The global gas and dust budget of the Large Magellanic Cloud: AGB stars and supernovae, and the impact on the ISM evolution

The origin of the dust in the ISM is less well understood. Dust forms in stellar ejecta at temperatures of about 1000-1500 K and at high densities. Different stars produce different types of dust. Red supergiants produce oxygen-rich (silicate) dust. AGB stars have two distinct chemical types: oxygen-and carbon-rich, depending on the abundance ratio of oxygen and carbon atoms within their atmospheres. Oxygen-rich stars form silicate dust, while carbon-rich stars yield amorphous carbon, graphite and SiC dust. Supernovae can produce both dust types, depending on the abundances in the different layers of the ejecta (Rho et al. 2008).

The current rate of ISM enrichment by dust and gas depends on the total stellar population, the initial mass function and the star-forming history (e.g. Salpeter 1955). Type II SNe are expected to dominate the enrichment in the early phases of galaxy evolution (e.g. Maeder 1992). Hirashita & Ferrara (2002) argue that dust grains injected from SNe into the ISM accelerate star formation in young galaxies. It takes more than 100 Myr for the first intermediate-mass stars to evolve onto the AGB (e.g. Vassiliadis & Wood 1993). Thus, dust and gas enrichment from AGB stars occurs later than from high-mass stars. Different galaxies, at different stages of this process, may be expected to show differences in gas-to-dust ratios, dust content, and, in consequence, ISM dust extinction curves.

In our Galaxy, the major dust sources are presumed to be AGB stars and SNe (Gehrz 1989). Some other sources, such as Wolf-Rayet stars and novae, also contribute dust to the ISM of the Milky Way, but only in small quantities (Gehrz 1989). The relative importance of AGB stars and SNe remains uncertain. Dwek (1998) suggests that SNe are important silicate dust sources, while most carbonaceous dust grains are from carbon-rich AGB stars. Jura & Kleinmann (1989) show that AGB stars are an important gas source within our Galaxy, except in the Galactic plane where SNe appear to dominate. The dust formation efficiency in SNe remains controversial (e.g. Sugerman et al. 2006), since SN shocks also destroy dust (Tielens 1994) and since dust production prior the explosion remains unclear.

There are also uncertainties in the dust formation efficiencies for AGB stars. This is expected to depend on the mass-loss rate, chemical type, and metallicity. For oxygen-rich stars mass-loss rates decrease towards lower metallicity (Wood et al. 1998;Bowen & Willson 1991;Marshall et al. 2004). In contrast, for carbonrich stars, although the metallicity dependence of massloss rates is still unclear (Habing 1996;van Loon 2000;Wachter et al. 2008;Mattsson et al. 2008), there does not appear to be any obvious metallicity dependence in the range from solar, down to one-twentieth of the solar metallicity (Groenewegen et al. 2007;Matsuura et al. 2007;Sloan et al. 2009). However, we would expect the mass-loss rates of carbon stars to differ between environments with different abundances of alpha elements. In these cases the initial abundance of oxygen in the carbon-star precursor would be vital in determining how much carbon must be dredged up before any was free to become dust.

Galaxy evolution models aim at fitting the current composition of the ISM by following the evolution of the stellar population over the lifetime of the galaxy. This requires a detailed knowledge of dust formation efficiencies, gas and dust expulsion efficiencies and nucleosynthesis in stellar interiors. Here we aim to measure the current integrated rate of gas and dust input into the ISM for one specific galaxy: the LMC. Recent observations by the Spitzer Space Telescope (hereafter Spitzer (Werner et al. 2004))

…(Full text truncated)…