Effect of chemical composition on the neutral reaction products produced during electron beam irradiation of carbon tetrachloride/water (ice) films

Article abstract of DOI:10.1039/b204016f

The neutral reaction products formed during electron beam irradiation of CCl₄/H₂O (ice) films were studied as a function of the film's initial CCl₄/H₂O ratio using a combination of reflection absorption IR spect

Full text of DOI:10.1039/b204016f

View Article Online / Journal Homepage / Table of Contents for this issue
Effect of chemical composition on the neutral reaction products
produced during electron beam irradiation of carbon tetrachloride/
water (ice) films
A. J. Wagner, C. Vecitis, G. M. Wolfe, C. C. Perry and D. Howard Fairbrother*
Department of Chemistry, The Johns Hopkins University, 3400 N. Charles Street, Baltimore,
MD 21218
Received 26th April 2002, Accepted 29th May 2002
First published as an Advance Article on the web 1st July 2002
The neutral reaction products formed during electron beam irradiation of carbon tetrachloride/water (ice) films
have been studied as a function of the film’s initial CCl4 : H2O ratio using a combination of reflection
absorption infrared spectroscopy, mass spectrometry and X-ray photoelectron spectroscopy. When the initial
CCl4 : H2O ratio was high, the dominant reaction products in the film were C2Cl4 and a partially chlorinated
carbonaceous film (CClx) formed as the result of carbon–carbon coupling reactions in the film. In these CCl4
rich films, chlorine is partitioned mainly into the gas phase while CO is the dominant carbon-containing gas
phase species. As the CCl4 : H2O ratio decreases, CO2 becomes an increasingly important reaction product at
the expense of species generated from carbon–carbon coupling reactions, while chlorine is increasingly
partitioned as HCl in the film, producing H3O+ and Clꢀ. The production of both H2 and O2 from electron
stimulated reactions associated with H2O are suppressed in CCl4/H2O films, although oxygen is more efficiently
quenched in the presence of CCl4 .
intermediates.20 A number of neutral gas phase products have
Introduction
also been detected during the interaction of <200 eV electrons
with amorphous D2O ice (D-ice) including atomic D and O, as
well as D2 and O2 .21–28 Analogous studies on the interaction
of low energy electrons (30–200 eV) with carbon tetrachloride
in solid argon at 12 K revealed the production of :CCl2 , CCl3 ,
ꢁCCl3+, CCl4+ and Cl3+ as well as C2Cl4 although the relative
yield of neutral products became more dominant at higher
incident electron energies.29,30
The electron stimulated reactions of CCl4 in aqueous solu-
tions at room temperature have been studied by Hoffman
and Choi.31 Results from these investigations revealed that
the initial electron transfer step to CCl4 involves both one
and two electron transfer reactions leading to the production
of trichloromethyl radical and dichlorocarbene species:
The interaction of electrons with surfaces and thin films is of
fundamental importance in many chemical and physical pro-
cesses including surface modification,1 oxidation,2 dielectric
breakdown,3 radiolysis4 and environmental remediation.5
Electron stimulated reactions are typically characterized by
non-thermal processes that include the formation of both neu-
tral and charged species. A dissociative electron attachment
(DEA) process involving electron transfer producing a tempor-
ary negative ion is generally believed to initiate reaction. Sub-
sequent decay of this negative ion produces a neutral and
stable anionic species.6 DEA has been studied extensively in
the gas phase as a means to elucidate information on the sym-
metry and electronic structure of excited states.7,8 DEA also
forms the basis for electron stimulated desorption which has
been used as a tool to determine structural characteristics from
submonolayer adsorbate coverages on metal surfaces.9
CCl4 þ eꢀ
! ꢁCCl3 þ ClꢀðaqÞðeꢀ
solvated electronÞ
ðaqÞ
ðaqÞ
ð1Þ
The chemical transformations that accompany the exposure
of water (ice) to reactive species such as electrons are of particu-
lar importance in interstellar chemistry/astrophysics where sur-
faces are continuously irradiated by planetary magnetospheric
electrons, cosmic rays and energetic ions from solar wind and
flares.10–14 The importance of electron stimulated reactions in
ice films has also been reported by Madey and Lu15–17 who dis-
covered enhancement factors of 102–104 in electron-induced
dissociation experiments for chlorofluorocarbons (CFCs)
adsorbed on top of ice films. This effect has been associated with
the efficient trapping of low-energy secondary electrons by
water to produce solvated electrons. These results suggest that
even in situations where the free electron density is low, elec-
tron-induced processes can play a significant role in the chem-
istry of organohalides adsorbed in or on top of ice films.
Previous studies have identified a number of charged species
including H+18 and Hꢀ(Dꢀ)19 during electron beam irradia-
tion of ice although hydroxyl radicals (ꢁOH), hydrogen atoms
(H) and hydrated electrons (eaqꢀ) are the dominant reaction
ꢁCCl3 þ eꢀ
! :CCl2 þ Clꢀ
ð2Þ
ðaqÞ
ðaqÞ
In oxygenated aqueous solutions, the trichloromethyl radical
reacts rapidly with dissolved oxygen:32
ꢁCCl3 þ O2 ! CCl3OOꢁ
ð3Þ
Decomposition of the peroxyl radical generates a phosgene
intermediate whose subsequent hydrolysis under ambient con-
ditions produces CO2 and HCl. In contrast, dichlorocarbene
can undergo a carbon–carbon coupling reaction to form tetra-
chloroethylene:33,34
:CCl2 þ :CCl2 ! C2Cl4
ð4Þ
or react with water to form carbon monoxide and hydrochloric
acid:35
:CCl2 þ H2O ! 2 HCl þ CO
ð5Þ
3806
Phys. Chem. Chem. Phys., 2002, 4, 3806–3813
This journal is # The Owner Societies 2002
DOI: 10.1039/b204016f

View Article Online
In this study, we focus on the role of the initial CCl4 : H2O
ratio in determining the product branching ratio for neutral
species produced in the film and those ejected into the gas
phase. In our initial studies, mass spectrometry in conjunction
with infrared spectroscopy identified CO2 , CO and HCl as the
neutral gas phase products while COCl2 , C2Cl4 and HCl were
produced as intermediates in the film.36 We have now focused
on the effect of chemical composition on the product partition-
ing during electron beam irradiation. In addition, experiments
on the evolution of CCl4 : H2O films during electron beam irra-
diation have been augmented by X-ray photoelectron spectro-
scopy (XPS). These results have relevance for environmental
remediation as well as atmospheric processes. In a more gen-
eral sense results from the present investigation are also impor-
tant in understanding the influence of co-adsorbates in electron
stimulated reactions in ice films.
