Home About ALMA Science with ALMA Evolved Stars - Stardust
Evolved Stars – Stardust

Molecules escape a dying star
Molecules escape a dying star

Large surveys of emission from the carbon monoxide (CO) molecule and other molecules show that evolved stars are losing mass in the form of stellar winds at relatively low outflow speeds like 20 km/s-1. The current observations suggest that, roughly speaking, the winds are isotropic and that the mass loss rates and outflow velocities are constant with time. Mass loss rates are derived by model fitting to observed CO profiles, to OH maser radii, to IR spectra and to HI and optical measurements. Generally, the values agree fairly well, so that in principle one should be able to study mass loss as a function of stellar parameters, in particular of location on the Hertzsprung-Russell (HR) diagram. As is the case with evolutionary studies in general, observations of very large numbers of stars will be required, perhaps of thousands of stars.

Winds from cool evolved stars are probably the dominant source of refractory dust grains in the interstellar medium. They are the starstuff from which we and our planet was formed. The grains manifest themselves through thermal emission extending from the far infrared through mm-wavelengths. At around 1 mm wavelength the emission is certainly optically thin, so that high-resolution maps of the thermal continuum from such winds will be an excellent tracer of the dust distribution. The high continuum sensitivity of the Atacama Large Millimeter/submillimeter Array (ALMA) will make possible direct imaging of the dust condensation zone for giant (AGB) stars within a few hundred parsecs at resolutions < 0.1". High-frequency performance of the ALMA is especially critical here, since the dust emission increases at least like f3 and since the angular resolution scales as f-1. Grain growth is expected to be most rapid at distances of a few x 1014 cm, so such observations will require the best possible resolution.

The ALMA will measure the angular sizes of circumstellar CO envelopes which will result in statistical studies of the distances to evolved (old) stars. Careful measurements of the angular extents of nearby evolved stars, whose distances may be known by independent means, will yield a typical linear size for the CO emitting regions; in these bloated aged stars, this can amount to thousands of times the distance from the Earth to the Sun. Statistical distance estimates of other stars can then be made by synthesis imaging with the ALMA. At resolutions of approximately 0.1", the ALMA can image CO envelopes well beyond the distance to the Galactic center. An interesting project will be to compare the distances measured in this manner with kinematic distances for the same objects, since the centroid of the CO profile is an excellent indicator of the stellar radial velocity, ALMA data will also provide the input for the kinematic distance determinations. Measurements of distances to a large numbers of these objects will indicate their spatial density and distribution in the Galaxy.

The high resolution available with the millimeter array will allow the detailed study of many such shells. This will allow the study of photochemistry in these environments; it will allow the observation of the shell kinematics and it will allow an examination of the evolutionary history of the star during its transition to the PN stage, since the molecular shell was emitted during the AGB phase and therefore contains information about that phase. Finally, the measurement of the shell masses of a large number of planetary nebulae, coupled with an examination of their luminosities and galactic kinematics, should allow a good value to be set for the upper progenitor mass limit for a white dwarf stars, or, conversely, for the lower mass limit for supernova progenitors.


Most of the observed transitions of the 125 known interstellar molecules lie in the mm/submm spectral region—here some 17,000 lines are seen in a small portion of the spectrum at 2mm. Spectrum courtesy B. Turner (NRAO).