elimination and H atom elimination channels are found to be the two major dissociation pathways in photodissociation of F₂CCH₂ at 157 nm excitation. Molecular hydrogen (H₂) elimination and double bond cleavage are also significant, while F atom elimination is a minor process. Product translational energy distributions for all dissociation channels have also been measured. All translational energy releases are peaked at energies away from zero, indicating that the dissociation of F₂CCH₂ at 157 nm excitation most likely occurs with exit barriers on the ground electronic potential surface through internal conversion from the initially excited electronic state. Branching ratios and averaged energy partitions for different channels have also been estimated.

Full text of DOI:10.1063/1.477780

Photodissociation dynamics of 1,1-difluoroethylene at 157 nm excitation
J. J. Lin, S. M. Wu, D. W. Hwang, Y. T. Lee, and X. Yang
Citation: The Journal of Chemical Physics 109, 10838 (1998); doi: 10.1063/1.477780
View online: http://dx.doi.org/10.1063/1.477780
View Table of Contents: http://scitation.aip.org/content/aip/journal/jcp/109/24?ver=pdfcov
Published by the AIP Publishing
Articles you may be interested in
Photodissociation dynamics of vinyl fluoride ( C H 2 C H F ) at 157 and 193 nm : Distributions of kinetic energy
and branching ratios
J. Chem. Phys. 125, 144315 (2006); 10.1063/1.2357946
Photodissociation dynamics of indole in a molecular beam
J. Chem. Phys. 123, 124303 (2005); 10.1063/1.2009736
Dynamics of OH formation in photodissociation of pyruvic acid at 193 nm
J. Chem. Phys. 118, 10093 (2003); 10.1063/1.1572133
Photodissociation dynamics of trifluoroethylene at 157 nm excitation
J. Chem. Phys. 109, 10719 (1998); 10.1063/1.477770
UV photodissociation dynamics of allyl radical by photofragment translational spectroscopy
J. Chem. Phys. 109, 5372 (1998); 10.1063/1.477156
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:
130.88.90.140 On: Tue, 09 Dec 2014 17:49:40

JOURNAL OF CHEMICAL PHYSICS
VOLUME 109, NUMBER 24
22 DECEMBER 1998
Photodissociation dynamics of 1,1-difluoroethylene at 157 nm excitation
J. J. Lin, S. M. Wu, and D. W. Hwang
Department of Chemistry, National Taiwan University, Taipei, Taiwan, Republic of China
Y. T. Lee
Department of Chemistry, National Taiwan University and Institute of Atomic and Molecular Sciences,
Academia Sinica, Taipei, Taiwan, Republic of China
X. Yanga)
Institute of Atomic and Molecular Sciences, Academia Sinica, Taipei, Taiwan, Republic of China
͑Received 19 June 1998; accepted 16 September 1998͒
Photodissociation of 1,1-difluoroethylene (F2CCH2) at 157 nm has been investigated using
photofragment translational spectroscopy. Five dissociation channels have been experimentally
observed; molecular HF elimination, H atom elimination, molecular hydrogen (H2) elimination, F
atom elimination, and double bond breaking. Molecular HF elimination and H atom elimination
channels are found to be the two major dissociation pathways in photodissociation of F2CCH2 at
157 nm excitation. Molecular hydrogen (H2) elimination and double bond cleavage are also
significant, while F atom elimination is a minor process. Product translational energy distributions
for all dissociation channels have also been measured. All translational energy releases are peaked
at energies away from zero, indicating that the dissociation of F2CCH2 at 157 nm excitation most
likely occurs with exit barriers on the ground electronic potential surface through internal
conversion from the initially excited electronic state. Branching ratios and averaged energy
partitions for different channels have also been estimated. © 1998 American Institute of Physics.
͓S0021-9606͑98͒00948-9͔
I. INTRODUCTION
ene molecules are good candidates for experimental studies
of this topic. At about 8 eV excitation, most of the fluoroet-
hylenes have significant absorption cross sections.3,4 This
will allow us to look at photodissociation processes at high
energy excitation in greater detail using the commercially
available F2 laser as the photolysis laser. Different fluori-
nated ethylenes actually have very interesting chemical prop-
erties. As the number of fluorine substitutions increases in
ethylene, the double bond ͑CvC͒ actually becomes weaker.
Figure 1 shows the double bond energies of the different
fluorinated ethylene molecules.5 It is therefore noteworthy to
look at the trends of the branching ratio of double bond
breaking relative to the other dissociation channels at the
VUV excitation in different fluorinated ethylenes.
Fluoroethylene molecules also serve as a good example
in studying molecular elimination processes, especially in
HF elimination processes. During the last decades, many ex-
perimental studies on the photodissociation of fluorinated
ethylenes have been conducted. Various methods were em-
ployed to excite the precursor molecules to energies at which
molecular fragmentation can take place. Photodissociation
by excitation to a higher electronic state, through absorption
of pulsed ultraviolet ͑UV͒ light from a flash lamp or laser,
were frequently used. Photofragmentation following UV ex-
citation is now believed to proceed largely from a highly
excited ground electronic state, formed by internal conver-
sion or intersystem crossing from the initially excited elec-
tronic state. The infrared multiphoton dissociation technique
was also applied to study the molecular dissociation of fluo-
roethylenes. In many of these experiments, the molecular
Photodissociation at very high energy excitation is inter-
esting and also of fundamental importance in chemistry.
