PRONAOS was born as a project in the mid eighties. The Far Infrared Space Telescope (now Herschel), a Cornerstone of the horizon 2000 plan of ESA, was the clear future of submillimetre astronomy, after several balloon experiments and the tremendous success of the IRAS satellite in the neighbour field of infrared astronomy. Guy Serra and other French scientists proposed to build a learning process in this new field around the PRONAOS project, a technical and scientific precursor to future submm space missions. This project was fully supported by CNES and by CNRS (IAS and CESR at the first place). PRONAOS (Projet National d'AstronOmie Submillimétrique) was based on the use of a two meter telescope aboard a stratospheric balloon-borne gondola. The size and the complexity of the instrument made PRONAOS a very unusual balloon-borne payload. Along the 15 years of the project, and three flights, a lot of learning processes took place. The main goals of this enterprise were reached. New and sometimes surprising scientific results came out from the observations. Teams learned critical scientific and technical lessons from this pioneering experiment. Some contributions to Herschel benefit directly from this experience, and the project Planck, a satellite of ESA, was born inside the PRONAOS team. This project was also a major professional and human experience for young (and also mature) scientists.

The spectroscopic properties in the FIR-mm range of analogue materials for interstellar refractory grains, such as silicates and carbon material, are reviewed. The mechanisms which determine the absorption and its dependence on composition, structure and temperature are discussed.

We have developed a dust emission model in order to reproduce the evolution of the far-infrared (FIR) to sub-millimetre (sub-mm) dust emission properties which is observed in the transition between the diffuse and the dense interstellar medium (ISM). We propose that the coagulation of grains into fluffy aggregates could be the process responsible for the observed dust evolution from the diffuse to the dense ISM in the Galaxy. In dense clouds, dust coagulation can explain the increase in grain size, the decrease in the grain temperature in excess of that predicted by extinction, the sub-mm emissivity excess and the decrease in the small grain abundance.

We have studied the effects of the laboratory-measured, temperature-dependent dust emissivity indices on the FIR/sub-mm emission properties of amorphous silicate and amorphous carbonaceous dust. When the experimentally-determined temperature dependence of the dust emissivity index is taken into account we find temperatures that can differ by as much as 1.5 K and emissivity indices that can differ by ~15%, with respect to the values obtained using published refractive index data as is to calculate these quantities. In addition, we find that the observed diffuse ISM dust temperature of ~17.5 K and emissivity index of 2 can seemingly be produced with a mixture of amorphous carbon and silicate grains only if temperature-dependent dust emissivity indices are adopted. We find that dust masses determined using the temperature-dependent dust emissivity indices are not significantly different from those derived using standard dust properties.

Interstellar gas is thought to be heated by photoelectrons ejected by small grains and PAHs. This process is reviewed and evidence for its importance is examined. In particular, the importance of the PAH charge for the photo electric effect and for the spectral characteristics of interstellar PAHs is discussed.

We present in this paper a review of an astrochemistry study concerning interactions between metals and molecules. The possibility of such an organometallic chemistry is based on the metal depletion in the interstellar medium (ISM) and the large abundance expected for polycyclic aromatic molecules (PAHs). Guy Serra was a pioneer of interstellar organometallic chemistry and contributed to promote this theory.

Strong and broad emission bands commonly attributed to polycyclic hydrocarbons dominate the mid-infrared spectrum of interstellar matter. Early photo- and spectrometry by the ISO and IRTS missions has proved that these bands are widespread in the diffuse interstellar medium. We have sampled the galactic distribution of unidentified infrared band emission in the diffuse ISM component. The remarkable insensitivity of the shape of the UIR band spectrum is seen in this sample; little variation with line strength or galactic position is seen suggesting either a homogeneous history of the UIR band carriers within the Milky Way, convergent evolution of grains suppressing any differences seen in younger populations or an emission mechanism highly insensitive to particle type and physical conditions.

The new astrochemistry experiment PIRENEA is presented. The set-up was designed to approach the physical conditions that prevail in the interstellar medium. It enables to perform original studies in particular when large molecules and nano-sized particles are involved. The experiment is now operational for science.

