APEX 1 mm line survey of the Orion Bar

Abstract

Unbiased molecular line surveys are a powerful tool for analyzing the physical and chemical parameters of astronomical objects and are the only means for obtaining a complete view of the molecular inventory for a given source. The present work stands for the first such investigation of a photon-dominated region. The first results of an ongoing millimeter-wave survey obtained towards the Orion Bar are reported. The APEX telescope in combination with the APEX-2A facility receiver was employed in this investigation. We derived the physical parameters of the gas through LVG analyses of the methanol and formaldehyde data. Information on the sulfur and deuterium chemistry of photon-dominated regions is obtained from detections of several sulfur-bearing molecules and DCN. Comment: APEX A&A special issue, accepted

Full text

Page 1

arXiv:astro-ph/0605714v2 23 Jun 2006
Astronomy & Astrophysics manuscript no. 5555
February 3, 2008
c ? ESO 2008
APEX 1mm line survey of the Orion Bar
S. Leurini1, R. Rolffs1, S. Thorwirth1, B. Parise1, P. Schilke1, C. Comito1, F. Wyrowski1, R. G¨ usten1, P. Bergman2,
K. M Menten1, and L.-Å. Nyman2
1Max-Planck-Institut f¨ ur Radioastronomie, Auf dem H¨ ugel 69, 53121 Bonn, Germany
e-mail: sleurini@mpifr-bonn.mpg.de
2European Southern Observatory, Casilla 19001, Santiago, Chile
ABSTRACT
Context. Unbiased molecular line surveys are a powerful tool for analyzing the physical and chemical parameters of astronomical objects and
are the only means for obtaining a complete view of the molecular inventory for a given source. The present work stands for the first such
investigation of a photon-dominated region.
Aims. The first results of an ongoing millimeter-wave survey obtained towards the Orion Bar are reported.
Methods. The APEX telescope in combination with the APEX-2A facility receiver was employed in this investigation.
Results. We derived the physical parameters of the gas through LVG analyses of the methanol and formaldehyde data. Information on the
sulfur and deuterium chemistry of photon-dominated regions is obtained from detections of several sulfur-bearing molecules and DCN.
Key words. ISM: individual objects (Orion Bar) - ISM: abundances - ISM: molecules
1. Introduction
Photon-dominated regions (PDRs) are commonly found inter-
stellar environmentswhoseglobalpropertiesaredeterminedby
intense far-ultraviolet (FUV) radiation emerging from nearby
youngOB stars (e.g. Hollenbach & Tielens 1997). Owing to its
proximity and nearly edge-on orientation, the Orion Bar PDR
has received particular attention as a template for studies of the
spatial stratification of various atomic and molecular species
from the highlypenetrated surface layers deep into the parental
molecular cloud. Targeted (sub)millimeter-wave investigations
indicate that, besides a considerable number of ubiquitous as-
tronomical molecules such as CO, CS, HCN, CH3OH, and
H2CO, other species can be found that are suggestive of a
unique PDR chemistry (Hogerheijde et al. 1995; Jansen et al.
1995; Fuente et al. 2003). One such example is the molec-
ular ion CO+(e.g. St¨ orzer et al. 1995), and another one ap-
pears to be CF+, which so far has only been detected toward
the Orion Bar (Neufeld et al 2006, this volume). PDRs are
also thought to show enhanced abundances of molecules car-
rying refractory elements due to grain breakup caused by the
FUV radiation field (Schilke et al. 2001). Additionally, obser-
vations of the Horsehead nebula suggest that at least part of the
molecularcarbon chain budget in PDRs may be producedfrom
UV-destruction of PAHs and carbonaceous grains (Pety et al.
2005).
Send offprint requests to: S. Leurini
In view of this, PDRs are attractive targets for unbiased
molecular line studies, which also help to derive the global pic-
ture of the physics and chemistry associated with them. Here,
we present the initial results from an unbiased molecular line
surveyobtainedbetween279and308GHzwiththeAPEXtele-
scope (G¨ usten et al. 2006, this volume).
2. Observation
Observations were conducted with the APEX 12-m telescope1
towards the Orion Bar at the position α2000 = 5h35m25.3s,
δ2000= −5◦24′34.0′′, correspondingto the “Orion Bar (HCN)”
position of Schilke et al. (2001), the most massive clump seen
in H13CN (Lis & Schilke 2003, hereafter LS03). Data were
taken in 2005, between October and December in the position-
switching mode, with the reference position at (600′′, 0′′). The
APEX-2Areceiver(Risacheret al.2006,this volume)was used
in combination with the Fast Fourier Transform Spectrometer
(Klein et al. 2006, this volume) providing a bandwidth of
1GHzandaresolutionof0.12kms−1. Thefrequencyrangebe-
tween 279 and 307.7 GHz was completely covered; additional
data were taken in selected frequency ranges from 318.5 GHz
to 361.5 GHz. Step-widths of approximately 500MHz were
1This publication is based on data acquired with the Atacama
Pathfinder Experiment (APEX). APEX is a collaboration between
the Max-Planck-Institut f¨ ur Radioastronomie, the European Southern
Observatory, and the Onsala Space Observatory.

