Twinkle, Twinkle, Double Star
By Steven N. Koppes

In the study of galactic genealogy, astronomers have much to learn from the twins, the binary stars.

"There are stars out there that are touching each other, and it's not apparent how they're created," said Scott Shaw, a UGA astronomy professor. "The two stars initially are not touching. We know that they're old, so they must evolve in such a way to bring them into contact with each other and develop a stable system."

Binary stars are a foreign concept to most earthlings, who expect to see just one rising star in the morning. In binary systems, two stars twirl around one another in a thermonuclear orbital dance. In contact binaries, the stellar couple physically embraces.

Binary star research is journeyman work - nothing so flashy that would land it on the cover of a major news magazine. But other astronomers will use the research as a valuable reference in their own work.

For starters, binary star measurements provide a way to define the physical parameters of stars, such as their brightness, temperature and size. More recently, binary stars are helping astronomers determine more accurate distances to the star clusters in which binaries are found. Someone also needs to nail down the basic physics of stars and star clusters, Shaw said. "A lot of our understanding of how stars form, congregate and evolve relies on basic data of clusters and binary stars."

Shaw has to sift through a lot of stars to uncover one binary system. Up to a thousand stars populate each cluster he searches, and it takes 10 clear nights to search a single cluster. More than half of all stars in a cluster are binaries, but Shaw's quarry, eclipsing binaries, are rarer. No more than 10 per thousand stars in a cluster are eclipsing as seen from Earth. If Shaw finds four eclipsing binaries in one cluster, that's a lot.

"It seems that some clusters form a lot of binary stars, and some don't, and we haven't figured out why yet," Shaw said. One clue: Contact binary stars appear to live in clusters that are at least a billion years old. "You don't find them in new clusters," he said.

Drawing upon a decade of research, Shaw has developed a possible scenario for contact binary star evolution. The scenario begins with near-contact binaries - two stars orbiting fairly close to each other. One star loses matter to the other as they begin to exchange mass, causing the two stars to come closer together. Eventually they come into contact.

"You've got to try to find out how you could evolve a system to represent essentially everything you see out there. That's what you're supposed to do," Shaw said. "We'll probably have to settle for 90 percent."

For more information, e-mail Scott Shaw at jss@juno.physast.uga.edu.
Illustration by William Hartmann, courtesy of University of Arizona