Most molecules have extensive absorption at high energy
excitation ͑8 eV͒, and yet photochemistry in this energy re-
gime has been very rarely studied. This is largely due to the
lack of strong photolysis light sources in the vacuum ultra-
violet ͑VUV͒ region. Only until recently are strong VUV
light sources becoming available.1 Bond selectivity in pho-
tochemical processes at such high energy excitation is also
an interesting topic in general. Bond selectivity in dissocia-
tion processes generally can result from two main sources:
selectivity due to energetics or selectivity due to purely dy-
namics. Selectivity due to energetics means that the weaker
the chemical bond, the easier it is to break. However, if it is
purely due to dynamics,2 the resulting selectivity will not be
decided by thermodynamics. It is not immediately clear
whether the photochemistry at extremely high energy elec-
tronic excitation, where every bond is breakable, is purely
determined by energetics or by pure dynamics, especially in
relatively large molecules. At about 8 eV excitation, where
strong photolysis lasers ͑157 nm͒ are already available, al-
most every chemical bond existed in nature is already break-
able. Many interesting questions can then be posed: At this
excitation of energy, how are these chemical bonds broken?
Are the bonds broken in a selective way, or are the dissocia-
tion pathways purely determined by energetics? Fluoroethyl-
a͒
Author to whom correspondence should be addressed; electronic mail:
xmyang@po.iams.sinica.edu.tw
0021-9606/98/109(24)/10838/9/$15.00
10838
© 1998 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:
130.88.90.140 On: Tue, 09 Dec 2014 17:49:40

J. Chem. Phys., Vol. 109, No. 24, 22 December 1998
Lin et al.
10839
FIG. 2. Energy diagram of possible photodissociation pathways from 1,1-
difluoroethylene at 157 nm excitation.
FIG. 1. Graph shows the CvC double bond energies of the fluorinated
ethylenes. The double bond energy decreases as the number of the substi-
tuted fluorine atoms increases.
ethylene.4 Five possible binary dissociation channels are
listed in the energy diagram—three simple bond rupture
channels and two molecular elimination channels:
F CvCH ϩh␯ 157 nm͒→F C:ϩ:CH
͑
2
2
2
2
hydrogen halide elimination channel was carefully studied
by observing infrared emission from specific rovibrational
levels. The results of these investigations carried out before
1980 can be found in several review articles,6–8 and refer-
ences therein. More recently, Setser, Arunan, and
Wategaonkar9 studied several three-center and four-center
HF and HCl eliminations from molecules formed by the re-
action of H atoms with radicals by detecting the steady-state
infrared ͑IR͒ emission from HF or HCl, respectively. Similar
experiments on vinyl fluoride were performed by Donaldson
et al.10,11 Leone and co-workers12,13 have used novel time-
resolved Fourier transform infrared ͑FTIR͒ spectroscopy to
investigate the dissociation of vinylfluoride, trans-1,2-
dichloroethylene and 1,1-chlorofluoroethylene. Gordon and
co-workers14–16 also investigated dichloroethylene isomers
and vinyl chloride using time-of-flight resonance-enhanced
multiphoton ionization, to detect the photofragment prod-
ucts. Recently, Hall and co-workers studied the photodisso-
ciation dynamics of 1,1-difluoroethylene at 193 nm excita-
tion using time-resolved FTIR emission.17 Photofragment
translational spectroscopy was employed to study vinyl chlo-
ride, trans- and cis-dichoroethylene, and 1,1-dichloroethy-
lene at 193 nm excitation.18 The photodissociation dynamics
of vinyl fluoride at 157 nm was studied recently using pho-
tofragment translational spectroscopy.19 Only the HF elimi-
nation channel was clearly identified. Photodissociation of
1,1- and 1,2-difluoroethylene at 193 nm was also studied
very recently.20 Experimental results show that the transla-
tional energy distributions of the HF elimination channel for
the 1,1- and 1,2-difluoroethylene are significantly different,
indicating very different dynamics for the two molecules. H2
and H elimination channels were also observed for both mol-
ecules. The translational energy distribution of H2 elimina-
tion displays a significant exit barrier in the dissociation pro-
cess, similar to that of ethylene photodissociation.
→FCvCH2ϩF
→F2CvCHϩH
→F2CvC:ϩH2
→FCvCHϩHF.
Triple dissociations are only energetically possible after HF
elimination from 1,1-difluoroethylene at 157 nm excitation.
From the energy diagram, the available energies for the triple
dissociation channels are very small. Therefore it would be
difficult to detect them experimentally. In this work, photo-
dissociation products are measured by photofragment trans-
lational spectroscopy. Product translational energy distribu-
tions are mapped out from time-of-flight spectra of
photodissociation products. Relative contributions of differ-
ent channels are also estimated, giving us a valuable set of
data for comparisons with other fluoroethylene molecules.