Page 2

2Leurini et al.: APEX 1mm line survey of the Orion Bar
used to cover each frequency setup twice to facilitate the side-
band assignment. The antenna temperature was converted to
the main-beam brightness temperature by using forward and
beam efficiencies of 0.97 and 0.74, respectively. The noise
level is of the order of 0.06 K–0.1 K, and the system temper-
ature around 160 K. The calibration was performed by using
the APECS software (Muders et al. 2006, this volume). The
pointing was checked on Mars and was found to be accurate
within a few arc-seconds.We estimate a calibrationuncertainty
of ∼ 30%.
3. Analysis of the data
The line identification was based on the Cologne Database
for Molecular Spectroscopy2(M¨ uller et al. 2001) and on the
JPL3(Pickett et al. 1998) line catalog and performed with
the XCLASS software. We identified 16 different species and
a number of their isotopologs; no strong unidentified lines
were found in the range surveyed. In the following para-
graphs, we focus on the excitation conditions of methanol
(CH3OH) and formaldehyde (H2CO), on the one hand, and
on the analysis of S-bearing molecules, on the other. Methanol
and formaldehyde are present in our dataset with a large num-
ber of transitions (Table 1), and their collisional rates are
available (Green 1991; Pottage et al. 2002). Thus, a more rig-
orous multi-line analysis is possible for these two species,
which delivers temperature and density estimates. As for the
S-bearing molecules in our dataset, some of them, namely
CS, H2S, SO, SO+, and SO2, have already been studied to-
wards this source (Hogerheijde et al. 1995; Jansen et al. 1995;
Fuente et al. 2003), whereas HCS+, H2CS and NS are being
analyzed here for the first time in this environment. A compre-
hensiveanalysiswill bepresentedoncethelinesurveyhasbeen
completed.
3.1. CH3OH and H2CO
The excitation analysis of CH3OH and H2CO was performed
byusingthemethoddescribedbyLeurini et al.(2004),recently
modified to treat H2CO as well. The procedure is based on
the simultaneous fit of all the lines with a synthetic spectrum
computed in the LVG approximation with the cosmic back-
groundas the only external radiation field. If lines are optically
thick, the model fits source size, temperature, H2density, and
CH3OH/H2CO column density; for optically thin lines, source
size and column density cannot be determined independently,
and beam-averaged column densities are provided.
To improve the signal-to-noise ratio in the weaker transi-
tions, we smoothed the data to a resolution of 0.5 km s−1. At
this resolution, each line has a single-peak Gaussian profile
(vlsr = 10.0 ± 0.2 km s−1, ∆v = 1.7 ± 0.3 km s−1), while, at
the original resolution of 0.12 km s−1, two peaks were detected