Fig. 1 RAIR spectra of a 13CCl4/H2O film with a CCl4 : H2O film
ratio of 0.472. Fig. 1(a) shows the as-deposited film with the O–H
stretch observed between 3600–3000 cmꢀ1 and 13CCl4 Fermi doublet
at 780 and 758 cmꢀ1. Fig. 1(b) shows the difference spectra after 5
min of electron beam exposure. New peaks are observed at 2340,
2272, 1758, 885, 729 and 1950–1560 cmꢀ1. The insert shows the cali-
bration curve used to determine the film’s initial CCl4 : H2O ratio from
RAIRS data. The IR CCl4 : H2O ratio was determined from the area
under the curve between 3725–2919 cmꢀ1 for H2O and between 798–
742 cmꢀ1 for CCl4 . The XPS ratio was determined from the Cl(2p)
and O(1s) peak areas after correcting for relative elemental sensitivities
and stoichiometry.
Experimental
Experiments were carried out in a vacuum chamber equipped
with capabilities for reflection absorption infrared spectro-
scopy (RAIRS), X-ray photoelectron spectroscopy (XPS) as
well as mass spectrometry (MS) and have been described in
previous publications.36,37 CCl4/H2O films were condensed
onto a polycrystalline Au (99.99%, Accumet) mirror mounted
on a copper sample holder attached to a ceramic feedthrough
coupled to a UHV manipulator. The sample was cooled by
passing liquid nitrogen into a stainless steel tube connected
to the feedthrough. This arrangement allowed for a tempera-
ture of 103 K to be maintained during experiments.
determined by measuring the Cl(2p) and O(1s) regions follow-
ing IR spectra and prior to electron beam exposure. XP areas
were corrected for relative elemental sensitivity,42 and the
Cl(2p) region was divided by 4 to correct for the CCl4 stoichio-
metry. The insert to Fig. 1 illustrates that there is a linear rela-
tionship between the CCl4 : H2O ratio determined from IR and
XPS measurements for the range of film concentrations
employed in the present investigation. This calibration curve
RAIR spectra were recorded at a resolution of 4 cmꢀ1 using
a narrow band mercury–cadmium telluride (MCT) detector
(700–4000 cmꢀ1). Spectra were either referenced to the Au sur-
face at 103 K before CCl4/H2O dosing, or in the case of differ-
ence spectra, to the CCl4/H2O film before electron beam
exposure. XP spectra were recorded using Mg Ka X-ray radia-
tion (1253.6 eV) at 15 kV and 300 W with a 45ꢂ take-off angle.
Regional XPS scans were acquired using a pass energy 71.55
has been used to convert IR ratios into quantitative CCl4
:
eV and 0.1 eV step .
ꢀ1 38 XPS fitting was done with 100% gaus-
H2O ratios to provide a measure of the film’s initial chemical
composition.
sian peak shapes using commercial software.39
Carbon tetrachloride (12CCl4 Aldrich (99.9%): 13CCl4 Cam-
bridge Isotopes (99%)) and water (millipore, deionized) were
stored in separate glass vacuum bulbs attached to a gas mani-
fold. Both CCl4 and H2O were subject to several freeze–pump–
thaw cycles prior to use. The vapor for each compound was
then expanded into an isolated volume prior to mixing in the
gas manifold, where the gas mixture typically stabilized for
ꢃ10 min prior to dosing. The CCl4 : H2O ratio of the deposited
film was varied by adjusting the CCl4 pressure contained in the
isolated volume prior to mixing. During dosing, the gas purity
was monitored by mass spectrometry.
Electrons were generated with a Specs 15/40 low energy
flood gun using 4 mA emission current and a 10 eV extraction
voltage. A sample bias of +200 V was employed to increase the
kinetic energy of the incident electrons and the rate of chemical
reactions in the film.
Effect of electron beam irradiation on CCl4/H2O
(ice) films
(i) RAIRS results
Fig. 1(a) shows the IR spectrum of a 13CCl4/H2O (ice) film
deposited on a Au mirror at 103 K. The spectrum is dominated
by a broad feature between 3600 and 3000 cmꢀ1 associated
with the O–H stretching mode of adsorbed H2O and the
13CCl4 Fermi doublet comprising the n3 stretching mode at
Fig. 1(b) shows the difference spectrum for a 13CCl4/H2O film
with an initial 13CCl4 : H2O ratio of 0.472 following 5 min of
electron beam exposure. New peaks are visible at 2339, 2272,
1758 and 885 cmꢀ1 previously identified with the production
of 12CO2 , 13CO2 , 13COCl2 and 13C2Cl4 respectively.36 A new
peak was also observed at 729 cmꢀ1 that shifted to 751 cmꢀ1
when 12CCl4 films were employed. This peak was also observed
during electron beam irradiation experiments on pure 13CCl4
or 12CCl4 films and has been assigned to the production of
CClx species.43 New broad IR bands are also observed
between 1950 –1560 cmꢀ1 and 2200–2000 cmꢀ1 due to the anti-
symmetric bending mode and an overtone band respectively
associated with the hydroxonium ion (H3O+) in an amorphous
ice film.44,45 Negative absorbance peaks between 3600–3000
cmꢀ1 and 780–758 cmꢀ1 are observed due to the electron sti-
mulated consumption of both H2O and CCl4 . The intensity
of IR peaks associated with various product species was found
780 cmꢀ1 and the n1 + n4 combination mode at 758 cmꢀ1 40
.
The negative absorbance seen at 939 cmꢀ1 has been previously
identified as an optical interference effect caused by the thick-
ness of the CCl4/H2O films employed in the present investiga-
36,41
˚
tion, estimated to be >300 A.