II. EXPERIMENT
The instrument used in the experimental investigation of
1,1-difluoroethylene photodissociation at 157 nm is a newly
built crossed molecular beam apparatus. A detailed descrip-
tion of the new machine can be found in Ref. 21. Briefly, the
new apparatus consists of three parts: a main chamber, a
source chamber, and a rotatable detector. The source cham-
ber generates molecular beams for studies of crossed mo-
lecular beam reactions or molecular photodissociations. The
apparatus can be set up in two different running modes for
photodissociation studies: heavy fragment detection and light
fragment detection ͑mainly H and H2). For heavy fragment
detection, the main chamber is where the laser beam and the
molecular beam interact with each other for photodissocia-
tion experiments. While for light atom detection ͑H and H2),
laser and molecular beams interact with each other in the
source chamber with the molecular beam perpendicular to
the detection axis. The interaction region is about 8 mm
away from the nozzle. Such close distance is used is to com-
pensate for low detection efficiency in the detector of the
The experiments described in this article aim to under-
stand the photodissociation dynamics of 1,1-difluoroethylene
at 157 nm. At 157 nm excitation, 1,1-difluoroethylene in-
volves mostly Rydberg excitation. Figure 2 shows the energy
diagram of possible dissociation channels of 1,1-difluoro-
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:
130.88.90.140 On: Tue, 09 Dec 2014 17:49:40

10840
J. Chem. Phys., Vol. 109, No. 24, 22 December 1998
Lin et al.
lighter fragments ͑H and H2) because of the shorter resi-
dence time in the ionizer. The photodissociation products are
detected using a differentially pumped universal detector. A
neat sample is used in this detection scheme. For heavy frag-
ment detection, the laser, the molecular beam, and the detec-
tion axes are in the same plane. Normally, a molecular beam
seeded in Ar is used. The F2CCH2 sample in the molecular
beam are then intercepted by an unpolarized 157 nm laser
beam, which is generated by a Lambda Physik LPX210I la-
ser with a NOVA tube laser cavity. About 1–3 mJ of 157 nm
laser light was normally used during the experiment to avoid
multiphoton effects. The laser pulse duration is about 15 ns.
The laser beam is focused to a 5 mm ͑width͒ϫ2 mm ͑height͒
spot size in the interaction region using a MgF2 lens with a
focal length of 100 cm. Since black deposits on the focus
lens, which reduce laser power to the interaction region, can
be easily formed by the 157 nm high power laser beam, a
differentially pumped laser beam path is used so that this
problem can be avoided. The whole laser beam path is in a
vacuum tube with a vacuum of about 1ϫ10Ϫ7 Torr or better.
The focusing lens can then be easily cleaned without chang-
ing the laser beam path and without venting the main cham-
ber of the apparatus. Photodissociation products recoiled
from the interaction region are detected by an electron im-
pact ionization detector. The electron impact ionizer is lo-
cated in the detection region where ultrahigh vacuum
(10Ϫ12 Torr) is maintained. The ions are then extracted from
the ionizer and guided into a quadrupole mass filter ͑Extrel͒
and counted by a Daly-type detector. A multichannel scaler
͑Turbo MCS by EG&G͒ is then used to record the product
time of flight spectra at different angles between the molecu-
lar beam and the detector axis.
The F2CCH2 sample was purchased from PCR, Co. The
sample is used without further purification. For heavy frag-
ment detection, the F2CCH2 molecular beam is generated by
expanding a 5% premixture of F2CCH2 in Ar through a com-
mercially available pulsed nozzle ͑General Valve Co.͒. The
nozzle is maintained at 120 °C during the experiment in or-
der to avoid possible clustering in the molecular beam.
Since the product time-of-flight ͑TOF͒ spectra from pho-
todissociation of F2CCH2 are measured in laboratory frame,
in order to obtain the center-of-mass ͑CM͒ translational en-
ergy distribution P(E), a laboratory-to-center of mass con-
version is required. This conversion is done using an already
available program CMLAB2 running on a Pentium PC. The
analysis program uses an iterative forward convolution
method which is described elsewhere.22 In this program, a
trial P(E) and CM angular distribution ͑␤ parameter͒ is used
to calculate the TOF spectrum for a photofragment mass at a
given molecular beam-detector angle using the known appa-
ratus parameters and the measured beam velocity distribu-
tion. The calculated TOF is compared with the experimental
TOF spectrum and the P(E) was then improved by adjusting
the distribution point by point on the computer screen until
satisfactory fits are achieved for TOF spectra measured at
different angles. In the experimental set up for heavy frag-
ment detection, the laser beam, the molecular beam, and the
detection direction are in the same plane, as pointed out
above. In this configuration, the photofragment anisotropy ͑␤
FIG. 3. TOF spectrum at m/eϭ1, ⌰ϭ90° from photodissociation of 1,1-
difluoroethylene. This spectrum was taken by accumulating signals over 100
k laser shots using the light mass detection scheme. The open circles are the
experimental data points, while the solid line is the fit to this spectrum using
the translational energy distribution shown in Fig. 5 ͑curve 1͒.
parameter͒ will couple to the angular distribution of products
even if an unpolarized laser beam is used as the photolysis
source. In this experiment, however, the anisotropy is not
considered.
