Spectroscopic investigations of high-power laser-induced dielectric breakdown in gas mixtures containing carbon monoxide

Article abstract of DOI:10.1021/jp712011t

Large-scale plasma was created in gas mixtures containing carbon monoxide by high-power laser-induced dielectric breakdown (LIDB). The composition of the mixtures used corresponded to a cometary and/or meteoritic impact into the Earth's early atmosphere. A multiple-centimeter-sized fireball was created by focusing a single 85 J, 450 ps near-infrared laser pulse into the center of a 15 L gas cell. The excited reaction intermediates that formed in various stages of the LIDB plasma chemical evolution were investigated by optical emission spectroscopy (OES) with temporal resolution. Special attention was paid to any OES signs of molecular ions. However, carbon monoxide cations were registered only if their production was enhanced by Penning ionization, i.e., excess He was added to the CO. The chemical consequences of laser-produced plasma generation in a CO-N₂-H₂O mixture were investigated using high resolution Fourier-transform infrared absorption spectroscopy (FTIR) and gas chromatography (GC). Several simple inorganic and organic compounds were identified in the reaction mixture exposed to ten laser sparks. H ₂¹⁸O was used to avoid possible contamination. The large laser spark triggered more complex reactivity originating in carbon monoxide than expected, when taking into account the strong triple bond of carbon monoxide causing typically inefficient dissociation of this molecule in electrical discharges.

Full text of DOI:10.1021/jp712011t

7162
J. Phys. Chem. A 2008, 112, 7162–7169
Spectroscopic Investigations of High-Power Laser-Induced Dielectric Breakdown in Gas
Mixtures Containing Carbon Monoxide
Svatopluk Civisˇ,† Dagmar Baba´nkova´,†,‡ Jaroslav Cihelka,†,‡ Petr Sazama,† and Libor Juha*,‡
J. HeyroVsky Institute of Physical Chemistry, Academy of Sciences of the Czech Republic,
DolejskoVa 3, 182 23 Prague 8, Czech Republic, and Institute of Physics, Academy of Sciences of the Czech
Republic, Na SloVance 2, 182 21 Prague 8, Czech Republic
ReceiVed: December 21, 2007; ReVised Manuscript ReceiVed: May 11, 2008
Large-scale plasma was created in gas mixtures containing carbon monoxide by high-power laser-induced
dielectric breakdown (LIDB). The composition of the mixtures used corresponded to a cometary and/or
meteoritic impact into the Earth’s early atmosphere. A multiple-centimeter-sized fireball was created by focusing
a single 85 J, 450 ps near-infrared laser pulse into the center of a 15 L gas cell. The excited reaction
intermediates that formed in various stages of the LIDB plasma chemical evolution were investigated by
optical emission spectroscopy (OES) with temporal resolution. Special attention was paid to any OES signs
of molecular ions. However, carbon monoxide cations were registered only if their production was enhanced
by Penning ionization, i.e., excess He was added to the CO. The chemical consequences of laser-produced
plasma generation in a CO-N2-H2O mixture were investigated using high resolution Fourier-transform infrared
absorption spectroscopy (FTIR) and gas chromatography (GC). Several simple inorganic and organic
compounds were identified in the reaction mixture exposed to ten laser sparks. H218O was used to avoid
possible contamination. The large laser spark triggered more complex reactivity originating in carbon monoxide
than expected, when taking into account the strong triple bond of carbon monoxide causing typically inefficient
dissociation of this molecule in electrical discharges.
I. Introduction
high content of carbon monoxide. A CO-dominant atmosphere
may not have existed on the early Earth for a long period of
time, due to the relatively high reactivity of CO with numerous
other components of the system, e.g., H2 and OH in the primitive
atmosphere and OH- in the early ocean.8 However, several
processes have been suggested and recognized,5 which could
lead to its temporary establishment at exactly the time when
the organic compounds necessary for later life should have been
formed. The lunar impact record shows a high flux of various
bodies on the Earth-Moon system until 3.8 billion years ago.
Thus the formation of the large amount of CO on Earth could
have been connected with frequent high-velocity impacts of
extraterrestrial bodies at that time. The CO production may have
occurred in the three following ways:5
The standard theories of chemical evolution in the Earth’s
early atmosphere presume that the organic molecules necessary
for life formation on Earth (like amino acids, nitrogen bases,
sugars, lipids, etc.) were formed by the action of, for example,
lightning discharge plasma or an extraterrestrial body impact.
The laboratory simulations of the high-energy-density processes
have been carried out in various models of the Earth’s early
atmosphere, from highly reducing1–3 to the slightly reducing
and neutral.4–6 The simulations used both laser plasma and
electric discharges; their comparison can be found in ref 7.
Chemical physics investigates primarily the relations between
the parameters of such laboratory plasma and its chemical
actions in mixtures simulating the chemistry of the Earth’s early
atmosphere. The identification of the reaction products and
intermediates, together with the measuring of their concentra-
tions in a time scale, gives us an idea of the mechanisms and
yields of these reactions in different gaseous mixtures. Geochem-
ists are provided in this way with information on the probability
of the formation of organic molecules via the high-energy-
content phenomena, within their considered models of the
Earth’s early atmosphere. These models can be assessed from
the point of view of life formation probability. In addition to
this, the purpose of the chemical-physics research into the
described phenomenon is also to find the conditions under which
the laboratory plasma is at its closest to the natural conditions.
(1) Comets could have supplied CO directly.9
(2) All the carbonaceous material contained in the extrater-
restrial body may have been oxidized by oxygen derived from
silicates, to produce CO in the impact plume.
(3) Meteoritic iron evaporated in the atmosphere could have
reduced atmospheric CO2 to CO.
Miyakawa et al.5 suggest that a CO-dominant atmosphere
could have been built up by impacts of extraterrestrial bodies
into the early Earth’s atmosphere. The CO-rich atmosphere
should have exhibited a transient character because of at least
three sinks for CO molecules:
(a) a variety of atmospheric reactions leading to organic
molecules.