It should be noted that in
the present investigation 13CCl4 was used in RAIR experi-
ments to avoid interference from 12CO2 background adsorp-
tion.
The insert in Fig. 1 shows a comparison of 13CCl4 : H2O
ratios determined from IR and XPS measurements for a num-
ber of 13CCl4/H2O films. IR ratios were calculated using the
integrated area between 3725–2919 cmꢀ1 for H2O and 798–
742 cmꢀ1 for 13CCl4 . Quantitative CCl4 : H2O XPS ratios were
Phys. Chem. Chem. Phys., 2002, 4, 3806–3813
3807

View Article Online
Fig. 2 Variations in the RAIR spectra for (i) 13CO2 at 2227 cmꢀ1; (ii)
H3O+ between 1950–1560 cmꢀ1 and 13COCl2 at 1758 cmꢀ1; (iii)
13C2Cl4 at 885 cmꢀ1; (iv) 13CClx at 729 cmꢀ1 as a function of the film’s
initial 13CCl4 : H2O ratio. Fig. 2(a)–(f) correspond to films of different
initial 13CCl4 : H2O ratios after 5 min of electron beam irradiation.
to scale with the thickness of the film, indicating that the reac-
tion products observed are the result of reactions within the
film rather than at the Au surface.
Fig. 2 shows the variation in the IR regions for the 13CO2 ,
H3O+, 13COCl2 , 13C2Cl4 and 13CClx products observed during
electron beam irradiation as a function of the films initial
13CCl4 : H2O ratio (shown as the numerical values in Fig.
2(i)). In the following discussion we shall refer to concentrated
films, where the initial 13CCl4 : H2O ratio is high (>ꢄ0.25),
and dilute films, where the initial 13CCl4 : H2O ratio is low
(<ꢄ0.1). In these experiments, the H2O content of the film,
measured by the integrated area of the O–H stretching region,
was maintained at a constant level, while the 13CCl4 concentra-
tion, measured by the area of the n3 stretching mode (780
cmꢀ1) and n1 + n4 combination mode (758 cmꢀ1), was varied.
Each spectrum was recorded as a difference spectrum after 5
min of electron beam irradiation ratioed against the initial
film, as shown in Fig. 1(b).
Fig. 3 Variation in the distribution of species within the film as a
function of the initial film composition. Fig. 3 shows the variations
in (a) 13C2Cl4/D13CCl4 (S) and 13CClx/D13CCl4 (N); (b) 13CO2/
D13CCl4 (S) and (c) H3O+/D13CCl4 (S). See text for details.
partitioning was confirmed by separate results obtained in
13CCl4/H2O films of different H2O thicknesses (measured by
the integrated area of the O–H stretching region). Thus CO2
was observed as the dominant reaction product in films where
the initial CCl4 : H2O ratio was low irrespective of the total
film thickness; similarly C2Cl4 was only detected in films con-
centrated with CCl4 .
Fig. 2(i) shows that the 13CO2 peak located at 2272 cmꢀ1
increases as the concentration of 13CCl4 in the H2O (ice) film
decreases. A similar trend is also observed in Fig. 2(ii) where
the broad H3O+ band is generally more prominent in dilute
films. In contrast, however, the 13C2Cl4 peak at 885 cmꢀ1
(Fig. 2(iii)), and the 13CClx peak at 729 cmꢀ1 (Fig. 2(iv))
decrease in intensity with decreasing 13CCl4 concentration in
the H2O(ice) film.
(ii) Mass spectrometry results
Fig. 4(a)–(e) shows the variation in the gas phase species pro-
duced during electron beam irradiation of 13CCl4/H2O(ice)
films as a function of the initial 13CCl4 : H2O ratio. Each spec-
trum was recorded after 5 min of electron beam irradiation.
All spectra in Fig. 4 exhibit peaks corresponding to 12CO
(m/q ¼ 28) and 12CO2 (m/q ¼ 44) from residual back-
ground gases in the vacuum environment.46 Consistent with
previous investigations, carbon dioxide (13CO2 ; m/q ¼ 45),
carbon monoxide (13CO; m/q ¼ 29) and hydrogen chloride
(H35/37Cl) as well as molecular hydrogen and oxygen are
observed as the neutral gas phase products of electron stimu-
lated reactions in 13CCl4/H2O (ice) films.36 Fig. 4(f) shows
the mass spectrum observed during electron beam irradiation
of a 13CCl4/D2O/H2O mixture. In addition to the production
of 13CO, 13CO2 , O2 and H2 , mass spectral features are also
observed at m/q ¼ 4 (D2), m/q ¼ 3 (HD), m/q ¼ 20 (D2O),
m/q ¼ 19 (DHO) and m/q ¼ 37, 39 (D35/37Cl).
Fig. 3 illustrates the partitioning of 13CCl4 into (i) 13C2Cl4 ,
13CClx , (ii) 13CO2 , and (iii) HCl/H3O+ as a function of the
film’s initial 13CCl4 : H2O ratio. The results shown in Fig. 3
were obtained after 5 min of electron beam irradiation,
compiled from a data set that includes the spectra shown in
Fig. 2. The partitioning of 13CCl4 into 13CO2 , HCl/H3O+,
13C2Cl4 , and 13CClx was determined by dividing the integrated
IR area of the product species by the corresponding change in
the CCl4 IR area (D13CCl4) during electron beam exposure.
