Temperature dependence of electron attachment to CH2ClBr: Competition between Cl- and Br- formation

Article abstract of DOI:10.1016/S0009-2614(03)00953-9

Dissociative electron attachment (DEA) to CH₂ClBr has been studied in a crossed electron/molecular beams experiment in the electron energy range between 0 and 2 eV and in the gas temperature range between 328 and 449 K. In this electron energy range we find the two negative fragment ions Cl^- and Br^- which are both formed from a prominent low energy resonant feature (≈0 eV). While the Br^- ion yield (recorded at ≈0 eV) considerably increases with the temperature, that of Cl^- slightly decreases resulting in an overall increase of ion formation at low electron energy. The temperature dependence is interpreted by the relative disposition of the involved potential energy surfaces. For the exothermic Br^- channel an activation energy of 107 ± 10 meV for the DEA reaction is obtained. Using absolute rates from swarm data the total cross section for the DEA to CH₂ClBr has been calibrated to absolute values leading to 5 × 10^-21m² at 0.5 eV.

Full text of DOI:10.1016/S0009-2614(03)00953-9

Chemical Physics Letters 375 (2003) 660–665
www.elsevier.com/locate/cplett
Temperature dependence of electron attachment to CH2ClBr:
competition between Clꢀ and Brꢀ formation
a,
a
b
S. Matejcik *, I. Ipolyi , E. Illenberger
a
Department of Plasmaphysics, Comenius University, Mlynska dolina F2, 84248 Bratislava, Slovak Republic
Institut f€ur Chemie – Physikalische und Theoretische Chemie, Freie Universit€at Berlin, Takustrasse3, D-14195 Berlin, Germany
b
Received 14 April 2003; in final form 3 June 2003
Published online
Abstract
Dissociative electron attachment (DEA) to CH2ClBr has been studied in a crossed electron/molecular beams ex-
periment in the electron energyrange between 0 and 2 eV and in the gas temperature range between 328 and 449 K. In
this electron energyrange we find the two negative fragment ions Cl ꢀ and Brꢀ which are both formed from a prominent
low energyresonant feature ( ꢁ0 eV). While the Brꢀ ion yield (recorded at ꢁ0 eV) considerablyincreases with the
temperature, that of Clꢀ slightlydecreases resulting in an overall increase of ion formation at low electron energy. The
temperature dependence is interpreted bythe relative disposition of the involved potential energysurfaces. For
the exothermic Brꢀ channel an activation energyof 107 ꢂ 10 meV for the DEA reaction is obtained. Using absolute
rates from swarm data the total cross section for the DEA to CH2ClBr has been calibrated to absolute values leading to
5 ꢃ 10ꢀ21 m2 at 0.5 eV.
Ó 2003 Elsevier Science B.V. All rights reserved.
1. Introduction
the kinetic energyof the electrons and the gas
temperature (i.e., rotational and vibrational exci-
tation of the target molecules). While the electron
energydependence is explained bythe resonant
nature of the attachment process and the associ-
ated Franck–Condon transition (with its limita-
tions in the verylow energyregion), the increase of
the DEA cross section with temperature (as ob-
served for the systems CF3Cl, CF2Cl2 and CHCl3
[3–7]) was attributed to the existence of an energy
barrier for DEA. On the other hand, there also
exist exothermic DEA reactions (CCl4, CHBr3)
where the cross section is independent on the gas
temperature as reported from beam experiments
[4,5] and also swarm experiments [8–10].
It is well known that the halo-methanes (with
the exception of some fluorocarbons) undergo ef-
fective dissociative electron attachment (DEA). In
the past, electron capture bythese molecules has
extensivelybeen studied using electron swarm and
crossed electron-molecular beams techniques [1,2].
Previous studies on compounds with one single
exothermic DEA channel [3–5] have demonstrated
that the cross sections can stronglydepend on both
* Corresponding author. Fax: +421-2-65429980.
E-mail address: matejcik@fmph.uniba.sk (S. Matejcik).
0009-2614/03/$ - see front matter Ó 2003 Elsevier Science B.V. All rights reserved.
doi:10.1016/S0009-2614(03)00953-9