In this article, we focus on the chemical manifestations of
the laser plasma in a model mixture of molecular gases with a
(b) oxidation to CO2 by hydroxyl radicals produced by water
vapor photolysis10,11
(c) the reaction in alkaline aqueous solutions producing formic
acid8
We have focused on the reaction described in (a). The
dissociation energy of carbon monoxide is largely due to the
* Author to whom correspondence should be addressed. Phone: +420-
266052741. Fax: +420-286890265. E-mail:juha@fzu.cz.
† J. Heyrovsky Institute of Physical Chemistry.
‡ Institute of Physics.
10.1021/jp712011t CCC: $40.75
2008 American Chemical Society
Published on Web 07/17/2008

Dielectric Breakdown in Gas Mixtures Containing CO
J. Phys. Chem. A, Vol. 112, No. 31, 2008 7163
triple bond between the carbon and oxygen atoms. The results
of experiments carried out so far show that the plasma initiation
of atmospheric reactions involving CO and producing organic
compounds (e.g., ref 12) seems to be less effective in compari-
son to the effect of the ionizing radiation (e.g., refs 5 and 13).
Our main aim has been to find out whether this holds also for
our method of initiation.6,14,15
The synchronization of the intensified CCD camera with the
laser pulse was accomplished using a photodiode signal. The
spectrum was integrated for a 1 s period after the trigger. Each
spectrum presented in this article corresponds to a single laser
shot gas exposure. Due to the risk of electrical interference on
the ICCD head occurring in the high-power laser hall, the
spectrometer was placed in a Faraday cage.
In this work, we study the chemical effects of large laser
sparks created by the focused beam of a high-power laser in
carbon monoxide and its mixtures with other molecular gases
which represent a model of Earth’s transient early atmosphere
rich in CO. Optical emission spectroscopy with temporal
resolution (tr-OES) was used for the observation of the early
phases of laser plasma evolution (the formation of transient
species such as atoms, radicals and molecular ions). In addition
to this, high-resolution FTIR spectrometry and gas chromatog-
raphy (GC-FID/TCD) were used for the identification of stable
products.
All above-mentioned apparatus and procedures are in more
details described in our previous papers6,14 and references cited
therein.
The absorption spectra were measured with a Bruker IFS 120
Fourier-transform infrared spectrometer (Bruker, Germany), in
the spectral interval of 1800-7000 cm-1 using an InSb
semiconductor detector and 0.01 cm-1 spectral resolution.
Detailed, high-resolution spectra were measured with a non-
apodized resolution of 0.0035 cm-1 in a narrow spectral interval
determined by interference filters. It was usually necessary to
accumulate 100 scans to obtain a sufficient signal/noise ratio.
For measurement purposes, the gaseous mixture studied
(before and after the laser spark generation) was transferred into
a previously evacuated glass IR absorption spectrophotometry
cell with a length of 30 cm, equipped with KBr windows. The
optimal pressure for the high-resolution spectra measurement
was chosen with regard to the minimal spectral broadening of
individual rotation-vibration lines (usually 450 Pa).
Gas chromatography (GC Hewlett-Packard 6090) was used
as a complementary technique for the analysis of the volatile
products of reactions initiated in the mixture by a large laser
spark. Two gaseous samples of the CO-N2-H2O mixture were
exposed to ten laser sparks and then trapped in liquid nitrogen.
After that, they were simultaneously injected with 10-port
valves. In the branch with a thermal conductivity detector
(TCD), a Hysep column (packed column) was used for the
retention and removal of organic compounds and water; columns
of HP-Plot Q (30 m × 0.53 mm × 40 µm film thickness) and
a Molecular sieve 5A (30 m × 0.53 mm × 25 µm film
thickness) were used for the separation of CO, CO2 and N2O.
HP-Plot Q separated CO2 from CO and N2O. The sixth port
valve was used to bypass the column Molecular sieve 5A during
the analysis of CO2 on TCD. After the analysis of CO2, the
mixture containing CO and N2O was separated with the column
molecular sieve 5A and monitored on the TCD. Following
conditions of the analysis were employed: Carrier gas He,
pressure 20 psi, splitless mode, loop volume 1 mL, bypass of
Molecular sieve 5A at 5.1-10.0 min, constant temperature 40
°C for 23 min, then temperature gradient 20 °C min-1 to 200
°C. In the second branch, a DG-200 column (packed column)
was used for water removal. C1-C4 alkanes and alkenes were
separated on an HP-Plot Al2O3 “M” column (50 m × 0.53 mm
× 15 µm film thickness, carrier gas: He 7.9 mL.min-1, constant
flow mode, split ratio: 1:50, loop: 250 µL, constant temperature
40 °C for 23 min, then temperature gradient 20 °C min-1 to
200 °C) and monitored by a flame ionization detector (FID).
II. Experimental Section
Laser-induced dielectric breakdown was achieved in the
molecular gaseous mixture using the high-power iodine pho-
todissociation laser system PALS. A single pulse of radiation
at a wavelength of 1.3152 µm (pulse duration of 400 ps and
energy of several hundred Joules) was extracted behind the
fourth amplifier of the system. The diameter of the laser beam
behind the fourth amplifier was 15 cm. One pulse was delivered
every 25 min. The fourth-amplifier beam was focused into a
gas cell by a plano-convex lens with a diameter of 15 cm and
a focal length of 25 cm. A pulse energy was typically of 100 J.
Power losses due to NIR beam reflections at the focusing lens
and the cell window did not exceed 15%.
For the static LIDB experiments, a 15 L glass cell was used.
The shape of the cell body was that of a cross with length and
width both of 40 cm, with four windows. The main-laser-beam
entrance windows were made of 4 cm thick glass, with a
diameter of 20 cm, and 1.5 cm thick inspection windows with
a diameter of 10 cm served for the collecting of the OES signals.