For example, Fig. 3(a) shows that the fraction of 13CCl4 con-
verted into 13C2Cl4 and 13CClx increases with increasing
13CCl4 concentration in the film. In contrast, the probability
of 13CCl4 conversion into 13CO2 (Fig. 3(b)), or HCl/H3O+
(Fig. 3(c)) in the ice film decreases with increasing 13CCl4 con-
centration in the film. It should be noted that no attempt was
made to analyze the variation in phosgene (13COCl2) area in
Fig. 2(ii) due to the significant background introduced by the
antisymmetric bending mode of the hydroxonium ions. The
validity of the approach used in Fig. 3 for determining reaction
The variation in the gas phase 13CO : 13CO2 and O2 : H2
ratios as well as the O2 and H2 signal levels during electron
beam irradiation are shown in Fig. 5 as a function of the film’s
initial 13CCl4 : H2O ratio. Unless noted, the results shown in
Fig. 5 were obtained after 5 min of electron beam irradiation,
and were compiled from a data set that includes the spectra
shown in Fig. 4. The 13CO : 13CO2 ratio47 is approximately
unity in dilute films but rises steadily with increasing 13CCl4
3808
Phys. Chem. Chem. Phys., 2002, 4, 3806–3813

View Article Online
Fig. 4 The mass spectra (0–60 amu) of neutral volatile species as a
function of the film’s initial 13CCl4 : H2O ratio, Fig. 4(a)–(e). For
m/q < 10, the QMS intensity was multiplied by 0.5. Fig. 4(f) shows
the volatile species produced during electron beam irradiation of a
13CCl4/D2O/H2O film. The spectra were recorded after 5 min of elec-
tron beam irradiation and have been normalized to the 12CO peak at
m/q ¼ 28 to compensate for variations in the mass spectrum intensity
between experiments.
Fig. 5 Variation in the distribution of gas phase species as a function
of the film’s initial 13CCl4 : H2O ratio. Fig. 5 shows the variations in (a)
13CO : 13CO2 ratio after 5 (S) and 10 min (X) of electron beam irradia-
tion, (b) O2 : 12CO (S) and H2 : 12CO (H) ratios and (c) O2 : H2 ratio, as
a function of the film’s initial CCl4 : H2O ratio. All intensities have
been corrected for the mass spectrometer’s intrinsic sensitivity to differ-
ent gases. The H2 signal has also been corrected for the small back-
ground signal at m/q ¼ 2 observed prior to electron beam irradiation.
concentration in the ice film. While the 13CO : 13CO2 ratio is
strongly influenced by the film’s initial 13CCl4 : H2O ratio
for a particular film, the 13CO : 13CO2 ratio is relatively insen-
sitive to the electron beam irradiation time (Fig. 5(a)). Figs. 4
and 5(b) also show that although the initial water content in
each of the 13CCl4 : H2O(ice) films is virtually constant, both
O2 and H2 production (normalized to the background 12CO
signal) during electron beam irradiation decrease with increas-
ing 13CCl4 concentration in the film. In dilute films, the O2 : H2
ratio is similar to that obtained during electron beam irradia-
tion of pure water (ice) films. As the 13CCl4 concentration in
the film is increased, however, the O2 : H2 ratio decreases shar-
ply, reaching a limiting value of ꢄ0.035 for films with initial
13CCl4 : H2O ratios in excess of ꢄ0.15.
(iii) XPS results
Fig. 6 shows the evolution of the C(1s) and Cl(2p) XPS regions
for a film with an initial 13CCl4 : H2O ratio of 0.173 during
electron beam irradiation. The initial film, composed of mole-
cular CCl4 and H2O shows a C(1s) region dominated by a sin-
gle peak centered at 290.2 eV due to CCl4 while the Cl(2p)
region is well fit by a single Cl(2p3/2/2p1/2) doublet (Cl(2p3/
2) at 200.0 eV) characteristic of a CCl species.42 Upon electron
beam irradiation, the C(1s) spectral envelope broadens
towards lower binding energy, culminating in the growth of
a new peak, centered at 285.3 eV, after approximately 4 min
of electron beam irradiation (Fig. 6). On the basis of separate
experiments where molecular CCl4 was deposited on a residual
carbonaceous film present on the Au substrate, this new
Fig. 6 Variation in the C(1s) and Cl(2p) XPS regions for a film with
an initial CCl4 : H2O ratio of 0.173 (determined from Fig. 1 – insert) as
a function of electron beam exposure time. It should be noted that no
peak fitting has been attempted in the C(1s) region. The composite fit
in the Cl(2p) region is comprised of an initial Cl(2p3/2/2p1/2) doublet
(2p3/2 peak at 200.0 eV) and the growth of a second Cl doublet (2p3/2
peak at 197.5 eV).
feature centered at 285.3 eV is identified with the production
of a largely dechlorinated carbonaceous species. The Cl(2p)
region also broadens to lower binding energies upon exposure
to electron beam irradiation (Fig. 6). The resulting envelope
Phys. Chem. Chem. Phys., 2002, 4, 3806–3813
3809

View Article Online
IR measurements indicating that the CCl4/H2O surface, mea-
sured by the Cl(2p) : O(1s) ratio, scales with the bulk ratio cal-
culated from RAIRS (Fig. 1). The subsequent interaction of
electrons with CCl4/H2O (ice) films generates a range of reac-
tive species, most notably ꢁCCl3 , :CCl2 , Hꢁ, ꢁOH as well as sol-
vated electrons (eꢀ(aq)). The products observed in this study
are a result of interactions between these species as well as par-
ent H2O and CCl4 molecules.
The results presented in this investigation and our previous
study36 have identified C2Cl4 , COCl2 , CO2 , HCl as well as a
partially chlorinated carbonaceous overlayer (hereafter
referred to as a CClx film) as the dominant neutral species pro-
duced in the ice film during electron-beam irradiation. In con-
trast CO, CO2 , O2 , H2 and HCl are observed as neutral gas
phase products. Erosion proceeds as these volatile species are
removed from the film and into the gas phase.
Effect of substrate temperature. Compared to electron beam
irradiation experiments carried out on aqueous solutions at
room temperature, the mobility of reactive species will be sig-
nificantly reduced in the present investigation. As a result,
reactions are expected to occur principally between reactive
intermediates generated as the result of electron stimulated
reactions (e.g. :CCl2) and other species in close proximity
within the ice film.
Fig. 7 Variation in the C(1s) and Cl(2p) XP spectral regions for films
of different initial CCl4 : H2O ratios after 10 min of electron beam irra-
diation. The composite fit in the Cl(2p) region is comprised of two
Cl(2p3/2/2p1/2) doublets (2p3/2 peaks at 200.0 and 197.5 eV). See text
for details. It should be noted that the C(1s) and Cl(2p) XPS areas were
normalized to compensate for variations in intensity.
