THE ASTROPHYSICAL JOURNAL, 503 : 959È975, 1998 August 20

( 1998. The American Astronomical Society. All rights reserved. Printed in U.S.A.

THEORETICAL AND LABORATORY STUDIES ON THE INTERACTION OF COSMIC-RAY PARTICLES

WITH INTERSTELLAR ICES. III. SUPRATHERMAL CHEMISTRYÈINDUCED FORMATION OF
HYDROCARBON MOLECULES IN SOLID METHANE (CH ),
4
ETHYLENE (C H ), AND ACETYLENE (C H )
2 4 2 2
R. I. KAISER1 AND K. ROESSLER2
Received 1997 March 25 ; accepted 1998 March 24

ABSTRACT
Methane, ethylene, and acetylene ices are irradiated in a ultra high vacuum vessel at 10 K with 9.0
MeV a-particles and 7.3 MeV protons to elucidate mechanisms to form hydrocarbon molecules upon
interaction of Galactic cosmic-ray particles with extraterrestrial, organic ices. Theoretical calculations
focus on computer simulations of ion-induced collision cascades in irradiated targets. Our experimental
and computational investigations reveal that each MeV particle transfers its kinetic energy predomi-
nantly through inelastic encounters to the target leading to electronic excitation and ionization of the
target molecules. Here electronically excited CH species can fragment to mobile H atoms and non-
mobile CH radicals. The potential energy stored 4in Coulomb interaction of the CH` ions release ener-
getic H and3 C atoms not in thermal equilibrium with the 10 K target (suprathermal 4species). Moderated
to 1È10 eV kinetic energy, these carbon atoms and those triggered by the elastic energy transfer of the
MeV projectile to the target are found to abstract up to two H atoms to yield suprathermal CH and
CH species. C and CH, as well as CH , can insert into a CwH bond of a CH molecule to form
2 2 (C H ), and ethane (C H ). HCCH either 4 loses H /2H to
methylcarbene (HCCH ), the ethyl radical
form acetylene, C H , rearranges to ethylene, C H , or adds two H atoms to form ethane, C H2 . C H
3 2 5 2 6 3
2 2an H atom, giving ethane and
2 4 ethylene, respectively. C H and C H are 2 found
6 2 to5
can abstract or lose
2 2 2 4
react with suprathermal H atoms to form C H and C H , respectively. Overlapping cascades and an
increasing MeV ion exposure transforms C 2H 3 (x \ 2,2 . .5. , 6) to even more complex alkanes up to
2 x
C H . These elementary reactions of suprathermal species to insert, abstract, and add in/to bonds
14 30
supply a powerful pathway to form new molecules in icy grain mantles condensed on interstellar grains
or in hydrocarbon rich bodies in our solar system even at temperatures as low as 10 K.
Subject headings : cosmic rays È ISM : molecules È methods : laboratory È molecular processes

1. INTRODUCTION minor amount to the interstellar matter, but temperatures

can rise up to 4000 K in the outer photosphere (Alksne,
The interstellar medium (ISM) contains B10% of the Alksnis, & Dzervitis 1991), and a more complex chemistry is
mass of our Galaxy and consists of gas (99%) as well as anticipated.
0.1È0.2 km ellipsoidal-shaped grain particles (1%), with Although the interstellar dust component embodies only
averaged number densities of 1 H atom cm~3 and 10~11 1% of the ISM mass, these predominantly silicate- and
grains cm~3, respectively (Hollenbach & Thronson 1987). carbonaceous-based grain nuclei play a key role in the for-
Its chemical composition is dominated by hydrogen and mation of new molecules (Hollenbach & Thronson 1987).
helium (93.38% H, 6.49% He), whereas oxygen, carbon, and Deep in the interior of dense molecular clouds, grain par-
nitrogen contribute only 0.11% (O : C : N B 6 : 3 : 1) (Sche†ler ticles e†ectively shield newly synthesized molecules in the
& ElsaŽsser 1988 ; Cowley 1985). The remaining elements gas phase from the destructive external Galactic UV radi-
furnish 0.02%. This elementary composition is reÑected in ation Ðeld with a Ñux / of about 108 photons cm~2 s~1 at
106 hitherto identiÐed detected interstellar radicals, atoms, energies greater than 6.2 eV. In addition, these grains hold
and molecules as well. These species are not spread homo- typical temperatures of 10 K in molecular clouds
geneously in the interstellar medium (ISM) but disclose pro- (Greenberg 1971 ; Tielens & Allamandola 1987) and mol-
nounced localization in large-scale structures such as ecules, radicals as well as atoms from the gas phase are
interstellar clouds and more localized circumstellar accreted on grain surfaces resulting in an icy mantle up to
envelopes of, e.g., dying carbon and oxygen stars. Di†use 0.1 km thick. Here solid mixtures containing H O, CO,
(hot) clouds hold number densities of n of about 10 mol CH OH, NH , H S, CH , H CO, OCS, OCN~, 2H , and
CO3 were identiÐed
cm~3 and mean translational temperatures T of 100 K, 3 2 unambiguously
4 2 2 on
via IR spectroscopy
while in dense (cold, dark, molecular) clouds typical sce- 2
interstellar grains (Whittet 1993 ; Schmitt 1994 ; Schutte
narios range n \ 102È106 cm~3 and T \ 10È40 K. Mol- 1998 ; Tielens et al. 1991b ; Chiar et al. 1994 ; Grim & Green-
ecules in the outÑow of carbon stars contribute only a berg 1987 ; Lacy et al. 1984 ; Tegler et al. 1995 ; Schutte et al.
1996 ; Palumbo, Tielens, & Tokunaga 1995 ; Schutte &
Greenberg 1997). Further, molecular hydrogen, H , the
1 Institut f uŽr Nuklearchemie, Forschungszentrum JuŽlich, 52425 JuŽlich, most abundant interstellar molecule, is thought 2to be
Germany. Present address : Institute of Atomic and Molecular Sciences, 1, formed predominantly on grain surfaces via recombination
Section 4, Roosevelt Road, Taipei, 106, Taiwan ;
kaiser=po.iams.sinica.edu.tw. of two physi-/chemisorbed H atoms (Tielens & Allamon-
2 Institut f uŽr Nuklearchemie, Forschungszentrum JuŽlich, 52425 JuŽlich, dola 1987) and inside ice layers thru suprathermal chem-
Germany. istry (Kaiser et al. 1997).
959
960 KAISER & ROESSLER Vol. 503

