Using the IRAM instruments (interferometer on Plateau de Bure and 30-m telescope on Pico Veleta) we have made numerous observations of molecular absorption lines in front of extragalactic millimeter wavelength radio sources. Observations of HCO+, CO and OH show that the lines of sight studied in this way sample the outer edges of molecular clouds or the diffuse clouds with highest column densities. Collisional excitation of the rotational levels is not significant in this density range for most molecular species, and accurate column densities may be derived by assuming radiative equilibrium with the cosmic microwave background. Using this technique we have measured column densities of CO, HCO+, H2CO, CN, HCN, HNC, CS, SO, H2S, C2H, and C3H2 in several lines of sight, intersecting about 20 individual clouds with CO column densities in the range 2 1014 to ∼ 1016 cm−2. These results confirm that complex molecules achieve dark-cloud abundances at low extinctions, either by formation in the gas phase or on grains.

Two-dimensional model calculations for inhomogeneous diffuse and translucent clouds are presented which include line and continuum radiative transfer, thermal balance and chemical equilibrium. The effects of an inhomogeous density distribution on the abundances of chemical species and line maps are discussed. Additional calculations are presented for the fluorescently pumped H2 visual and infrared lines which emanate from spherical diffuse clouds. The results indicate that these lines yield sensitive probes of the ambient cloud density when combined with co-spatial HI observations.

It is the intention with the present review to summarize the role of molecules in cool stars. Which molecules do we find there, which are the most important ones, what do we know about them, what more would we like to know? – and what are the current main problems and progresses we meet when including molecules in models of cool stars ?

The circumstellar gas/dust envelopes of stars on and beyond the asymptotic giant branch harbor an impressive number of different molecular species and their isotopic variants. Some of these molecules are of photospheric origin, but the majority have been produced through various chemical processes in the envelope. This chemistry depends primarily on the chemical composition of the gas (i.e., C/O<1 or >1), but the conditions also change as the central star evolves. In this review we present circumstellar molecular abundance and isotope ratio estimates, and discuss how they can be used in the study of the late evolution of stars, which in extreme cases may become heavily obscured.

We present chemical models of the inner circumstellar envelope of a typical carbon-rich AGB star. The effect of pulsation-driven shocks on the gas close to the stellar photosphere is considered. The chemistry of dust condensation nuclei formation is described and applied to the gas layers close to the star. We derive formation yields for polycyclic aromatic hydrocarbon (PAH) species and their dimers and discuss their role in the condensation of dust. Models of the chemistry of silicon- and sulfur-bearing molecules in the inner envelope are also presented.

We review the properties of the molecular envelopes of planetary nebulae with an emphasis on the on-going chemistry which takes place in these regions.

During the last decade there have been enormous steps forward in our understanding of the molecular gas properties of other galaxies, due to the huge increase in sensitivity and spatial resolution of instruments in the millimeter and submillimeter wavelength ranges. In particular, the emphasis has shifted from trying to detect CO in a variety of objects to studying the gas distributions and kinematics in a variety of species and transitions. For nearby galaxies, we are now able to study the physical and chemical processes occurring in the molecular gas over GMC size scales. Here I discuss a few examples of progress made in the study of galaxy nuclei.

The number of identified high-redshift far-infrared hyperluminous sources (LFIR ≥ 1013L⊙) is now approaching ~20 (see e.g. Rowan-Robinson 1996). In about half of them, at z > 3.5, evidence for strong far-infrared emission has been obtained from its redshifted detection in the mm continuum. Most of the latter are bright radio-quiet QSOs (Omont et al. 1996b). The most prominent one, BR1202-0725 at z = 4.7, was first detected at 1.25mm with the IRAM 30m telescope (McMahon et al. 1994) with a flux of ∼ 10-15 mJy. Its submm detection at JCMT (Isaak et al. 1994) supports dust emission because of a submm-mm spectral index ≥ 3.

In the past three years, observations of extragalactic water masers have demonstrated the extraordinary power of spectral line radio interferometry to probe the inner regions of active galaxies on subparsec scales. In the case of the active galaxy NGC 4258, the observations of Miyoshi et al. (1995) have established the presence of a thin, slightly warped, and Keplerian circumnuclear accretion disk that orbits a central object of mass 3.6 × 107M⊙. In addition to their immense value as kinematic probes, extragalactic water masers also provide valuable constraints upon the physical and chemical conditions in the masing circumnuclear gas, yielding an estimate of the mass accretion rate through the circumnuclear disk of NGC 4258. These recent observational and theoretical developments are reviewed here.

Molecular abundances derived in a sample of CO complexes in the Magellanic Clouds are discussed. With a possible exception of HCO+, the chemical compositions observed in this sample show a striking uniformity. The data indicate that the molecular concentrations are down by a factor of 5 in the SMC, relative those observed in a cloud associated with the H II region N159 in the LMC. A similar difference seems to exist within the LMC, between the 30 Doradus region and the N159 area. An estimate of the electron abundance in the N159 cloud is presented, based on a recent detection of DCO+.

Searches for molecular gas in high redshift quasar absorption-line galaxies are reviewed. Despite a great deal of effort, firm evidence has been found in only a few, atypical cases — in clouds which are very close to a luminous quasar, or in objects which have been magnified by gravitational lensing. Recent observations suggest that normal galaxies at z ∼ 3 had star-formation rates about ten times that of the present-day Milky Way. Hence, detections of molecular gas by means of the millimeter transitions of CO will require more sensitive observations than have been attempted to date.

Astrochemistry makes enormous demands on laboratory and theoretical chemistry to obtain line frequencies, molecular parameters and rate coefficients for chemical reactions under interstellar conditions. The astrophysical needs for such large data bases continue to increase. For example, spectral line surveys at submillimeter wavelengths routinely contain thousands of rotational lines, whereas high resolution infrared and optical data require large amounts of information on vibrational and electronic transitions. The astrophysical modeling, both of interstellar clouds and of the envelopes and photospheres of stars, has improved to the stage that highly accurate molecular data are essential.