III. RESULTS
TOF spectra of photofragments from the F2CCH2 pho-
todissociation at 157 nm were measured using the mass fil-
tered photofragment translational spectrometric detector, as
described above. Signals at m/eϭ63, 62, 50, 45, 44, 43, 31,
25, 24, 20, 19, 14, 13, 12, 2, 1 were detected from photodis-
sociation of F2CCH2 . After detailed analyses of the experi-
mental data, all five dissociation channels listed in the energy
diagram ͑Fig. 2͒ were observed. No experimental evidence
for triple dissociations is found for 1,1-difluoroethylene at
157 nm excitation. Multiphoton effects have been checked
carefully during the experiment. All results shown here
should have little multiphoton effect. In the next few para-
graphs, the experimental data and detailed analyses will be
described.
A. Atomic hydrogen elimination: F2CCH2؉h␯
˜F2CCH؉H
H atom elimination from the 1,1-difluoroethylene at 157
nm excitation has been experimentally observed. The TOF
spectrum of the H atom (m/eϭ1) product was measured at
the perpendicular direction of the molecular beam using the
light mass detection scheme described above. Figure 3 shows
the TOF spectrum of the H atom product from the photolysis
of F2CCH at 157 nm. The heavy fragmentation partner of the
H atom elimination channel is the F2CCH radical (m/e
ϭ63), which would also go through dissociative ionization
to form daughter ions in the electron bombardment ionizer,
for example, F2CCϩ, CF2ϩ , CFϩ, etc. Figure 4 shows the
TOF spectra at mass 63 (F2CCHϩ) at two different labora-
tory angles. Since TOF spectra of H and F2CCH fragments
are from the same dissociation channel, total linear momen-
tum should be conserved in this process. Figure 5 shows the
translational energy distributions used to model the TOF
spectra at m/eϭ1 ͑H, curve 1͒ and m/eϭ63 (F2CCH, curve
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:
130.88.90.140 On: Tue, 09 Dec 2014 17:49:40

J. Chem. Phys., Vol. 109, No. 24, 22 December 1998
Lin et al.
10841
FIG. 6. TOF spectrum at m/eϭ2, ⌰ϭ90° from photodissociation of 1,1-
difluoroethylene. This spectrum was taken by accumulating signals over 100
k laser shots using the light mass detection scheme. The open circles are the
experimental data points, while the solid line is the fit to this spectrum using
the translational energy distribution shown in Fig. 7.
tional energy, and subsequently higher internal excitation
͑internally hot͒. The internally hot F2CCH radicals might go
through complete dissociative ionization much more easily
than the internally cold ones, which could make the disso-
ciative ionization patterns of the hot and cold radicals sig-
nificantly different. In the H atom elimination channel, the
internally hot F2CCH (m/eϭ63) might not survive at all in
the electron bombardment ionizer. Therefore, the slower part
of the TOF spectrum at m/eϭ63 is missing. Curve 3 in Fig.
5 represents the translational energy distribution of the
F2CCH radical products which are internally hot and go
through dissociation ionization in the ionizer. Hence in the
fitting of the daughter ion masses from the dissociation ion-
ization of the F2CCH product radicals, the translational en-
ergy distributions for both the cold and hot radicals were
used. This approach works well in fitting the TOF spectra of
all daughter ion masses cracked from the F2CCH radical
product due to the H atom elimination channel. This indi-
rectly supports the cold and hot radical assumption.
FIG. 4. TOF spectra at m/eϭ63, ⌰ϭ7.5° ͑upper trace͒ and 20° ͑lower
trace͒. The spectrum at 7.5° was taken by accumulating signals over 100 k
laser shots and the spectrum at 20° over 25 k laser shots. The open circles
are the experimental data points, while the solid lines are the fits to the
spectra using the translational energy distribution shown in Fig. 5 ͑curve 2͒.
2͒ using the CMLAB2 program. The two translational energy
distributions actually show a significant difference in the
lower energy part of the distribution, which should be ex-
actly the same due to total linear momentum conservation. It
seems that some of the lower energy part of the translational
energy distribution derived from the m/eϭ63 data is miss-
ing. The lower translational energy part of the m/eϭ63 data
represents the F2CCH product radicals with lower transla-
In Fig. 5, the translational energy distribution of the H
atom elimination channel ͑curve 1͒ is peaked at ϳ5 kcal/
mol, indicating that a small barrier exists in the dissociation
process. This result implies that H atom elimination most
likely occurs through the ground potential energy surface.
B. Molecular hydrogen elimination: F2CCH2؉h␯
˜F2CC:؉H2
The molecular hydrogen elimination channel from the
F2CCH2 molecule at 157 nm excitation was also observed
experimentally. TOF spectra at m/eϭ2 and 62 show the
existence of the molecular hydrogen elimination channel.
Figure 6 shows the TOF spectrum of the H2 product from the
photolysis of the F2CCH2 molecule using the light mass de-
tection configuration. The translational energy distribution
used to model the TOF spectrum of H2 is shown in Fig. 7.
The heavy fragment of the H2 elimination channel should be
the difluorovinylidene radical (F2CC:, m/eϭ62). The TOF
spectrum at m/eϭ62 (F2Cϩ2 ) ͑Fig. 8͒ shows the contribution
FIG. 5. Translational energy distributions used to fit the TOF spectra of the
dissociation products from the H atom elimination channel. The distribution
shown in curve 1 here is used to fit the H atom product TOF spectrum,
which should be the complete translational distribution of the H atom elimi-
nation process. The distribution of curve 2 is used to fit the m/eϭ63
͑F2CCH radical͒ TOF spectrum, representing the cold F2CCH radical frag-
ments which survive in the electron impact ionizer. The distribution of curve
3 is the subtraction of curve 2 from curve 1, representing the hot F2CCH
radical fragments that do not survive in the electron impact ionizer.