in the 51,5 → 41,4H2CO and in the 6−1 → 5−1 CH3OH-E
transitions. Non-Gaussian profiles were found in other transi-
tions as well. By fitting the 6−1 → 5−1CH3OH-E line with
2http://www.cdms.de
3http://spec.jpl.nasa.gov
Table 1. Line parameters of the observed transitions
transitionfrequency
[MHz]
281526.929
290623.405
291237.767
291380.488
291384.264
300836.635
351768.645
289939.477
290069.824
290110.666
290248.762
290307.376
290307.643
303366.890
304208.350
305473.520
307165.940
289209.068
291485.935
293912.087
342882.850
298690.453
341350.229
296550.064
301286.124
304077.844
340714.155
344310.612
346528.481
301361.501
301736.791
304307.645
342946.335
343322.111
348534.225
299700.097
300098.057
345823.532
346220.774
300505.560
Elow
[K]
32.06
20.96
68.09
126.95
126.95
33.45
45.57
47.93
40.39
34.82
55.87
60.72
57.07
2.32
6.96
13.93
23.21
34.70
34.98
35.27
49.37
43.01
57.34
50.66
56.50
47.55
64.89
70.96
62.14
38.92
39.03
71.60
74.13
126.83
88.47
39.82
39.93
54.20
54.33
154.48
?
TMBdv
[K km s−1]
10.7 ± 0.5
5.9 ± 0.4
2.4 ± 0.3
3.4 ± 0.3
3.7 ± 0.3
9.8 ± 0.5
10.1 ± 0.9
0.5 ± 0.1
0.9 ± 0.2
1.2 ± 0.2
0.3 ± 0.2
0.3a± 0.2
H2CO 41,4→ 31,3
H2CO 40,4→ 30,3
H2CO 42,3→ 32,2
H2CO 43,2→ 33,1
H2CO 43,1→ 33,0
H2CO 41,3→ 31,2
H2CO 51,5→ 41,4
CH3OH-E 60→ 50
CH3OH-E 6−1→ 5−1
CH3OH-A 60→ 50
CH3OH-E 61→ 51
CH3OH-E 62→ 52
CH3OH-E 6−2→ 5−2
CH3OH-A 11→ 10
CH3OH-A 21→ 20
CH3OH-A 31→ 30
CH3OH-A 41→ 40
C34S 6 → 5
C33S 6 → 5
CS 6 → 5
CS 7 → 6
HCS+7 → 6
HCS+8 → 7
SO 76→ 65
SO 77→ 66
SO 78→ 67
SO 87→ 76
SO 88→ 77
SO 89→ 78
SO+ 2Π1/213/2 → 11/2e
SO+ 2Π1/213/2 → 11/2f
H2CS 91,9→ 81,8
H2CS 100,1,0→ 90,9
H2CS 102,9→ 92,8
H2CS 101,9→ 91,8
NS2Π1/213/2 → 11/2e
NS2Π1/213/2 → 11/2f
NS2Π1/215/2 → 13/2e
NS2Π1/215/2 → 13/2f
H2S 33,0→ 32,1
0.6 ± 0.2
0.8 ± 0.2
0.9 ± 0.2
1.4 ± 0.3
4.6 ± 0.7
1.6 ± 0.3
40.7 ± 2.2
33.1 ± 1.9
0.3 ± 0.1
0.6 ± 0.1
3.7 ± 0.5
4.0 ± 0.6
7.2 ± 1.3
3.6 ± 0.7
3.0 ± 1.1
7.2 ± 1.2
0.6 ± 0.1
0.4 ± 0.1
0.6 ± 0.1
0.4 ± 0.1
0.4 ± 0.1
0.8 ± 0.2
0.4 ± 0.1
0.4± 0.1
0.7 ± 0.2
0.8 ± 0.2
0.5 ± 0.1
ablended with the 6−2→ 5−2CH3OH-E line.
the GAUSS method of CLASS, we found two velocity com-
ponents at vlsr = 10.89 ± 0.06 and vlsr = 9.82 ± 0.05 km s−1,
with ∆v = 0.62 ± 0.13 and ∆v = 0.63 ± 0.06 km s−1. These
values resemble what LS03 found for the H13CN clumps 1, 5
and8, whichare all withinourbeam.Forthe sake ofsimplicity,
one velocity component was used both for the CH3OH and the
H2CO analysis, as derived from the smoothed CH3OH data.
High density (n ∼ 7 × 106cm−3) and a moderately high
temperature (T ∼ 70 K) are needed to reproduce the CH3OH
spectrum. The temperature is constrained to this value by the
non-detection of high-excitation (Elow ≥ 80 K) lines (Fig. 1).
The highdensity valuewe derivedsuggeststhat CH3OH comes
from small clumps. Therefore, guided by LS03, we used a

