X-ray and Observational Astronomy

News June 2003: Major white dwarf study published

A sample of some of the high quality data used in this study, part of the HST STIS spectrum of REJ0558-373. The data is shown below, the modelled spectrum offset above. Absorption lines due to oxygen, silicon and iron dominate here.	The abundance of silicon for each white dwarf plotted as a function of temperature (10,000 to 80,000 degrees). As for other elements, high abundances are associated with the hottest white dwarfs. Some lower temperature white dwarfs have anomalously high silicon abundance also.	The ratio of iron to nickel abundance in the sample. The average value is ~20, the cosmic value, but theory predicts a value around 1.0.

A white dwarf is the end-point of stellar evolution for any stars less massive than 8 solar masses. After nuclear burning of hydrogen to helium in the central regions of the star ends due to exhaustion of the hydrogen fuel, the star shrinks to about the size of the earth, no longer having the internal support due to the high central temperature. At this size, the surface of the white dwarf will initially be very hot (more than 100,000 degrees C). It will slowly cool by radiation, possibly remaining visible for a time-scale comparable to the age of the universe. The strong ultra-violet light from the very hot surface of the white dwarf can illuminate gas ejected when it was a red giant, at the end of its hydrogen-burning phase. This results in a planetary nebula, nothing to do with planets, but often spectacularly beautiful. The vast majority of normal stars will end up as white dwarfs.

The long life of white dwarfs makes them a good tool for understanding the ancient history of their environment, if you know how the appear now you can work out what the were like earlier. This can be useful in determining the early history of the Galaxy and other stellar systems such as globular clusters. Of course, it is essential to know precisely how white dwarfs age in order to achieve this.

Professor Martin Barstow, at the University of Leicester, together with others in the USA and Germany, has just published a major study of hot white dwarfs. This makes a state-of-the-art comparison of the best observations with theory. The IUE, FUSE and HST far-ultraviolet spectra of 25 hydrogen-rich white dwarfs were systematically compared with sophisticated stellar atmosphere models to measure the trace abundances of the heavier elements. These models include all the known physics affecting the light from the white dwarfs, and they can be used to predict a spectrum for a given set of physical parameters (such as the mass and temperature of the white dwarf). Chemical abundances are measured by comparing the depth of the absorption lines they cause in the model spectra with those seen in the stars themselves (see the figure top left).

This work has resulted in abundance measurements for the elements carbon, nitrogen, oxygen, silicon, iron and nickel in the 25 stars. The temperatures of the white dwarfs were found to be more important than their masses as a predictor of the abundances of these elements, higher temperatures are associated with higher abundances (see the figure top centre). This is in accord with the prediction of radiative levitation. The high gravitational fields of white dwarfs should lead to rapid settling of heavy elements (all elements are heavier than the majority constituent - hydrogen), however their high temperature means that these heavy elements are pushed towards the surface by the pressure of the intense radiation within the star (this does not happen to the same extent for hydrogen because the heavier element atoms have more electrons, and so are more subject to the radiation pressure). Abundances as high as 10 atoms per million hydrogen atoms are seen at temperatures in the range 50-80,000 degrees. (These abundances are ~1000 times lower than seen in our solar system for carbon and nitrogen, but coincidently are similar for iron and nickel.)

In spite of the general agreement between the observations and the theoretical models, some problems have been revealed. The precise measured abundances are not well predicted, some cooler white dwarfs have pure hydrogen atmospheres while others do not, and the measured ratio of iron to nickel abundance is similar to the cosmic average value - 20 times that predicted by the theory (see figure top right).

While it is possible that accretion of gas from either an unseen companion star or the surrounding interstellar medium could account for some of these anomalies, there is no direct evidence for this and there is evidence against for some of the stars. A search for companion stars is currently being made.

More details on the Leicester group's white dwarf research can be found here. The paper describing the research highlighted here is available from the arXiv service.

Other news items are available here.

Back to group page.

Last updated: 2003 June 04 by Julian Osborne