The windows were mounted onto the glass body with stainless-
steel flanges and sealed with viton rings. The cell was mounted
on a 1 cm thick dural platform for easier manipulation and
positioning. The cell’s body was equipped with two vacuum
valves (ACE Glass, USA) for gas handling.
The clean glass cell (the main body of cell was oven-annealed
at 450 °C) was evacuated to a pressure of 3 × 10-5 Torr and
then filled with high-purity N2 and CO gases up to the
atmospheric pressure (1:1, CO, Linde 99.99%; N2, Messer
99.996%). Twenty milliliters of deionized water (or 2 mL of
18O labeled water (97%, Cambridge Isotope Laboratories) in
the case of isotopic experiments) was poured into the cell during
the gas filling procedure with help of a syringe. After filling,
the cell was closed and transported to the laser facility.
The optical emission from LIDB plasmas entered directly into
the entrance slit of the spectrometer; i.e., no focusing lens was
used because of the high intensity of laser spark light emission.
Two ruled gratings were switched in the spectrometer (MS257,
Oriel), one for preview (150 lines/mm, width of measured
spectra ) 400 nm) and the second for high-resolution measure-
ment (1200 lines/mm, width of measured spectra ) 60 nm).
The dispersed spectrum was detected by an intensified CCD
camera (ICCD; iStar 720, Andor), with a resolution of 0.08 nm/
pixel for the high-resolution grating. Spectral lines in a distance
down to 0.2 nm have been distinguished if a very narrow
entrance slit is used to prevent saturation of the camera.
III. Results and Discussion
A. High-Resolution FTIR Absorption Spectroscopy. The
15 L cell was filled with a mixture of carbon monoxide and
nitrogen up to atmospheric pressure and 20 mL of deionized
water was poured into its bottom. The outer jacket of the cell
was heated with a heating strip to 50 °C; therefore the partial
pressure of water vapor inside the cell was 12.3 kPa. A small
sample of the mixture (1 mL) was transferred into an IR
absorption cell and the blank absorption spectra were measured
using FTIR; see Figure 1a. The spectra showed rotation-vibration

7164 J. Phys. Chem. A, Vol. 112, No. 31, 2008
Civisˇ et al.
Figure 3. Comparison of the FTIR absorption spectra of (a) a N2O
standard and (b) the CO-N2-H2O mixture trapped in liquid nitrogen
after its exposure to ten laser shots. The inset shows the comparison
of the rotation structure of the ν1 fundamental ro-vibration band in the
real sample and in the N2O standard.
Figure 1. High-resolution FTIR absorption spectra of (a) an unirra-
diated CO-N2-H2O gaseous mixture, (b) the same mixture after ten
laser sparks, and (c) the CO2 band composed of spectral components
belonging to the C16O18O and C16O2 isotopomers.
Figure 4. Comparison of the FTIR absorption spectra of (a) a C2H6
standard and (b) the CO-N2-H2O mixture trapped in liquid nitrogen
after its exposure to ten laser shots. The inset shows the comparison
of the rotation structure of the ν1 fundamental ro-vibration band in the
real sample and in the C2H6 standard.
Figure 2. The FTIR absorption spectrum of the CO-N2-H2O mixture
exposed to ten laser sparks and then trapped in liquid nitrogen.
hydrocarbons, apart from rotation-vibration bands of carbon
monoxide (CO) and carbon dioxide (CO2), were identified in
the spectrum. From the hydrocarbons, there were namely ethane
(Figure 4), ethylene (Figure 5), acetylene (Figure 6), and acetone
(Figure 7). All compounds were identified on the basis of their
comparison with the spectra of laboratory standards.
(ro-vibration) bands of carbon monoxide, both fundamental and
second harmonic frequencies, then two combination bands of
water vapor, and also the third harmonic frequency of carbon
dioxide.16,17 The appearance of the carbon dioxide band was
due to the presence of small carbon dioxide impurities (186.5
ppm) in the carbon monoxide gas.
The four strongest bands seen in the spectrum belonged to
acetylene. Two of these are assigned to the fundamental rotation-
vibration frequencies, specifically ν1 and ν3, at 3374 and 3289
cm-1, respectively. The remaining two bands at 4090 and 2642
cm-1 belong to higher harmonic or combination frequencies.16,17
Only one rotation-vibration band of ethane at 2985 cm-1 was
identified, belonging to its fundamental vibration frequency ν10.
For ethylene, four ro-vibration bands were identified, all
belonging to the fundamental frequencies, ν1, ν5, ν9 and ν11,
with centers at 3026, 3103, 3106 and 2989 cm-1 (refs 16 and
17).
The cell was then transported to the PALS facility and ten
laser sparks were generated in the gaseous mixture. After the
LIDB exposure, a sample of the gaseous mixture was again
transferred from the irradiation cell into the IR cell. The IR
absorption spectra (Figure 1b) of the exposed gas were measured
using the FTIR spectrophotometer. The spectra showed a
significant increase in intensity of the carbon dioxide rotation-
vibration band caused by the action of laser sparks in the CO-
N2-H2O gaseous mixture.
To remove the excess nitrogen and carbon monoxide from
the gaseous mixture, the mixture was slowly pumped with a
membrane pump through a nitrogen freezing trap. FTIR
absorption spectra of the trapped and then evaporated gases were
measured (Figure 2). Two rotation-vibration bands of nitrous
oxide (N2O) (Figure 3), and various bands of several simple
Although the formation of small hydrocarbon molecules is
in agreement with previous findings, N2O has not yet been
observed in nitrogen fixation experiments conducted with laser
plasma (see, for example, ref 18 and, for a review, ref 15).

Dielectric Breakdown in Gas Mixtures Containing CO
J. Phys. Chem. A, Vol. 112, No. 31, 2008 7165
Figure 5. Comparison of the FTIR absorption spectra of (a) a C2H4
standard and (b) the CO-N2-H2O mixture trapped in liquid nitrogen
after its exposure to ten laser shots. The inset contains the comparison
of the rotation structure in the real sample and in the C2H4 standard.