This ice mantle is processed by the cosmic-ray-induced (PAHs). Our present investigations focus on methane, ethyl-
internal UV radiation present even in the deep interior of ene, and acetylene as model compounds to elucidate
dense clouds (/ \ 103 photons cm~2 s~1) (dÏHendecourt & detailed mechanisms to form intermediate sized neutral
Allamandola 1986 ; Allamandola, Sandford, & Valero 1988 ; hydrocarbons such as C H (x \ 1, . . . , 6) and C H (x \ 4,
Grim et al. 1989 ; Shalabiea & Greenberg 1994) and through 2 x 3 x
6, 8) before extending studies to astrophysically relevant
particles of the Galactic cosmic radiation Ðeld, leading to interstellar ice mixtures. The choice of MeV particles in our
new molecules synthesized in the solid state (Johnson, Lan- experiments characterizes the Ñux distribution maximum of
zerotti, & Brown 1984 ; Geiss et al. 1992 ; Simpson 1983 ; Galactic cosmic-ray particles peaking at about 8È15 MeV
Johnson 1990 ; Brown, Lanzerotti, & Johnson 1982). The (see discussion in Paper I), whereas experimental doses rep-
particle component consists of 97%È98% protons (p, H`) resent equivalent irradiation by the interstellar cosmic-ray
and 2%È3% helium nuclei (a-particles, He2`) in the low- particle Ðeld of about 1 ] 109 yr. These mechanisms are
energy range of 1È10 MeV, with / \ 10 particles cm~2 s~1 investigated by (1) variation of the energy transfer from the
(Johnson 1990 ; Lanzerotti & Johnson 1987) and higher implanted ion to the CH target, the so-called averaged
energies up to the PeV limit (/ \ 10~12 particles cm~2 4
linear energy transfer (LET), from 4.0 ^ 0.2 keV km~1 (7.3
s~1). Since, however, typical carbon-hydrogen and carbon- MeV protons) to 45 ^ 5 keV km~1 (9 MeV a-particles) ; (2)
carbon bond strengths in organic molecules range between increasing the absorbed energy per target molecule, dose D,
3 and 10 eV, the cosmic-ray particles are too energetic to to 30 eV ; (3) elucidating di†usion limited chemistry of
form stable chemical bond as implanted into the icy mantle. atoms, CH, CH , as well as CH radicals via substitution of
CH by CD ; 2and (4) changing
But upon interaction with the solid target, each cosmic-ray 3 the carbon to hydrogen
particle releases its excess energy to the target atoms in 4 4
ratio from 4 : 1 (CH ) via 2 : 1 (C H ) to 1 : 1 (C H ). Further,
CH /CD mixtures4 are irradiated
successive collisions via elastic and inelastic interactions 2 4 to di†er between
2 2
inter-
4 4
(Kaiser & Roessler 1997). Here the elastic process transfers and intra-molecular reactions, while 13CH targets are
energy to the nuclei of the target atoms igniting primary 4
selected to verify on line and in situ Fourier transform infra-
knock-on particles (PKOs ; Ðrst generation of knock-on red (FTIR) spectroscopy and quadrupole mass spectrom-
particles) if this amount is larger than the binding energy of etry (QMS) data.
the atom. MeV a-particles, for example, generate carbon 2. EXPERIMENTAL APPROACH AND
PKOs with kinetic energies up to 10 keV. These knock-on
COMPUTATIONAL MODEL
particles can transfer their energy in consecutive encounters
to the target atoms resulting in a collision cascade of sec- The simulation experiments were performed in a UHV
ondary, tertiary, etc., knock-on atoms. Moderated to about chamber equipped with a closed cycle helium refrigerator
1È10 eV, the so-called chemical energy range, these atoms and an attached silver wafer (Kaiser, Gabrysch, & Roessler
are not in thermal equilibrium with the 10 K target (hence : 1995). Ice layers are prepared by depositing gases on a
nonequilibrium or suprathermal particles) and can react cooled wafer, keeping the cold Ðnger at 10 K for 60 minutes,
Ðnally with the target molecules via elementary steps of heating the ices to 35 K with 0.005 K s~1, equilibrating at
bond insertion, addition to double or triple bonds, or 35 K for 1 hr, and cooling down to 10 K with 0.005 K s~1.
hydrogen abstraction (StoŽcklin 1969 ; Roessler 1991). This temperature program ensures a well-deÐned low-
The power of these suprathermal reactions to form new temperature modiÐcation of all ices. Further, a constant
molecules at temperatures even as low as 10 K is based in LET is crucial to guarantee target-depthÈinvariant reaction
their ability to overcome reaction barriers in the entrance mechanisms, and the target thickness must be limited to
channel easily since suprathermal species can impart their 5 ^ 1 km, well below the range of 7.3 MeV protons and 9
excess kinetic energy into the transition state of the reac- MeV a-particles with about 930 and 130 km. Hydrocarbon
tion. Even reactions endothermic at 10 K are feasible and free UHV conditions of about 4 ] 10~10 mbar are supplied
extend the synthetic power of this reaction class further by a membrane, molecular drag, and cryopump proved to
beyond thermal processes of di†usion controlled chemistry be essential to exclude organic contamination and restrict
on interstellar grains (Roessler 1991). These unique aspects condensation of residual gases onto the silver waver to one
of suprathermal reactions result in reaction rate constants k monolayer in B10 hr. All targets are irradiated at 10 K with
orders of magnitude larger than their thermal counterparts. ion Ñuxes / of 9.0 MeV a-particles and 7.3 MeV protons
Detailed calculations on the reactions of, e.g., 1 eV supra- adjusted to / (a) \ 127 nA cm~2 and / (p) \ 111 nA cm~2
thermal hydrogen atoms with H O and CH to form H to limit the temperature increase of the frost surface to
2 4 k (1 eV H,2 14 ^ 1 K. After the irradiation, all targets are warmed up to
and OH as well as CH , respectively, depict
3
H O) \ 1.7 ] 10~11 cm3 s~1 and k (1 eV H, CH ) \ 5.0 293 K. Isotopically pure gases are supplied by MSD iso-
]210~11 cm3 s~1 versus thermal rate constants of k4(293 K, topes [13CH (99.9% 13C) ; CD (99.9% D) ; 12CH (99.9%
12C)] and are4 used without further
4 puriÐcation. 4
H ] H O) \ 9.6 ] 10~27 cm3 s~1 and k (293 K, H
] CH 2) \ 2.5 ] 10~19 cm3 s~1 (Heyl 1990), a di†erence of The solid state is monitored on line and in situ with a
up to 164 orders of magnitude. Fourier transform infrared (FTIR) spectrometer (4000È400
These Ðndings engaged our interest to elucidate chemical cm1, NICOLET) in absorption-reÑection. Table 1 compiles
and physical e†ects upon Galactic cosmic-ray MeV ion those species and their absorption coefficients whose con-
interaction with frozen organic samples, resulting in the centration proÐles of which have been quantiÐed via the
formation of new molecules in interstellar ices. Kaiser & Lambert-Beer relation (1) :
Roessler (1997, hereafter Paper I) and Kaiser et al. (1997,
I(l8 ) \ I (l8 )e~v(’)n , (1)
hereafter Paper II) of this series disclosed precise mechanis- 0
tic information on the synthesis of the most primitive atom with the intensity of the IR beam after, I(l8 ), and before
and molecule, atomic and molecular hydrogen, as well as absorption, I (l8 ), at a wavenumber l8 , the wavenumber-
0
complex species such as polycyclic aromatic hydrocarbons dependent absorption coefficient v(l8 ) (in units of cm~2) and
No. 2, 1998 INTERSTELLAR ICES. III. 961
TABLE 1
INTEGRAL ABSORPTION COEFFICIENTS A , INTEGRATION LIMITS, AND MODES OF QUANTIFIED SPECIES
exp
DURING THE IRRADIATION EXPERIMENTS

Integration limits A
Species Vibration (cm~1) exp
(cm) Reference

CH . . . . . . . . . . . . . . . . . . . . . . . . . . . l 629È590 (2.5 ^ 0.8) ] 10~17 1, 2

CH3 . . . . . . . . . . . . . . . . . . . . . . . . . . . l2 1349È1249 (6.2 ^ 0.6) ] 10~18 3
4 4
l ]l 4483È4246 (3.5 ^ 0.5) ] 10~18 3
3 4
l ]l 4231È4168 (1.6 ^ 0.4) ] 10~18 3
1 4
3l 3860È3830 (2.0 ^ 0.3) ] 10~19 3
l ]4l 2887È2784 (2.3 ^ 0.4) ] 10~18 3
2 4
C H ......................... l 3269È3209 (3.2 ^ 0.3) ] 10~17 4
2 2 3
l ]l 1449È1359 (4.6 ^ 0.5) ] 10~18 4
4 5
l 809È729 (1.4 ^ 0.2) ] 10~17 4
l5 1990È1960 (1.0 ^ 0.3) ] 10~19 4
2
C H ......................... l 2987È2951 (1.0 ^ 0.2) ] 10~18 5
2 4 11
l 1023È837 (1.5 ^ 0.3) ] 10~17 5
C H ......................... l7 1374È1360 (6.0 ^ 0.6) ] 10~19 3
2 6 l6 830È817 (1.9 ^ 0.2) ] 10~18 3
9
l 1480È1445 (2.7 ^ 0.3) ] 10~18 3
8
CH (n \ 4, 5, 6, 7) . . . . . . l (CH ) 2965È2949 (2 ^ 1) ] 10~17 3
n 2n`2 las(CH3) 2921È2923 (3.8 ^ 1.4 ] 10~17 6
las(CH 2) 2855È2845 (6 ^ 3) ] 10~18 6
ls(CH2) 2875È2865 (3.0 ^ 1.5) ] 10~18 6
s 3
CH D (x ] y \ 4) . . . . . . . . . . pro CH (1.5 ^ 0.2) ] 10~18 6
x y pro CD (1 ^ 0.3) ] 10~18 3
CH CD . . . . . . . . . . . . . . . . . . . . . las(CH ) 2980È2970 (1 ^ 0.2) ] 10~17 3
3 3 las(CD3) 2200È2400 (3.6 ^ 0.5) ] 10~18 7
CD . . . . . . . . . . . . . . . . . . . . . . . . . . . l 3 2345È2329 (7 ^ 2) ] 10~18 7
4 l ] 3l 3150È3070 (8 ^ 3) ] 10~20 8
13l 4 2935È2920 (6 ^ 2) ] 10~20 8
l ]4l 3261È3220 (4 ^ 1) ] 10~19 8
3l 4 1010È980 (4 ^ 1) ] 10~18 8
C D ......................... l4 2433È2368 (1.5 ^ 0.5) ] 10~17 8
2 2 3
l 548È539 (0.7 ^ 0.3) ] 10~17 8
5
C D ......................... l 748È606 (0.8 ^ 0.2) ] 10~17 8
2 4 7
C D ......................... l 2235È2226 (1.0 ^ 0.5) ] 10~17
9 5,
2 6 7
l 2087È2077 9(2 ^ 1) ] 10~18 5,
5
l 605È583 (1 ^ 0.3) ] 10~18
9 5,
9
REFERENCES.ÈSnelson (1970) ; 2. Wormhould & McCurdy (1989) ; 3. Pearl et al. (1991) ; 4. Pearl 1994,
private communication ; 5. Cowieson et al. (1981) ; 6. Bellamy (1968) ; 7. Kondo & Saeki (1973) ; 8.
Addepalli & Rao (1976) ; 9. Smit, van Straten, & Visser (1978) ; 9. Nyquist et al. (1952).

the number of absorbing species (per cm2), n. Reformulating incorporated in the computations to extract individual, cali-
equation (1) with A(l8 ) 4 log [I (l8 )/I(l8 )] gives brated components of gas mixtures.
0 Physical processes triggered by the ion implantation into
A(l8 ) \ v(l8 )n/ln 10 . (2) a CH target are simulated with the MARLOWE program
4
Integrating from l8 to l8 yields (Robinson 1990) extended for quantum mechanical and
1 2 relativistic treatments of our MeV ions (Paper I). This code
/’2 A(l8 ) dl8 (ln 10) cos 62.5 calculates elastic and inelastic energy transfers from an
n \ ’1 ] , (3)
A 2 impinging particle to the target atom(s) of the condensed
exp hydrocarbon ices.
with the integrated absorption /’2 A(l8 ) dl8 in cm~1 and the
integral absorption coefficient A’1 \ /’2 v(l8 ) dl8 in cm. The
factor cos 62.5 accounts for the exp angle ’between
1
the surface 3. RESULTS
normal of the (111) silver waver and the IR beam, while a 3.1. Computer Simulations
division by 2 corrects for the ingoing and outgoing IR 3.1.1. Elastic Energy T ransfer
beam. This procedure is accurate within ^10%È20%.
The gas phase is probed through a calibrated quadrupole The MARLOWE calculations show that implanted a-
mass spectrometer (QMS). Data processing is performed particles and protons transfer less than 1% of the energy via
via matrix interval algebra (Kaiser et al. 1995). BrieÑy, m/e elastic encounters to the CH target. Our result is consistent
ratios are chosen to result in an inhomogeneous system of with previous investigations4 depicting a larger fraction of
linear equations including the measured ion current (right- inelastic versus elastic energy loss if the kinetic energy of the
hand vector), partial pressures (unknown quantity), and projectile in keV amu~1 is larger than the projectileÏs mass
calibration factors of fragments of individual gaseous in atomic mass units (Roessler 1991). This energy transfer
species determined in separate experiments. Since all quan- engages high energetic C, as well as H atoms holding them-
tities are provided with experimental errors, matrix interval selves kinetic energies up to 17 and 28 keV, respectively.
arithmetic, i.e., an IBM high-accuracy arithmetic sub- Since these energies are still 3 orders of magnitude too large
routine deÐning experimental uncertainties as intervals, is to form stable chemical bonds, the kinetic energy has to be
962 KAISER & ROESSLER Vol. 503