We present ISOCAM observations (5-18 μm) of illuminated surfaces of molecular clouds. The emission properties of the transiently heated small particles, which dominate the emission observed by ISOCAM, are analysed in relation with the spatial structure of the material. Striking spatial variations of the infrared colour (5-8.5 μm/12-18 μm) are detected. Spectroscopic observations show that they are due to variations of the intensity of the aromatic features (especially at 7.7 μm) relatively to the continuum emission increasing towards long wavelengths. Compared to the intensity of the continuum emission, the intensity of the aromatic features are significantly fainter at the illuminated surfaces of dense structures than in low density regions surrounding these dense structures. This effect could be the signature of photo-chemical evolution, size segregation due to grain dynamics in uni-directional radiation fields, or abundance variations of very small particles.

Recent CO surveys and measurements of the magnetic field intensity and direction have renewed the description of molecular clouds. They appear as a large self-similar hierarchy of structures with a unique power law mass spectrum over 9 orders of magnitude in masses. No lower threshold has yet been observed for this hierarchy in non star forming clouds. At the opposite, in clouds which have started to form stars, the self-similarity breaks down at the scale of the dense cores. The whole hierarchy is fractal. Molecular clouds appear to be in virial balance. The magnitude of the external pressure terms is large in non star forming complexes and very likely due to magnetic stresses. The need for an external pressure decreases as the star formation activity increases and the equality 2T+Ω=0 is reached only for the most massive substructures of GMCs, on the verge of (or already) forming stars. The gas motions are turbulent and likely trans-Alfvénic, given the observed magnetic field intensities. The whole hierarchy of scales is characterized by an invariant, ρvl3/l, the transfer rate of kinetic energy. Molecular cloud cores appear to be supercritical and a static magnetic field cannot stabilize them against gravitational instability. MHD turbulence captures several of the characteristics of molecular clouds and is able, as long as it is fed, to prevent clouds from collapsing. Its dissipation, alike in hydrodynamical turbulence, is concentrated in bursts, which are sheetlike structures.

The MidCourse Space eXperiment (MSX) surveyed the Galactic plane within 5° latitude in four mid-infrared spectral bands. A set of full resolution (20'') 1.5°×1.5° images on 6'' pixel centers has been created in each spectral band by co-adding all the survey data. A lower (1.2') resolution atlas of 10°×10° images provide large-scale panoramas of the plane. A new class of objects has been identified in the images, infrared dark clouds, which are silhouetted against the mid-infrared background emission from the interstellar medium in the Galactic plane. The IRAS ISSA plates indicate that these clouds are dark out to 100 μm. Submillimeter emission traces the form of the dark cloud and reveals cores indicative of class 0 protostars.

Wide area sub-millimeter mapping of nearby molecular clouds allows for the study of large scale structures such as the Integral Shaped Filament in the Orion A cloud. Examination of these regions suggests that they are not equilibrium isothermal structures but rather require significant, and radially dependent, non-thermal support such as produced by helical magnetic fields Also observed in the large area maps are dense condensations with masses typical for stars. The mass distribution of these clumps is similar to the stellar initial mass function; however, the clumps appear stable against collapse. The clumps are clustered within the cores of molecular clouds and restricted to those locations where the molecular cloud column density is high (AV > 4). As well, the typical sub-millimeter clump reveals little or no emission from isotopes of CO, likely indicating that the combination of high density and low temperatures within the clumps provides an environment in which these molecules freeze-out onto dust grain surfaces.

Infrared Astronomy Group of the Tata Institute of Fundamental Research (TIFR) has an ongoing programme of high resolution mapping of star forming regions in two trans-IRAS far infrared bands using a 1 metre balloon-borne telescope. The telescope has been flown many times from TIFR Balloon Facility, Hyderabad, India. The photometer consists of two bolometer arrays cooled to 0.3 K by a closed cycle liquid 3He refrigerator. The effective wavelengths of the two bands are ~140 and 200 μm and the angular resolution of the MEM deconvolved maps is ~1'.2. These bands are suitable for the study of cold sources (temperature ≤30 K). Using this system many Galactic star forming regions have been mapped. Using simultaneous maps in the two bands, maps of dust temperature and optical depth have also been obtained. Our maps are combined with other available data and radiation transfer calculations have been made. In this paper we will describe the telescope and give a sample of the results obtained for some of the star forming regions in the Galaxy.

