[Paper Review] The nature of the circumstellar CO_2 emission from M giants

N. Ryde, Kimmo Eriksson+1 more|arXiv (Cornell University)|Nov 20, 1998

Astronomy and Astrophysical Research3 references5 citations

TL;DR

This paper investigates the origin of circumstellar CO₂ emission in M-type giant stars using ISO infrared spectra. It proposes a two-layer model—warm, dense gas near the star and cold, extended outflowing material—explaining spectral band differences and emission strengths, with band widths enabling temperature estimates despite non-LTE conditions.

ABSTRACT

The 13-16 um region observed by the Infrared Space Observatory (ISO) of several ABG stars are discussed. We present and analyse spectra of seven M giants which show carbon dioxide features. To explain the features of the bands, we suggest they originate from two different circumstellar layers, one being a warm and high density layer close to the star, possibly making the 15 um band optically thick, and the other being a large, cold and optically thin layer extending far out in the wind. This could explain the difference in temperatures of the different bands found in the analysis of the spectra and the number of molecules needed for the emission. It is demonstrated that in spite of the bands probably not being formed in vibrational LTE, the temperatures can be estimated from the widths of the bands.

Motivation & Objective

• Understand the physical origin of CO₂ emission features in the 13–16 μm range observed in M-type giants.
• Address discrepancies in derived temperatures and molecule numbers from spectral features.
• Determine whether CO₂ emission arises from a single or multiple circumstellar layers.
• Assess the validity of using band widths to estimate temperatures under non-LTE conditions.
• Reconcile observed emission strengths with the required number of CO₂ molecules in the circumstellar environment.

Proposed method

• Analysis of ISO short-wavelength infrared spectra (13–16 μm) of seven M giant stars with CO₂ features.
• Modeling of CO₂ emission using a two-component circumstellar structure: a warm, dense inner layer and a cold, extended outer layer.
• Use of band profile widths to estimate excitation temperatures, even when vibrational non-LTE conditions prevail.
• Comparison of observed band shapes and depths with synthetic spectra to constrain physical parameters.
• Evaluation of optical depth effects, particularly for the 15 μm band, to assess whether it is optically thick.
• Estimation of required CO₂ column densities based on observed emission strengths and model assumptions.

Experimental results

Research questions

• RQ1 What causes the observed differences in effective temperatures derived from various CO₂ bands in M giants?
• RQ2 Can the observed CO₂ emission be explained by a single circumstellar layer, or is a multi-layer model necessary?
• RQ3 To what extent can band width measurements provide reliable temperature estimates in non-LTE conditions?
• RQ4 Why is the 15 μm CO₂ band often optically thick, and what does this imply about the inner circumstellar environment?
• RQ5 How do the required CO₂ molecule numbers compare with the observed emission intensities, and what does this imply about the mass-loss geometry?

Key findings

• The CO₂ emission arises from two distinct circumstellar layers: a warm, dense inner layer close to the star and a cold, extended outer layer.
• The 15 μm CO₂ band is likely optically thick, indicating high column density in the inner region.
• Despite non-LTE conditions, band widths provide reliable estimates of excitation temperatures.
• The model explains the observed range of temperatures across different CO₂ bands through spatial and thermal separation of emitting regions.
• The required number of CO₂ molecules is consistent with the observed emission only when both warm and cold layers are included.
• The two-layer model reconciles discrepancies between derived temperatures and molecule numbers, supporting a complex, structured mass-loss environment.