Abstract

Massive stars and the crystallinity of interstellar silicates

Sofia Wallström (ASIAA)

Peter Scicluna (ASIAA) Ciska Kemper (ASIAA)

Crystalline silicates are observed in the ISM of other galaxies (Spoon et al., 2006); in particular, starburst galaxies with crystallinities up to 95% have been observed (Aller et al., 2012). As the timescale for amorphisation by cosmic rays is relatively short (e.g. Bringa et al., 2007), the formation of crystalline material must be associated with massive stars or the starburst itself. We identify three possible sources of silicates injected into the ISM by massive stars: First, the condensation of silicates in the gradual winds of pre-Supernova phases, in particular Red Supergiants; second, the relative fast direct condensation of silicates in Supernova, and finally, the ejection of silicates entrained in jets coming from massive YSOs. The formation and processing of large quantities of crystalline dust is difficult to explain: most massive stellar sources of dust have relatively low crystallinities (e.g. up to a few percent in Red Supergiants (Harwit et al., 2001), and similar in Luminous Blue Variables (Guha Niyogi et al., 2014) and B[e] stars (Kastner et al. 2010), although very few sources have detections) and crystalline silicates have not been detected in any supernova remnants to date. Only a small number of supernova remnants have been targeted for mineralogy studies (Bouchet et al. 2006, Rho et al. 2009, Arendt et al. 2014), while the reported detection of crystalline forsterite in SNR G292.0+1.8 by Ghavamian et al. (2009) appears to be foreground emission. A possible environment where crystalline silicates may form is stable, long-lived discs, where material can be processed at high temperatures and ejected in a disc wind. Such discs might be found around massive interacting binaries or massive YSOs. In particular, massive YSOs provide a means to process existing dust through high-temperature or shock annealing (Harker & Desch, 2002), and recent work has suggested that accretion efficiency of massive stars may be as low as 10% (Haemmerlé et al., 2016), implying that 90% of the natal cloud material will be re-injected into the ISM. This provides a means to recycle existing interstellar dust and enrich its crystalline fraction without altering the chemistry of the dust. We will conduct a mid-infrared spectroscopic survey of evolved massive stars, supernova remnants and massive star-forming regions throughout the Local Group, with a particular focus on the high-metallicity environments of M31 and M33, but not limiting ourselves to these environments. Mid-infrared spectroscopy has been shown to be extremely useful in disentangling crystalline and amorphous silicates (Molster & Kemper, 2005). The observations will be interpreted using a robust radiative-transfer model-fitting framework to constrain the mineralogy of the dust produced in each of the environments in question. By comparing these constraints to identical analyses of existing observations of Galactic and Magellanic sources, we will have a homogeneous sample of crystalline silicate producers which can be used to identify the most likely culprits in distant starbursts.

Mode of presentation: poster