released even further to the solid methane target resulting in 3.2. Quadrupole Mass Spectrometry (QMS)
the generation of collision cascades (see ° 1). Averaging over 3.2.1. 10 K Irradiation
104 simulated cascades, we Ðnd total numbers of about
0.2/0.07 C (7.3 MeV H`) and 3È4 H/1.2 C atoms (9.0 MeV During the MeV particle irradiation of the methane
a) per impinging MeV particle in a 5 km CH target. An targets only atomic as well as molecular hydrogen at
4 m/e \ 1 (H), 2 (H ) (CH -target), 2 (D), 4 (D ) (CD target)
increasing hydrogen to carbon ratio from 1 (C H ) via 2 2 4 2 4
2 2 and 1 (H), 2(D/H ), 3 (HD), and 4 (D ) (CD /CH target)
(C H ) as compared to 4 (CH ) studied earlier results 2 2 4 4
2 4 4 could be detected. Their formation mechanisms are eluci-
in an enhanced ratio of H knock-ons versus C knock-ons,
i.e., [H knock-ons] /[C knock-ons] \ 2, [H dated in Paper II. No sputtered CH (x \ 1, . . . , 4) or higher
C2H2 C2H2 x
knock-ons] /[C knock-ons] \ 3, and [H masses were detected within the 1 amol detection limit of
C2H4 C2H4 the mass spectrometer. During the irradiation phase of the
knock-ons] /[C knock-ons] \ 10.
CH4 CH4 C H and C H no atomic or molecular hydrogen was
2 2 2 4
3.1.2. Inelastic Energy T ransfer identiÐed.
The inelastic energy transfer dominates the moderation 3.2.2. W arming-up Phase to 293 K
of the MeV projectiles in the solid hydrocarbon targets and Molecular ions of alkanes show pronounced maxima
results in generation of excitons (electronic excitations in upon annealing the irradiated methane samples to 293 K
the solid state). Grigorev, Pshezhetskii, & Trakhtenberg (Fig. 1). At about 20 K, the sublimation of the remaining
(1985) determined the Ðrst CH exciton level (triplet) to 5.4 CH /CD matrix releases matrix isolated alkane molecules
4 (singlet) ranges at 10.9 eV. 4 C 4D . Correcting for the ionization cross section,
eV, while the second exciton level up to
The singlet state can be stabilized through phonon inter- 11 24]/[alkane] ratios of about 103È104. These pat-
we Ðnd [CH
actions to the triplet state, transfer its excess energy to the 4 do not go hand in hand with increasing
terns, however,
surrounding matrix or decay to methyl radicals, CH , and molecular weights of sublimating products as expected from
H atoms (asterisk indicates electronically excited) : 3
sublimation temperatures increasing by 14È16 K per addi-
tional CH or CD unit in alkanes. Here the sublimation
sequence is2 reversed
CH* ] CH ] H . (4) 2 : m/e \ 180 (C D ) is detected Ðrst,
4 3
while perdeutero propane C D is 11 24
monitored last. During
Although no detailed exciton calculations on solid ethyl- 3 8
the warm-up phase, the outer layers of the target, which
ene and acetylene have been performed yet, electronically were heated during the irradiation up to 15 K, are sub-
excited C H molecules very likely relax to C H and H liming prior to those layers close to the silver target, holding
2 2C H can fragment to C H radicals2 and H :
atoms, while
2 4 2 3 the 10 K temperature even during the actual irradiation.
[C H ]* ] C H ] H , (5) These Ðndings strongly indicate that C D is synthesized
2 2 2 in target layers directly exposed to the11ion24beam, but that
[C H ]* ] C H ] H . (6) C D is synthesized in target regions close to the Ag-(111)
2 4 2 3 3 8 Hence, the increasing molecular weight of alkanes
wafer.
Besides formation of excitons, the inelastic energy trans- coincides with production zones of higher temperatures and
fer can ionize the target circa 10~16 to 10~15 s after the infers a temperature-dependent component to their forma-
actual inelastic encounter (Moshammer 1990) if this process tion mechanism.
transfers more than the Ðrst ionization potential (12.7 eV) to Besides a qualitative identiÐcation of alkanes up to
the CH molecule. This mechanism generates energetic elec- C H (9 MeV a-particle irradiation) and C H (proton
4
trons (d-electrons) released almost perpendicular to the
14 30
experiments), data analyses via matrix algebra
8 18 reveals
MeV ion trajectory. Utilizing a model from Katz and production of unsaturated C -species methylacetylene,
3
coworkers to calculate the maximum range of these elec- CH CCH, allene, H CCCH , propylene and cyclopropane,
3 2 2
trons (Katz et al. 1971), we calculate an upper limit of a 50 Ó both C H isomers, and 13C substituted counterparts upon
heating3the
6 irradiated target to 293 K (Table 2). Most strik-
radius perpendicular to the trajectory of the implanted ion.

FIG. 1a FIG. 1b
FIG. 1.ÈDetected alkanes during matrix sublimation of CD (m/e \ 20) after irradiation at 10 K at dose D \ 30 eV ; m/e-52, 58, 84, 100, 116, 132, 148,
4 molecular ions. (a) Survey, (b) magniÐcation of m/e [ 20. The x-axis has been converted
164, and 180 (per-deutero propane ] per-deutero undecane) deÐne
from time to temperature scale.
No. 2, 1998 INTERSTELLAR ICES. III. 963
TABLE 2
HYDROCARBONS DETECTED IN THE WARM-UP PHASE OF IRRADIATED CH , C H , AND C H TARGETSa
4 2 4 2 2
Target Hydrocarbons

CH , C H , C H , n [ C H , i [ C H , C H (dimethylpropane),
Alkanes . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4 [ 2C 6H ,3 C8H , C 4H 10, C H 4 , (C
n/i 10 H 5 ), 12
(C H ), (C H ),
5 12 6 14 7 16 8 18 9 20 10 22 11 24
[(C H )], [(C H )], [(C H )]
Unsaturated hydrocarbons . . . . . . 12 26 13 28 14 30
C H , C H , C H (allene, methylacetylene), C H (cyclopropane, propene)
2 2 2 4 3 4 3 6
a Molecules in parentheses were not detected in proton-irradiated methane targets, and those in brackets were not
in irradiated acetylene targets.

ing, C H , C H , C H , C H , C H , C H pristine sample (Table 3) with the irradiated ice clearly

9 20 10 22 11 24 12 26 13 28 14 30
were not synthesized in proton-irradiated methane targets, shows absorptions of acetylene (C H ), ethylene (C H ),
2 2 2 4
most likely due to a LET in this system 1 order of magni- and ethane (C H ) the features of which are compiled in
2 6
tude less as compared to a-particleÈirradiated methane Table 4. Open shell species, vinyl (C H , 898 cm~1), ethyl
2 3
samples. Since Ðnancial limitations restricted calibration of (C H , 534 cm~1), carbene (CH , 609 cm~1), and methyl
2 5 2
the deuterated molecules to D , CD , and C D (x \ 2, 4, radicals (CH , 609 cm~1) are present as well. Further, ali-
2 4 2 x 3
6), no information on C D products was available. phatic absorptions of symmetric and asymmetric stretching
3 x vibrations of methyl (CH ) and methylene (CH ) groups
3.3. Fourier T ransform Infrared Spectroscopy (FT IR) 3 2
can be assigned. The position of skeleton vibrations of
The FTIR spectra are analyzed in three steps. First, we alkanes infers predominantly branched hydrocarbons since
investigate the absorptions qualitatively and assign their absorption patterns for the (CH ) chain unit are only
n H , propane, C H ,
carriers. Hereafter the temporal developments upon MeV detected for n \ 1, 2, 3, i.e., ethane2 C
and n-butane, C H . Likewise, the 2Ðne
ion implantation of molecules and radicals with one and 6 structure of3 the
8
two carbon atoms are investigated quantitatively as out- 4 10
symmetric deformation mode of the CH group strongly
lined in ° 2. Finally, these data are employed to calculate 3
supports the existence of tertiary butyl groups, wC(CH ) .
production rates k(i, j ) in units of synthesized molecules i 33
Further, substituted cycloalkanes such as cyclobutanes,
per MeV projectile j in the solid target at 10 K. To calculate pentanes, as well as hexanes are detected based on their
destruction rates per unit length, numbers given in the fol- skeleton vibrations. Besides aliphatic absorptions, oleÐnic
lowing section must be divided by the sample thickness, i.e., groups are identiÐed through their xCH stretching vibra-
4.0 ^ 0.4 km (CH targets), 5.5 ^ 1.5 km (CD targets), 2
4 targets), and 3.0 ^ 0.1 km 4 (C H tions, while mono- and disubstituted acetylene derivatives
2.5 ^ 0.3 km (C H are assigned via CwH and CyC stretching modes. Bom-
2 4 2 2
targets). The error bars of the destruction/formation rates bardments of a-particles show qualitatively identical
are based on the Ñuctuating ion beam intensity (^10%È absorption patterns. The FTIR spectra of 9 MeV a-
20%) and the infrared integral absorption coefficients particleÈirradiated 13CH targets underline these assign-
(^10%È30%), see Table 1. ments and depict absorption4 patterns of similar per-13C-
3.3.1. Qualitative Analyses substituted groups (Fig. 3, Tables 5 and 6).
3.3.1.1 12CH and 13CH Targets 3.3.1.2. CD Targets
4 4 4
The e†ects of the 7.3 MeV proton irradiation of the In analogy to 12CH È and 13CH Èirradiated targets,
4
12CH target are displayed in Figure 2. A comparison of the CD samples show per-deutero ethane,4 ethylene, acetylene,
4 4

