By pinpointing individual suns in the glare of the most tightly packed cluster of stars in our galaxy, the Hubble telescope has unveiled hints of either a massive black hole or another remarkable phenomenon: a "core collapse" driven by the intense gravitational pull of so many stars in such a small volume of space.
Astronomers used the telescope's sharp images to count an extraordinary number of stars in the ancient globular cluster M15, about 37,000 light-years from Earth. Hubble spied hundreds of stars in a tiny area at the center of this cluster. Careful analysis of the distribution of these and thousands of neighboring stars suggest that at some point in the distant past, the stars converged on M15's core, like bees swarming to their hive. An alternate scenario also could explain the pileup of stars at M15's core: a black hole that may have formed early in the cluster's history. The black hole would have gradually gained mass as more stars spiraled inward. The black-and-white picture shows the cluster's central region; the color image is a close-up of the core.
Credit: P. Guhathakurta (UCO/Lick Observatory, UC Santa Cruz), B. Yanny (Fermi National Accelerator Lab), D. Schneider (Pennsylvania State Univ.), J. Bahcall (Inst. for Advanced Study), and NASA.
By pinpointing individual suns in the glare of the most tightly packed cluster of stars in our galaxy, the Hubble Space Telescope has unveiled hints of either a massive black hole or another remarkable phenomenon: a "core collapse" driven by the intense gravitational pull of so many stars in such a small volume of space.
A team of astronomers used the telescope's sharp images to count an extraordinary number of stars in the ancient globular cluster M15, about 37,000 light-years away. Hubble spied hundreds of stars in a tiny area at the center of M15, whereas earthbound telescopes see a single blur of light. Careful analysis of the distribution of these and thousands of neighboring stars suggest that at some point in the distant past, the stars converged on M15's core, like bees swarming to their hive. This runaway collapse, long theorized by researchers but never seen in such detail, may have lasted a few million years- a flash in the 12-billion-year life of the cluster.
Thanks to the laws of physics, the core probably stopped collapsing before many of the stars collided. Rather, stars near the center would have settled into an uneasy cosmic waltz, both attracted to each other by gravity and repelled by close encounters that slingshot them through space.
An alternate scenario also could explain the pileup of stars at M15's core: a black hole that may have formed early in the cluster's history. The black hole would have gradually gained mass as more stars spiraled inward. If it exists, it would now be several thousand times more massive than our sun.
The study, which will appear in the January 1996 issue of the Astronomical Journal, was led by Puragra Guhathakurta of UCO/Lick Observatory, UC Santa Cruz. Coauthors are Brian Yanny of the Fermi National Accelerator Laboratory, Donald Schneider of Pennsylvania State University, and John Bahcall of the Institute for Advanced Study in Princeton. All of the astronomers were associated with the Institute for Advanced Study when the research began.
A precise reading of the speeds at which stars move near M15's core would reveal whether the stars are packed so tightly because of the influence of a single massive object, or simply by their own mutual attraction. Stars would orbit more quickly in the grip of a black hole's gravitational field. Such measurements are time consuming but possible with the Space Telescope.
"It is very likely that M15's stars have concentrated because of their mutual gravity," Guhathakurta says. "The stars could be under the influence of one giant central object, although a black hole is not necessarily the best explanation for what we see. But if any globular cluster has a black hole at its center, M15 is the most likely candidate."
The team began using Hubble to observe the centers of globular clusters in 1991 and now has data on about twenty clusters, but the images of M15 are by far the most stunning. Hubble's Wide Field Planetary Camera 2 (WFPC2) probed M15 in April 1994, four months after astronauts installed corrective optics to sharpen the telescope's blurry focus.
"I first started thinking about this observation in 1970," says Bahcall. "I never expected that Hubble would see things as clearly as it does. The results are so exciting that they are a dream come true."
Bahcall and astrophysicist Jeremiah Ostriker of Princeton University first proposed in 1975 that M15 might harbor a black hole. While distinguished by its extreme density of stars, M15 is in other respects similar to the rest of the dozens of globular clusters that freckle space in and around our Milky Way. Each cluster is like a miniature galaxy, with 100,000 to one million stars in a compact spherical blob. The largest and closest-including M15, in the constellation Pegasus-are visible to the naked eye on dark nights as faint hazy patches.
Globular clusters contain almost no gas or dust and show few signs of recent star formation. Astronomers believe they are primordial remnants, left over from the birth of the Milky Way. As such, they are ideal laboratories for studying how stars evolve. Cluster stars also provide a limit on the age of the universe, independent of the expansion of the universe itself.
Stars at the core of M15 may be crowded closer together than anywhere in the Milky Way except in the galaxy's hidden heart. Attempted studies of this exotic locale with ground-based telescopes proved frustrating. Atmospheric blurring washed out the interesting details at the core. Astronomers used Hubble before its repair mission to examine M15, but even after correcting the distorted images they could not discern the true distribution of the innermost stars. In contrast, the latest WFPC2 photos of the inner 22 light-years of the cluster revealed about 30,000 distinct stars. That's a fraction of M15's population, but far more stars than scientists had ever imaged in such a small region of a globular cluster.
The astronomers used the Planetary Camera (the highest-resolution part of WFPC2) to study M15's core. The closer they looked toward the core, the more stars they found. This increase in stellar density continued all the way to within 0.06 light-years of the center-about 100 times the distance between the sun and Pluto.
"Detecting separate stars that close to the core was at the limit of Hubble's powers," Yanny says. Beyond that point, even Hubble's eagle eye could not reliably resolve individual stars or locate the exact position of the core. However, the researchers suspect that stars jam together ever more tightly inside that radius. The team plotted the distribution of the stars as a function of distance from the core. Computer simulations helped them include stars they may have missed when bright stars drowned out faint ones in the Hubble images. The resulting pattern matches the predictions of Bahcall and others for what would happen under the influence of a central black hole. But the pattern also is consistent with a core collapse, known as a "gravothermal catastrophe. "Astronomers think the cores of about 20 percent of all globular clusters may have collapsed in this way.
For a gravothermal catastrophe to occur, globular clusters must transfer energy from the inner parts of the cluster to outer regions. As this happens, stars near the core lose some of the energy of their random ("thermal") motions. Several billion years might pass before the stars become too lethargic to resist the gravitational pull of their neighbors. At that point, they begin to collapse inward as a group.
"It's a catastrophe in the sense that once it starts, this process can run away very quickly," Guhathakurta says. "But other processes could cause the core to bounce back before it collapses all the way." The major such process, researchers believe, is the powerful jolt of new motion that binary-star systems can impart to a third star that wanders too close-effectively spreading the stars out again.
University of California, Santa Cruz
Pennsylvania State University
Space Telescope Science Institute