Astronomers peer into stellar sandstorms

An international team of astronomers has found a potential explanation of how red-giant stars loose the bulk of their mass towards the end of their lives – a process currently not fully understood. Using new observational techniques, the researchers looked at the dust shells surrounding these dying stars, which gave them information about what causes the powerful "superwind" of dust grains that leads stars to lose their mass. Much of this star dust comprises of silicates, which go on to form planets such as the Earth.

      An artist's impression of the sandstorm surrounding one of the red-giant stars observed.

Celestial lifestyles

Towards the end of their lives, intermediate-mass stars – stars with masses ranging from 0.6–10 solar masses – eject the bulk of their outer envelope in a slow, dense wind. This "superwind" occurs over a period of 10,000 years, is a 100 million times stronger than the solar wind and removes almost half the mass of the star, leaving behind a fading stellar remnant.

The problem with this model, though, is explaining just how so much mass is lost in the "superwind". This is because it is difficult to observe gas and dust that is very close to its parent star. Current theory suggests that the "superwinds" occur because of the acceleration of the minute dust grains in the shells surrounding the stars. These grains are said to absorb starlight, which transfers momentum to them and causes the dust to blow away from the star. The problem with this model is that at the grain size estimated, light from the star would cause the dust to sublimate before it could be pushed away.

Unmasking stellar dust

In the new work published in Nature, the team, led by Barnaby Norris from the University of Sydney, Australia, looked at three red-giant stars and their dust shells using the European Southern Observatory's Very Large Telescope in Chile. The researchers used a technique known as "aperture-masking polarimetric interferometry" to look at the red giants plus other dust-free stars to verify their detection methods. Team member Albert Zijlstra, of Manchester University's Jodrell Bank Observatory in the UK, explains that using an "aperture mask" inside an infrared instrument along with a polarimeter is a combination that had not been previously used for this purpose. "The aperture mask turns a single telescope into a collection of much smaller telescopes, which can then be used as an interferometer. This gives excellent, reproducible image quality at the cost of reduced sensitivity," he explains. Using the observed data, the team developed a model to determine the dust-shell radius and the amount of light scattered by the shell at each wavelength.

A simulated image produced from the data of the dust shell around the red-giant star W Hydrae, shown in two polarizations of light: horizontal and vertical. This is equivalent to viewing the dust shell through polarized sunglasses, with the glasses held horizontally (left image) and vertically (right image). The image of the star in the centre of the dust shell is not from these observations and is for illustrative purposes only. The Sun (a single dot) is included for scale. 

Closer and bigger

The researchers found that the dust exists a lot closer to the stars than previously thought – less than two stellar radii. They also found that the grains are much larger in size than expected, being almost a micrometre across or about 300 nm in radius – this is quite large for stellar-wind particles. At these sizes, the grains are transparent to starlight, and so would not be sublimated by the intense radiation from the light. Although transparency suggests that the grains would again not be propelled away to form the wind, the researchers say that the acceleration occurs as a result of photon scattering rather than absorption. These large grains are driven out at speeds of 10 km s^–1, creating a virtual "stellar sandstorm".