FIG. 2.ÈFTIR spectra of the 12CH target before (top) and after (middle) a 7.3 proton irradiation at 10 K. Bottom spectrum was recorded at 60 K after
sublimation of CH and C H (x \ 1, . 4. . , 6). Absorptions and assignments are compiled in Tables 3 and 4.
4 2 x
 TABLE 3
FUNDAMENTAL AND COMBINATION BANDS OF SOLID 12CH AT 10 K
4
Wavenumber
Vibration (cm~1)
l ]l ....... 4349, 4324, 4328, 4314.2, 4302.5, 4299.5, 4307.8, 4283.3
3 4
l ]l ....... 4527.1, 4537
l2 ] l3 . . . . . . . 4201.2, 4203.5, 4206.2
1 4
l ] 2l . . . . . . 4113.7, 4120.6
2 4
3l . . . . . . . . . . . . 3890.6, 3847.6, 3851.8, 3843.1
4
l ............. 3114.5, 306.4, 3028, 3023.3, 3019.5, 3014.5, 3003, 3000
l3 ] l . . . . . . . 2864.6, 2831, 2824.8, 2816.7, 2815.4, 2805.7
2 4
2l . . . . . . . . . . . . 2642, 2614.3, 2599.1, 2594.4, 2591.9
4
l ............. 1596.1, 1350.7, 1309.2, 1303.5, 1298.6, 1295.5, 1293.6, 1290
4

TABLE 4
SELECTED ABSORPTIONS OF 7.3 MeV PROTONÈIRRADIATED 12CH AT 10 Ka
4
Carrier and Wavenumber
Assignment (cm~1)
HwCyCwR . . . . . . . . . . . . . . . . . . . l (3278, 3230, 3293, 3258), l (2136, 2134)
RHCxCxCHR@ . . . . . . . . . . . . . . . lCH(sp) (1934) C.C
C/C/C
CH xCxCRR@ . . . . . . . . . . . . . . . . l (1923, 1918)
CH2xCxCHR@ . . . . . . . . . . . . . . . . lC/C/C(1943)
2 (sp2) . . . . . . . . . . . . . . . . . . . . . .
xCH lC/C/C
(3098), l (1643, 1650), d (1417, 1419), d (890, 897)
wCH2 . . . . . . . . . . . . . . . . . . . . . . . . . . .lCH(2964, 2962,C/C2958, 2954), l (2884),
ip d (1472), op
d (1457
3 as1453, 1448), d (1386, 1366)s b as
wC(CH ) w . . . . . . . . . . . . . . . . . . . . s
d (1386, 1366), l (1172), l (1157), l (842)
wCH w3 2(sp3) . . . . . . . . . . . . . . . . . . . l s (2958, 2934, 2917),
2 3
l (2853), 4
d (1457)
w(CH2 ) wCH (n \ 1, 2) . . . . . . as
rocking s 749, n b\ 2)
(782, n \ 1 ; 743,
2n
wCHw(sp3) . . 3. . . . . . . . . . . . . . . . . . . l(2898)
Substituted Cycloalkanes . . . . . . Skelet vibrations [n \ 6 : 1026, 1007, 993, 975, 970, 898( ?)] ;
C H [n \ 6, 5, 4, 3( ?)] . . . . . . . . n \ 5 : 939, 935, 898 ( ?), 891 ( ?) ; n \ 4 : 926, 807 ; n \ 3 : 817 ( ?)
n 2n
RwCyCwR@ . . . . . . . . . . . . . . . . . . . l (2167, 2214, 2227, 2239, 2248)
C.C
Mono, o, m, p, 1.2.4-, . . . . . . . . . . . Ring vibrations : 1607, 1593, 1581, d (870, 867 : 1 free H ; 855,
op
1.3.5-substituted 842 : 2 free H), d (1110, 1086, 1073, 1066, 1053, 1044, 1039,
ip
Benzenes 1026, 1039, 1026, 993, 975, 970 : p and 1.2.4. substituted)
xCH (sp2) . . . . . . . . . . . . . . . . . . . . . . l (3090, 3081.7, 3067)
2 CH
xCH(sp2) . . . . . . . . . . . . . . . . . . . . . . . . l (3019)
CH
a R and R@ indicate organic groups.

FIG. 3.ÈFTIR spectra of the 13CH target before (top) and after (middle) a 9 MeV a-particle irradiation at 10 K. Bottom spectrum was recorded at 60 K
after sublimation of 13CH and 13C H 4 (x \ 1, . . . , 6). Absorptions and assignments are compiled in Tables 5 and 6.
4 2 x
 INTERSTELLAR ICES. III. 965
TABLE 5 3.3.1.3. 12CH /CD Targets
4 4
SELECTED ABSORPTIONS OF SOLID Isotopically mixed species are prominent tracers for
13CH AT 10 K recombination and H-D-exchange reactions at 10 K (Paper
4
II). FTIR spectra depict absorption of CH D (563 cm~1),
Wavenumber 2
Vibration (cm~1) CHD (513 cm~1), and CHD (998.2 cm~1 ; 1284.7 cm~1)
2 3
with rising dose. Most important, the mobility of CH ,
l ]l ............ 4283 2
l3 ] l4 . . . . . . . . . . . .
CD , CH , and CD radicals starts between 60 and 75 K,
1 3
4193 2 3 3
3l . . . . . . . . . . . . . . . . 3825 well below our upper temperature limit of 15 K on the
4 target surface. In strong coincidence, no scrambled recom-
bination products C H D (CH ] CD ) and C H D
2 2 2 2 2 2 3 3
(CH ] CD ) were detected, indicating that even reactions
as well as methyl, vinyl, and ethyl radicals after the ion 3 3
of neighboring carbon-containing hydrocarbon radicals are
bombardment (Fig. 4, Tables 7 and 8). In strong contrast to
° 3.3.1.1, however, aliphatic methyl and methylene absorp- not detectable.
tions show less pronounced absorption features. This 3.3.1.4. C H Targets
2 4
Ðnding correlates strongly with our QMS data : only per- The FTIR spectra of the irradiated C H target are dis-
2 4
deutero alkanes up to C D could be probed in the deu- played in Figure 5 and shows modes of ethane, C H , [t
11 24 2 6 9
(822 cm~1), t (1464 cm~1), t (1374 cm~1)], acetylene mol-
terated samples, while hydrocarbons up to C H were
14 30 8 6
ecules, C H , [t (3389 cm~1), t (3271 cm~1), t (736
sampled in the CH system.
4 2 2 1 3 5
TABLE 6
FTIR ABSORPTIONS OF 9 MeV a-PARTICLEÈIRRADIATED 13CH AT 10 K
4
Carrier and Wavenumber
Assignment (cm~1)

Hw13Cy13CwR . . . . . . . . . . . . . . . . . . l (3264, 3251, 3248, 3220)

x13CH (sp2) . . . . . . . . . . . . . . . . . . . . . . . lCH(sp)
(3084, 3077, 3035), l (1643, 1650), d (1417, 1419), d (890, 897)
2 CH C/C ip op
w13CH . . . . . . . . . . . . . . . . . . . . . . . . . . . . l (2969, 2962, 2958), l (2880), d (1464, 1454), d (1361, 1360)
3
w13C(13CH ) w ................... as s
d (1386, 1366), l (1135), l (824, as820, 816) s
32 s 3 4
w13CH w(sp3) . . . . . . . . . . . . . . . . . . . . . l (2951, 2930), d (1433)
w(13CH2 ) w13CH n \ 1, 2 . . . . . . as b
rocking (764 n \ 1 ; 745, n \ 2)
2n 3
w13CHw(sp3) . . . . . . . . . . . . . . . . . . . . . . l(2880)
13Rw13Cy13Cw13R@ . . . . . . . . . . . . . l (2156, 2177, 2087, 2201)
Substituted benzenes . . . . . . . . . . . . . . C.C 1028, 1026, 835, 793
1030,
13CH (carbene) . . . . . . . . . . . . . . . . . . . . . l (1108)
2 2
13CH radical . . . . . . . . . . . . . . . . . . . . . . l (604)
13C H3 radical . . . . . . . . . . . . . . . . . . . . . l2(531)
13C2H5 radical . . . . . . . . . . . . . . . . . . . . . l9(892)
13C2H3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . l7(3261), l (735)
13C2H2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . l3(956) 5
13C2H4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . l7(1460), l (1366), l (820)
2 6 (sp2) . . . . . . . . . . . . . . . . . . . . . . .
x13CH l8 (3076) 6 9
2
x13CH(sp2) ........................ CH
l (3034)
CH