S. Matejcik et al. / Chemical Physics Letters 375 (2003) 660–665
661
2. Experimental
The studyof temperature effects becomes par-
ticularlyinteresting in cases where competitive
exothermic reaction channels are available. Spanel
et al. [11] performed a coordinated swarm/crossed
The present experiments were performed on the
newlyestablished crossed electron/molecular beam
apparatus at Bratislava. The experimental setup
has previouslybeen described in detail [5] and only
a brief description will be given here. The electron
beam is formed bymeans of a trochoidal electron
monochromator (TEM). In the course of the
present experiments the instrument was operated
at an electron energyresolution of about 60 meV.
Calibration of the electron energyscale and
estimation of the electron energyresolution was
established using the well known electron attach-
ment process SFꢀ6 /SF6.
The molecular beam was produced in an effu-
sive molecular beam source (EMBS) which is
temperature controlled. The beam is formed by
effusing the gas through a channel (0.5 mm di-
ameter and 4 mm long) and an external aperture.
In the case of CH2BrCl the vapour is introduced
into the EMBS via a precision leak valve. The
vapour pressure inside the EMBS (typically 1 Pa)
is measured byan absolute pressure gauge. For
this molecular beam source we know the variation
of the gas densityin the beam with the gas tem-
perature [5]
beams studyon the compound CCl Br with re-
3
spect to the temperature effect for the total DEA
cross section but also that for the two competitive
channels Clꢀ and Brꢀ with a verygood agreement
between the beam and swarm experiments. A
further swarm studyof DEA to the chloro-bromo
methanes CHCl2Br, CHClBr2 and CCl2Br2 re-
vealed interesting effects concerning the tempera-
ture dependence of the rate coefficient and also the
product ion distribution [12].
In this contribution we studyDEA to the
chloro-bromo methane CH2ClBr in the electron
energyrange from about 0 to 2 eV and in the
gas temperature range from 328 to 449 K. We
have chosen CH2ClBr since the relevant DEA
channels
e ð0 eVÞ þ CH2ClBr ! Clꢀ þ CH2Br
þ 0:18 eV
e ð0 eVÞ þ CH2ClBr ! Brꢀ þ CH2Cl
þ 0:41 eV
ð1Þ
ð2Þ
n ꢄ Tꢀ0:5
;
ð3Þ
are exothermic by0.18 and 0.41 eV, respectively.
These values are obtained from the bond dis-
sociation energies D(Cl–CH2Br) ¼ 3.43 eV and
D(Br–CH2Cl) ¼ 2.95 eV [13], and the well known
electron affinities EA(Cl) ¼ 3.61 eV and EA(Br) ¼
3.36 eV [14].
In the literature we find an electron attachment
studyto CH 2ClBr using a swarm technique by
Sunagawa et al. [15], however, without mass
spectrometric identification of the products. The
overall rate coefficient for DEA was measured as
a function of the mean electron energyapplying
the microwave pulsed radiolysis method. The
thermal attachment rate coefficient was estimated
as 7.1(ꢂ 0.2) ꢃ 10ꢀ9 cm3 sꢀ1. Using a deconvolu-
tion procedure the total value for the DEA cross
section was obtained as 4.4 ꢃ 10ꢀ20 m2 at 0.1 eV
[14]. To our knowledge no crossed beams studies
for this molecule are available in the literature
so far.
and the temperature dependencies of the cross
section have to be corrected accordingly. The
profile of the molecular beam does not change
with the temperature in the present experiment.
Negative ions formed within the intersection
between the electron beam and the molecu-
lar beam are extracted bya weak electric field
(1 V mꢀ1) and focused onto a quadrupole mass
spectrometer (QMS). The mass analysed negative
ion signal is then detected as a function of the
electron energyat different target gas tempera-
tures. A spatial discrimination of the ions exists
in present experiment, i.e., onlyions from very
small spatial angle are extracted into mass spec-
trometer. This discrimination does not depend on
the mass of the molecule and on the kinetic en-
ergyof the molecule. For this reason we do not
expect discrimination effect concerning ration
Clꢀ/Brꢀ.

