The dust condensation sequence in red super-giant stars

Context: Red super-giant (RSG) stars exhibit significant mass loss through a slow and dense wind. They are often considered to be the more massive counter parts of Asymptotic Giant Branch (AGB) stars. While the AGB mass loss is linked to their strong pulsations, the RSG are often only weakly variable. Aim: To study the conditions at the base of the wind, by determining the dust composition in a sample of RSG. The dust composition is thought to be sensitive to the density, temperature and acceleration at the base of the wind. Method: We compile a sample of 27 RSG infrared spectra (ISO-SWS) and supplement these with photometric measurements to obtain the full spectral energy distribution (SED). These data are modelled using a dust radiative transfer code. The results are scrutinised for correlations. Results: We find (1) strong correlations between dust composition, mass-loss rate and stellar luminosity, roughly in agreement with the theoretical dust condensation sequence, (2) the need for a continuous (near-)IR dust opacity and tentatively propose amorphous carbon, and (3) significant differences with AGB star winds: presence of PAHs, absence of ’the’ 13 micron band, and a lack of strong water bands. Conclusions: Dust condensation in RSG is found to experience a similar freeze-out process as in AGB stars. Together with the positive effect of the stellar luminosity on the mass-loss rate, this suggests that radiation pressure on dust grains is an important ingredient in the driving mechanism. Still, differences with AGB stars are manifold and thus the winds of RSG deserve separate studies.

💡 Research Summary

The paper investigates the dust formation and wind‑driving mechanisms of red super‑giant (RSG) stars by analysing a homogeneous sample of 27 ISO‑SWS infrared spectra complemented with broadband photometry to construct full spectral energy distributions (SEDs). Using a one‑dimensional dust radiative‑transfer code, the authors fit each SED with a mixture of candidate dust species—silicates (magnesian, phosphates, olivine), metal oxides (FeO, Fe₂O₃, etc.), and carbonaceous components (amorphous carbon, PAHs). Stellar parameters (effective temperature, luminosity, distance) are taken from the literature, while the mass‑loss rate is derived from the best‑fit model.

The analysis reveals several robust correlations. First, the mass‑loss rate scales positively with stellar luminosity, confirming that higher radiation pressure on dust grains can more efficiently accelerate the wind. Second, dust composition varies systematically with mass‑loss rate: at low rates (~10⁻⁶ M⊙ yr⁻¹) the spectra are dominated by high‑temperature condensates such as Al‑Ca silicates, whereas at higher rates the contribution of cooler silicates and metal oxides grows. This trend mirrors the theoretical dust condensation sequence proposed for asymptotic giant branch (AGB) stars, where early‑forming refractory grains “freeze‑out” and later‑forming species condense as the outflow expands and cools.

A notable finding is the requirement for a continuous near‑infrared opacity that cannot be reproduced by silicates alone. The authors tentatively assign this opacity to amorphous carbon, which provides a smooth absorption component across the 1–5 µm region. This suggestion departs from the classic AGB paradigm, where carbonaceous dust appears only in carbon‑rich stars, and implies that RSG atmospheres may host a modest amount of carbon dust even when the overall chemistry is oxygen‑rich.

Spectroscopically, the RSG sample exhibits clear polycyclic aromatic hydrocarbon (PAH) emission features, a hallmark absent in most AGB spectra. Conversely, the characteristic 13 µm band (often attributed to Al₂O₃) and strong water vapor bands that dominate many AGB SEDs are either very weak or missing in the RSG data. These differences point to a hotter, less water‑rich circumstellar environment in RSGs, likely a consequence of their larger radii, higher surface gravities, and weaker pulsations.

The authors also identify a threshold mass‑loss rate (~10⁻⁵ M⊙ yr⁻¹) beyond which the dust optical depth becomes large enough for radiation pressure to overcome gravity, providing a quantitative support for the dust‑driven wind hypothesis in RSGs. However, because RSGs display only modest variability, the initiation of dust formation may rely on a combination of radiative cooling, turbulent motions, and possibly magnetic fields, rather than the strong pulsation‑induced shock waves that dominate AGB mass loss.

In summary, the study demonstrates that red super‑giants share the fundamental dust‑condensation sequence and radiation‑pressure‑driven wind physics with AGB stars, yet they also possess distinct chemical signatures (PAHs, lack of 13 µm and water bands) and require an additional near‑IR opacity component, plausibly amorphous carbon. These findings underscore that while the overall framework of dust‑mediated mass loss applies across the upper‑HR diagram, the detailed wind physics of RSGs warrants dedicated observational and theoretical work, especially high‑resolution infrared spectroscopy and time‑domain monitoring to capture the interplay between stellar luminosity, dust formation, and wind acceleration.