FIG. 4.ÈFTIR spectra of the CD target before (top) and after (middle) a 9 MeV a-particle irradiation at 10 K. Bottom spectrum was recorded at 60 K
after sublimation of CD and C D (x4 \ 1, . . . , 6). Absorptions and assignments are compiled in Tables 7 and 8.
4 2 x
966 KAISER & ROESSLER Vol. 503
TABLE 7 TABLE 9
SELECTED ABSORPTIONS OF SOLID CD AT 10 K SELECTED ABSORPTIONS OF SOLID
4
C H AT 10 K
Wavenumber 2 4
Vibration (cm~1) Wavenumber
Vibration (cm~1)
2l . . . . . . . . . . . . 4479.5, 4477, 4505
3
l ] 2l . . . . . . 4222(]l ), 4217.9(]l ), 4195 l ................... 3113
l3 ] l 4. . . . . . . 1 (]t ), 3287(]t
3236, 3277 1 ) l9 ] l . . . . . . . . . . . . 3073
3 4 L L 2 12
l ]l ....... 3092.2, 3086.6 l .................. 1438
1 4 12
3l . . . . . . . . . . . . 2930, 2926.7, 2925 l ................... 951
4 7
l ............. 2255, 2249, 2238, 2232, 2217.8, 2290(]t )
3 ............
2l 1975, 1973.6, 2002(]t ) L
4 L
l ............. 1000, 998, 996, 993, 989.6, 987.6, 985, 982
4
TABLE 10
TABLE 8 FTIR ABSORPTIONS OF 9 MeV a-PARTICLEÈIRRADIATED C H
2 4
AT 10 K
FTIR ABSORPTIONS OF 9 MeV a-PARTICLEÈIRRADIATED
CD AT 10 K Carrier and Wavenumber
4
Assignment (cm~1)
Carrier and Wavenumber
Assignment (cm~1) HwCyCwR . . . . . . l (3231)
CH(sp)
xCH (sp2) . . . . . . . . . l (3085, 3073), l (1642, 1647),
DwCyCwR . . . . . . l (2395, 2336, 2348, 2340), 2 CHd (1419) d (893)C/C
CD(sp) ip op
wCD . . . . . . . . . . . . . . l (2218), d(1062) wCH . . . . . . . . . . . . . . l (2963), l (2871), d (1472), d (1457),
wCD3w(sp3) . . . . . . las(2116), l (2032, 2028, 2017) 3 asd (1375) s b as
2 as s s
C D ............... l (2232), l (2081) wCH w(sp3) . . . . . . l (2928), l (2852)
C2D6 . . . . . . . . . . . . . . . l7(722), l 5(1069) 2
wCHw(sp3) ....... as
l(2898) s
2 4 7 12
C D ............... l (2427), l (543) RwCyCwR@ . . . . . . l (2167, 2214, 2227, 2239, 2248)
2 2 3 5
C D ...............
2 3
l (705)
7 xCH (sp2) . . . . . . . . . lC.C (3240)
C D ............... l (2197), l (2046) 2 CH
2 5 1 3
CD . . . . . . . . . . . . . . . . l (460), l (2372), l (1024)
3 2 3 4
C D ............... l (847)
2 2 2
TABLE 11
cm~1)], as well as ethyl radicals, C H , [t (542 cm~1)]. SELECTED ABSORPTIONS OF SOLID
2 5CH 9 targets could C H AT 10 K
C H radicals as detected in irradiated 2 2
not be sampled. This Ðnding is consistent 4with previous
2 3
Wavenumber
matrix isolation studies, showing that C H is only stable in Vibration (cm~1)
noble gas and methane matrices. As 2the3 radiation dose
increases, acetylenic, oleÐnic, and even aliphatic structures l .................. 3373
l1 . . . . . . . . . . . . . . . . . . 3270/3272
were probed (see Tables 9 and 10). l3 . . . . . . . . . . . . . . . . . . 1963
3.3.1.5 C H Targets
2 2 l2 ] l . . . . . . . . . . . . 1331
In strong contrast to C H and CH irradiation experi- l4 . . . . .5. . . . . . . . . . . . . 736
l5 . . . . . . . . . . . . . . . . . .
ments, CwH stretching 2 vibrations
4 4 monosubstituted 4
631
of

FIG. 5.ÈFTIR spectra of the C H target before (top) and after (middle) a 9 MeV a-particle irradiation at 10 K. Bottom spectrum was recorded at 75 K
2 H4 . Absorptions and assignments are compiled in Tables 9 and 10.
after sublimation of the remaining C
2 4
No. 2, 1998 INTERSTELLAR ICES. III. 967

FIG. 6.ÈFTIR spectra of the C H target before (top) and after (middle) a 9 MeV a-particle irradiation at 10 K. Bottom spectrum was recorded at 75 K
2 H
after sublimation of the remaining C 2 . Absorptions and assignments are compiled in Tables 11 and 12.
2 2

acetylenic and oleÐnic species dominate the FTIR spectra methyl radicals to carbene and atomic hydrogen as shown
(Fig. 6, Tables 11 and 12). As already established in QMS previously (Paper II) :
analysis, the reduced H : C ratio of 1 as compared to 2 in the
CH ] CH ] H . (5)
ethylene target hinders formation of higher alkanes. Like- 3 2
wise, no hydrogen rich species C H and C H were probed In addition, synthesized closed shell molecules acetylene,
except weak absorptions of C H2 . 6 2 5
C H , ethylene, C H , and ethane, C H , as well as ethyl
2 4
3.3.2. Quantitative Analyses (C2 H2 ) and vinyl 2 (C4 H ) radicals and2 6 their perdeutero
2 5 2 3
analogs contribute 86%È91% to all newly formed molecules
Figures 7È11 compile the temporal development of inte- containing at least two carbon atoms. Here the production
grated absorptions of selected CH(D) (x \ 2, 3, 4) and rates (Tables 13È15) give valuable insights in the formation
C H(D) (x \ 2, 3, 4, 5, 6) species during x the irradiation at
2 x mechanism of these C2 species. As expected from a
10 K and consecutive equilibration period at 10 K. Here- hydrogen-rich target (C : H \ 1 : 4), fully saturated alkanes,
after all molecules/radicals containing two carbon atoms in this case ethane, hold large production rates of C2
will be referred to as C2 species. Considering these graphs, species with 670 ^ 270 synthesized C H molecules per 9
we calculate the destruction rates of the CH /CD mol- 2 6 followed by the
MeV a-projectile into the CH target
ecules, as well as production rates of selected4 molecules/
4 4
oleÐne C H (280 ^ 150) and the acetylenic C2 species
radicals during the early irradiation stage. 13CH -irradiated 2 4
targets show the same temporal concentration 4 proÐle
within the error limits of 10%È20% as compared to the TABLE 13
12CH system and hence are not included. PRODUCTION AND DESTRUCTION (CH ) RATES OF SELECTED SPECIES
4 10 K OF a/CH , p/CH ,
The4 response of the CH and CD systems upon the DURING MeV ION BOMBARDMENT AT
4 4
MeV particle bombardment 4is governed4 by the formation of
AND a/CD SYSTEMS
4
methyl, CH , and carbene, CH , radicals as well as their Target j/Species i a/CH p/CH a/CD
3 2 4 4 4
deuterated counterparts as the result of a dominating elec-
CH . . . . . . . . . . . . . . [1320 ^ 300 [110 ^ 15 [1040 ^ 310
tronic energy loss (see ° 3.1.2), and secondary radiolysis of CH4 . . . . . . . . . . . . . . 320 ^ 150 37 ^ 15 60 ^ 40
C H3 . . . . . . . . . . . . . 670 ^ 270 122 ^ 62 160 ^ 110
C2H6 . . . . . . . . . . . . .
TABLE 12
280 ^ 150 32 ^ 13 42 ^ 20
FTIR ABSORPTIONS OF 9 MEV a-PARTICLEÈIRRADIATED C H AT 10 K C2H4 . . . . . . . . . . . . . 87 ^ 60 11 ^ 3 13 ^ 10
2 2 2 2
Carrier and Wavenumber
Assignment (cm~1)
TABLE 14
HwCyCwR . . . . . . . . . . . . . . . . . . . . l (sp)(3283), 3281, 3229, 3219), RATIO OF PRODUCTION RATES OF C2-SPECIES
CHl (2137, 2116, 2110, 2087) CALCULATED FROM TABLE 13
dC/C.(642)
CH xCxCRR@ . . . . . . . . . . . . . . . . . l HhC (1926, 1918) C2-Species Ratio
2 ............................
xCH lC/C/C(3084), dip(1407, 1418)
2
xCH(aromatic/sp2) ............. lCH(sp2)
(3040-30), l (1643, 1650), k(C H , a/CH )/k(C H , p/CH ) . . . . . . 9^7
CHd (1417, 1419),
C/Cd (890, 897)
k(C2H4,
6 a/CH 4)/k(C 2H 6, p/CH 4) . . . . . .
12 ^ 10
ip 1280, 1265, 1233,
op 1093, 1081,
k(C H , a/CH4 )/k(C2 H4 , p/CH4 ) . . . . . .
2
di-, tri- and tetra-substituted . . . 1365, 10 ^ 8
benzenes 1036, 1004, 997, 975, 929, 855,756 k(C2H2 , a/CH4)/k(C2D2 , a/CD4) . . . . . . 10 ^ 8
CH xCxCHR@ . . . . . . . . . . . . . . . . l (1958) k(C2H4 , a/CH4)/k(C2D6 , a/CD4) . . . . . . 11 ^ 9
CH,2 sp3 ( ?) . . . . . . . . . . . . . . . . . . . . . . . C/C/C ; 2984, 2933, 2924
3000-2700 k(C2H4 , a/CH4)/k(C2D4 , a/CD4) . . . . . . 25 ^ 24
2 2 4 2 2 4
968 KAISER & ROESSLER Vol. 503

FIG. 7.ÈTemporal development of integrated absorptions of selected species in the p/CH system : (a) CH (t ] t ) ; (b) CH (t ) ; (c) CH (t ) ;
4 4 1 4 2 2 3 2
(d) C H (t ) ; (e) C H (t ) ; ( f ) C H (t ), C H (t ), C H (t ).
2 3 7 2 5 9 2 6 8 2 4 7 2 2 5

C H (87 ^ 60). Comparing these numbers with those in the MARLOWE calculations reveal a dominant inelastic
2 p/CH
the 2 system, we Ðnd qualitatively the same trend, i.e., energy loss, the formation mechanism of the C2 species is
4 rates of C H [ C H [ C H . However, the
production expected to be strongly correlated with this energy transfer
2 6of ethane,
2 4 ethylene,
2 2 and acetylene
absolute production rates process (see ° 4). Finally, it is worth mentioning that the
are about a factor of 10 larger in the a/CH system than in production rates of the C2 species in the a/CH system is
4
the second one (p/CH ). This order of magnitude correlates larger by 1 order of magnitude as compared to4 irradiated
4
strongly with the ratio of the linear energy transfers (LET) CD samples. This Ðnding very likely points to an isotope
of both systems, LET(a/CH )/LET(p/CH ) \ 10 ^ 1. Since 4 in the formation of ethane, ethylene, and acetylene as
e†ect
4 4
No. 2, 1998 INTERSTELLAR ICES. III. 969

FIG. 8.ÈTemporal development of integrated absorptions of selected species in the a/CH system : (a) CH (t ] t ) ; (b) CH (t ) ; (c) CH (t ) ;
4 4 1 4 2 2 3 2
(d) C H (t ) ; (e) C H (t ) ; ( f ) C H (t ), C H (t ), C H (t ).
2 3 7 2 5 9 2 6 8 2 4 7 2 2 5

already documented in an enhanced production of H A quantitative analysis of the C H system reveals an

