ATHENS, Ga. -- When low-mass stars called red supergiants die, they fade away on a wimpy wind-or so scientists have thought. New research, just published, suggests that the exact opposite may be true. These stars, in fact, may die with a bang and not with a whimper.
The study, published today in the journal Science, may lead researchers to a new understanding of red supergiants, which are studied to resolve issues in nucleosynthesis, stellar structure and the evolution of stars.
"This discovery was really a gigantic surprise," said Dr. Michael Duncan, a research professor of chemistry at the University of Georgia. "One of the beauties of doing fundamental science is that you never quite know where it may lead."
Other authors of the paper published in Science are Gerard Meijer, Gert von Helden and Deniz van Heijnsbergen of the FOM Institute for Plasma Physics in Nieuwegein, A. G. G. M. Tielens of the University of Groningen and S. Hony and L. B. F. M. Waters of the University of Amsterdam, all in the Netherlands.
During their death throes, low mass stars turn into red supergiants, which are more properly called asymptotic giant branch stars or AGBs. Actually a stage of development rather than a specific kind of star, the AGB phase is a relatively short stage during which low-mass stars become their brightest but experience heavy mass loss that leads them rapidly to the planetary-nebula phase and a final cooling to white dwarfs. (White dwarfs are extremely hot, Earth-sized objects that fade and cool for billions of years until they become black, cold cinders.)
Scientists have been studying AGB stars for a long time, but research has been accelerated recently through use of the Hubble Space Telescope and the European Space Agency1s Infrared Space Satellite.
Duncan's involvement in the discovery was the kind of scientific serendipity that often leads to unexpected breakthroughs. His area, the study of gas-phase metal clusters, has lately taken a huge step forward due to a collaboration with Gerard Meijer, whom he met at a scientific meeting at Ohio State University in 1998, and Meijer's colleagues in The Netherlands..
"He was talking about the free-electron laser called FELIX [Free-Electron Laser for Infrared Excitation] that had been built at the FOM Institute, and I happened to ask him if it had ever been used to study gas-phase metal clusters," said Duncan. "From that, our collaboration was born."
There are probably no more than 20 free-electron lasers in the world, and only five in the U. S. (Priorities for use at the U. S. machines is largely for medical science or industrial applications.) FELIX is the only one optimized for measuring infrared signals or "spectra" of chemicals, and seemed a perfect match for the metal-cluster experiments.
The study of metal clusters has been around only for about two decades. Duncan helped initiate the field because of a laboratory accident when he was a graduate student in chemistry at Rice University with Richard Smalley. Duncan and a fellow graduate student were working on a molecular beam experiment when an accidentally misaligned laser vaporized part of the apparatus.
Not realizing what had happened, they looked at a diagnostic tool called a mass spectrometer and saw a signal obviously associated with metallic compounds. Only after studying the problem did they realize what had happened. That accident, however, lead to a new way to produce large, regular molecules called metal clusters-most of which exist only for milliseconds and are thus devilishly hard to study.
(The Smalley group later used the same equipment and repeated the experiments on carbon and discovered a form of the element called carbon-60. Shaped in panels like the geodesic dome invented by architect Buckminster Fuller, the C60 forms were named "buckyballs." The team that discovered them was award the Nobel Prize for Chemistry in 1996.)
"After meeting Meijer, we realized his team had the free-electron laser, and I had the pulsed molecular beam machine and experience working with metal clusters, and we needed to find a way to make them work together," said Duncan. Luckily, Meijer received at about that time a large grant from the Dutch government, and so the team in the Netherlands was able to construct the molecular beam machine that Duncan had been using to study metallic clusters and mate it with the free-electron laser. The actual construction of the beam machine was overseen by Gert von Helden and Deniz van Heijnsbergen.
The result was a machine that could detect the infrared spectra of gas-phase metals and thus give important clues to how they are structured. The new apparatus worked beautifully, and when Duncan visited the lab last summer, the team achieved the first direct infrared spectra of these clusters ever done. Their work was published in the journal Physical Review Letters (PRL, 83, 4983, 1999).
These spectra in themselves will likely open a new era in the study of how gas-phase metals are structured, but a chance meeting with other Dutch scientists at the FOM Institute initiated a startling discovery that led the research from the lab to the stars.
"These astronomers were visiting the FELIX lab and hearing about work on polyaromatic hydrocarbons, which are important in the composition of interstellar space," said Duncan. "It just so happened that our work on gas-phase metals was on a machine nearby, and they asked what it was. Meijer and von Helden showed them the machine and the spectra we had. That's when their jaws dropped."
The astronomers, led by Alexander Tielens of the University of Groningen, realized immediately that the infrared spectra that the group had elicited from their study of titanium carbide nanocrystals corresponded almost exactly to spectra of unknown origin seen again and again in AGB stars. The discovery created a problem, however.
Meteorites containing micrometer-sized graphite grains with embedded titanium carbide (TiC) grains have been discovered on Earth. Isotopic analysis has identified AGB stars as the birthplace of these grains, though there had been no direct link. Astronomers believe that as AGB stars begin to die, newly synthesized elements such as TiC are mixed to the surface where they spread over the galaxy in a wind, most often in the form of stardust.
The problem lies in the fact that the abundance of titanium in low-mass stars is so low that "high densities are required just to get a high enough collision gains to grow to the sizes observed in graphite stardust." For some 20 years, scientists have thought that a so-called "superwind" phase takes place when these stars exhibit a dramatic loss of mass. But the superwind phase, despite its name, has been considered a relatively modest event in which the star's remaining stellar envelope is blown away.
The identification of the infrared spectra around AGB stars as gas-phase titanium carbide, however, changes that picture. Because of the low amounts of titanium in the stars and the apparent large amount in the ejecta, the event creating them must be cause by something that releases tremendous energy over a relatively short period of time. Or as the authors write, "The TiC identification suggests that rather than with a wimpy wind, low-mass stars end their lives with (almost) a bang."
Studies of the infrared spectra of AGB stars is just beginning to take off. A conference in France in 1998 reported what it called the "first mature results" of this research. The new study should add fuel to the fires of speculation about how stars are formed, how they live-and ultimately, how they die.
(Writers/Editors: Dr. Duncan will be out of the country for 10 days starting April 14, but he will respond to requests for interviews. Please leave a message with his secretary Mamie Watson, at 706-542-1749, and he will return your call.)