Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://www.mrao.cam.ac.uk/yerac/marechal/marechal.ps Äàòà èçìåíåíèÿ: Wed Feb 22 23:13:38 1995 Äàòà èíäåêñèðîâàíèÿ: Tue Oct 2 03:25:29 2012 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: annular solar eclipse

Molecular oxygen in interstellar clouds
By P r i s c i l l a Ma r e c ha l
e­mail: priscila@mesion.obspm.fr
DEMIRM, Observatoire de Paris, FRANCE
The prospects for detection of interstellar O2 depend on its abundance and its excitation. Our
results shown the importance of the C=O ratio for the emmissivity of O2 and the role of molecular
oxygen in the thermal balance. O2 has not yet been detected in interstellar clouds but we
obtained the first tentative detection of 16 O 18 O.
1. Introduction
According to chemical models, O 2 is one of the most important carriers of oxygen in
molecular clouds, predicted to be as abundant as CO. This fact encourages attempts
to observe O 2 despite low line strengths of submillimeter magnetic dipole transitions of
this molecule. The balloon­borne telescope PRONAOS­SMH (Beaudin et al. (1994))
should observe the 368 GHz (N; J) : (3; 2) ! (1; 1) line of 16 O 2 . To prepare the mission,
we must select as well as possible molecular clouds which should have high intensity
in this line. Thus, I inserted the processes of radiative and collisional transfer between
rotational levels of this molecule in a model which works out chemical and thermal
balance in molecular clouds (Warin et al. (1995)). In addition I briefly present results
obtained during an observing session made at POM­2 telescope (Castets et al. (1988))
to detect the 234 GHz (N; J) : (2; 1) ! (0; 1) line of the isotopomer 16 O 18 O (Pagani et
al. (1994), Pagani et al. (1995)).
2. Modelling the chemistry and the rotational excitation of oxygen
2.1. The steady state model
The abundances of chemical species and the temperature profile throughout a molecular
cloud are obtained by solving, in a self­consistent way, a radiative transfer equation for
UV photons --- taking into account absorption by both dust and gas --- the chemical
balance equations and a thermal balance equation. The atomic and molecular spectral
lines are obtained by adding the statistical equilibrium equations which lead to the fine
structure and the rotational population of the species. The complete process is iterative
and is illustrated in Figure 1.
The chemistry takes into account 136 chemical species and about 2,800 reactions. In
the case of O 2 , the main production route is the reaction:
OH+ H \Gamma! O 2 + H (2.1)
O 2 is mainly destroyed by photodissociation, photoionization and collisions with C + and
He + :
O 2 + hš \Gamma! O+ O (2.2)
O 2 + hš \Gamma! O +
2 + e (2.3)
C + + O 2 \Gamma!
ae
CO + +O
O + +CO (2.4)
He + + O 2 \Gamma!
ae
O + +O + He
O +
2 + He (2.5)
1

2 Priscilla Marechal: Molecular oxygen in interstellar clouds
Figure 1. The steady state model of interstellar clouds
Figure 2. a) rotational levels of O2 , b) radiative transitions and Einstein coefficients for the
first levels (N = 1; 3)
2.2. Level population of oxygen
To obtain the rotational population of O 2 , we take into account the following processes:
spontaneous emission, stimulated emission and absorption of the ambient background
radiation, excitation and de­excitation by collisions with H 2 and He. The model takes
into account the 24 first rotational levels of O 2 .