2
versus D and CH versus CD in these targets. enhanced destruction rate of the C 2H 2 molecules as com-
2 3 3
pared to the CH system, i.e., 4030 2^ 830
2 and 1320 ^ 300,
Compared to the a/CH system, the production rates of
C H and C H in hold 4the same magnitude within the
4
and only a minor production rate of 3 ^ 2 C H molecules.
2 6 2error
2 limits in irradiated C H targets, i.e.,
This pattern correlates with formation of a2 solid
4
experimental residue
240 ^ 130 and 540 ^ 350 (a/C H ) versus2 670
4 ^ 270 and
upon annealing the C H target to 293 K. Neither CH nor
87 ^ 60 (a/CH ). 2 4 2 2 at identical doses, pressures,
C H samples irradiated 4
4 2 2
970 KAISER & ROESSLER Vol. 503

FIG. 9.ÈTemporal development of integrated absorptions of selected species in the a/CD system : (a) CD (2t ) ; (b) CD (t ) ; (c) CD (t ) ; (d) C D (t ) ;
4 4 2 2 2 3 2 2 3 7
(e) C D (t ) ; ( f ) C D (t ), C D (t ), C D (t ).
2 5 9 2 6 5 2 4 7 2 2 3

LETs, and temperatures show these residues at room tem- (Paper II). Since the LET in all systems is governed by
perature (Paper I), indicating that a critical C : H ratio is inelastic energy transfers to the solid target, the exciton
essential to form these residues. formation dominates followed by a decay to [CH . . . . H]
4. DISCUSSION and [CD . . . . D] radical pairs. This pair is matrix3isolated
3
by surrounding CH molecules, and its fate determined by
4.1. Formation of Molecules with one Carbon Atom 4 energy of the H/D atom. If the di†u-
the remaining kinetic
The synthesis of CH and CH radicals is strongly corre- sion barrier can be passed, the radical pair separates to
lated to the formation3of atomic2 and molecular hydrogen mobile H/D atoms but nonmobile CH radicals. Otherwise,
3
No. 2, 1998 INTERSTELLAR ICES. III. 971

FIG. 10.ÈTemporal development of integrated absorptions of selected species in the a/C H system : (a) C H (t ] t ) ; (b) C H (t )
2 2 2 2 2 4 2 4 12

a reverse reaction to CH takes place. These processes are carbon atoms prior to reaction explicitly. The resonance
4
well documented by detection of HD as a recombination rule, however, predicts that only carbon atoms in their 3P
product of H and D atoms as well as probing CHD mol- and 1D states are formed ; potential carbon ions are neu-j
ecules formed thru recombination of CD and H atoms 3 in tralized2before chemical forces take over (StoŽcklin 1969, and
3
CH /CD targets at doses D [ 9 eV. Likewise, the references therein). The insertion of a carbon atom into the
4
increased4production rates of CH versus CD of 320 ^ 150 CwH bond of a single CH molecule to internally excited
3 with a higher
3 H yield as 4 ]* (reaction 1, Fig. 12), rep-
and 60 ^ 40, respectively, together methylcarbene [HwCwCH
compared to D underline the role of an enhanced2 H atom resents the primary reaction 3step in strong analogy to gas-
2
di†usion coefficient versus D atoms. Here the separation of phase reactions studied (StoŽcklin 1969 ; Dubrin, McKay, &
the [CH . . . . H] radical pair to methyl radicals and atomic Wolfgang 1964 ; McKay & Wolfgang 1961). C(3P ) reacts to
j singlet
hydrogen3 is facilitated owing to the lower mass of the H triplet methylcarbene, while C(1D ) leads to the
atom. To a minor amount, secondary dissociation of CH 2
counterpart. As expected from a short lifetime of less than
3
forms H and CH . 1 ps, no reactive methylcarbene intermediate could be
2
sampled via FTIR spectroscopy. Methylcarbene undergoes
4.2. Formation of Molecules with T wo Carbon Atoms H elimination (singlet methylcarbene ; reaction 2) or
2
fragments to two H atoms (triplet carbene ; reaction 3) to
4.2.1. C H
2 2
form an acetylene molecule. Here the key role of H
Since the experiments were performed under bulk condi- atoms/H molecules is well documented in terms of an
tions, we cannot probe the spin state of the suprathermal 2
isotope e†ect in an enhanced production rate of an C H
2 2
versus C D of 25 ^ 24. Since no isotopically scrambled
2 2
HCCD was detected in CH /CD irradiated samples,
recombination of CH and CD4 to form
TABLE 15 4
acetylene can be
PRODUCTION AND DESTRUCTION RATES OF SELECTED SPECIES DURING clearly ruled out.
MEV ION BOMBARDMENT AT 10 K OF a/CH , a/C H , AND
a/C H SYSTEMS 4 2 4 4.2.2. C H
2 2 2 4
In the solid state, methylcarbene can transfer its excess
Target j/Species i a/CH a/C H a/C H
4 2 4 2 2 energy to the surrounding matrix and thermalizes (reaction
CH . . . . . . . . . . . . . . [1320 ^ 300 ... ... 4) ; alternatively, singlet methylcarbene can undergo
C H4 . . . . . . . . . . . . . 670 ^ 270 240 ^ 130 0 [1, 2]-H-migration to form ethylene, reaction 5 which can
C2H6 . . . . . . . . . . . . . 280 ^ 150 [320 ^ 95 3^2 be stabilized itself to form a thermal ethylene molecule,
C2H4 . . . . . . . . . . . . . 87 ^ 60 540 ^ 350 [4030 ^ 830
2 2 reaction 6. Previous isotope studies showed that a two-
972 KAISER & ROESSLER Vol. 503

undergoes either CwH bond rupture to C H via reaction

2 4
11 or thermalized to C H radicals (reaction 12). The
2 5
crucial role of the H atom in the formation of C H is
2 4
expressed in isotope e†ects as found in our experiments : the
ratio of the ethylene production rates in CH compared to
4
CD holds 11 ^ 9 and can be explained in terms of a
4
reduced rate constant in the rearrangement of methyl-
carbene to ethylene (reaction 5) and/or through a hindered
di†usion of D atoms versus H to form C D /C H
2 4recom-
(reaction 8). Finally, we would like to point out that 2 4
bination of two CH radicals to form an ethylene molecule
does not contribute2 since no isotopically mixed H CCD
2 2
(CH ] CD ) could be sampled. This Ðnding strongly cor-
2 2
relates to an onset of a CH di†usion at 60 K as compared
2
to our target temperature of 15 K.
4.2.3. C H
2 6
The synthesis of C H can proceed through insertion of
suprathermal carbene2 formed
6
via reaction 13 into a CwH
bond of a methane molecule followed by thermalization
(reactions 14È15). An H atom addition to C H can
contribute to ethane as well (reaction 16). Here2 the
5 iso-
topic e†ect expressed in the ratio of the production rates of
C H versus C D , i.e., 10 ^ 8, strongly indicates a prefer-
2 6 H atom 2addition
6
ential to C H since the suprathermal
pathway via CH insertion is2 not
5 expected to show an
2
isotope e†ect. In strong coincidence with these results and
°° 4.2.1 and 4.2.2, no recombination radical-radical recom-
bination could be sampled, here of two CH radicals to
C H . Again, the di†usion of CH starts at 3about 70 K,
FIG. 11.ÈTemporal development of integrated absorptions of selected
species in the a/C H system : (a) C H (t ) ; (b) C H (t ) ; (c) C H (t ) ; 2 6 higher than our 14 K target. 3
(d) C H (t ). 2 4 2 4 7 2 2 5 2 5 9 much
2 6 8
The inability of the MeV particle irradiation of C H to
step process via CwH bond rupture in methylcarbene to H synthesize C H molecules at 10 K and the extreme 2 2low
and vinyl radicals (reaction 7) followed by H atom addi- 2 6
production rate of C H is well reÑected in the C : H ratio of
tion can synthesize C H as well (reaction 8 ; Taylor, one in our sample. 2This 4 trend is consistent with reduced
2 4
Ache, & Wolf 1975). In addition, suprathermal CH radicals production rates k of C H and C H in irradiated, carbon-
rich samples, i.e., k(C2 H6 //a/CH2 ) [
formed upon H atom abstraction by suprathermal carbon 4 k(C H //a/C H ) [
2 6 4 2 [6 0 2 4and
atom via reaction 9 can insert into a CH methane bond to k(C H //a/C H ) \ 670 ^ 270 [ 240 ^ 130
2 6 2 2
build up a C H radical (reaction 10). This intermediate k(C H //a/CH ) [ k(C H //a)C H ) \ 280 ^ 150 [ 3 ^ 2.
2 5 2 4 4 2 4 2 2

FIG. 12.ÈCompilation of mechanisms leading to formation of C H species (x \ 2, 3, 4, 5, 6)