662
S. Matejcik et al. / Chemical Physics Letters 375 (2003) 660–665
3. Results and discussions
the electron energydistribution function: the
measured ion yield is a convolution of the electron
energydistribution function f ðE; UÞ and the cross
section rðEÞ
In the energyrange from about 0 to 2 eV the
fragment ions Brꢀ and Clꢀ are the onlynegative
ions observable from CH2ClBr. Fig. 1 presents the
ion yields for both negative ions recorded at a gas
temperature of 328 and 449 K. Note the loga-
rithmic scale for the ion intensity. The intensities
are in arbitraryunits but within the same scale for
both fragment ions and temperatures. From Fig. 1
it is immediatelyobvious that at verylow energies
Z
1
IðUÞ ¼
rðEÞf ðE; UÞdE;
ð4Þ
0
where IðUÞ is the ion yield at the acceleration
voltage U, rðEÞ is the cross section for the DEA
reaction and f ðE; UÞ is the distribution function
for the electron energywhich is directlyconnected
to the acceleration voltage U.
ꢀ
(close to 0 eV) the intensityof the Br signal in-
creases and that of the Clꢀ signal decreases upon
heating. The total ion yield, on the other hand,
increases with temperature. Both ion yields peak at
energies close to 0 eV. The shape of the Brꢀ curve
exhibits an additional shoulder near 0.1 eV. The
thermal cracking studied bypositive ion mass
Fig. 2 compares the energydependence of the
total DEA cross section obtained from the swarm
technique with the present results. The absolute
values of the total DEA cross section was obtained
from the pulse radiolysis swarm experiment [15] by
measuring the rate constant at different electron
temperatures and applying a deconvolution pro-
cedure. The cross section curve from the present
experiment was obtained bytaking the sum of Cl ꢀ
and Brꢀ formation at a gas temperature of 328 K
(which approximatelycorresponds to that in the
swarm experiment). The present cross section
curve was then calibrated to the absolute value at
E ¼ 0:1 eV. Fig. 2 indicates that the beam experi-
ment exhibits a weaker energydependence result-
ing in a lower value at energies below the
calibration point and vice versa. Apart from the
spectrometry(reported in [5] for molecules CCl
and CHCl3) was not observed for CH2ClBr in the
measured temperature range.
4
It should be noted that near threshold (0 eV),
the shape of the ion yields are strongly affected by
Fig. 2. The overall cross sections for DEA to CH2ClBr by
Sunagawa and Shimamori [15] and present study. The error
bars of the present data are equivalent to the size of the data
points at low electron energies and to the statistical scatter of
the data at higher electron energies.
Fig. 1. (a) Brꢀ and (b) Clꢀ ion yields for DEA to CH2ClBr
measured at the two different temperatures 328 and 449 K.