Figure 7. Comparison of the FTIR absorption spectra of (a) standard
acetone,(CH3)2Cd16O,and (b) the CO-N2-H218O mixture trapped in
liquid nitrogen after its exposure to ten laser shots.
investigation. In the Earth’s contemporary atmosphere, nitrous
oxide is considered24–27 to be formed through the following
reactions:
1
(
)
O D + N2 + M f N2O + M
(4)
(5)
(6)
N + NO2 f N2O + O
O3* + N2 f N2O + O2
+
N A3Σ + O f N O + O
(7)
(
2
)
µ
2
2
However, reactions 6 and 7 could not take place in the
CO-N2-H2O mixture because of the negligible formation rate
of O2 and/or O3 under our experimental conditions.
We may consider that N2O is produced by some secondary
processes, i.e., eqs 2–5, in the late stage of the chemical
evolution of the LIDB plasma, because this molecule may be
easily decomposed photochemically as well as thermally. NO
and NO2 are among the products of N2O photolysis; they were
not detected among the variety of products of LIDB exposure
to the CO-N2-H2O mixture. Therefore, N2O is probably
formed in the late phase of the laser spark expansion into the
cold molecular gases when VUV emission of LIDB plasma
becomes weak.
The formation of N2O, which in conventional plasmachem-
istry is typical for glow and corona discharges in air19 and
CO2-N2,28 could be explained in the present study by (a) a
low oxygen content and strong oxygen bonding in the irradiated
system and (b) by the expansion of the LIDB plasma into a
large volume of cold molecular gas. The laser spark in our case
is produced in the center of the 15 L gas cell so that the hot
plasma is cooled down slowly, in a large volume. In smaller
reactors, the plasma is quenched by collectors and/or walls
located close to the plasma core; see for example ref 12. The
much larger volumes of relatively cold plasma produced in the
15 L cell cause a product pattern more typical for a low
temperature discharge plasma than for a high temperature one.
Plasma quenching by a gas, not by a solid surface, is also closer
to an evolution of a lightning channel and/or impact plasma
occurring in a real planetary atmosphere.
Figure 6. Comparison of the FTIR spectra of (a) a C2H2 standard and
(b) the CO-N2-H2O mixture trapped in liquid nitrogen after its
exposure to ten laser shots. Insets show the comparison of the rotation
structure of (c) the ν3 fundamental and (d) ν1 + ν5 combination ro-
vibration bands in the real sample and in the C2H2 standard.
Nevertheless, its formation was reported for example in a corona
discharge in moist air.19 This product was not expected to be
found in the reaction mixture because of its low dissociation
energy D(N2-O) ) 1.7 eV.20,21 According to the intuitive
stoichiometry argument, N2O could be formed by the addition
of an oxygen atom to a nitrogen molecule. However, the reaction
(refs 18 and 20-22) occurs differently:
O + N2 f NO + N
(1)
Oxygen atoms are created in the mixture preferentially by the
water dissociation because the triple bond in CO is much
stronger than the single OH bond. Thus N2O formed by the
addition of an oxygen atom to a nitrogen molecule should be
N218O, if it is produced in the reaction mixture containing
isotopically labeled water. However, only N216O was indicated
in FTIR spectra measured in the present work.
The simple reaction paths leading to N2O could be described
by the following equations:
( )
CO* + N2 f N2O + C s
(2)
(3)
High-resolution FTIR spectroscopy makes it possible to carry
out a laser-plasma chemical experiment with stable isotope
labeling. H218O ([18O] ) 99.9%) was introduced into the mixture
of CO and N2 gases in the 15 L irradiation cell, instead of water
with a natural isotope composition. The mixture was exposed
NO* + NO f N2O + O
Reaction 3 is well-known (see, for example, ref 20, p 175,
and ref 23) whereas (2) has not yet been subjected to detailed

7166 J. Phys. Chem. A, Vol. 112, No. 31, 2008
Civisˇ et al.
(CH3)2Cd16O and H218O has been observed under these
conditions. Thus conclusive evidence is given that the acetone
found in the LIDB-exposed reaction mixture was formed due
to the laser-plasmachemical reaction of water with other
components of the mixture.
A detailed description of the individual spectral bands of
isotopically labeled acetone in the range of ro-vibration bands
ν1 and ν20 requires detailed ab initio calculations and will be
the subject of a following study. Within the framework of the
experiments with 18O-labeled water (H218O), bands of C16O18O,
C18O2, C18O and H216O molecules were observed next to the
dominant basic rotation-vibration bands of C16O2, C16O and
H218O in the spectra of the studied mixture after exposure.
(Examples may be seen in Figure 1c; assignment and interpreta-
tion can be found in ref 29) Carbon dioxide formation is
considered to be primarily governed by the reaction11 between
carbon monoxide molecules and hydroxyl radicals produced by
dissociation of water vapor:
Figure 8. Comparison of gas chromatograms (FID signal) taken for
(a) a mixture of standard hydrocarbons and (b) the CO-N2-H2O
mixture trapped in liquid nitrogen after its exposure to ten laser shots.
CO + OH f CO2 + H
However, the carbon monoxide disproportionation reaction31
may also give carbon dioxide:
(8)
TABLE 1: Gaseous Products of LIDB-Initiated Reactions in
CO-N2-H2O Analyzed by Gas Chromatography
a
concentration CO2 N2O CH4 C2H6 C2H4 C3H8 C3H6 C4H10
( )
ppm
559.5 0.81 0.24 0.1
1099 1.59 0.17 0.14 0.13 0.32 0.16 0.21
0.1
0.16 0.09 0.08
2CO f CO2 + C s
(9)
10-6 g/L
The study of the processes that can lead to the formation of
all the above-described isotopomers will be the subject of our
subsequent paper. Both the reactions of isotope exchange and
of laser-plasma chemical type are observed. The latter are
studied in relation to the delivered energy of the laser pulses.