2 x
No. 2, 1998 INTERSTELLAR ICES. III. 973

Data analyses of the alkane distribution support this con- eV, much smaller than the threshold dose to overlapping
clusion as well, and higher alkanes C H to C H were collision cascades. This Ðnding strongly suggests that
only synthesized in the C H system. 12 26 14 30 alkanes and cycloalkanes are formed inside a single colli-
2 4 sion cascade at lower doses and not through a successive
4.2.4. C H /C H
2 3 2 5 insertion of thermal carbene, CH , into aliphatic CwH
Besides pathways already outlined in °° 4.2.1È4.2.3, H 2
bonds like the reaction sequence
atom addition to in situ synthesized C H and C H con-
2 2 2 4 CH ] CH ] CH wCH , (6)
tributes to C H (reaction 17) and C H (reaction 8), 2 4 3 3
2 3 2 5
respectively. These pathways have been conÐrmed in our CH wCH ] CH ] CH wCH wCH , (7)
experiments bombarding solid C H and C H at 10 K 3 3 2 3 2 3
2 2 2 4 CH wCH wCH ] CH ] i/nwC H , etc . (8)
with 9 MeV a-particles. Both reactions, however, hold
3 2 3 2 4 10
entrance barriers of about 10.5 and 11.5 kJ mol~1, respec-
As shown in ° 3, the alkane formation correlates with the
tively. Recalling that our target temperature is less than 15
temperature proÐle of the irradiated target, and per-deutero
K and that 100 kJ mol~1 is equivalent to about 10,000 K,
undecane, C D , is predominantly formed in target zones
only suprathermal H/D atoms react via reactions 17 and 8. 11 24
heated to 15 K, while lower alkanes such as per-deutero
4.3. Formation of Molecules with T hree Carbon Atoms butanes, C D , are preferentially synthesized in cooler
4 10
The mechanism to form linear C3 species methyl- regions. Since the di†usion coefficient of the D atoms and
acetylene, CH CCH, allene, H CCCH , and propane, can D molecules are enhanced by a factor of 3È5 in the 15 K
3 recent crossed 2 molecular
2 2
regions as compared to 10 K zones, an increasing D recom-
be inferred from beam experi-
ments of thermal atomic carbon C(3P ) with C H (Kaiser, bination to D and a di†usion to the vacuum-CD surface
j 2Here
4 C(3P ) 2 4
Lee, & Suits 1996, and references therein). boundary followed by a D release into the gas-phase
2
reacts with C H under addition to the n-bond to form ja results in a carbon enrichment in these outer 15 K layers.
2 4 Therefore, a multicenter reaction mechanism, similar to the
reactive cyclopropylidene intermediate (reaction 1 Fig.
13). This internally excited intermediate undergoes ring formation of polycyclic aromatic hydrocarbons (PAHs ;
opening to allene (reaction 2). Since the crossed beam Paper I), could contribute to alkanes and cycloalkanes.
experiments were performed under single collision condi- Detailed three-dimensional collision cascades indicate
tions, the allene intermediate cannot be stabilized in a third ““ tube-like ÏÏ energized zones inside the irradiated target,
body reaction and fragments to atomic hydrogen and a where this synthesis might take place. At higher doses and
propargyl radical, C H . In our solid state experiments, overlapping cascades, however, a stepwise synthesis
3 3 of allene can be transferred to through equations (6)È(8) might contribute to alkane/
however, this excess energy
the surrounding CH matrix, resulting in a ““ thermal ÏÏ cycloalkane formation as well. Here, however, the inserting
4 3. Further, H migration in allene CH must be suprathermal since no di†usion of thermal
allene molecule, reaction 2
to methylacetylene followed by thermalization could carbene at 10 K takes place (see ° 4.2.2).
account for methylacetylene molecules, reaction 4. Since the
energy of reactive C atoms in the present experiments is up 4.5. T he Source of Suprathermal H and C Atoms
to 20 times higher as compared to the crossed beam reac- Based on our MARLOWE calculations, the elastic
tions, insertion in a single CwH bond of ethylene forming energy loss generates only 0.22 ^ 0.04 H/0.07 ^ 0.02 C (7.3
vinylcarbene (reaction 5) followed by H migration can con- MeV H`) and 3.17 ^ 0.52 H/1.15 ^ 0.25 C atoms (9.0 MeV
tribute to allene (reaction 6) and/or methylacetylene a) per impinging MeV particle in a 5 km CH target. As
4 C atoms
(reaction 7) as well. Finally, the propane molecules are very outlined in the previous paragraphs, suprathermal
likely formed through two successive H addition reactions are crucial to synthesize C2 species ; thermal radical recom-
to C H intermediates or insertion of suprathermal CH . bination were found to play no role in these processes. A
3 we4 would like to point out that we neglect the dis- 2 comparison of the number of knock-on carbon atoms with
Again,
cussion of di†erent spin states and a potential role of inter- the sum of synthesized acetylene, ethylene, and ethane mol-
system crossing in the solid state since our MARLOWE ecules, i.e., 1037 ^ 340 (9.0 MeV a) and 165 ^ 78 (7.3 MeV
calculations give no explicit information on the spin dis- H`), clearly demonstrate that a second source of supra-
tribution. thermal C atoms besides direct knock-on particles must
exist to account for the discrepancy between theory and
4.4. Formation of Molecules with Four and More experiment.
Carbon Atoms Hitherto, we contributed only to the elastic energy trans-
Our FTIR data clearly indicate an onset of aliphatic fer generating suprathermal knock-on atoms and neglected
wCH and wCH group absorption at doses as low as 0.1 inelastic processes. As mentioned above, the energy loss to
3 2

FIG. 13.ÈCompilation of mechanisms leading to formation of C H species (x \ 4, 6, 8)

3 x
974 KAISER & ROESSLER Vol. 503

the electronic system of the target and secondary electrons CH /CD model compounds could hold far-reaching con-
4 4
ionize single CH molecules within a radius of about 50 Ó sequences to ““ realistic ÏÏ interstellar ices composed of H O,
2
perpendicular to4 the trajectory of the implanted ion. The CO, CH OH, NH , H S, CH , H CO, OCS, OCN~, H ,
3 3 2 4 2 2
stored potential energy, here via Coulomb interaction, can and CO (see ° 1). As a typical example, we consider Ðrst the
2
be released into kinetic energy of CH` fragments such as postulated synthetic route to interstellar methanol,
4
C`, CH`, as well as CH` (Roessler 1991) to result into CH OH, molecules on grain surfaces. Models assume a
2 3
additional, suprathermal species. As quoted above, these methanol synthesis through four successive H addition
ions are neutralized prior to reaction and very likely rep- steps (reactions [11]È[14]) through the hydroxyformyl
resent the missing suprathermal C source. radical (11), formaldehyde/hydroxycarbene (12), and
hydroxymethyl (13) on interstellar grains (Schutte et al.
4.6. Comparison to Photolysis of CH T argets
4 1996, and references therein) :
A comparison of our MeV experiments with a recent
study of UV photon processing of CH ices at 10 K by CO ] H ] HCO , (11)
Gerakines, Schutte, & Ehrenfreud (1997) 4 o†ers further HCO ] H ] H CO/HCOH , (12)
insights in the crucial role of suprathermal C atoms in 2
hydrocarbon rich ices. These UV irradiation experiments H CO/HCOH ] H ] CH OH , (13)
2 2
indicate that 38% of the CH was converted to methyl CH OH ] H ] CH OH . (14)
4 2 3
radicals (0.01%), C2 species ethylene (2.6%) and ethane
(8%), the C3 species propane, and higher hydrocarbons Recent laboratory work, however, indicates reaction (12) is
holding structural units RwCH , RwCH wR, too slow to form signiÐcant formaldehyde (Hiraoka et al.
3 where ““ R ÏÏ2indi- 1994). Suprathermal chemistry could resolve this puzzle :
R CxCH , R CxCRH, as well as HCCR
2 an organic
2 2 group. Most important neither acetylene, Galactic cosmic-ray particle induced nonequilibrium O and
cates
C H , nor vinyl radicals, C H , could be sampled. Since, C atoms released in interstellar ice composita can insert
2 2 2 3formed in our MeV irradia- into a CwH bond of CH (eq. [15] below) and a OwH
however, C H molecules were 4 [16] below) to form an inter-
2 2 bond of a H O molecule (eq.
tion through insertion of suprathermal carbon atoms and 2
C H via addition of H atoms to C H , we must conclude nally excited (*) methanol and formaldehyde as found in
2 3UV exposure of CH cannot generate2 2 laboratory experiments (Schmitz, Nebeling, & Roessler
that suprathermal C
4 1987 ; Nebeling 1988) :
atoms. Here UV photons interact in single quantum pro-
cesses with a CH molecule and are unable to ignite colli- O ] CH ] [CH OH*] ] CH OH , (15)
sion cascades to 4form suprathermal C atoms. In strong 4 3 3
coincidence, Gerakines and coworkers postulated the for- C ] H O ] [HCOH*] ] [H CO*] ] H CO . (16)
2 2 2
mation of ethylene and ethane through reaction of UV Since the excess energy can be released via phonon inter-
photon-induced CH and CH radicals : action with the solid target, the suprathermal pathway pre-
2 3
CH ] CH ] C H , (9) sents a temperature-independent, one-step reaction to form
2 4 2 6 H CO as well as CH OH even at temperatures as low as 10
CH ] CH ] C H , (10) K.2 This approach is 3extendible to higher hydrocarbons as
2 2 2 4
well, and insertion of suprathermal O into a CwH bond of
2 CH ] C H . (11)
3 2 6 ethane, C H , gives excited ethanol, C H OH, which could
2 6
relax via phonon interaction as well : 2 5
Since CH as well as CH are not mobile at 10 K, reactions
2 3 to neighboring radicals in the
(9)È(11) must be restricted
O ] C H ] [C H OH*] ] C H OH . (17)
UV-processed CH targets. 2 6 2 5 2 5
4 This solid state oxygen and carbon chemistry could rep-
5. ASTROPHYSICAL IMPLICATIONS resent a strong alternative to hitherto assumed di†usion
Our experiments demonstrate explicitly that MeV ionÈ controlled reactions on interstellar grain composita (Millar,
induced suprathermal carbon atoms and nonequilibrium Herbst, & Charnley 1991). Since the cosmic-ray particles
chemistry inside a single collision cascade present a power- induce not only suprathermal C and O atoms, but N and S
ful tool to synthesize a broad product spectrum from atoms in interstellar ice composita as well, these supra-
atomic and molecular hydrogen, H and H , via C2 species thermal species react with molecules in the solid state, and
C H , C H , C H , C H , and C H 2 , C species
an unprecedented product variety can be expected. It is
CH CCH, H CCCH , C H , and C H , as 6well as
2 2 2 3 2 4 2 5 2 3 alkanes
worth mentioning that the cosmic-ray particle penetrates
3 2 2 3 6 3 8
and polycyclic aromatic hydrocarbons (PAHs) as complex the micrometer-sized interstellar grain core, and supra-
as coronene. The synthetic power of the implanted ions is thermal Fe, Mg, and Si atoms will be generated. These
even stronger as compared to UV photons since the last one species o†ers unique synthetic power to form
cannot generate knock-on carbon atoms as well as collision organometallic compounds. Once these molecules are
cascades. Therefore, reactions in UV-processed samples are formed on interstellar grains through suprathermal chem-
dominated by di†usion-limited reactions, e.g., equations istry, equilibrium sublimation through heating of grains
(9)È(11), between neighboring radicals at 10 K. Supra- and their icy mantles by young stellar object embedded
thermal H atoms, however, can be generated in UV- inside dense clouds, as well as explosive grain ejections of
irradiated samples as well considering a CwH bond energy grains storing a critical concentration of radicals redistrib-
of 4.4 eV in CH , and a UV photon of, e.g., 6.0 eV. ute these newly formed molecules from the grains into the
The versatile4concept of suprathermal chemistryÈinduced gas phase (dÏHendecourt et al. 1982).
formation of hydrocarbon molecules via insertion and addi- Even an order of magnitude calculation shows that
tion of non-equilibrium C, CH, and CH species as well as although the internal UV Ñux in dense, molecular clouds
H abstraction processes elucidated from 2MeV irradiation of (/ \ 103 cm~2 s~1) ranges 2 orders of magnitude higher
No. 2, 1998 INTERSTELLAR ICES. III. 975