S. Matejcik et al. / Chemical Physics Letters 375 (2003) 660–665
663
deviation at the highest electron energy, the
agreement is reasonable. At that point it should be
noted that in comparing cross sections between
beam and swarm experiments one has to be aware
of the totallydifferent conditions under which
both experiments are carried out. While the swarm
experiment was performed under a gas pressure of
about 100 mbar (corresponding to a collision fre-
quencyof the order of 10 9 per single molecule), the
present beam experiment operates under true sin-
gle collision conditions. Collisions between the
molecules give access to stabilization routes for the
temporarynegative ion. This can affect, e.g., its
autodetachment lifetime which itself depends on
the electron energy. Further problems may arise
from discrimination effects in the mass spectrom-
eter but also the deconvolution procedure, which
was used to obtain the cross section from the rate
coefficient.
Returning to Fig. 1 it can be seen that byin-
creasing the gas temperature from 328 to 449 K
the Brꢀ ion yield at the low energy peak (close to
0 eV) is increased byalmost three times from 4000
to 11 000 U. In contrast to that the Clꢀ intensity
decreases from 1570 to 600 U resulting in an
overall increase of the ion intensity. These tem-
perature dependences over the entire range are
explicitlyshown in Figs. 3a and b, note that the
experimental points are not corrected for the
temperature dependence of the gas density.
Fig. 3. The temperature dependencies of the (a) overall and Brꢀ
ion yield (b) Clꢀ ion yield at low electron energies. The signal is
not corrected on the decrease of the gas number densitywith
the temperature.
transitions from vibrationallyexcited states ex-
tend the Franck–Condon region thus allowing
transitions to the repulsive ionic potential energy
curve at lower electron energies and hence higher
dissociation probabilities. This results in the ap-
pearance of a threshold peak and an increase of
the cross section. This effect is the more pro-
nounced the stronger autodetachment competes
to dissociation in the transient anion. From a
thermodynamic point of view the energy of the
crossing point Rc can be viewed as the activation
energy for the dissociative attachment of thermal
electrons.
The temperature dependence of the DEA cross
section can be explained bythe relative disposition
of the potential energysurfaces involved in DEA.
For the present system CH2ClBr electron capture
generates the transitoryion in a repulsive state
along the C–Br and C–Cl axis. This is directly
related to the virtual molecular orbital (MO) in-
volved in low energyelectron capture having
rꢅ(C–Cl) and rꢅ(C–Br) antibonding character. We
ꢀ
maythus approximate DEA into Br and Clꢀ by
means of two dimensional potential energycurves.
The temperature dependence can then be described
bythe population of vibrational levels and the
competition between dissociation and autode-
tachment in the transient negative ion [8,9].
In the case where the potential curve of the
anion crosses that of the neutral at distances
larger than the equilibrium distance of the neutral
On the other hand, if the ionic potential energy
curve crosses left of the equilibrium distance, one
expects an decrease of the DEA cross section with
temperature. In this case, the negative ion is no
longer accessible from vibrationallyexcited states
and the ground state is depopulated upon heating.
This situation is analogous to that of the inverted
region in electron transfer reactions [16].

664
S. Matejcik et al. / Chemical Physics Letters 375 (2003) 660–665
In this spirit we interpret the present results by
means of DEA via the schematic potential energy
curves shown in Fig. 4. Both negative ion curves
indicate bound states although the existence of a
parent negative ion CH2ClBrꢀ has not been re-
ported so far. If bound states exists observation of
CH2ClBrꢀ is unlikelyunder collision free condi-
tions (lack of effective stabilization mechanisms)
but electron attachment to CH2ClBr clusters can
give access to the stabilized parent anion. The
shoulder near 0.1 eV on the Brꢀ yield can be in-
terpreted as a Franck–Condon transition near the
equilibrium distance and the 0 eV peak due to a
transition from a vibrational excited state near the
level of the crossing point between the two po-
tential curves (hot band transition). Note that the
intensityof the 0 eV peak can reach appreciable
values in spite of a moderate population of excited
vibrational levels. This is due to the reciprocal
energydependence of the attachment cross section
and also the autodetachment lifetime which
stronglyvaries with the electron energy.
Fig. 5. Arhenius plot for the Brꢀ channel to obtain the acti-
vation energyfor the DEA reaction.
energysurfaces and the application of the Franck–
Condon principle. This picture has its limitations
in the case of verylow electron energies.
We hence mention a further mechanism, which
can generate structures in low energyion yields. In
a collaborative research using the high-resolution
laser photoelectron attachment (LPA) technique
und the present beam technique [17] DEA to
CH2Br2 resembled a sharp peak at an energynear
72 MeV. This is close to the onset for excitation of
one quantum of the symmetric CBr2 stretch vi-
bration m3 ¼ 1, spectroscopicallylocated at 72.9(5)
MeV. Similarlyat the onset for the asymmetric
CBr2 stretch vibration m9 (at 81.0 ꢂ 2 meV) a weak
upward cusp was observed. The structures in the
DEA cross section of the CH2Br2 molecule at the
thresholds for vibrational excitation were conse-
quentlyinterpreted as a vibrational Feshbach
resonance (VFR) [5], i.e., coupling of the attach-
ment process with the inelastic scattering channels.
We can hence not exclude that the presentlyob-
served structure is related to a VFR.
Fig. 5 shows that the temperature dependence
of the Brꢀ yield in the lnðIÞ=T ꢀ1 representation
shows a reasonablylinear behaviour (Arhenius
plot). From the slope of the line we derive the
ꢀ
activation energyfor Br formation as Ea(Brꢀ) ¼
107 ꢂ 10 meV.
It should be mentioned that the above inter-
pretation is based on the Born–Oppenheimer ap-
proximation i.e., the use of localized potential
Acknowledgements
This work was supported bythe Volkswagen-
Stiftung (I/73593) (Germany), the Slovak Grant
agencyVEGA (No. 1/8313/01) and Austria-Slo-
vakia Contract No. 21s17.
Fig. 4. Schematic two dimensional potential energycurves for
the exothermic DEA channels in CH2ClBr (see the text).