Another consequence of the low temperatures employed in
the present study is that hydrolysis reactions are not signifi-
cant. For example, phosgene is rapidly hydrolyzed to CO2
and HCl at room temperature:48,49
can be well fit using the initial Cl(2p3/2/2p1/2) doublet asso-
ciated with CCl species (Cl(2p3/2) at 200.0 eV) along with a
new Cl(2p3/2/2p1/2) doublet with the Cl(2p3/2) peak centered
at 197.5 eV. Based on literature assignments, this new doublet
is indicative of Clꢀ production.42
COCl2 þ H2O ! 2 HCl þ CO2ðk ¼ 70 sꢀ1 at 50 ꢂCÞ ð6Þ
Fig. 7 shows comparative XPS spectra of the C(1s) and
Cl(2p) regions for films of different initial CCl4 : H2O ratios
following 10 min of electron beam exposure. The evolution
of both the C(1s) and Cl(2p) regions is found to be sensitive
to the film’s initial chemical composition. For example, during
electron beam irradiation of pure CCl4 (Fig. 7(a)) only a single
doublet corresponding to CClx species is observed in the
Cl(2p) region. In contrast, electron beam irradiation of CCl4
films containing H2O produces Clꢀ species (Fig. 7(b)–(d)).
As the concentration of CCl4 in the film decreases, the ratio
of Clꢀ to CClx species increases. Analysis of the C(1s) region
also reveals that for a given period of electron beam irradia-
tion, the distribution of carbon-containing species is influenced
by the film’s initial chemical composition, with the relative
contribution from graphitic species being most pronounced
in concentrated films where the initial CCl4 : H2O ratio is high-
est.
In the present investigation experiments revealed that in the
absence of electron beam irradiation the phosgene concentra-
tion is constant. This illustrates that under the low temperature
conditions (103 K) prevalent in this study, hydrolysis does not
contribute to the loss of COCl2 or the production of CO2 .
Electron stimulated CCl4 dissociation in ice films. C–Cl bond
cleavage in CCl4 co-adsorbed in an ice film can be initiated
either by a dissociative electron attachment (DEA) process:
CCl4 þ eꢀ ! ꢁCCl3 þ Clꢀðk ¼ 1:3 ꢅ 1010
CCl3 þ eꢀ ! :CCl2 þ Clꢀ
or a hydrogen abstraction reaction:
M
ꢀ1sꢀ1
Þ
ð1Þ
ð2Þ
H þ CCl4 ! HCl þ ꢁCCl3ðk ¼ 3:2 ꢅ 107 Mꢀ1sꢀ1
Þ
ð7Þ
In general, the evolution of the near surface region indicates
that changes in the C(1s) and Cl(2p) regions are most pro-
nounced at short irradiation times. Indeed for irradiation times
in excess of ꢄ10 min, a steady state distribution of carbon and
chlorine containing species was generally observed, based on
the observation of constant C(1s) and Cl(2p) spectral envel-
opes (see Fig. 6). This steady-state composition of carbon
and chlorine containing species as judged by XPS measure-
ments was, however, sensitive to the film’s initial CCl4 : H2O
ratio (Fig. 7). Once attained, the steady state distribution of
carbon and chlorine was maintained until films were largely
depleted of the initial CCl4 and H2O, evidenced by changes
in the RAIR spectra and the appearance of XPS intensity asso-
ciated with the Au(4f) substrate.
The initial step in the destruction of CCl4 is expected to pro-
ceed via dissociative electron attachment based on these rela-
tive rate constants and the fact that hydrogen atoms must be
generated from the reaction of incident and secondary elec-
trons with adsorbed water molecules.
C–Cl bond cleavage can also proceed as a result of electron
stimulated CCl4 bond dissociation:29
CCl4 þ eꢀ ! ꢁCCl3 þ ꢁCl þ eꢀ
ð8Þ
If this reaction pathway was significant we would expect that
Cl2 would be a significant desorption product, particularly in
CCl4 rich films. In contrast, our experimental results indicate
that Cl2 was only detected as a minor desorption product
(Cl2/HCl ꢄ 0.01) during electron beam irradiation of pure
CCl4 films. Consequently, we postulate that Clꢀ rather than
Cl is the most important chlorine containing reactive species
produced during the initial dissociation of CCl4 . Furthermore,
since HCl is observed as the dominant chlorine containing des-
orption product during electron beam irradiation experiments
of pure 13CCl4 it appears that Clꢀ reactions with the chamber
walls also contribute to HCl production.
Discussion
Upon deposition at 103 K, the film consists of an amorphous
structure consisting of both molecular CCl4 and H2O rather
than a segregated overlayer. This is supported by XPS and
3810
Phys. Chem. Chem. Phys., 2002, 4, 3806–3813

View Article Online
Role of dichlorocarbene (:CCl2). Fig. 3(a) shows that the
fractional conversion of CCl4 into C2Cl4 increases with the
film’s initial CCl4 : H2O ratio. This is a result of the fact that
C2Cl4 is produced from the bimolecular coupling of the
dichlorocarbene (:CCl2) intermediate, thus:50
(ꢁCCl3),32,50,53 involving either hydroxyl radicals or molecular
oxygen:
ꢁCCl3 þ ꢁOH=O2 !! OCCl2 þ HCl
ð11Þ
Production of CO/CO2. Carbon monoxide can be produced
from the hydrolysis of the dichlorocarbene (reaction (5)) or
from the electron or hydrogen atom mediated decomposition
of phosgene:54,55
:CCl2 þ :CCl2 ! C2Cl4
ð4Þ
The probability of a bimolecular encounter between two car-
bene species leading to C2Cl4 production is expected be highest
in films where the initial CCl4 : H2O ratio is greatest, consistent
with our experimental observations.
Dichlorocarbene (:CCl2) can also react with H2O in the film
to produce HCl and CO:35
COCl2 þ eꢀ=H ! ꢁCOCl þ Clꢀ=HCl ! CO
ð12Þ
In contrast CO2 production can arise from either electron or
hydrogen atom mediated phosgene decomposition:54,55
:CCl2 þ H2O ! CO þ 2 HCl
This is proposed to be the dominant route for CO production
in films concentrated with CCl4 .