Priscilla Marechal: Molecular oxygen in interstellar clouds 3
Diffuse (AV =1) Translucent (AV =4) Dense (AV =10)
nH = 5 10 2 cm \Gamma3 nH = 10 3 cm \Gamma3 nH = 10 4 cm \Gamma3
T = 50 K T = 25 K T = 10 K
OH (cm \Gamma2 ) 3.24 10 12 1.04 10 15 1.21 10 15
H2O (cm \Gamma2 ) 3.45 10 11 3.62 10 14 5.63 10 15
O (cm \Gamma2 ) 1.73 10 17 4.94 10 17 6.79 10 17
CO (cm \Gamma2 ) 8.76 10 13 1.92 10 17 6.95 10 17
O2 (cm \Gamma2 ) 8.55 10 10 1.22 10 15 1.70 10 17
O2/H2 8.6 10 \Gamma11 3.0 10 \Gamma7 1.7 10 \Gamma5
O2/CO 9.8 10 \Gamma4 6.4 10 \Gamma3 0.24
I(368 GHz) (K km s \Gamma1 ) 6.65 10 \Gamma8 9.12 10 \Gamma4 7.42 10 \Gamma2
18 OH (cm \Gamma2 ) 6.54 10 9 2.45 10 12 2.59 10 12
H 18
2 O (cm \Gamma2 ) 6.97 10 8 8.19 10 11 1.15 10 13
18 O (cm \Gamma2 ) 3.46 10 14 1.14 10 15 1.61 10 15
C 18 O (cm \Gamma2 ) 1.41 10 11 2.42 10 14 1.16 10 15
16 O 18 O (cm \Gamma2 ) 3.48 10 8 5.43 10 12 6.83 10 14
16 O 18 O/H2 3.4 10 \Gamma13 1.3 10 \Gamma9 6.7 10 \Gamma8
16 O 18 O/C 18 O 2.4 10 \Gamma3 2.2 10 \Gamma2 0.59
I(234 GHz) (K km s \Gamma1 ) 2.12 10 \Gamma9 3.75 10 \Gamma5 4.81 10 \Gamma3
Table 1. Oxygen­bearing molecules in ``standard'' molecular clouds models: AV represents
the visual extinction througout the whole cloud
The ground electronic state of O 2 is a 3 \Sigma state with two unpaired electrons. So its
rotational levels are described by the rotational quantum number N and the total angular
momentum quantum number J as illustrated by Figure 2a:
J = N \Gamma 1; N ; N + 1 (2.6)
O 2 consists of two identical atoms obeying to the Bose­Einstein statistics so that ro­
tational levels with even value of N do not occur. Since O 2 is homopolar, it has no
permanent electric dipole moment. Hence the radiative transitions obey to selection
rules of magnetic dipole transitions as shown in Figure 2b for the first levels of the
molecule:
\DeltaN = 0; \Sigma2 and \DeltaJ = 0; \Sigma1 (2.7)
The collisional rates are taken from Black & Smith (1984) who derived their values
from experimental data about collisions between O 2 and N 2 . These rates are however
very uncertain and inclusion in the model of collision rates deduced from theoretical
calculations (Corey et al. (1986)) are in progress.
2.3. Results
Table 1 shows the results of the model for three standard clouds: a diffuse one, a trans­
lucent one and a dense one. The most abundant oxygen­bearing species are O, CO and
O 2 . The 368 GHz line of O 2 is intended to be observed by PRONAOS balloon and the
234 GHz line of 16 O 18 O is the one we observe at POM­2. In fact, only dense clouds can
be observed by those telescopes.
Figure 3 displays the intensity profile of the 368 GHz line of O 2 computed by the model
for dense clouds and for some values of the gas density and the C=O ratio. The density
does not affect so much the emissivity of O 2 . The integration time for a 3oe detection is
between 20 minutes and 1 hour with actual characteristics of PRONAOS­SMH receiver

4 Priscilla Marechal: Molecular oxygen in interstellar clouds
Figure 3. The 368 GHz line of O2 in dense clouds: a) for different values of nH , b) for
different values of the C=O ratio
Figure 4. Atomic and molecular cooling rates in a dense cloud with nH = 10 4 cm \Gamma3 .
(T rec ' 250 K). The C=O ratio is a parameter which causes dramatic changes on the
abundance and the emissivity of O 2 . The integration time varies between a few minutes
for C=O=0.1 and several years for C=O=1.
The cooling of a molecular cloud, plotted in Figure 4, occurs through fine structure
emission of atoms and ions and rotational transitions of molecules. The model takes into
account cooling by C + , C, O, CO, 13 CO, C 18 O, O 2 and 16 O 18 O. For a standard C=O
ratio of 0.4, the cooling by O 2 in the center of dense clouds, shown in Figure 5, is about
30%. It could be as great as 80% for C=O = 0:1 and lower than 1% for C=O = 1.