B. Gas Chromatography. The gas chromatography analysis
of LIDB-initiated mixture was used to confirm and quantify
the above-mentioned FTIR findings. The results of GC analysis
of products formed by ten laser sparks in the CO-N2-H2O
mixture, which were subsequently trapped in liquid nitrogen
are shown in Figure 8 (FID signal) and summarized in Table
1. The high sensitivity of the method employed enables detection
of ppm of CO, CO2, and N2O and tenths of ppm of C1-C4
hydrocarbons. In addition to CO, a high concentration of CO2
and a medium concentration of N2O were found in the LIDB-
transformed mixture. Together with the FTIR spectrometry
results, there is conclusive evidence of the efficient formation
of these gases in CO-N2-H2O mixture exposed to a large laser
spark. Signal of the flame ionization detector at GC analysis of
the transformed mixture and model mixture of C1-C4 alkanes
and alkenes standards is shown in Figure 8. Peaks belonging to
methane, ethane, ethylene, propane, propene, and butane were
identified in the chromatogram of the mixture exposed to laser
sparks. This indicates that a complex mixture of hydrocarbons
was formed during LIDB of the CO-N2-H2O gaseous mixture.
In comparison with the FTIR absorption spectroscopy, the gas
chromatography is the more sensitive method. Because of higher
FTIR spectrophotometry detection limits, not all compounds
detected by GC were identified by FTIR.
a 186.5 ppm (366.3 × 10-6 g/L) of CO2 as an impurity from the
pressure cylinder of CO.
to ten laser sparks and then investigated by the high-resolution
FTIR absorption spectroscopy. A sample was frozen and after
removing the excess nitrogen and carbon monoxide, it was
quantitatively transferred into an IR absorption cell. The spectra
of most of the organic compounds were identical to the previous
spectra with unlabeled water, with the exception of the
2950-3020 cm-1 range, which is dominated by ν1 and ν20 bands
of acetone.16,17 A reference spectrum of acetone with a natural
isotopic composition was measured with a resolution of 0.0035
cm-1. This spectrum was compared with the spectrum of acetone
formed in a CO-N2-H218O mixture exposed to ten laser sparks.
The comparison of the two spectra in Figure 7 revealed the
following band shifts in the sample obtained from the mixture
with 18O labeled water: ν1 at 3018.5 cm-1 (CH3 d-stretch a1
symmetry16) exhibits a red shift of 0.5 cm-1, ν20 at 2972 cm-1
(CH3 d-stretch b2 symmetry16) shows a blue shift of 0.5 cm-1
,
which both belong to (CH3)2Cd18O, i.e., acetone labeled with
the heaviest stable isotope of oxygen.29 This could indicate that
the acetone found in the irradiated mixture was synthesized from
H218O. The calculated shift of νCO (1732 cm-1) frequency is
about 29 cm-1 (ref 30) for the (CH3)2Cd18O molecule and the
calculations predict very small shifts (about 3 cm-1) for the
other carbonyl bands (carbonyl out-of-plane band). Therefore
a weak influence of 18O atom on the CH3 symmetrical stretch
vibration is expected. This is in good agreement with the shift
of 0.5 cm-1 observed in our experiments.
C. Time-Resolved Optical Emission Spectroscopy. Among
the possible causes15 of chemical changes in a laser-plasma-
chemical system, (a) high energy photons (VUV, XUV, and
X-ray radiation) and (b) highly charged ions and fast electrons,
emitted from the hot core of a laser spark, may directly and
efficiently induce the ionization of molecules in the gas
surrounding the primary LIDB plasma. The surprising result of
our previous spectroscopy investigations was that only the
following transient species (namely those listed in Table 2) were
detected:
However, there is also an unlikely possibility that the acetone
was an impurity appearing in the sample during its preparation
and handling and was labeled by an isotope exchange with
H218O. To test this hypothesis, we mixed 0.2 mL of acetone
and 0.5 mL of H218O (99.9%) in a glass ampule. The ampule
was sealed, heated, and kept at 80 °C for 180 h. After 180 h,
the ampule was opened and the mixture was analyzed using
high-resolution FTIR spectrophotometry in the same way
as described above. No isotope exchange between the

Dielectric Breakdown in Gas Mixtures Containing CO
J. Phys. Chem. A, Vol. 112, No. 31, 2008 7167
TABLE 2: LIDB-Produced Chemical Species Detected in
CO/CO2-N2-H2O with Different Diagnostic and Analytical
Techniques
analytical/diagnostic
technique (ref)
gas
mixture
identified
species
HPLC/MS
(ref 6)
CO2-N2-H2O
glycine, alanine,
serine, asparagine
alanine
CO-N2-H2O
CO2-N2-H2O
otical ES (refs
6 and 14)
CN
CO-N2-H2O
CO-N2
CN, C2, C3
X-ray ES
(refs 14 and 32)
N
O
6+, N5+, O7+
,
6+, C5+, C4+
FTIR
(present work)
GC/FID
(present work)
GC/TCD
CO-N2-H2O
CO-N2-H2O
CO-N2-H2O
N2O, C2H2, C2H4,
C2H6, C3H6O
CH4, C2H4, C2H6,
C3H6, C3H8, C4H10
CO2, N2O
(present work)
Figure 9. Ultraviolet emission spectrum of the late evolution (t > 1
µs) of a large laser spark in the CO-He (1:100) mixture at atmospheric
pressure.
(A) in the early stage of laser spark evolution (t ∼ τpulse),
highly charged atomic ions in the hot, dense plasma of laser
spark core14,32
(B) in the late stage of laser spark evolution (t . τpulse), small
radicals, single charged atomic ions, and neutral atoms.