S. Matejcik et al. / Chemical Physics Letters 375 (2003) 660–665
665
[9] D. Smith, N.G. Adams, E. Alge, J. Phys. B: At. Mol. Opt.
Phys. 17 (1984) 461.
References
[10] L.G. Christophorou, J.K. Olthoff, Adv. Atom. Mol. Opt.
Phys. 44 (1999) 155.
[1] L.G. Christophorou (Ed.), Electron–Molecules Interac-
tions and their Applications, Academic, Orlando, 1984.
[2] L.G. Christophorou, J.K. Olthoff, M.V.V.S. Rao, J. Phys.
Chem. Ref. Data 26 (1997) 1.
ꢀ
ꢀ
ꢂ
ꢀꢁ
€
[11] P. Spanel, D. Smith, S. Matejcık, A. Kindler, T.D. Mark,
Int. J. Mass. Spect. Ion. Proc. 167/168 (1997) 1.
[12] P. Spanel, D. Smith, Int. J. Mass Spect. 205 (2001) 243.
[13] S.-H. Lee, Y.-J. Jung, K.-H. Jung, Chem. Phys. 260 (2000)
143.
ꢀ
[3] A. Kiendler, S. Matejcık, J.D. Skalny, A. Stamatovic, T.D.
ꢀꢁ
Mark, J. Phys. B 29 (1996) 6217.
ꢁ
€
[4] S. Matejcik, G. Senn, P. Scheier, A. Kiendler, A. Stama-
[14] D.R. Lide (Ed.), CRC Handbook of Chemistryand
Physics, 78th edition, CRC Press, Boca Raton, New York,
1997–1998.
€
tovic, T.D. Mark, J. Phys. Chem. 107 (1997) 8955.
[5] S. Matejcık, V. Foltin, M. Stano, J.D. Skalny, Int. J. Mass
Spect. 223–224 (2003) 9.
[15] T. Sunagawa, H. Shimamori, J. Chem. Phys. 107 (1997)
7876.
[6] Hahndorf, L. Lehr, E. Illenberger, J. Manz, Chem. Phys.
Lett. 231 (1994) 460.
[16] R.A. Marcus, Angew. Chem. 105 (1993) 1161.
[17] A. Schramm, M.-W. Ruf, M. Stano, S. Matejcik, I.I.
Fabrikant, H. Hotop, J. Phys. B: At. Mol. Opt. Phys. 35
(2002) 1.
[7] I. Hahndorf, E. Illenberger, Int. J. Mass Spect. Ion Proc.
167/168 (1997) 87.
[8] L.M. Warman, M.C. Sauer, Int. J. Radiat. Phys. Chem. 3
(1971) 273.