In contrast to C2Cl4 no evidence was found for C2Cl6 pro-
duction in the film, evidenced by the absence of any IR inten-
sity at 778 cmꢀ1 (d(CCl3)) during electron beam irradiation
experiments carried out using 12CCl4 . This lack of C2Cl6 pro-
duction is also consistent with previous low temperature (12 K)
ð5Þ
COCl2 þ eꢀ=H ! ꢁCOCl þ Clꢀ=HCl
ð13Þ
ð14Þ
ꢁCOCl þ ꢁOH !! CO2 þ HCl
or as a result of reactions between CO and hydroxyl or oxygen
radicals before CO escapes from the ice film:56,57
CO þ ꢁOH=O ! CO2 þ H
ð15Þ
ꢁ
Figs. 2–5 show that as the fraction of water co-adsorbed in the
film increases, CO2 becomes increasingly significant both in the
film and as a gas phase desorption product. This is consistent
with the proposed routes for CO2 production since both
involve sequential reactions involving a carbon and oxygen
containing species. This is expected to represent a more impor-
tant reaction pathway in dilute films where the relative H2O
(and thus O2/OH) concentration in the film are greatest. In
contrast CO is the dominant carbon-containing gas phase spe-
cies observed (Figs. 4 and 5) in films concentrated with CCl4
due to the reaction of the dichlorocarbene with water (reaction
(5)).
studies on electron irradiated CCl4 where CCl3 , :CCl2 and
C2Cl4 were observed but not C2Cl6 .29
CClx (x < 2) film production. IR (Figs. 1, 2 and 3) and XPS
(Figs. 6 and 7) measurements indicate that a partially chlori-
nated CClx (x < 2) carbonaceous film is produced during elec-
tron beam irradiation. The formation of a CClx (x < 2) film is
hypothesized to result from carbon–carbon coupling reactions
involving the products of electron stimulated carbon–carbon
coupling reactions. For example:
C2Cl4 þ eꢀ ! ꢁC2Cl3 þ Clꢀ
ð9Þ
ð10Þ
2ꢁC2Cl3 ! ꢁC4Cl6 etc:
Fate of HCl and Clꢀ. The production of HCl in the film can
lead to ionization by reactions with water molecules, even at
103 K:44
The role of carbon–carbon coupling reactions in generating
this CClx film is supported by the fact that the relative yield of
CClx species is greatest in CCl4 rich films (Figs. 2, 3 and 7).
The formation of surface bound CClx species has been pre-
viously observed by Yates et al. in a study of CCl4 decomposi-
tion on CaO.43 Ejection of CClx (x < 4) species has also been
observed in plasma studies of Cl2/Ar etching of SiC51 and dur-
ing electron irradiation of CCl4 on clean and hydrogen precov-
ered Si(100).52 In contrast, no evidence of CClx ejection was
found during electron beam irradiation of CCl4/H2O (ice)
films. Under the influence of continued electron beam irradia-
tion the chlorine content in the CClx film was observed to
decrease due to electron stimulated C–Cl bond cleavage, ulti-
mately producing a graphitic overlayer (Fig. 6).
HClðgÞ þ H2OðsÞ ! H3OþðaqÞ þ ClꢀðaqÞ
ð16Þ
This process is largely responsible for the production of hydro-
xonium ions (H3O+) (Figs. 1 and 2) and chloride ions (Figs. 6
and 7) during electron beam irradiation. Figs. 2, 3(c) and 7
show that in concentrated CCl4 films H3O+ and Clꢀ produc-
tion is a relatively minor reaction channel due to the lack of
available water for solvation/ionization. As a result, most of
the chlorine is removed from the film and ejected into the
gas phase as Clꢀ and HCl. In contrast, in films where the initial
CCl4 : H2O ratio is low HCl reacts readily with H2O in the
film44,45 leading to the production of H3O+(aq) and Clꢀ(aq),
evidenced by the large fraction of CCl4 associated with
H3O+ production (due to HCl) in dilute films as well as the
dominance of the Clꢀ peak in the Cl(2p) XPS region (Figs.
2, 3(c) and 7).
The CClx structure also exhibits greater thermal stability
compared to molecular CCl4 , evidenced by the presence of a
residual carbon and chlorine XPS signal intensity after the
sample has been warmed to room temperature. In contrast,
pure CCl4 films thermally desorb intact, leaving a Au XPS sig-
nal associated with the underlying mirror. The reactivity of the
It should be noted that H3O+ production can also accom-
pany electron beam irradiation of pure water.58 In separate
experiments, however, no observable increase in the broad
H3O+ peak between 1950–1560 cmꢀ1 was observed during
electron beam irradiation of pure ice at 103 K. This result indi-
cates that the vast majority of H3O+ observed in the present
investigation can be associated with the production and subse-
quent ionization of HCl in the film.
ꢁ
CClx overlayer towards H and OH species generated by elec-
tron beam irradiation of H2O was examined separately by fol-
lowing the change in Au substrate XPS signal after a H2O film
(of a thickness similar to the ones used in Figs. 1–3) was depos-
ited on top of a CClx overlayer and exposed to electron beam
irradiation. These results indicated that there was little change
in the Au XPS signal intensity after the H2O overlayer was
depleted by electron beam irradiation, indicating that the CClx
overlayer is not significantly eroded by H and OH species on
the time scale of the experiments reported in this investigation.
H2O/HDO production. Fig. 4 also shows that H2O is pro-
duced during electron stimulated reactions of CCl4 : H2O mix-
tures. In principle, H2O can be ejected directly into the gas
phase as the result of electron stimulated desorption although
the presence of significant HDO in Fig. 4(f) suggests that radi-
cal-radical reactions in the film between hydrogen atoms and
Production of COCl2. The production of phosgene is domi-
nated by reactions associated with the trichloromethyl radical
Phys. Chem. Chem. Phys., 2002, 4, 3806–3813
3811

View Article Online
hydroxyl radicals, e.g.:
OH þ D ! HDO
also play a significant role in the production of gas phase water
observed under the influence of electron beam irradiation.
mally stable well above 103 K, evidenced by its continued pre-
sence after the sample has been warmed to room temperature.