We did not observe any signs of molecular ions, formed by
the photoionization, charge-transfer and/or impact-ionization
processes, in the optical-emission spectra of the laser spark in
CO/CO2-N2-H2O gases. Therefore we conducted an experi-
ment to test our experimental layout and OES diagnostics to
detect molecular ions when their formation was enhanced by
adding helium gas to satisfy the conditions of the Penning
effect.33
Penning ionization experiments have already been carried out
in our laboratory to increase the abundance of molecular ions
in electrical discharges; e.g., see the earlier study focused on
H3+ ions.34 In the present study we are benefiting from the fact
that He has significantly higher ionization potential (24.6 eV)
than CO (13.99 eV, ref 35). Thus a collision of an electronically
highly excited helium atom with a carbon monoxide molecule
leads to the ionization of the molecule.
Figure 10. Time-resolved OES spectra of LIDB in the He-CO mixture
(100:1) at a total pressure of 33.3 kPa.
its d3∆2-a3Π1 (11,2) band), the atomic line of He and probably
the 1s22s5p-1s22s7s line of C2+ at 515 nm.
The studied CO gas was diluted with a high amount of helium
(approximately 1:100) and irradiated by an 85 J laser pulse.
The time-resolved UV-vis emission spectra were measured in
a single-shot mode on the microsecond time scale with a spectral
resolution of 0.08 nm/pixel using the high-resolution grating.
The gas mixtures were investigated at atmospheric pressure and
at a pressure of 33.3 kPa. During the exposure in the time
interval of 1 µs to 1 s, an emission spectrum was measured in
the 220-280 nm range (Figure 9). The spectrum revealed
several bands, of which the band at 230 nm belonged to the
B2Σ+ f X2Σ+ (0,1) electron transition of the CO+ ion (ref 36);
two intensive lines at 248 and 251 nm belonged to the carbon
and helium atoms, respectively. The latter coincided with the
C+ line, which also appeared at 275 nm.
Figure 11 shows the temporal evolution of the He line at
501.6 nm and the CO+ and CO2+ bands at 492.5 nm, recorded
in 10 µs intervals. The 501.6 nm He line (1s2s-1s3p) shows
very fast collision relaxation, as seen in Figure 10 and Figure
11. The corresponding lifetime of the line is 3.5 µs. In contrast
to that, the lines corresponding to CO+ a CO2+ appear in the
spectrum after 10 µs.
The chemical processes in He/CO plasma that can lead to
the formation and the decay of CO+ can be described in the
following way:
He + efast f He* + eslow
He* + CO f CO++e- + He
(10)
(11)
The next step was to carry out detailed measurements in the
480-530 nm range, where the A2Π1/2-X2Σ+ (0,0) comet-tail
emission subband of CO+ should be located according to ref
37. This band is overlapped by the 13Σ+-X3Π (0,0) band of
CO2+ around 492 nm. In addition, this spectral range also
showed lines of neutral CO and atomic He at 504.8 nm, which
coincides with the carbon atom line around 505.2 nm.
Time-resolved spectra in 10 µs steps were measured afterward
in the 490-530 nm range; see Figure 10. The most distinctive
lines in the first 10 µs are the triplet transition of CO (probably
Belic et al.38 discussed one electron ejection from CO+ ions
via Franck-Condon transitions, which leads to purely dissocia-
tive states or to doubly charged CO2+ ions that are metastable
against dissociation; a similar mechanism may be considered
for double charge ion formation:
He* + CO+ f CO2+ + e- + He
(12)
The different additional dissociative channels to be considered
are

7168 J. Phys. Chem. A, Vol. 112, No. 31, 2008
Civisˇ et al.
e- + CO+ f C+ + O+ + 2e-
f C2+ + O + 2e-
f O2+ + C + 2e-
e- + CO2+ f C+ + O+ + e-
(13)
(14)
(15)
(16)
The formation of a pair of singly charged ions - symmetrical
dissociation (13), the formation of double charged atomic ions
produced together with a neutral C or O atom by asymmetrical
dissociation (14) and (15). Belic et al.38 presented experimentally
obtained absolute cross-reaction measurements for electron
impact ionization reaction, asymmetrical dissociative ionization
channels (14) and (15). Kim and Irikura39 calculated total
ionization cross-sections for several positive ions including CO+.
For many ions the theory was in good agreement with
experimental data measured by Belic et al.38 but in the case of
the CO+ ion, the experimentally obtained data were much lower
than the theory. This large difference between the theory and
experiment on CO+ is in all probability a strong indication of
the dominance of the dissociative ionization channel (13), which
was not included in the interpretation of Belic’s experiments.38
Figure 11. Temporal evolution of line intensities of He [1s2s-1s3p],
CO+, O+ [2s22p2(3P)3p-2s22p2(3P)3d] and C+ [2s2p(3P°)-2s2p(3P°)3p]
electronic transition in an expanding laser spark produced by a high-
power laser system in the He-CO mixture (100:1) at a total pressure
of 33.3 kPa.
In our time-resolved spectra, we observed a composed band
at 492.5 nm, which may be assigned to the A2Π-X2Σ+ bands
of CO+ and the 13Σ+-X3Π bands of CO2+ ions. Cossart et
al.37 provided a hypothesis about the predissociation of upper
13Σ+ through the spin -orbit coupling with a repulsive 13Σ-
state. Larsson et al.40 calculated that the predissociation lifetime
of 13Σ+ is shorter than 1 µs. The Cossart hypothesis is based
on the fact that the transition lifetimes for the A2Π-X2Σ+ bands
of CO+ are about 3 µs (see ref 41), whereas the corresponding
lifetimes for the 13Σ+-X3Π bands of the CO2+ band should be
about twice as long. In the experiments of Belic et al.,38 the
products of CO2+ dications needed about 9 µs to reach the
detector. Therefore, only the dications with a lifetime larger
than this value were detected.