Page 3

Leurini et al.: APEX 1mm line survey of the Orion Bar3
Fig.1. Spectrum of the 6k→ 5kCH3OH band. Overlaid on the
data, in blue, is the synthetic spectrum corresponding to our
best fit.
source size of 5′′and derived N(CH3OH-E) = 7 × 1014cm−2
and N(CH3OH-A) = 8 × 1014cm−2. Our estimate of n in
the clumps agrees with previous studies (e.g. Young Owl et al.
2000; LS03); Tkin is also similar to previous results for the
clump gas (see discussion in LS03) and to theoretical mod-
els (Gorti & Hollenbach 2002). The H2column density (9 ×
1022cm−2) was derived from our observations of the C17O(3-
2) line, assuming17O/16O∼ 1790 (Wilson & Rood 1994) and
Trot=70 K. The methanol column density then corresponds to
a fractional abundance averaged over the beam of 10−9, typical
ofdarkclouds.The3σconfidencelevelforTkinrangesbetween
50 and 75 K, and (5–20)×106cm−3for nH2.
The parameters derived from H2CO differ significantly
from the ones of CH3OH, as high temperature (T ∼ 150 K)
and moderate density (n ∼ 5 × 105cm−3) are needed to re-
produce the data. Since all H2CO lines are optically thin, we
derived a beam-averaged column density of 9 × 1013cm−2for
ortho-H2CO and 3 × 1013cm−2for para-H2CO, correspond-
ing to a fractional abundance of 10−9. Fuente et al. (1996)
found the same value, although at another position in the
Bar. Our analysis infers a Tkin in agreement with the re-
sults of Batrla & Wilson (2003) from NH3, but higher than
the one derived by Hogerheijde et al. (1995), 85 ± 30 K,
also with formaldehyde, but toward a different position. The
density derived is substantially higher than that proposed by
Hogerheijde et al. (1995) for the interclump medium (3 ×
104cm−3), but it agrees with the value inferred by Simon et al.
(1997) based on CN. While the nH2is well-constrained (1–
5×105cm−3), the temperature we derived is a lower limit.
The infrared pumping through an internal radiation field
would probably affect our results for both molecular species.
However, since no model of the IR field is available for the
Bar, no internal radiation field was taken into account.
Our data suggest that CH3OH is found mainly in the
clumpy medium, while H2CO traces the interclump mate-
rial. Both methanol and formaldehyde can form on grain sur-
faces (Watanabe & Kouchi 2002). It therefore seems remark-
able that they trace different environments. We speculate that
in the interclump gas, methanol, once released from grain
surfaces, is photodissociated to form formaldehyde (one of
the possible photodissociation products, Le Teuff et al. 2000).
Alternatively, our observations may indicate a different for-
mation mechanism for the two species, surface chemistry for
CH3OH (as no gas-reaction can efficiently form it, Luca et al.
2002), and gas phase chemistry for H2CO (via the neutral–
neutral reaction CH3+ O, e.g. Le Teuff et al. 2000). More de-
tailed modeling, includingrelease of molecules from grain sur-
faces in the PDR, is clearly needed.
3.2. Sulfur-bearing species
The excitation analysis of the S-bearing molecules listed in
Table 1 was performed by fitting all the observed transitions
with a synthetic spectrum computed under the LTE assump-
tion, in the manner described by Comito et al. (2005). Since
the kinetic temperature, the source size, and the column den-
sity are degenerate parameters in the optically thin limit, we
fixed Tkinto the value derived from H2CO and fit the beam-
averaged column density of each species. For SO2, OCS, and
HCS, upper limits to the column densities were derived, based
on the non-detections of lines in our survey range. Results are
given in Table 2. The uncertainties on the column densities are
derived with a χ2analysis and correspondto the 3σ confidence
level. CS and its isotopologs are not included in our analysis,
because of the anomalous33S/34S ratio of ∼ 3 we observed(for
comparison, see Chin et al. 1996). Observations of other rota-
tional transitions are needed to constrain this ratio reliably.
Sternberg & Dalgarno (1995, hereafter SD95) studied the
chemistry of PDRs produced in a dense molecular cloud ex-
posed to intense far-UV radiation fields. The density they em-
ploy (nH = 106cm−3) is higher than the one relevant here for
the interclump gas; discrepancies in the results may be due to
this. Their chemical network does not include NS and H2CS,
but all the other S-bearing molecules observed here are found
there.
According to SD95, the SO+/SO ratio (observed 0.2 from
Table 2) can be used as diagnostic of the different chemical
zones in the cloud and changes between 0.23 at Av= 1.5 and
0.0028 at Av = 3. Assuming that the Bar is edge-on, based
on the distance between the ionization front and our position
(∼ 30′′), and assuming an H2 density of 105cm−3, we de-
rived a visual extinction of ≥ 10 mag (Frerking et al. 1982).
Given the complexgeometryof the Orion Bar, which has edge-
on (but oblique) and face-on parts (Hogerheijde et al. 1995), it
is likely that together with the molecular material at this ex-
tinction, a hotter, more diffuse layer of gas, corresponding to a
face-on surface layer, contributes to the emission we observe.
SD95 predict SO2to be more abundant than SO and SO+at
high Av, while our upper limit indicates that SO2is less abun-
dant. Similar to the result found for SO+/SO, our estimates of
SO2/SO and SO2/SO+suggest that these species have a signif-
icant contribution from a hotter layer of gas with lower extinc-
tion than the molecular layer.
Fuente et al. (2003) report different column densities for
SO+and SO2; however, their SO+/SO2ratio ranges between
1.2 at the IR front and 0.4 at (20′′,-20′′) from it, which is sim-
ilar to our value of 0.3. Jansen et al. (1995) found values of
N(H2S) between 8.9 × 1012cm−2and 3.9 × 1014cm−2across

