Constraints on the H₂O formation mechanism in the wind of carbon-rich AGB stars^⋆

¹ Department of Earth and Space Sciences, Chalmers University of Technology, Onsala Space Observatory, 439 92 Onsala, Sweden
e-mail: lombaert@chalmers.se
² KU Leuven, Instituut voor Sterrenkunde, Celestijnenlaan 200D 2401, 3001 Leuven, Belgium
³ University of Amsterdam, Astronomical Institute “Anton Pannekoek”, PO Box 94249, 1090 GE Amsterdam, The Netherlands
⁴ Universidad de Alcalá de Henares, Departamento de Física y Matemáticas, Campus Universitario, 28871 Alcalá de Henares, Madrid, Spain
⁵ Johns Hopkins University, Department of Physics and Astronomy, 3400 North Charles Street, Baltimore, MD 21218, USA
⁶ Group of Molecular Astrophysics, Instituto de Ciencia de Materiales de Madrid, CSIC, C/Sor Juana Inés de La Cruz N3, 28049 Cantoblanco, Madrid, Spain
⁷ Vrije Universiteit Brussel, Department of Physics and Astrophysics, Pleinlaan 2, 1050 Brussels, Belgium
⁸ Koninklijke Sterrenwacht van België, Ringlaan 3, 1180 Brussels, Belgium
⁹ University of Vienna, Department of Astrophysics, Türkenschanzstraße 17, 1180 Wien, Austria

Received: 23 July 2015
Accepted: 12 January 2016

Abstract

Context. The recent detection of warm H₂O vapor emission from the outflows of carbon-rich asymptotic giant branch (AGB) stars challenges the current understanding of circumstellar chemistry. Two mechanisms have been invoked to explain warm H₂O vapor formation. In the first, periodic shocks passing through the medium immediately above the stellar surface lead to H₂O formation. In the second, penetration of ultraviolet interstellar radiation through a clumpy circumstellar medium leads to the formation of H₂O molecules in the intermediate wind.

Aims. We aim to determine the properties of H₂O emission for a sample of 18 carbon-rich AGB stars and subsequently constrain which of the above mechanisms provides the most likely warm H₂O formation pathway.

Methods. Using far-infrared spectra taken with the PACS instrument onboard the Herschel telescope, we combined two methods to identify H₂O emission trends and interpreted these in terms of theoretically expected patterns in the H₂O abundance. Through the use of line-strength ratios, we analyzed the correlation between the strength of H₂O emission and the mass-loss rate of the objects, as well as the radial dependence of the H₂O abundance in the circumstellar outflow per individual source. We computed a model grid to account for radiative-transfer effects in the line strengths.

Results. We detect warm H₂O emission close to or inside the wind acceleration zone of all sample stars, irrespective of their stellar or circumstellar properties. The predicted H₂O abundances in carbon-rich environments are in the range of 10^-6 up to 10^-4 for Miras and semiregular-a objects, and cluster around 10^-6 for semiregular-b objects. These predictions are up to three orders of magnitude greater than what is predicted by state-of-the-art chemical models. We find a negative correlation between the H₂O/CO line-strength ratio and gas mass-loss rate for Ṁ_g> 5 × 10^-7 M_⊙ yr^-1, regardless of the upper-level energy of the relevant transitions. This implies that the H₂O formation mechanism becomes less efficient with increasing wind density. The negative correlation breaks down for the sources of lowest mass-loss rate, the semiregular-b objects.

Conclusions. Observational constraints suggest that pulsationally induced shocks play an important role in warm H₂O formation in carbon-rich AGB stars, although photodissociation by interstellar UV photons may still contribute. Both mechanisms fail in predicting the high H₂O abundances we infer in Miras and semiregular-a sources, while our results for the semiregular-b objects are inconclusive.