As a result, these species represent an increasing fraction of the
remaining carbon-containing species as the irradiation process
continues. Fig. 7 also shows that the distribution of carbon
containing species after a fixed period of electron beam irradia-
tion time is sensitive to the film’s initial CCl4 : H2O ratio. This
is a reflection of the fact that in concentrated CCl4 films elec-
tron stimulated reactions lead predominantly to the formation
of a CClx overlayer. In dilute CCl4 films, however, this path-
way becomes less important.
ð17Þ
Hydrogen and oxygen production. The yield of hydrogen and
oxygen are both attenuated in films concentrated with CCl4
(Figs. 4 and 5). The production of hydrogen during electron
beam irradiation of ice has been ascribed to a variety of differ-
ent reactions, including fast proton transfer:21
H
ꢀ þ H2O ! H2 þ ꢀ OH
ð18Þ
and bimolecular recombination:11
Conclusions
H þ H ! H2
ð19Þ
In films concentrated with CCl4 the electron stimulated reac-
tions of carbon tetrachloride in water (ice) films are dominated
by carbon–carbon coupling reactions leading to the formation
of a thermally stable partially chlorinated carbonaceous over-
layer (CClx) and C2Cl4 . Under these conditions carbon mon-
oxide is the dominant volatile carbon-containing species
produced while chlorine is lost predominantly into the gas
phase. As the initial CCl4 : H2O ratio in the film decreases,
the production of CO2 becomes increasingly important at the
expense of the CClx overlayer and C2Cl4 . In these water rich
films chlorine is partitioned as HCl in the film, producing
H3O+ and Clꢀ. The distribution of gas phase hydrogen and
oxygen produced from electron stimulated reactions of H2O
are also influenced by CCl4 with the production of oxygen
being effectively quenched in films concentrated with CCl4 .
Support for these mechanisms is provided by the observation
of mass 4 (D2), 3 (HD) and 2 (H2) during electron stimulated
reactions in CCl4/D2O/H2O films (Fig. 4(f)). In concentrated
CCl4 films, the nascent probability of these reactions will be
reduced. Furthermore, a large fraction of the water will be
consumed by dichlorocarbene hydrolysis (reaction (5)). Both
of these factors are expected to contribute to the decrease in
H2 produced in CCl4 rich films. Although the detailed elemen-
tary reaction steps responsible for molecular oxygen produc-
tion during electron stimulated reactions of ice are not
known,25 it can be reasonably assumed to involve a multi-body
collision event involving oxygen containing species whose
probability will also decrease in more CCl4 rich films.
Fig. 5(c) also shows that the oxygen/hydrogen (O2 : H2)
ratio decreases sharply as a function of increasing CCl4 con-
centration in the film. Thus, for all films with an initial CCl4
: H2O ratio >ꢄ0.1, the production of oxygen is effectively
quenched. This suggests that O2 production is more signifi-
cantly attenuated in the presence of CCl4 . This is postulated
to arise in part from the additional consumption of oxygen
by reactions in the film involving carbon-containing free radi-
cals, most notably:
Acknowledgements
Support for this research was provided by the National Science
Foundation (# CHE-0089168) as part of the Collaborative
Research Activities in Environmental Molecular Science in
Environmental Redox-Mediated Dehalogenation Chemistry
at the Johns Hopkins University.
ꢁCCl3 þ O2 !! OCCl2
ð20Þ
An alternative explanation for the continued production of H2
even in films concentrated with CCl4 is based on the greater
mobility anticipated for hydrogen species (H or Hꢀ) responsi-
ble for H2 production (eqns. (18) and (19)). Irrespective of the
detailed mechanistic explanation, results from the present
investigation suggest that hydrogen production from water
(ice) under the influence of electron beam irradiation is an
important process even in the presence of co-adsorbates, while
O2 production is more likely to be quenched.
References
1
R. L. Clough, Nucl. Instrum. Methods Phys. Res., Sect. B, 2001,
185, 8.
2
3
4
H. D. Ebinger and J. J. T. Yates, Surf. Sci., 1998, 412/413, 1.
L. Sanche, IEEE Trans. Dielectr. Electr. Insul., 1997, 4, 507.
B. Boudaiffa, P. Cloutier, D. Hunting, M. A. Huels and L.
Sanche, Science, 2000, 287, 1658.
5
6
7
8
T. I. Balkas, Int. J. Radiat. Phys. Chem., 1972, 4, 199.
G. J. Schulz, Rev. Mod. Phys., 1973, 45, 423.
M. Jungen, J. Vogt and V. Staemmler, Chem. Phys., 1979, 37, 49.
R. Azria, Y. L. Coat, G. Lefevre and D. Simon, J. Phys. B, 1979,
14, 175.
Time dependence. The observed product intensities in both
IR and mass spectral analysis are also sensitive to electron
beam exposure, increasing at short irradiation times before
decreasing upon prolonged periods of irradiation as the film
is depleted of material.36 The product partitioning, however,
is much more sensitive to changes in the film’s initial chemical
composition. For example, Fig. 5 illustrates that only small
changes were observed in the CO : CO2 gas phase ratio during
electron beam irradiation, although this same ratio is strongly
influenced by changes in the film’s initial chemical composi-
tion. This result is qualitatively consistent with IR measure-
ments (not shown) indicating that the average CCl4 : H2O
composition in the film does not change significantly during
electron beam irradiation.
9
N. J. Sack and T. E. Madey, Surf. Sci., 1996, 347, 367.
10 R. L. Hudson and M. H. Moore, Icarus, 1997, 126, 223.
11 R. L. Hudson and M. H. Moore, Icarus, 1999, 140, 451.
12 R. E. Johnson, Rev. Mod. Phys., 1996, 68, 305.
13 R. L. Hudson and M. H. Moore, J. Geophys. Res., Planets, 2001,
106, 33 275.
14 J. F. Cooper, R. E. Johnson, B. H. Mauk, H. B. Garrett and
N. Gehrels, Icarus, 2001, 149, 133.
15 Q.-B. Lu and T. E. Madey, J. Phys. Chem. B, 2001, 105, 2779.
16 Q.-B. Lu and T. E. Madey, Surf. Sci., 2000, 451, 238.
17 Q.-B. Lu and T. E. Madey, J. Chem. Phys., 1999, 111, 2861.
18 J. O. Noell, C. F. Melius and R. H. Stulen, Surf. Sci., 1985, 115,
119.