Priscilla Marechal: Molecular oxygen in interstellar clouds 5
Figure 5. Cooling by O2 in dense clouds for different C=O ratio.
C 18 O 16 O 18 O
offset position column density/ column density/ O2=CO ratio
cm \Gamma2 cm \Gamma2
(0; 0) 1:3 10 15 5:4 10 14 0:2
(2:5 0 ; \Gamma1 0 ) 1:2 10 15 ¸ 6:7 10 14 ¸ 0:23
(4 0 ; \Gamma1 0 ) 2:5 10 15 ¸ 4:9 10 14 ¸ 0:1
Table 2. Comparison of the O2/CO ratio in 3 different positions in L134N
3. The observations of the 234 GHz line of 16 O 18 O in L134N
Observations were made in several runs at POM­2, a 2.5 meter antenna, on Plateau de
Bure (France). The instrument works with an SIS mixer of temperature T rec (DBS) =
65 \Sigma 5 K. The observations are made with the autocorrelator backend 18 MHz bandwidth
(78 kHz = 0:1 kms \Gamma1 resolution) using frequency switching by \Sigma4 MHz.
In addition to its low line strength, 16 O 18 O is a rare isotopomer species. So we need
very long time of integration --- over 100 hours of effective time on the source --- to get
a 3oe detection towards L134N.
Figure 6 shows the result of the observations of 3 positions in the cold dense molecular
cloud, L134N. The central position in L134N (Pagani et al. (1994)) shows three peaks at
1.3, 2.5 and 3.4 km s \Gamma1 . At the two other positions of L134N (Pagani et al., submitted),
there are only two peaks at 1.3 and 2.8 km s \Gamma1 and their spacing is different. So they are
not a fine structure emission or a symmetric top triplet from another species. Moreover,
the O 2 =CO ratio deduced from observations (Table 2) are consistent with our predicted
value of about 0.24 for a standard dense model (Table 1). Those facts would reinforce
the case for molecular oxygen detection.
4. Conclusion
The modelling of emissivity of O 2 shows that the detection of the 368 GHz line of O 2
in dense clouds is feasible by a project like PRONAOS­SMH provided the C=O ratio is

6 Priscilla Marechal: Molecular oxygen in interstellar clouds
Figure 6. Observations of the (N;J) : (2; 1) ! (0; 1) transition of 16 O 18 O in the dark cloud
L134N. Axes represent the usual LSR velocity (km s \Gamma1 ) and brightness temperature (K).
about, or less than, the standard value of 0.4. The balloon might fly during twenty or
thirty hours and such clouds could be detected between a few minutes and one hour.
We have also seen that O 2 has a non­negligible role in thermal balance of dense clouds.
Models do not generally take this fact into account.
The low signal­to­noise ratio and the presence of several lines prevented us from reach­
ing definite conclusions but we have some evidence for 16 O 18 O detection in L134N. This
source certainly stands out as a very peculiar source for its chemico­physical behaviour
for which the observations still reveal exciting possibilities.
REFERENCES
Beaudin et al., 1994, 24 th European Microwave Conference proceedings, Vol. III, 1955.
Black J.H. & Smith P.L., 1984, ApJ, 277, 562.
Castets A., Lucas R., Lazareff B. et al., 1988, A&A, 194, 340.
Corey, Alexander & Schaeffer, 1986, J.Chem.Phys., 85, 2726.
Pagani L., Langer W.D. & Castets A., 1994, A&A, 274, L13.
Pagani L., Marechal P., Castets A., Langer W.D., 1995, submitted.
Warin S., Benayoun J.J., Viala Y.P., 1995, in preparation.