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:
130.88.90.140 On: Tue, 09 Dec 2014 17:49:40

10842
J. Chem. Phys., Vol. 109, No. 24, 22 December 1998
Lin et al.
FIG. 7. Translational energy distribution for the H2 elimination channel
from photodissociation of 1,1-difluoroethylene at 157 nm.
FIG. 9. TOF spectrum at m/eϭ50, ⌰ϭ15° from photodissociation of
F2CCH2 at 157 nm. The spectrum was taken by accumulating signals over
100 k laser shots. The open circles are the experimental data points, while
the solid lines are the fits to the spectra using the CMLAB2 program. Curve 1
is the contribution attributed to the double bond cleavage process using the
translational energy distribution shown in Fig. 11, while curve 2 is the
contribution from the H atom elimination channel in which the cold F2CCH
fragments dissociatively ionizes to F2Cϩ in the electron impact ionizer.
Curve 3 is the overall fit the spectrum combining the above two contribu-
tions.
of the H2 elimination channel calculated using the transla-
tional energy distribution in Fig. 7. In addition, the contribu-
tion to the TOF at m/eϭ62 from the dissociative ionization
of the F2CCH radical in the ionizer is also clear. The two
curves to fit the H atom elimination channels representing
the hot F2CCH fragment and the cold F2CCH fragment are
shown in Fig. 8. The translational energy shown in Fig. 7 is
a peaked distribution with its peak at about 15 kcal/mol,
indicating that the H2 elimination channel has an exit barrier
in the dissociation process. This result implies that the H2
elimination process likely occurs on the ground electronic
state through internal conversion from the initially excited
electronic state, similar to the H atom elimination process.
The H2 elimination from the F2CCH2 molecule is a
three-center elimination process. The reverse barrier height
for a three-center H2 elimination from ethylene on the
ground electronic surface is calculated to be about 12
kcal/mol.23 The result here is somewhat consistent with the
three-center elimination process, which implies that when H2
leaves the F2CCH2 molecule the C2F2 radical should be still
in the vinylidene structure.
C. CvC double bond cleavage: F2CCH2؉h␯
˜F2C:؉CH2
The CvC double bond fission from the excited F2CCH2
molecule at 157 nm excitation was also observed. The TOF
spectra of the two fragments: mass 50 and mass 14 are
FIG. 8. TOF spectra at m/eϭ62, ⌰ϭ7.5° ͑upper trace͒ and ⌰ϭ15° ͑lower
trace͒. The spectrum at 7.5° was taken by accumulating signals over 500 k
laser shots and the spectrum at 15° over 500 k laser shots. The open circles
are the experimental data points, while the solid lines are the fits to the
spectra using the translational energy distribution shown in Fig. 7. Curve 1
is the contribution of the H2 elimination channel, while curves 2 and 3 are
the contributions from the so-called cold and hot F2CCH fragments from the
H elimination channel. Curve 4 is the overall fit to the spectrum.
FIG. 10. TOF at m/eϭ14 (CH2), ⌰ϭ20° from the photodissociation of
F2CCH2 at 157 nm. The spectrum was taken by accumulating signals over
750 k laser shots. The open circles are the experimental data points, while
the solid line is the fit to the TOF spectrum from the translational energy
distribution shown in Fig. 11 using the CMLAB2 program.
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:
130.88.90.140 On: Tue, 09 Dec 2014 17:49:40

J. Chem. Phys., Vol. 109, No. 24, 22 December 1998
Lin et al.
10843
FIG. 11. Translational energy distribution for the double bond cleavage
channel of 1,1-difluoroethylene at 157 nm.
FIG. 13. Translational energy distribution used to fit the TOF spectra at
m/eϭ45 for F atom elimination from 1,1-difluoroethylene at 157 nm.
shown in Figs. 9 and 10. The feature in the TOF spectrum of
mass 14 (:CH2) is assigned to the double bond cleavage
channel, the translational energy distribution ͑Fig. 11͒ de-
rived from the TOF spectrum of mass 14 should be the ki-
netic energy distribution for the CvC bond breaking pro-
cess. Using this translational energy distribution, the fast
peak shown in the TOF spectrum of mass 50 (F2C:ϩ) can be
momentum matched as the fragment partner of the CH2 frag-
ment using the CMLAB2 program, indicating that the fast peak
in the TOF spectrum at mass 50 is indeed from the F2C:
fragment of the double bond cleavage process. The slower
peak in the TOF spectrum of m/eϭ50 is due to the disso-
ciative ionization of the F2CCH fragment in the ionizer from
the hydrogen elimination channel. The translational energy
distribution ͑Fig. 11͒ obtained for the double bond cleavage
also peaks at about 10 kcal/mol. This result indicates that the
double bond cleavage processes in the F2CCH2 photodisso-
ciation of 157 nm has an exit barrier, implying that the dis-
sociation of the CvC bond breaking may not be an event
occurring on the initially excited electronic state. It likely
occurs on the ground electronic potential energy surface.