19 W. C. Simpson, L. Parentaeu, R. S. Smith, L. Sanche and T. M.
Orlando, Surf. Sci., 1997, 390, 86.
20 F. T. Mak, S. R. Zele, W. J. Cooper, C. N. Kurucz, T. D. Waite
and M. G. Nickelsen, Wat. Res., 1997, 31, 219.
21 G. A. Kimmel, T. M. Orlando, C. Vezina and L. Sanche, J. Chem.
Phys., 1994, 101, 3282.
Although the product distribution is not significantly influ-
enced by irradiation time, Fig. 6 shows that the fraction of car-
bon-containing species in the CClx carbonaceous film increases
during electron beam irradiation. This can be ascribed to the
fact that unlike CO or CO2 species, the CClx overlayer is ther-
3812
Phys. Chem. Chem. Phys., 2002, 4, 3806–3813

View Article Online
22 G. A. Kimmel, R. G. Tonkyn and T. M. Orlando, Nucl. Instrum.
Methods Phys. Res., Sect. B, 1995, 101, 179.
23 G. A. Kimmel, T. M. Orlando, P. Cloutier and L. Sanche, J. Phys.
Chem. B, 1997, 101, 6301.
24 T. M. Orlando, G. A. Kimmel and W. C. Simpson, Nucl. Instrum.
Methods Phys. Res., Sect. B, 1999, 157, 183.
25 M. T. Sieger, W. C. Simpson and T. M. Orlando, Nature, 1998,
394, 554.
26 W. C. Simpson, L. Parenteau, R. S. Smith, L. Sanche and T. M.
Orlando, Surf. Sci., 1997, 390, 86.
27 G. A. Kimmel and T. M. Orlando, Phys. Rev. Lett., 1995, 75,
2606.
41 J. L. Blanchard and J. T. Roberts, Langmuir, 1994, 10, 3303.
42 The Handbook of X-ray Photoelectron Spectroscopy; Perkin-Elmer
Corporation, Eden Prairie, MN, 1979.
43 O. B. Koper, E. A. Wovchko, J. A. Glass, Jr., J. J. T. Yates and
K. J. Klabunde, Langmuir, 1995, 11, 2054.
44 S. F. Banham, J. R. Sodeau, A. B. Horn, M. R. S. McCoustra and
M. A. Chesters, J. Vac. Sci. Technol., A, 1996, 14, 1620.
45 S. B. Barone, M. A. Zondlo and M. A. Tolbert, J. Phys. Chem. B,
1999, 103, 9717.
46 The presence of 12CO and 13CO2 as background gases was con-
firmed by the fact that the 12CO : 12CO2 ratio remained approxi-
mately constant for all of the films used in this investigation.
47 The 13CO : 13CO2 ratios have been corrected for the mass spectro-
meter’s intrinsic sensitivity to different gases.
28 G. A. Kimmel and T. M. Orlando, Phys. Rev. Lett., 1996, 77,
3983.
29 S. Suzer and L. Andrews, Chem. Phys. Lett., 1988, 150, 13.
30 A. Richter, H. Meyer, T. Muller, H. Specht and A. Schweig,
J. Mol. Struct., 1997, 436–437, 359.
31 W. Choi and M. R. Hoffmann, J. Phys. Chem., 1996, 100, 2161.
32 J. Monig, D. Bahnemann and K.-D. Asmus, Chem.-Biol. Interac-
tions, 1983, 47, 15.
33 W. Choi and M. R. Hoffmann, Environ. Sci. Technol., 1995, 29,
1646.
34 P. D. Hooker and K. J. Klabunde, Environ. Sci. Technol., 1994,
28, 1243.
35 E. A. Robinson, J. Chem. Soc., 1961, 1663.
36 A. J. Wagner, C. Vecitis and D. H. Fairbrother, J. Phys. Chem. B,
2002, 106, 4432.
37 S. R. Carlo, J. Torres and D. H. Fairbrother, J. Phys. Chem. B,
2001, 105, 6148.
38 AugerScan, AugerScan; 2.4 edn., RBD Enterprises Inc, Bend,
Oregon, USA, 1998.
39 N. Fairley, CASA-XPS; 1.0 edn., Casa Software Ltd, Cheshire,
UK, 2000.
48 W. H. Manogue and R. L. Pigford, Am. Inst. Chem. Eng. J., 1960,
6, 494.
49 R. Mertens, C. von Sonntag, J. Lind and G. Merenyi, Angew.
Chem., Int. Ed. Engl., 1994, 33, 1259.
50 I. Hua and M. R. Hoffmann, Environ. Sci. Technol., 1996, 30, 864.
51 P. Leerungnawarat, H. Cho, S. J. Pearton, C.-M. Zetterling and
M. Ostling, J. Electron. Mater., 2000, 29, 342.
52 K. H. Junker, G. Hess, J. G. Ekerdt and J. M. White, J. Vac. Sci.
Technol., A, 1998, 165, 2995.
53 W. Choi and M. R. Hoffmann, Environ. Sci. Technol., 1997, 31,
89.
54 M. Koch, D. R. Cohn, R. M. Patrick, M. P. Schuetze, L. Brom-
berg, D. Reilly, K. Hadidi, P. Thomas and P. Falkos, Environ. Sci.
Technol., 1995, 29, 2946.
55 H. Nichipor, E. Dashouk, A. G. Chmielewski, Z. Zimek and S.
Bulka, Radiat. Phys. Chem., 2000, 57, 519.
56 J. E. Roser, G. Vidali, G. Manico and V. Pirronello, Astrophys. J.,
2001, 555, L61.
57 N. Watanabe and A. Kouchi, Astrophys. J., 2002, 567, 651.
58 P. Gehringer and H. Eschweiler, Ozone Sci. Eng., 1999, 21,
523.
40 S. Abramovitz and J. Comeford, J. Spectrochim. Acta, 1964, 21,
1479.
Phys. Chem. Chem. Phys., 2002, 4, 3806–3813
3813