Context. The inner layers of circumstellar envelopes around asymptotic giant branch stars are sites where a variety of processes such as thermochemical equilibrium, shocks induced by the stellar pulsation, and condensation of dust grains determine the chemical composition of the material that is expelled into the outer envelope layers and, ultimately, into interstellar space. Aims. We aim at studying the abundances, throughout the whole circumstellar envelope of the carbon star IRC +10216, of several molecules formed in the inner layers in order to constrain the different processes at work in such regions. Methods. Observations towards IRC +10216 of CS, SiO, SiS, NaCl, KCl, AlCl, AlF, and NaCN have been carried out with the IRAM 30-m telescope in the 80−357.5 GHz frequency range. A large number of rotational transitions covering a wide range of energy levels, including highly excited vibrational states, are detected in emission and serve to trace different regions of the envelope. Radiative transfer calculations based on the LVG formalism have been performed to derive molecular abundances from the innermost out to the outer layers. The excitation calculations include infrared pumping to excited vibrational states and inelastic collisions, for which up-to-date rate coefficients for rotational and, in some cases, ro-vibrational transitions are used. Results. We find that in the inner layers CS, SiO, and SiS have abundances relative to H 2 of 4 × 10 −6 , 1.8 × 10 −7 , and 3 × 10 −6 , respectively, and that CS and SiS have significant lower abundances in the outer envelope, which implies that they actively contribute to the formation of dust. Moreover, in the inner layers, the amount of sulfur and silicon in gas phase molecules is only 27% for S and 5.6% for Si, implying that these elements have already condensed onto grains, most likely in the form of MgS and SiC. Metal-bearing molecules lock up a relatively small fraction of metals, although our results indicate that NaCl, KCl, AlCl, AlF, and NaCN, despite their refractory character, are not significantly depleted in the cold outer layers. In these regions a few percent of the metals Na, K, and Al survive in the gas phase, either in atomic or molecular form, and are therefore available to participate in the gas phase chemistry in the outer envelope.

Context. The15 N isotopologue abundance ratio measured today in different bodies of the solar system is thought to be connected to 15 N-fractionation effects that would have occurred in the protosolar nebula. Aims. The present study aims at putting constraints on the degree of 15 N-fractionation that occurs during the prestellar phase, through observations of D, 13 C, and 15 N-substituted isotopologues towards B1b. Molecules both from the nitrogen hydride family, i.e. N 2 H + , and NH 3 , and from the nitrile family, i.e. HCN, HNC, and CN, are considered in the analysis. Methods. As a first step, we modelled the continuum emission in order to derive the physical structure of the cloud, i.e. gas temperature and H 2 density. These parameters were subsequently used as input in a non-local radiative transfer model to infer the radial abundance profiles of the various molecules. Results. Our modelling shows that all the molecules are affected by depletion onto dust grains in the region that encompasses the B1-bS and B1-bN cores. While high levels of deuterium fractionation are derived, we conclude that no fractionation occurs in the case of the nitrogen chemistry. Independently of the chemical family, the molecular abundances are consistent with 14 N/ 15 N ∼ 300, a value representative of the elemental atomic abundances of the parental gas. Conclusions. The inefficiency of the 15 N-fractionation effects in the B1b region can be linked to the relatively high gas temperature ∼17 K, which is representative of the innermost part of the cloud. Since this region shows signs of depletion onto dust grains, we cannot exclude the possibility that the molecules were previously enriched in 15 N, earlier in the B1b history and that such an enrichment could have been incorporated into the ice mantles. It is thus necessary to repeat this kind of study in colder sources to test such a possibility.

Aims. This work deals with the rotational excitation of ortho/para-H 2 O with para/ortho-H 2 so that thermalized de-(excitation) rate coefficients up to 1500 K for the first 45th level of ortho/para-H 2 O are provided. Results are available in BASECOL with state-to-state rate coefficients, their fitting coefficients, and effective rate coefficients. In addition, we provide a routine that combines all data in order to create thermalized rate coefficients. Methods. Calculations were performed with the close coupling (CC) method over the whole energy range, using the same 5D potential energy surface (PES) as the one employed in previous papers. The current CC results were compared with thermalized quasi-classical trajectory (QCT) calculations using the same PES and with previous quantum calculations obtained between T = 20 K and T = 140 K with a different PES. The relative strengths of water excitation rate coefficients when water is excited with ortho-H 2 versus para-H 2 was also analyzed. Results. For collision with para-H 2 , the rotation-rotation process is found to be the dominant process for inelastic transfer for some water transitions, implying that calculations must include the j 2 = 2 level. An important result of this paper is that j 2 = 1 and j 2 = 2 effective rate coefficients are very similar so that either j 2 = 1 or j 2 = 2 need to be calculated for astrophysical applications. In addition, at high temperature ratios of j 2 = 2 (1) over j 2 = 0, effective rate coefficients converge towards one to within a few percent. This study confirms that j 2 = 3 effective rate coefficients are within 20% to j 2 = 1 effective rate coefficients. Conclusions. For astrophysical applications, these results imply that future collisional excitation of light molecules with H 2 should be carried out with para-H 2 , including j 2 = 2, so as to obtain correct effective j 2 = 0 effective rate coefficients and using the j 2 = 2 effective rate coefficients for all excited j 2 effective rate coefficients. In contrast, collisional excitation of heavy molecules with H 2 might be restricted to para-H 2 with j 2 = 0 and to ortho-H 2 with j 2 = 1, using the j 2 = 1 rate coefficients for all excited j 2 effective rate coefficients. These conclusions should simplify the future methodological choice for collisional excitation calculations applied to interstellar/circumstellar media.

Aims. Our objective is to obtain the best possible set of rotational (de)-excitation state-to-state and effective rate coefficients for temperatures up to 1500 K. We present state-to-state rate coefficients among the 45 lowest levels of o-H 2 O with H 2 ( j 2 = 0) and Δ j 2 = 0, +2, as well as with H 2 ( j 2 = 2) and Δ j 2 = 0, −2. In addition and only for the 10 lowest energy levels of o-H 2 O, we provide state-to-state rate coefficients involving j 2 = 4 with Δ j 2 = 0, −2 and j 2 = 2 with Δ j 2 = +2. We give estimates of effective rate coefficients for j 2 = 6, 8. Methods. Calculations are performed with the close coupling (CC) method over the whole energy range, using the same 5D potential energy surface (PES) as the one employed in our latest publication on water. We perform comparisons with coupled states (CS) calculations, with thermalized quasi-classical trajectory (QCT) calculations using the same PES and with previous quantum calculations obtained between T = 20 K and T = 140 K with a different PES. Results. We find that the CS approximation fares extremely badly even at high energy for j 2 different from zero. Comparisons with thermalized QCT calculations show large factors at intermediate temperatures and factors from 1 to 3 at high temperature for the strongest rate coefficients. Finally we stress that scaled collisional rate coefficients obtained with He cannot be used in place of collisional rate coefficients with H 2 .