Oxygen is the most abundant element in the universe after hydrogen and helium and thus plays an important role in the chemistry of interstellar clouds. Since the first chemical models of interstellar clouds dealing with the simplest molecules and assuming steady-state chemical equilibrium (Herbst & Klemperer 1973, Viala & Walmsley 1976), it has been recognized, and confirmed by more recent models, that molecular oxygen O2 should have an abundance comparable to that of CO deep inside the clouds. It must be however noted that a whole series of special models show a dramatic drop in O2 abundance in many situations. This is for instance the case in early stages of time dependent models (e.g. Millar et al. 1996), in models in which turbulent mixing occurs between the envelope and the core of the clouds (e.g. Chièze & Pineau des Forêts 1989), in the high ionization phase of models exhibiting two stable chemical phases (e.g. Le Bourlot et al. 1995) and in clouds showing a very highly clumpy structure (Spaans 1996, private communication). Thus the observation of molecular oxygen to measure its abundance should allow us to test the interstellar chemical models and to give some insight on the total amount of gas phase oxygen in molecular clouds.
Due to the presence of a large quantity of O2 in the terrestrial atmosphere, detection of 16 O2 in our Galaxy requires the launch of balloon- or satellite-borne heterodyne receivers. There are however two possibilities to try to detect O2 from the ground by getting rid of the opacity due to the atmospheric lines. The first one consists in observing extragalactic sources with redshift large enough for the lines to sit far away in the wings of the atmospheric lines. Several attempts have been made towards such extragalactic sources (e.g. Combes et al. 1991, Combes & Wiklind 1995), all leading to negative results. The second way is to detect its 18 O-substituted isotopomer. Indeed, lines connecting even N levels of 16 O18 O do not exist for the main 16 O2 isotope and are consequently not blocked by atmospheric lines. Several attempts to observe the 16 O18 O (N, J): (2, 1)-(0, 1) line at 234 GHz have been led without success (Goldsmith et al. 1985, Liszt & Vanden Bout 1985, Combes et al. 1991, Fuente et al. 1993, Maréchal et al. 1997a, hereafter MPLC), apart from one possible case in L134N for which a 3 tentative detection has been reported (Pagani et al. 1993). In some of the sources for which 16 O18 O has not been detected, our observations have permitted to establish the lowest upper limits of the 16 O18 O/C18 O ratio obtained up to now for galactic sources, which really begin to be in contradiction with standard chemical model predictions (MPLC).
The main purpose of this paper is to make theoretical predictions of the emissivities of the 16 O18 O lines accessible from the ground and to interpret the low upper limits of the 16 O18 O column densities deduced from the observations. To do this, in the same way as we have computed the O2 abundance and the emissivities of its rotational lines (Paper I: Maréchal et al. 1997b, hereafter MVB), the processes of radiative and collisional transfer between the rotational levels of 16 O18 O have been included in a steady-state model which works out chemical and thermal balances in molecular clouds. The paper is organized as follows : the determination of the spectroscopic parameters of 16 O18 O in its ground vibrational level needed to compute its rotational population is presented in Sect. 2, together with a brief description of its chemistry. The emissivities of 16 O18 O lines computed for various cloud parameters such as the total visual extinction, the density, the ultraviolet radiation field or the C/O elemental ratio are presented in Sect. 3. The interpretation of the observations is done in Sect. 4. The isotopic fractionation of the oxygen-bearing species has also been investigated and is presented in Sect. 5. Sect. 6 summarizes our main conclusions.
© European Southern Observatory (ESO) 1997
Online publication: March 26, 1998