µs, respectively (see Figure 11). The simultaneous detection of
both lines halfway through the 492.5 nm line dissociation
supports the symmetrical dissociation theory. Unfortunately, a
similar dissociation process and the same fragments can be
expected both with CO+ and CO2+ (13) and (16). On the basis
of the experimentally obtained lifetimes and their comparison
with the calculations of transition moments for CO+ and CO2+
(ref 37), the corresponding lifetimes for the 13Σ+-X3Π bands
of CO2+ in the same spectral region should be about twice as
long as the result of transition moment calculation for CO+.
We would therefore be inclined to conclude that the band at
492.5 nm belongs to the dication of CO2+. However, it is
necessary to take into account that the ionization energy of the
helium atom is only 24.6 eV, i.e., slightly lower than the first
ionization threshold of CO+ (27.3 eV). The question remains
whether the highly excited helium atoms have sufficient energy
for the Penning ionization of CO+ to CO2+ (12). However, the
CO+ ions in the considered system are also in a state of rather
high electronic excitation (see Figure 9), thus from the energetic
point of view, this type of ionization is possible.
These results show that we are able to detect molecular ions
if they are present in the laser spark system. The experiment
raised the question of why we do not detect any molecular ions
by means of their molecular emission spectra in the case of a
helium-free mixture, i.e., when the Penning ionization is turned
out. The explanation which assumes that the rates of photo-,
impact-, and charge-transfer ionization are too low to exceed
the OES detection limit of formed species seems unlikely. More
likely is the explanation that the direct molecular ionization
processes occur immediately after the optical breakdown of the
gaseous mixture. In this case, a narrower temporal window for
collecting OES records should be applied shortly after the laser
pulse passes through the gas. However, strong atomic, ionic
and blackbody emission from the core of LIDB plasma could,
in some systems, make this measurement impossible.
Based on the above-given facts, the processes of carbon
monoxide molecular ion formation in the CO-He plasma are
represented by eqs 11 and 12. The ion extinction mechanism is
described by eqs 13–16. The intensity of He, CO2+, O+ and
C+ lines against time in the microsecond scale, as extracted
from the optical emission spectra of LIDB plasma in the visible
spectral range, is presented in Figure 11. The lifetime of the
He I excited state [1s2s-1s3p], is estimated from the fit of the
exponential decay which corresponds to 3.4 µs. During this time,
helium exchanges the energy by collisions with molecules of
CO. The single ionization threshold of the CO determined by
Belic et al.38 corresponds to 14 eV. In the experiment carried
out by G. Dawber et al.,42 the CO2+ appearance potential (in
3
the Π, V ) 0 state) relative to the V ) 0 of the ground X1Σ
state of neutral CO was determined to be 41.29 eV. The
difference between this value and the CO single ionization
threshold gives the value of 27.3 eV for the CO+ ionization
threshold.
In our spectra, after 10 µs, a band with an approximate
lifetime of 12.9 (1.6) µs appears at 492.5 nm, which could
belong to CO+ or CO2+. The theoretically obtained and
theoretically predicted longer lifetimes support the hypothesis
of 13Σ+-X3Π bands of CO2+
.
IV. Concluding Remarks
As has been mentioned above, Kim and Irikura39 proved that
the dissociative channel CO+ (13), originally rejected by Belic
et al.,38 plays an important role. The atomic lines of O+ and
C+ are created in this symmetrical dissociation process. In our
case also, the band dissociation at 492.5 nm produces O+ (515.9
nm) and C+ (512.2 nm) with lifetimes of 18.9(2.2) and 27.1(5.4)
The results of our earlier experiments with large laser sparks
(ref 6 Table 2) indicated that carbon monoxide in model
atmosphere produced a less complex product pattern for the
amino-acid formation than the gas mixtures containing other
possible constituents of our historical planetary atmospheres,

Dielectric Breakdown in Gas Mixtures Containing CO
J. Phys. Chem. A, Vol. 112, No. 31, 2008 7169
(12) Miyakawa, S.; Sawaoka, A. B.; Ushio, K.; Kobayashi, K. J. Appl.
Phys. 1999, 85, 6853.
e.g., carbon dioxide. However, the results of our present work
(Table 2), focused on the formation of small-molecule volatile
products, are encouraging, in terms of both the quality and the
quantity of such products formed in the mixtures containing
CO, which are the laboratory simulations of the possible
transient CO-rich types of the planetary atmosphere.
In such an atmosphere, not only the nitrogen fixation,
following reaction paths that result in the formation of nitrous
oxide, but also the synthesis of volatile organic compounds could
have occurred. In addition, carbon dioxide production is
efficiently initiated by high-energy density phenomena in an
atmosphere composed of carbon monoxide, nitrogen and water
vapor. Thus the transient atmosphere could have been quickly
transformed by lightning and related phenomena into the
standard oxidation composition.
Optical emission spectra do not indicate the formation of
molecular ions connected with the formation and evolution of
the large laser spark in the CO-rich mixture. Although it could
testify to the key role of thermal and shock waves in the
chemistry of the spark under the given experimental conditions,
the secondary ionization processes cannot yet be rejected. In
the present work, the generation of molecular ions (CO+ and
CO2+) by Penning ionization satisfactorily proved that our
experimental layout and diagnostics are able to investigate such
phenomena. A systematic search for molecular ions, conducted
closely to the LIDB core in time as well as in space, is required
to give the final answer to this question.
(13) Kobayashi, K.; Kaneko, T.; Saito, T. AdV. Space Res. 1999, 24,
461.
(14) Baba´nkova´, D.; Civisˇ, S.; Juha, L.; Bittner, M.; Cihelka, J.; Pfeifer,
M.; Ska´la, J.; Bartnik, A.; Fiedorowicz, H.; Mikolajczyk, J.; Ryc´, L.;
ˇ
Sedivcova´, T. J. Phys. Chem. A 2006, 110, 12113.
(15) Baba´nkova´, D.; Civisˇ, S.; Juha, L. Prog. Quantum Electron. 2006,
30, 75.
(16) http://cfa-www.harvard.edu/hitran/.