We review the submillimetric PRONAOS observations in the direction of the IC 1318a nebular located at the edge of an ionized region in the Cygnus X complex. The extended emission measured in the PRONAOS four broad photometric bands extending from 200 up to 1100 μm enables a full characterization of the interstellar dust big grains component radiating at thermal equilibrium, leading to a temperature and column density mapping.

The ISOGAL survey discovered a population of narrow (down to 10 arcsec in size) absorbing filaments and globules which appear as dark features in the 7 μm and 15 μm plates. These infrared dark clouds turn out to be the highly condensed parts of Giant Molecular Clouds (GMC's) kiloparsecs away from the Sun. We report here follow-up observations at radio-mm wavelengths conducted at the IRAM 30-m telescope, and present the first results of the physico-chemical analysis of the gas associated to the dense dust revealed in mid-IR.

The maximum luminosity of protostars forming in molecular clouds is investigated as a function of the parent cloud mass on the basis of a rich cloud sample searched for in the literature. In total, we gathered 499 molecular clouds among the published data, out of which 243 clouds are found to be associated with candidates for protostars selected from the IRAS point source catalog. A diagram of the maximum stellar luminosity in each cloud and the parent cloud mass shows that the young stars in the clouds associated with regions are apparently more luminous than those in clouds away from HII regions over the entire cloud mass range investigated (1 < MCL/M⊙ < 106). In addition, we found that there are well-defined upper and lower limits in the maximum stellar luminosity distribution with the lower limit having a steeper dependence on the cloud mass (LMAX ∝ MCL1.5) than the upper one (LMAX ∝ MCL0.8). All these features could be naturally accounted for if we assume that the luminosity function of protostars is controlled by the cloud mass and the external pressure imposed on the cloud surface. We introduce a simple model for the stellar luminosity function as a function of these quantities.

L134N is a cold, starless cloud, very high above the galactic plane, close to us and well delineated in continuum dust emission maps. This cloud is considered to be representative of oxygen rich dark clouds (with the presence of SO, SO2, NO, ...). It is thus a good reference together with TMC-1 to test astrochemical models. Thanks to ISO, SCUBA and near IR wide field cameras, the detailed study of the dust has become possible in such cold and dark clouds. In parallel, progress in radiotelescope receiver sensitivity now allows to map weak lines on large surfaces. We have thus started a project to study both dust and a few gaseous key species (CO, CS, SO and N2H+) to address several questions. We want to assess the quantity of dust and gas all over the cloud, study possible C18O and/or C17O depletion towards dense cores, evaluate the structure of the gas, the abundance of CS and SO to possibly estimate the chemical age of the cloud (time dependent models show that the CS/SO ratio diminishes with time) and evaluate the rare isotope abundances, especially 17O and 34S in a first step. To constrain the molecular abundances with the highest possible confidence, we have observed several transitions for each species and each isotopomer. Though we have observed far less species than Dickens et al. [1], we have done it on a larger area, including thus the strongest C18O peak and two other peaks, with a better signal-to-noise ratio. Most of the data are already acquired. We present here preliminary results.

We provide insight into the origin of the far-IR to mm emission from the Galaxy by presenting a decomposition of the emission where we separate the contributions from dust in the atomic gas and dust associated with quiescent molecular gas, as a function of Galacto-centric distance. We find that the sub-mm dust emissivity per hydrogen change by a factor 3 between the diffuse ISM and molecular clouds, first observed on localized clouds in the Solar Neighborhood, applies on large scale to the Molecular Ring. The decomposition leaves an emission excess in the sub-mm with a very cold effective temperature, concentrated in the plane but with a brightness independent of Galactic longitude. This emission might be associated with interstellar matter in the outer Galaxy not traced by H I nor CO emission.