D. F atom elimination: F2CCH2؉h␯˜FCCH2؉F
The C—F bond rupture channel was also observed from
the photolysis of F2CCH2 at 157 nm. The TOF spectra at
m/eϭ45, ⌰ϭ20°, 30° (FCCH2 radical͒ are shown in Fig.
12. The feature shown in these spectra is assigned to the F
atom elimination process. The translational energy distribu-
tion derived from the TOF spectra at m/eϭ45 is shown in
Fig. 13. Because of the dissociative ionization in the ionizer,
the TOF spectrum at m/eϭ19 (Fϩ) is very complicated
FIG. 14. TOF spectrum at m/eϭ19 (Fϩ), ⌰ϭ20°. The spectrum was taken
by accumulating signals over 50 k laser shots. The open circles are the
experimental data points, while the solid lines are the modeled fits to the
spectrum using CMLAB2 program. Curve 1 is the contribution from the pri-
mary HF product ͑HF elimination͒ cracking to Fϩ in the ionizer, curve 2 is
from the primary F product F atom elimination͒, curve 3 is from the primary
FCCH product ͑HF elimination͒ cracking to Fϩ, curve 4 is resulted from the
primary CF2 product cracking to Fϩ, curve 5 is due to the primary F2CC:
product cracking to Fϩ, and curve 6 is the overall fit.
FIG. 12. TOF spectra at m/eϭ45 (FCCHϩ2 ), ⌰ϭ20° ͑upper trace͒ and
⌰ϭ30° ͑lower trace͒. The spectrum was taken by accumulating signals over
100 k laser shots. The open circles are the experimental data points, while
the solid lines are the fits to the spectra using the translational energy dis-
tribution, shown in Fig. 13, for the F atom elimination channel.
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:
130.88.90.140 On: Tue, 09 Dec 2014 17:49:40

10844
J. Chem. Phys., Vol. 109, No. 24, 22 December 1998
Lin et al.
FIG. 15. TOF spectrum at m/eϭ20 (HFϩ), ⌰ϭ20°. The open circles rep-
resent the experimental data points, while the solid line is the fit to the
spectrum using the translational energy distribution shown in Fig. 16.
FIG. 17. TOF spectrum at m/eϭ44 (HCCFϩ), ⌰ϭ15°. The spectrum was
taken by accumulating signals over 500 k laser shorts. The open circles
represent the experimental data points, while the solid lines are the modeled
fits to the spectrum. Curve 1 is the contribution from the FCCH product ͑HF
elimination͒ calculated using the translational energy distribution shown in
Fig. 16. Curve 2 is from the F2CCH product ͑H atom elimination͒ cracking
to the FCCHϩ ion in the ionizer. Curve 3 is the overall fit to the spectrum.
since other primary photodissociation products which con-
tain F atom would also eventually contribute to the TOF
spectrum at m/eϭ19 (Fϩ) ͑see Fig. 14͒. The fitting to the
TOF spectrum in Fig. 14 shows that there is some contribu-
tion from the F atom elimination channel, or at least some
feature in the spectrum is consistent with the F atom elimi-
nation channel. The contribution from the F atom elimination
channel is calculated using the translational distribution de-
rived from the TOF spectrum at m/eϭ45 (FCCH2 frag-
ment͒. Other channels also exist in the TOF spectrum at
m/eϭ19. The TOF spectrum in Fig. 14 is modeled using a
total of five different contributions ͑see Fig. 14 caption͒ in-
cluding F atom elimination. The fitting to the TOF spectrum
of mass 19 only serves the purpose to show that it is possible
to model this complicated TOF spectrum using the known
existing channels, including the HF elimination channel
which will be described in the next paragraph. The transla-
tional energy distribution for the F atom elimination derived
from the TOF spectra at mass 45, shown in Fig. 13, is also
very similar to those of other dissociation channels. The dis-
tribution is peaked at approximately 13 kcal/mol, indicating
a possible dissociation mechanism from the ground potential
energy surface.
E. Molecular HF elimination: F2CCH2؉h␯˜FCCH؉HF
Molecular HF elimination was also detected from the
photolysis of 1,1-difluoroethylene at 157 nm. Most of the
previous studies on photodissociation of fluorinated ethyl-
enes involve detection of HF products using the IR emission
technique. Nascent vibration state distributions of HF have
been mapped out using this method. In this study, the TOF
spectra of the photodissociation products ͑FCCH and HF͒
were measured. Figure 15 shows the TOF spectrum at m/e
ϭ20 (HFϩ), ⌰ϭ30°. The feature shown in this spectrum is
attributed to the HF elimination channel. The translational
energy distribution ͑see Fig. 16͒ for the HF elimination can
then be derived from the TOF spectrum in Fig. 15. Figure 17
shows the TOF spectrum at m/eϭ44, ⌰ϭ15°. The fast peak
shown in the spectrum can clearly be modeled as the
momentum-matched partner to the HF fragment, indicating
that the fast peak in Fig. 17 is from the FCCH photofrag-
ment, while the slower peak is mainly due to the dissociative
FIG. 18. TOF spectrum at m/eϭ43 (CCFϩ), ⌰ϭ15°. The spectrum was
taken by accumulating signals over 250 k laser shorts. The open circles are
the experimental data points, while the solid lines are the fits to the spec-
trum. Curve 1 is the contribution from the primary FCCH ͑mass 44͒ product
͑HF elimination͒ cracking to FCCϩ in the ionizer, curve 2 is resulted from
the F2CC product (H2 elimination͒ cracking to FCCϩ, curve 3 is from the
F2CCH product H atom elimination͒, and curve 4 is the overall fit.
