G. C. Sloan: Dust production in Local Group galaxies

Dust production in Local Group galaxies

Main Collaborators

Albert Zijlstra, Eric Lagadec (Manchester Univ.)
Peter Wood (Australian National Univ.)
Kathleen E. Kraemer (Air Force Research Lab., Hanscom AFB)
Jeronimo Bernard-Salas (Cornell)
Mikako Matsuura (National Astronomical Obs. of Japan)
Martin Groenwegen (Univ. of Leuven)
Jacco van Loon (Univ. of Keele)
(There are many more astronomers involved in our projects.)

Introduction
Carbon stars
Future
References

Introduction

The Spitzer Space Telescope has proven to be a sensitive and powerful telescope, making it possible to observe large numbers of individual evolved stars in nearby galaxies in the Local Group. To understand why this capability is so important, it helps to know what astronomers mean when they say "metallicity." To astronomers, every element heavier than helium is a metal. It might seem odd to think of carbon, nitrogen, or oxgyen as metals, but when an astronomer talks about metallicity, they are talking about the abundances of everything in the Universe heavier than the lightest two elements, which really boils down to C, N, and O, since they dominate what is left.

As a galaxy evolves, each succeeding generation of stars produces metals in its core through fusion reactions, then regurgitates this metal-enriched material back into space. Thus, each new generation of stars starts with a higher metallicity than the previous one. In galaxies like the Milky Way, the disk is relatively metal-enriched, because stars have been actively forming since the Milky Way formed 13-14 billion years ago. But other galaxies, including most of those near our own, have much lower rates of star formation. With fewer stars forming to produce heavier elements, they have remained relatively metal-poor. As a result, these nearby galaxies have abundances similar to the primordial abundances of all galaxies when they had just formed.

By studying how dying stars in these nearby metal-poor galaxies eject mass and produce dust, we can understand how these processes worked early in the history of the Universe. These processes of mass loss and dust production depend critically on the abundances the star formed with in the first place, so we have a complicated feedback loop, with the metallicity affecting how stars enrich space, which affects the metallicity of succeeding generations of stars. Using Spitzer to study Local Group galaxies, we hope to break the feedback loop. This project is ongoing, and we are still working to analyze our data, but the initial round of papers has provided some thought-provoking clues.

The Local Group galaxies which have drawn the most attention are the Large Magellanic Cloud (LMC) and the Small Magellanic Cloud (SMC).

Galaxy Metallicity Distance

Milky Way solar 0-30 kpc

LMC 1/2 solar 50 kpc

SMC 1/5 solar 60 kpc

Galaxy	Metallicity	Distance
Milky Way	solar	0-30 kpc
LMC	1/2 solar	50 kpc
SMC	1/5 solar	60 kpc

Carbon stars

As explained on the page about AGB stars, stars on the asymptotic giant branch can become carbon rich. The minimum initial mass for a star to become a carbon star depends on metallicity. The lower the metallicity, the more stars can become carbon stars. Several different programs used Spitzer to study evolved stars in the LMC and SMC, and the first realization we all made was that our samples contained a lot of carbon stars. As a result, the initial round of papers, concentrated on carbon-rich objects, with a couple of exceptions.

Kathleen Kraemer published the very first paper from these samples (Kraemer et al. 2005). She found two R Corona Borealis stars in the LMC. R CrB stars are carbon-rich, post-AGB objects, and they are extremely rare. To find two in a sample of only 36 sources is unusual. The next paper was one of the exceptions. Jacco van Loon led it, and he reported a possible young stellar object in the LMC (van Loon et al. 2005). The other exception was the next paper (Sloan et al. 2006a), which focused on HV 2310, an oxygen-rich Mira in the LMC with some of the most unusual circumstellar dust I had yet seen. The dust shows evidence of warm crystalline grains which affect the 10 µm silicate feature.

The next seven papers were all on carbon-rich objects. I published the next paper on the carbon stars in the SMC sample chosen by Mike Egan, Kathleen Kraemer, and me (Sloan et al. 2006b). This paper was quickly followed by the LMC sample from Peter Wood's program (Zijlstra et al. 2006) and Mikako Matsuura's analysis of the carbon-rich gas in these stars (Matsuura et al. 2006). Then the various groups published two back-to-back papers on post-AGB objects (Kraemer et al. 2006 and Bernard-Salas et al. 2006). Next came Martin Groenwegen's radiative transfer models of the spectra (Groenewegen et al. 2007) and Eric Lagadec's study of the SMC carbon stars in Peter Wood's program (Lagadec et al. 2007). The recent paper by Leisenring et al. (2008) expands the sample of carbon stars to include those observed by Ciska Markwick-Kemper.