than the cosmic-ray particle Ñux distribution maximum of important, the experimental identiÐcation of acetylene,
/ \ 101 cm~2 s~1 for 1È10 MeV protons and a-particles, C H , and ethane, C H , formed via suprathermal chem-
2 2 2 6
each MeV particle can generate about 100 suprathermal istry could resemble a pathway to form acetylene and
species in a 0.2 km thick icy layer and 0.3 km thick grain ethane found in comet C/1996 B2 Hyakutake (Mumma et
core. The Ñux advantage of the internal UV Ðeld deep inside al. 1996 ; Mumma 1996 ; Hudson & Moore 1997).
molecular clouds is clearly eliminated by the ability of a
single MeV particle to induce collision cascades with up to The authors want to thank the cyclotron crew of the
100 suprathermal species. Therefore, an opening from compact cyclotron CV 28 at the Forschungszentrum JuŽlich
““ photon-only universe ÏÏ to process interstellar grains to for performing the irradiation and G. StoŽcklin, director of
include MeV ion synthesis is strongly encouraged. Institut fuŽr Nuklearchemie, for Ðnancing the experimental
setup. R. I. K. thanks especially P. Jung, Institut fuŽr Fest-
6. SOLAR SYSTEM IMPLICATIONS koŽrperchemie, Forschungszentrum JuŽlich, for switching
In our solar system, solid CH was identifed unam- irradiation time at the compact cyclotron CV-28, the Deut-
4
biguously on Triton, Uranus, Neptune, Pluto and its moon sche Forschungsgemeinschaft (DFG) for Ðnancial support
Charon. Therefore, MeV ions as well as keV projectiles to present parts of this work on Radiation-E†ects in Insula-
originating in the solar radiation Ðeld holding a Ñux / \ 2È tors (REI-7, Nagoya, Japan), and B. Krebs, WestfaŽlische
3 ] 108 cm~2 s~1 at 1 AU (AU \ distance between Earth Wilhelms UniversitaŽt MuŽnster, for administrative guidance
and Sun) are expected to modify these ices as well. Most and advice.

REFERENCES
Addepalli, V. B., & Rao, N. R. 1976, Indian J. Pure Appl. Phys., 14, 117 Lanzerotti, L. J., & Johnson, R. E. 1987, in Ion Beam ModiÐcations of
Alksne, Z. K., Alksnis, A. K., & Dzervits, U. K. 1991, Properties of Galactic Insulators, ed. P. Mazzoldi & E. Arnold (Amsterdam : Elsevier), 631
Carbon Stars (Malabar : Orbit Books) MacKay, C., & Wolfgant, R. 1961, J. Am. Chem. Soc., 83, 2399
Allamandola, L. J., Sandford, S. A., & Valero, G. J. 1988, Icarus, 76, 225 Millar, T. J., Herbst, E., & Charnley, S. B. 1991, ApJ, 369, 147
Bellamy, L. J. 1968, Advances in Infrared Group Frequencies (London : Moshammer, R. 1990, Ph.D. thesis, Fachhochschule Darmstadt
Chapman & Hall) Mumma, M. J. 1996, Nature, 383, 581
Brown, W. L., Lanzerotti, L. J., & Johnson, R. E. 1982, Science, 218, 525 Mumma, M. J., et al. 1996, Science, 272, 1310
Chiar, J. E., Adamson, A. J., Kerr, T. H., & Whittet, D. C. B. 1994, ApJ, Nebeling, B. 1988, Rep. JuŽlich 2245
426, 240 Nyguist, I. M., Mills, I. M., Person, W. B., & Crawford, B. 1957, J. Chem.
Cowieson, D. R., Barnes, A. J., Orville-Thomas, W. J., & Ram, J. 1981, Phys., 26, 552
Spectrosc., 10, 224 Palumbo, M. E., Tielens, A. G. G. M., & Tokunaga, A. T. 1995, ApJ, 449,
Cowley, C. R. 1985, An Introduction to Cosmochemistry (Cambridge : 674
Cambridge Univ. Press) Pearl, J., Ngoh, M., Ospina, M., & Khanna, R. 1991, J. Geophys. Res., 96,
dÏHendecourt, L. B., & Allamandola, L. J. 1986, A&AS, 64, 453 17477
Dubrin, J., MacKay, C., & Wolfgang, R. 1964, J. Am. Chem. Soc., 86, 959 Robinson, M. T. 1990, Nucl. Instrum. Meth., B67, 396
Geiss, J., et al. 1992, in COSPAR Colloq. Ser., Vol. 3, ed. E. Marsch & Roessler, K. 1991, in Solid-State Astrophysics, ed. E. Bussoletti & G. Straz-
R. Schwenn (New York : Pergamon), 20 zulla (Amsterdam : North-Holland)
Gerakines, P. A., Schutte, W. A., & Ehrenfreud, P. 1996, A&A, 312, 289 Sche†ler, H., & ElsaŽsser, H. 1988, Physics of the Galaxy and Interstellar
Greenberg, J. M. 1971, A&A, 12, 240 Matter (Berlin : Springer)
Grigorev, E. I., Pshezhetskii, S. Y., & Trakhtenberg, L. I. 1985, Khimila Schmitt, B. 1994, in AIP Conf. Proc., Molecules and Grains in Space, ed.
Vysika Energika, 19, 41 I. Nenner (New York : AIP), 735
Grim, R. J. A., & Greenberg, J. M. 1987, ApJ, 321, L91 Schmitz, G., Nebeling, B., & Roessler, K. 1987, Rep. JuŽlich 2119
Grim, R. J. A., Greenberg, J. M., de Groot, M. S., Baas, F., Schutte, W. A., Schutte, W. A. 1998, in The Cosmic Dust Connection, ed. J. M. Greenberg
& Schmitt, B. 1989, A&AS, 78, 191 (New York : AIP), in press
Heyl, M. 1990, Rep. JuŽl-2409 Schutte, W. A., et al. 1996, A&A, 309, 633
Hiraoka, K., et al. 1994, Chem. Phys. Lett., 229, 408 Schutte, W. A., Gerakines, P. A., Geballe, T. R., van Dishoeck, E. F., &
Hollenbach, D. J., & Thronson, H. A. 1987, Interstellar Processes Greenberg, J. M. 1996, A&A, 309, 633
(Dordrecht : Reidel) Schutte, W. A., & Greenberg, J. M. 1997, A&A, in press
Hudson, R. L., & Moore, M. H. 1997, Icarus, in press ÈÈÈ. 1991, A&A, 244, 190
Johnson, R. E. 1990, Energetic Charged Particle Interactions with Atmo- Shalabiea, O. M., & Greenberg, J. M. 1994, A&A, 290, 266
spheres and Surfaces (Berlin : Springer) Simpson, J. A. 1983, in Introduction to the Galactic Cosmic Radiation, ed.
Johnson, R. E., Lanzerotti, L. J., & Brown, W. L. 1984, Adv. Space Res., 4, R. Behrisch (Amsterdam : Reidel), 1
41 Smit, W. M. A., van Stalen, A. J., & Visser, T. 1978, J. Mol. Struct., 48, 177
Kaiser, R. I., Gabrysch, A., Eich, G., & Roessler, K. 1997, ApJ, 484, 487 Snelson, A. 1970, J. Phys. Chem., 74, 537
(Paper II) StoŽcklin, G. 1969, Chemie heier Atome (Weinheim : Verlag Chemie)
Kaiser, R. I., Gabrysch, A., & Roessler, K. 1995, Rev. Sci. Instrum., 66, 3058 Taylor, K. K., Ache, H. J., & Wolf, A. P. 1975, J. Am. Chem. Soc., 97, 5970
Kaiser, R. I., Jansen, P., Petersen, K., & Roessler, K. 1995, Rev. Sci. Tegler, S. C., Weintraub, D. A., Rettig, T. W., Pendleton, Y. J., Whittet,
Instrum., 66, 5226 D. C. B., & Kulesa, C. A. 1995, ApJ, 439, 279
Kaiser, R. I., Lee, Y. T., & Suits, A. G. 1996, J. Chem. Phys., 105, 8705 Tielens, A. G. G. M., & Allamandola, L. J. 1987, in Interstellar Processes,
Kaiser, R. I., & Roessler, K. 1997, ApJ, 475, 144 (Paper I) ed. D. J. Hollenbach & H. A. Thronson (Dordrecht : Reidel), 397
Katz, R., Ackerson, B., Hormayoonfar, M., & Sharma, S. C. 1971, Radi- Tielens, A. G. G. M., Tokunaga, A. T., Geballe, T. R., & Baas, F. F. 1991b,
ation Res., 47, 402 ApJ, 381, 181
Kondo, S., & Saeki, S. 1973, Spectrochim. Acta, A29, 1753 Whittet, D. C. B. 1993, in Dust and Chemistry in Astronomy, ed. T. J. &
Lacy, J. H., Baas, F., Allamandola, L. J., Persson, S. E., McGregor, P. J., D. A. Williams (Bristol : IOP Publ. Ltd.), 1
Lonsdale, C. J., Geballe, T. R., & van de Bult, C. E. P. 1984, ApJ, 276, Wormhould, J., & McCurdy, K. E. 1989, Chem. Phys. Lett., 156, 47
543