Giants in the globular cluster omega Centauri: dust production, mass-loss and distance
McDonald, Iain × van Loon, Jacco Th Decin, Leen Boyer, Martha L Dupree, Andrea K Evans, Aneurin Gehrz, Robert D Woodward, Charles E #
Wiley-blackwell publishing, inc
Monthly notices of the royal astronomical society vol:394 issue:2 pages:831-856
We present spectral energy distribution modelling of 6875 stars in omega Centauri, obtaining stellar luminosities and temperatures by fitting literature photometry to state-of-the-art MARCS stellar models. By comparison to four different sets of isochrones, we provide a new distance estimate to the cluster of 4850 +/- 200 (random error) +/- 120 (systematic error) pc, a reddening of E(B V) = 0.08 +/- 0.02 (random) +/- 0.02 (systematic) mag and a differential reddening of Delta E(B-V) < 0.02 mag for an age of 12 Gyr. Several new post-early-AGB candidates are also found. Infrared excesses of stars were used to measure total mass-loss rates for individual stars down to similar to 7 x 10(-8) M-circle dot yr(-1). We find a total dust mass-loss rate from the cluster of 1.3 +/- (0.8)(0.5) x 10(-9) M-circle dot yr(-1), with the total gas mass-loss rate being > 1.2 +/- (0.6)(0.5) x 10(-6) M-circle dot yr(-1). Half of the cluster's dust production and 30 per cent of its gas production comes from the two most extreme stars - V6 and V42 - for which we present new Gemini/T-ReCS mid-infrared spectroscopy, possibly showing that V42 has carbon-rich dust. The cluster's dust temperatures are found to be typically greater than or similar to 550 K. Mass-loss apparently does not vary significantly with metallicity within the cluster, but shows some correlation with barium enhancement, which appears to occur in cooler stars, and especially on the anomalous RGB. Limits to outflow velocities, dust-to-gas ratios for the dusty objects and the possibility of short-time-scale mass-loss variability are also discussed in the context of mass-loss from low-metallicity stars. The ubiquity of dust around stars near the RGB tip suggests significant dusty mass-loss on the RGB; we estimate that typically 0.20-0.25 M-circle dot of mass-loss occurs on the RGB. From observational limits on intracluster material, we suggest the dust is being cleared on a time-scale of less than or similar to 10(5) yr.