The latest paper in the series (as of July, 2008, Sloan et al. 2008) takes the first look at the oxygen-rich sources in the Magellanic samples, and it looks back a the carbon stars, using Martin Groenewegen's models, which showed that the color indices we developed for the carbon stars can be used to directly measure how much dust they produce.

Major findings:

As stars grow more metal-poor, those stars producing silicates and other oxygen-rich dust produce less of it. We even see fewer molecules that lead to dust, namely SiO.
Carbon stars, on the other hand, are completely different. Try as we might, we can't see any evidence that the amount of dust produced by a carbon stars depends on metallicity. This finding, if confirmed, is key. It implies that the most metal-poor galaxies will be producing plenty of carbon-rich dust but hardly any oxygen-rich dust, which means that the dust in the early Universe should hae a much different chemistry from galaxies like the Milky Way today.

Other important findings:

As carbon stars grow more metal-poor, they produce less SiC and MgS dust relative to the amorphous carbon which dominates the dust shell.
As stars grow more metal poor, the absorption bands from gaseous acetylene (C₂H₂) grow stronger.
Mikako's study revealed that acetylene dominated the gas, with little or no evidence for HCN, which is often seen in Galactic carbon stars.
The post-AGB objects revealed dust in transition from amorphous carbon to polycyclic aromatic hydrocarbons (PAHs) as the mass-loss from the central stars exposes the hot core, which then heats up the surrounding dust.

Future

Our focuses for the future are three-fold. First, we are now analyzing data from other Local Group galaxies which are even more metal poor. Mikako Matsuura has already published the first of these papers, on carbon stars in the Fornax Dwarf Spheroidal Galaxy (Matsuura et al. 2007). She finds that the carbon stars in this galaxy, which is even more metal-poor than the SMC, look a lot like SMC stars.

Second, we are now examining in more detail the stars in our Magellanic Cloud samples which are not carbon stars. The first of these papers (by Sloan et al. 2008, as noted above) has already come out and has us pretty excited. We're now working on the oxygen-rich stars in the other samples.

Third, we will soon have a paper ready on the dust around evolved stars in globular clusters, which are metal-poor like many of the small galaxies in the Local Group, but have been stripped of their gas, which has prevented them from making any stars except for those they originally formed with.

References

Bernard-Salas, J., et al. 2006, “The Spitzer-IRS spectrum of SMP LMC 11” ApJ Letters, 652, L29.

Groenewegen, M.A.T., et al. 2007, “Luminosities and mass-loss rates of carbon stars in the Magellanic Clouds,” MNRAS, 376, 313.

Kraemer, K.E., et al. 2005, “R CrB candidates in the Small Magellanic Cloud: Observations of cold, featureless dust with the Spitzer Infrared Spectrograph,” ApJ Letters, 631, L147.

Kraemer, K.E., et al. 2006, “A post-AGB star in the Small Magellanic Cloud observed with the Spitzer Infrared Spectrograph,” ApJ Letters, 652, L25.

Lagadec, E., et al. 2007, “Spitzer mid-infrared spectra of AGB stars in the Small Magellanic Cloud,” MNRAS, 376, 1270.

Leisenring, J.M., et al. 2008, “Effects of metallicity on the chemical composition of carbon stars,” ApJ, 681, 1557.

Matsuura, M., et al. 2006, “Spitzer observations of acetylene bands in carbon-rich AGB stars in the Large Magellanic Cloud,” MNRAS, 371, 415.

Matsuura, M., et al. 2007, #147;Spitzer Space Telescope spectral observations of AGB stars in the Fornax dwarf spheroidal galaxy,” MNRAS, 382, 1889.

Sloan, G.C., et al. 2006a, “The unusual silicate dust around HV 2310, an evolved star in the Large Magellanic Cloud,” ApJ, 638, 472.

Sloan, G.C., et al. 2006b, “Mid-infrared spectroscopy of carbon stars in the Small Magellanic Cloud,” ApJ, 645, 1118.

Sloan, G.C., et al. 2008, “The Magellanic zoo: Mid-infrared Spitzer spectroscopy of evolved stars and circumstellar dust in the Magellanic Clouds,” ApJ, in press.

van Loon, J.Th., et al. 2005, “ESO-VLT and Spitzer spectroscopy of IRAS 05328-6827: A massive protostar in the Large Magellanic Cloud,” MNRAS, 364, 71.

Zijlstra, A.A., et al. 2006, “A Spitzer mid-infrared spectral survey of mass-losing carbon stars in the Large Magellanic Cloud,” MNRAS, 370, 1961.