FIG. 16. Translational energy distribution for the HF atom elimination
channel from photodissociation of 1,1-difluoroethylene at 157 nm excita-
tion.
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:
130.88.90.140 On: Tue, 09 Dec 2014 17:49:40

J. Chem. Phys., Vol. 109, No. 24, 22 December 1998
Lin et al.
10845
TABLE II. Energy partition of different dissociation channels of 1,1-
difluoroethylene at 157 nm excitationa
Total
Etran
/Eint
ratio
Dissociation channel
͑157 nm excitation͒
available Averaged Averaged
energy Etot
Etrans
Eint
44.1
55.9
65.1
66.0
136.3
11.0
19.6
16.7
18.5
25.4
33.1
36.3
48.4
47.5
110.9
25/75
35/65
26/74
28/72
19/81
F2CvCH2→F2C:ϩ:CH2
F2CvCH2→FCvCH2ϩF
F2CvCH2→F2CvCHϩH
F2CvCH2→F2CvC:ϩH2
F2CvCH2→FCwCHϩHF
aThe averaged translational energies (Etrans) are experimentally measured
values, while the averaged internal energy is calculated by subtracting the
Etrans from the total available energy (EintϭEtotϪEtrans). All energies are in
kcal/mol.
FIG. 19. TOF spectrum at m/eϭ31 (CFϩ), ⌰ϭ25°. The spectrum was
taken by accumulating signals over 100 k laser shorts. The open circles
represent the experimental data points, while the solid lines are the fits to the
spectrum. Curve 1 is the contribution from the primary FCCH product ͑HF
elimination͒ cracking to CFϩ in the ionizer, curve 2 is resulted from the CF2
product ͑double bond rupture͒ dissociatively ionizing to CFϩ, curve 3 is
from the F2CCH product ͑H atom elimination͒, and curve 4 is the overall fit.
the estimation, the detection efficiencies of different ob-
served ions are assumed to be the same. There are also other
sources of errors in the fittings of some TOF spectra since
the correlations between the contributions from different
channels are rather significant. The anisotropy of the disso-
ciation processes will also likely increase the errors in calcu-
lating the branching ratios. In this experiment, however, this
effect is considered to be small since all dissociation pro-
cesses most likely occur from the ground potential energy
surface through internal conversion, which will wash out
most of the anisotropy in the dissociation processes. Due to
the above sources of error in estimation, the branching ratios
obtained here should be regarded as estimated values rather
than accurate measurement results. Nevertheless, these esti-
mated branching ratios should provide a good base for a
rough picture of the important dissociation channels. From
Table I, it is clear that the major dissociation channels are the
H atom elimination and the HF elimination channels. The H2
elimination and the double bond breaking channels are also
rather significant. The contribution from the F atom elimina-
tion channels is, however, minor.
ionization of the F2CCH product from the H atom elimina-
tion channel. TOF spectra at mass 43 ͑Fig. 18͒ and mass 31
͑Fig. 19͒ were also measured. The different features in these
spectra can all be attributed to dissociative ionization in the
electron impact ionizer from different primary dissociation
channels ͑see figure captions in Figs. 18 and 19͒. The trans-
lational energy release for the HF elimination channel also
has a peaked distribution, similar to those of other dissocia-
tive channels. The peak of the translational energy distribu-
tion is at about 20 kcal/mol. Therefore it is clear that all
dissociation processes of 1,1-difluoroethylene at 157 nm ex-
citation likely occur through the ground state potential, not
through direct dissociation on the initially excited electronic
state potential.
IV. DISCUSSION
A. Branching ratio
B. Energy partition
Branching ratios for photodissociation processes can
give us very important information on the dynamics of pho-
todissociation in general. In this work, TOF spectra for all
primary product masses and cracking masses from the pho-
todissociation of 1,1-difluoroethylene were carefully mea-
sured. Laser power for each TOF spectrum was also rather
well controlled and measured. By accumulating the contribu-
tions of the different fragment ions to each channel, the
branching ratios of each individual channel were estimated.
The estimated branching ratios are listed in Table I. During
From the translational energy distributions of all differ-
ent channels, all averaged translational energy releases can
be determined. Assuming p(E) is the translational energy
distribution of a dissociation process, the averaged energy
Eave can be calculated using the following formula:
͐
Ep E͒dE
͑
E
aveϭ
.
͐
p E͒dE
͑
Table II shows the averaged kinetic and internal energy re-
lease, and their ratios. From Table II, it is clear that the
product internal energy partition in photodissociation of 1,1-
difluoroethylene at 157 nm excitation are always signifi-
cantly larger than the translational energy release, with the
internal energy release ranging from 19% to 35% of the total
available energy.
TABLE I. Branching ratios for different dissociation channels of 1,1-
difluoroethylene at 157 nm excitation.