Page 4

4Leurini et al.: APEX 1mm line survey of the Orion Bar
Fig.2. H13CN(4-3) and DCN(4-3)lines. For clarity, the H13CN
line has been translated by 0.2 along the y-axis.
the Bar, with our value of 1.3 × 1013cm−2lying in between.
Discrepancies in the absolute values of the column densities
may depend on the source sizes, on temperatures, and on the
total H2column density at the observed position.
Table 2. Column densities of S-bearing molecules
Species
NS
SO
SO+
H2S
H2CS
HCS+
N [cm−2]
(1.3 ± 0.3)1013
(1.0 ± 0.2)1014
(2.0 ± 0.5)1013
(2.5 ± 1.0)1013
(2.5 ± 0.4)1013
(2.5 ± 0.7)1012
Species
SO2
HCS
OCS
N [cm−2]
< 3 1013
< 4 1014
< 4 1013
4. Detection of a deuterated molecule : DCN
Molecular deuteration studies have been extremely powerful
in probing the physical history of sources. The deuterium
isotopic ratio in molecules can indeed be enhanced in low-
temperature environments compared to the cosmic D/H ratio
(∼10−5, Linsky 2003). This enhancement proceeds initially
from the transfer of deuterium from the main reservoir HD to
two reactive ions (H2D+and CH2D+) through exothermic re-
actions with H+
ences therein). The endothermicity of the backward reactions
(respectively by 220K and 370K) ensures that the H2D+/H+
and CH2D+/CH+
peratures. Deuterium is then channeled to other molecules by
gas-phasereactionswiththese twoionsandbygrainsurfacere-
actions. Highly-deuterated molecules can also be found in hot
gas, where they are then out-of-equilibrium fossils of an ear-
lier cold phase (e.g. in hot corinos around low-mass protostars,
where they are believed to be remnants of the prestellar cold
and depleted chemistry; Parise et al. 2004, 2006).
The DCN(4-3) transition at 289.645GHz (Eup=34.56K)
was detected in oursurvey,andfoundto be very bright(Fig. 4).
This detection, together with the detection of DCO+in the
Horsehead nebula (Pety et al. 2006, in preparation), is opening
up the study of fractionationprocesses in PDRs. The H13CN(4-
3) transition(Eup=41.76K) was also detectedat 345.340GHz.
The H13CN emission was shown from interferometricobserva-
3and CH3+(Roberts & Millar 2000, and refer-
3
3ratios are significantly enhanced at low tem-
tions (LS03) to be mostly associated with clumps in the PDR.
TheDCN emissionis thus verylikelytoalso beassociatedwith
clumps,as DCN wouldnot survivethehightemperaturesofthe
interclump gas. The linewidths of DCN(4-3) and H13CN(4-3)
are comparable and non-Gaussian profiles are detected in both
cases. We thus assume that H13CN and DCN trace the same
region.As the energies of the two lines are very similar, the de-
rived abundanceratio is nearly independentof the temperature.
Assuming LTE and a12C/13C ratio of 70, we derived beam-
averagedDCN andH13CN columndensitiesbetween8.4×1011
and1.7×1012cm−2, andbetween1.7×1012and2.8×1012cm−2,
respectively, in the range of temperatures 30–150 K. This cor-
responds to a DCN/HCN ratio of 0.7-0.9%, which is an inter-
mediate value between that observed in warm gas (Orion Hot
Core, 0.1%, Schilke et al. 1992), on the one hand, and in dark
clouds (L134, 5%, Turner 2001) or cold (30-50K) gas of the
OMC1 ridge region (1-6% Schilke et al. 1992), on the other.
In hot cores/corinos, where the gas has not yet had time to
return to equilibrium (which would require timescales of ∼ 104
yrs), abundant deuterated molecules can be considered as rem-
nants of earlier cold prestellar chemistry, most probably stored
in the ices on grain surfaces and then released when the proto-
star heats its surroundings. In contrast, the gas in the shielded
clumps of the bar might be in steady state. In this case, the
observed deuterated molecules have to be present-day product
molecules. The detection of DCN in this relatively warm envi-
ronment, where H2D+would not survive, is in this sense quite
strikingandis a stronghintof theefficiencyofthe fractionation
processes occurring via the CH2D+channel, which survives
higher temperatures than H2D+, due to the higher exothermic-
ity of the CH+
thus appears a unique reference to test the fractionation reac-
tions involving CH2D+. Knowing the spatial distribution of the
DCN/HCN isotopic ratio would then be an interesting tool for
understanding the effect of extinction on the fractionation pro-
cesses.
3reaction with HD (Turner 2001). The Orion Bar
Acknowledgements. BP is grateful to the Alexander von Humboldt
Foundation for a Research fellowship
References
Batrla, W. & Wilson, T. L. 2003, A&A, 408, 231
Chin, Y.-N., Henkel, C., Whiteoak, J. B., Langer, N., &
Churchwell, E. B. 1996, A&A, 305, 960
Comito, C., Schilke, P., Phillips, T. G., et al. 2005, ApJSuppl.
S., 156, 127
Frerking, M. A., Langer, W. D., & Wilson, R. W. 1982, ApJ,
262, 590
Fuente, A., Rodr´iguez-Franco, A., Garc´ia-Burillo, S., Mart´in-
Pintado, J., & Black, J. H. 2003, A&A, 406, 899
Fuente, A., Rodriguez-Franco, A., & Martin-Pintado, J. 1996,
A&A, 312, 599
Gorti, U. & Hollenbach, D. 2002, ApJ, 573, 215
Green, S. 1991, ApJS, 76, 979
Hogerheijde, M. R., Jansen, D. J., & van Dishoeck, E. F. 1995,
A&A, 294, 792
Hollenbach, D. J. & Tielens, A. G. G. M. 1997, ARA&A, 35,
179