(17) Guelachvili, G.; Rao, K. N. Handbook of Infrared Standards;
Academic Press; New York, 1986.
(18) Navarro-Gonza´lez, R.; McKay, C. P.; Mvondo, D. N. Nature 2001,
412, 61.
(19) Bhetanabhotla, M. N.; Crowell, B. A.; Coucouvinos, A.; Hill, R. D.;
Rinken, R. G. Atmos. EnViron. 1985, 19, 1391, and references cited therein.
(20) Okabe, H. Photochemistry of Small Molecules; Wiley; New York,
Chichester, Brisbane, Toronto, 1978 (see also references cited therein).
(21) Yung, Y. L.; DeMore, W. B. Photochemistry of Planetary Atmo-
spheres; Oxford University Press; New York, Oxford, 1999. (see also
references cited therein).
(22) Yung, Y. L.; McElroy, M. B. Science 1979, 203, 1002.
(23) Borucki, W. J.; Chameides, W. L. ReV. Geophys. Space Phys. 1984,
22, 363.
(24) Malcolme-Lawes, D. J. Nature 1974, 247, 540.
(25) Zipf, E. C. Nature 1980, 287, 523.
(26) Prasad, S. S. Nature 1981, 289, 386.
(27) Iannuzzi, M. P.; Jeffries, J. B.; Kaufman, F. Chem. Phys. Lett. 1982,
87, 570.
(28) Nna-Mvondo, D.; Navarro-Gonza´lez, R.; Raulin, F.; Coll, P. Orig.
Life EVol. Biosph. 2005, 35, 401, and references cited therein.
(29) Pinchas, S.; Laulicht, J. Infrared Spectra of Labeled Compounds;
Academic Press; London, 1972; chapter 7 and references cited therein.
(30) Davis, R. E.; Grosse, D. J. Tetrahedron 1970, 26, 1171.
(31) Essenhigh, K. A.; Utkin, Y. G.; Bernard, C.; Adamovich, I. V.;
Rich, J. W. Chem. Phys. 2006, 330, 506, and references cited therein.
(32) Dav´ıdkova´, M.; Juha, L.; Bittner, M.; Koptyaev, S.; Ha´jkova´, V.;
Acknowledgment. This work was funded by the Grant
Agency of the Czech Republic, grant No. 203/06/1278. During
our experiments, the PALS facility was operated with the partial
financial support of the Czech Ministry of Education, grant
LC528.
ˇ
Kra´sa, J.; Pfeifer, M.; St´ısova´, V.; Bartnik, A.; Fiedorowicz, H.; Mikolajczyk,
J.; Ryc, L.; P´ına, L.; Horva´th, M.; Baba´nkova´, D.; Cihelka, J.; Civisˇ, S.
Radiat. Res. 2007, 168, 382.
(33) Penning, F. M. Naturwiss. 1927, 15, 818.
(34) Civis, S.; Kubat, P.; Nishida, S.; Kawaguchi, K. Chem. Phys. Lett.
2006, 418, 448.
References and Notes
(35) Krisnamurthi, V.; Nagasha, K.; Marate, V. R.; Mathur, D. Phys.
ReV. A 1991, 44, 5460.
(36) Cossart, D.; Cossart-Magos, C. Chem. Phys. Lett. 1996, 250, 128.
(37) Cossart, D.; Robbe, J. M. Chem. Phys. Lett. 1999, 311, 248.
(38) Belic, D. S.; Yu, D. J.; Siari, A.; Defrance, P. J. Phys. B 1997, 30,
5535.
(39) Kim, Y. K.; Irikura, K. K. J. Res. Natl. Inst. Stand. Technol. 2001,
105, 285.
(40) Larsson, M.; Baltzer, P.; Svensson, S.; Wannberg, B.; Martensson,
N.; Naves de Brito, A.; Correia, N.; Keane, M. P.; Carlsson-Goethe, M.;
Karlsson, L. J. Phys. B 1990, 23, 1175.
(41) Huber, K. P.; Herzberg, G. Molecular Spectra and Molecular
Structure IV. Constants of Diatomic Molecules; Van Nostrand Reinhold;
New York, 1979.
(42) Dawber, G.; McConkey, A. G.; Avaldi, L.; McDonalds, M. A.;
King, G. C.; Hall, R. I. J. Phys. B 1994, 27, 2191.
(1) Miller, S. L. Science 1953, 117, 147.
(2) Miller, S. L. J. Am. Chem. Soc. 1955, 77, 2351.
(3) Urey, H. C. Proc. Nat. Acad. Sci. U.S.A. 1952, 38, 351.
(4) Schlesinger, G.; Miller, S. L. J. Mol. EVol. 1983, 19, 376.
(5) Miyakawa, S.; Yamanashi, H.; Kobayashi, K.; Cleaves, H. J.; Miller,
S. L. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 14628.
(6) Civisˇ, S.; Juha, L.; Baba´nkova´, D.; Cvacˇka, J.; Frank, O.; Jehlicˇka,
J.; Kra´likova´, B.; Kra´sa, J.; Kuba´t, P.; Muck, A.; Pfeifer, M.; Ska´la, J.;
Ullschmied, J. Chem. Phys. Lett. 2004, 386, 169.
(7) Scattergood, T. W.; McKay, C. P.; Borucki, W. L.; Giver, L. P.;
Van Ghyseghem, H.; Parris, J. E.; Miller, S. L. Icarus 1989, 81, 413.
(8) Van Trump, J. E.; Miller, S. L. Earth Planet. Sci. Lett. 1973, 20,
145.
(9) Cottin, H.; Gazeau, M. C.; Raulin, F. Planet. Space Sci. 1999, 47,
1141, and references cited therein.
(10) Bar-Nun, A.; Hartman, H. Orig. Life 1978, 9, 93.
(11) Bar-Nun, A.; Chang, S. J. Geophys. Res. 1983, 88, 6662.
JP712011T