Dissociation channel
Branching ratio͑%͒
F2CvCH2ϩh␯→F2C:ϩ:CH2
F2CvCH2ϩh␯→FCvCH2ϩF
F2CvCH2ϩh␯→F2CvCHϩH
F2CvCH2ϩh␯→F2CvC:ϩH2
F2CvCH2ϩh␯→FCwCHϩHF
10
2
25
11
52
V. CONCLUSION
Photodissociation dynamics of 1,1-difluoroethylene at
157 nm excitation have been studied using photofragment
translational spectroscopy. Five dissociation channels have
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:
130.88.90.140 On: Tue, 09 Dec 2014 17:49:40

10846
J. Chem. Phys., Vol. 109, No. 24, 22 December 1998
Lin et al.
been experimentally observed. The main photodissociation
pathways are the HF elimination process and the H atom
elimination process. The H2 elimination and the double bond
breaking channels are also significant, while F atom elimina-
tion is a minor process in the photodissociation. The kinetic
energy releases for the five dissociation channels observed in
this work all show distributions peaked at energies higher
than zero, indicating that the photodissociation processes for
the difluoroethylene are not direct dissociation processes
from the initially excited electronic state. The excited 1,1-
difluoroethylene molecules most likely go through some type
of internal conversion from the initially excited electronic
state to the ground electronic potential energy surface, then
dissociate into different dissociation products. Experimental
results show that all dissociation channels experience some
exit barriers. Looking at the averaged kinetic energy releases
for different dissociation channels, it is clear that the internal
energy partitions are all significantly larger than the transla-
tional energy partitions.
radicals and molecules provided in M. R. Zachariah, P. R. Westmoreland,
D. R. Burgess, Jr., W. Tsang, and C. F. Melius, J. Phys. Chem. 100, 8737
͑1996͒, and references therein.
6 E. Zimir and R. D. Levine, Chem. Phys. 52, 253 ͑1980͒.
7 M. J. Berry, in Molecular Energy Transfer, edited by R. D. Levine and J.
Jortner ͑Wiley, New York, 1976͒, p. 114.
8 B. E. Holmes and D. W. Setser, in Physical Chemistry of Fast Reactions,
edited by I. W. M. Smith ͑Plenum, New York, 1980͒, Vol. 2, p. 83.
9 E. Arunan, S. J. Wategaonkar, and D. W. Setser, J. Phys. Chem. 95, 1539
͑1991͒.
10 D. J. Donaldson, D. G. Watson, and J. J. Sloan, J. Chem. Phys. 82, 1873
͑1985͒.
11 D. J. Donaldson and J. J. Sloan, J. Chem. Phys. 82, 1873 ͑1985͒.
12 D. J. Donaldson and S. R. Leone, Chem. Phys. Lett. 132, 240 ͑1986͒.
13 T. R. Fletcher and S. R. Leone, J. Chem. Phys. 88, 4720 ͑1988͒.
14 G. X. He, Y. A. Yang, Y. Huang, and R. J. Gordon, J. Phys. Chem. 97,
2186 ͑1993͒.
15 Y. Huang, Y. A. Yang, G. X. He, and R. J. Gordon, J. Chem. Phys. 99,
2752 ͑1993͒.
16 Y. Huang, Y. A. Yang, G. He, S. Hashimoto, and R. J. Gordon, J. Chem.
Phys. 103, 5476 ͑1995͒.
17 G. E. Hall, J. T. Muckerman, J. M. Presses, R. E. Weston, Jr., G. W.
Flynn, and A. Persky, J. Chem. Phys. 101, 3679 ͑1994͒.
18 M. Umemoto, K. Seki, H. Shinohara, U. Nagashima, N. Nishi, M. Ki-
noshita, and R. Shimida, J. Chem. Phys. 83, 1657 ͑1985͒.
19 K. Sato, S. Tsunashima, T. Takayanagi, G. Fijisawa, and A. Yokoyama,
Chem. Phys. Lett. 242, 401 ͑1995͒.
ACKNOWLEDGMENTS
This work is supported by the Academia Sinica and the
National Research Council of ROC.
20 B. A. Balko, J. S. Zhang, and Y. T. Lee, J. Phys. Chem. A 101, 6611
͑1997͒.
21 J. Lin, D. W. Huang, S. Harich, Y. T. Lee, and X. Yang, Rev. Sci. In-
strum. 69, 1642 ͑1998͒.
1 X. Yang, J. Lin and Y. T. Lee, Lambda Physik Highlights 50, 4 ͑1994͒.
2 Y. Wen, M. Dulligan, and C. Wittig, J. Chem. Phys. 101, 5665 ͑1994͒.
3 G. Belanger and C. Sandorfy, J. Chem. Phys. 55, 2055 ͑1971͒.
4 G. Belanger and C. Sandorfy, Chem. Phys. Lett. 3, 661 ͑1969͒.
5 All bond energies are estimated from the heat of formation of related
22 X. Zhao, Ph.D. thesis, University of California, Berkeley, 1990; J. D.
Myers, Ph.D. thesis, University of California, Berkeley, 1992.
23 A. H. H. Chang, A. M. Mebel, X. M. Yang, S. H. Lin, and Y. T. Lee, J.
Chem. Phys. 109, 2748 ͑1998͒.
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:
130.88.90.140 On: Tue, 09 Dec 2014 17:49:40