Page 5

Leurini et al.: APEX 1mm line survey of the Orion Bar5
Jansen,D. J., Spaans,M., Hogerheijde,M. R., &vanDishoeck,
E. F. 1995, A&A, 303, 541
Le Teuff, Y. H., Millar, T. J., & Markwick, A. J. 2000, A&AS,
146, 157
Leurini, S., Schilke, P., Menten, K. M., et al. 2004, A&A, 422,
573
Linsky, J. L. 2003, Space Science Reviews, 106, 49
Lis, D. C. & Schilke, P. 2003, ApJ, 597, L145, LS03
Luca, A., Voulot, D., & Gerlich, D. 2002, in WDS’02
Proceedings of Contributed Papers, Part II, Safrankova (ed),
Matfyzpress, 294
M¨ uller, H. S. P., Thorwirth,S., Roth, D. A., & Winnewisser, G.
2001, A&A, 370, L49
Parise, B., Castets, A., Herbst, E., et al. 2004, A&A, 416, 159
Parise,B.,Ceccarelli,C.,Tielens,A.,etal.2006,A&A,inpress
Pety, J., Teyssier, D., Foss´ e, D., et al. 2005, A&A, 435, 885
Pickett, H. M., Poynter, R. L., Cohen, E. A., et al. 1998, J.
Quant. Spectrosc. Radiat. Transfer, 60
Pottage, J. T., Flower, D. R., & Davis, S. L. 2002, J. Phys. B:
At. Mol. Phys., 35, 2541
Roberts, H. & Millar, T. J. 2000, A&A, 361, 388
Schilke, P., Pineau des Forˆ ets, G., Walmsley, C. M., & Mart´ ın-
Pintado, J. 2001, A&A, 372, 291
Schilke,P.,Walmsley,C.M.,PineauDes Forets,G.,etal.1992,
A&A, 256, 595
Simon, R., Stutzki, J., Sternberg, A., & Winnewisser, G. 1997,
A&A, 327, L9
Sternberg, A. & Dalgarno, A. 1995, ApJS, 99, 565, SD95
St¨ orzer, H., Stutzki, J., & Sternberg, A. 1995, A&A, 296, L9
Turner, B. E. 2001, ApJS, 136, 579
Watanabe, N. & Kouchi, A. 2002, ApJ, 571, L173
Wilson, T. L. & Rood, R. 1994, ARA&A, 32, 191
Young Owl, R. C., Meixner, M. M., Wolfire, M., Tielens,
A. G. G. M., & Tauber, J. 2000, ApJ, 540, 886