The Cataclysmic Death of Stars

Ever since he was a teenager, Stan Woosley has had a love for chemical elements and a fondness for blowing things up. Growing up in the late 1950s in Texas, "I did everything you could do with potassium nitrate, perchlorate, and permanganate, mixed with a lot of other things," he says. "If you mixed potassium nitrate with sulfur and charcoal, you got gunpowder. If you mixed it with sugar, you got a lot of smoke and a nice pink fire." He tested his explosive concoctions on a Fort Worth golf course: "I screwed the jar down tight and ran like hell."
kaboomWoosley, now an astronomer at the University of California at Santa Cruz, has graduated to bigger explosions—much bigger. Woosley studies some of the most powerful explosions since the birth of the universe: supernovae, the violent deaths of stars.

The universe twinkles with these cataclysms. They happen every second or so, usually in some unimaginably remote galaxy, blazing as bright as hundreds of billions of stars and creating a fireball that expands and cools for months.

We're lucky that they rarely strike close to home. The last supernova in our own galaxy exploded in 1604, rivaling Jupiter's brightness in the night sky and deeply impressing Johannes Kepler, the pioneering astronomer. A nearby supernova—within a few light-years—would bathe the Earth in lethal radiation.

Yet the legacy of supernovas is as close as our own bodies. The carbon in our cells, the oxygen in the air, the silicon in rocks and computer chips, the iron in our blood and our machines—just about every atom heavier than hydrogen and helium—was forged inside ancient stars and strewn across the universe when they exploded billions of years ago. Eager to understand our origins and, in some cases, simply wild about things that go bang, astronomers have been struggling for decades to understand why stars that shine peacefully for millions of years suddenly blow up.

Lately they've had two big breaks. One is a revelation about potent blasts of high-energy gamma rays that come from distant points in the heavens. For decades astronomers have puzzled over their origins, but space probes recently clinched the answer, which Woosley proposed more than a decade ago: Many gamma-ray bursts are the early warning signals from supernovas, emitted minutes before the explosion.

The link offers a glimpse of events leading up to the actual explosion—another mystery. There, too, researchers have made headway. Looking not at the heavens but at computer models of supernovas, some think they have figured out what may trigger the final cataclysm. The missing element may be unimaginably powerful reverberations—the sound of a star singing its own swan song.

For astronomers, there's usually no rush to study something before it vanishes. "The universe usually evolves as slowly as watching paint dry," says one. But these days, hundreds of astronomers keep cell phones and beepers close by so they can rush to work like doctors on call. They're waiting for word from a spacecraft called Swift.

Swift, launched in 2004, scans the skies for gamma rays. When it detects a burst, it swivels its telescopes toward the source to get a good fix and detect the afterglow—the lingering point of light that marks the spot where a burst originated. It also sends an alert to earthbound astronomers, who can take a closer look with bigger telescopes.

Early on February 18, 2006, Swift recorded an outpouring of gamma rays from somewhere toward the constellation Aries. Within three minutes, the satellite had determined the position of the burst and broadcast an alert. Two days later, astronomers at a telescope in Arizona reported that the burst came from a small, nearby galaxy, only a fraction as far away as usual.

Astronomers had already traced a connection between bursts and supernovas. But this burst was so close, and Swift had spotted it so quickly, that scientists hoped it would help confirm what they suspected: A gamma-ray burst is an exploding star's opening act.

After an unusually long flood of gamma rays and x-rays, lasting more than half an hour rather than the typical few seconds, the February 18 burst gave way to visible and infrared light. Within three days this afterglow was fading away—and then the supernova grabbed the spotlight.

Astronomers at the Very Large Telescope in northern Chile were watching the afterglow dwindle when they noticed a brightening. The star had exploded just a minute or so after the burst, but most of its energy was invisible ultraviolet and x-ray radiation. Its visible light had brightened more slowly, and now it was finally outshining the afterglow. For the first time, astronomers had seen a gamma-ray burst evolve into a supernova from the very beginning.

Eighteen days after the supernova flared into view, astronomers were still watching. Atop Palomar Mountain in southern California, the observatory dome's twin shutters slid open under patchy clouds, letting a sliver of night sky fall onto the caged mirror of the 200-inch (508-centimeter) Hale Telescope. Caltech astronomer Avishay Gal-Yam had two hours before the supernova would dip too low in the sky for the telescope to see it.

Still more luminous than a billion suns, the supernova outshone the combined light from all the stars in its home galaxy, glowing white-hot from the radioactive decay of unstable nickel atoms forged in the explosion. Gal-Yam pointed to a computer screen showing a squiggly line—the glow broken down into its component colors, or wavelengths. Each dip in the line represented a wavelength of light absorbed by a different element—silicon, cobalt, calcium, iron—in the debris of the star.

Destruction and creation were conjoined on the screen. The elements revealed there, like those from countless earlier supernovas, will eventually find their way into new stars and perhaps new planets, Gal-Yam said. He added: "I'm just really happy to be observing this."

The star had begun its race to destruction long before that night on Palomar, when it began to lose a lifelong fight against gravity. Gravity is responsible for setting newborn stars aflame, by squeezing atoms of hydrogen in the star's core so tightly that they fuse to make helium. The fusion generates light and heat and also exerts pressure that allows the core to withstand the enormous weight of the star's outer layers.

But when the core consumes all of its hydrogen, gravity compresses it. The temperature of the shrinking core rises to about a hundred million degrees, hot enough for helium nuclei to fuse and make carbon. The new surge of energy keeps the core from collapsing much further.

For an isolated star no heavier than the sun, there is little more to the story. The star burns all of its helium and shrivels. It turns into a white dwarf about the size of Earth, aging and cooling indefinitely—unless it lies close enough to another star to steal its neighbor's outer layers of hydrogen. If enough material falls onto the white dwarf, the siphoned fuel ignites a thermonuclear explosion. As the detonation spreads, the entire star blows up in what is known as a type 1a supernova—a giant nuclear bomb.

The supernova blossoming over Palomar was a different kind: not a thermonuclear blast but a star's catastrophic collapse. This is the only kind of supernova that can unleash a gamma-ray burst, and it is the inevitable fate of a star more than eight times as massive as the sun.

Such heavyweight stars always lose their battle with gravity. With the crushing weight of the star's outer layers bearing down on its core, the fusion reactions don't stop at carbon. The star continues to cook lighter nuclei into progressively heavier elements, but each nuclear reaction runs its course faster. The transformation from carbon to oxygen takes 600 years, from oxygen to silicon 6 months, from silicon to iron a day. Once the star's core turns to solid iron—a sphere no bigger than Earth that weighs as much as the sun—its fate is sealed. In less than a second, the star will explode.

Iron marks the end of the road because unlike lighter elements, iron atoms consume rather than create energy when they fuse. Fusion can no longer provide the energy to support the star's outer layers, and the core simply implodes. Usually the result is a neutron star, a stellar cinder so dense a teaspoon would weigh more than a billion tons. In the most massive stars the collapse leaves only a voracious pit called a black hole.

At this point, Woosley believes—before the collapse somehow turns into an explosion—some supernovas unleash a blast of gamma rays. Woosley's interest in these bursts goes back decades, when they were so mysterious that over a hundred more or less serious ideas about their cause were in play, from "starquakes" to the exhaust plumes of alien spacecraft. But his fascination deepened in the early 1990s, when a spacecraft called the Compton Gamma-Ray Observatory showed that gamma-ray bursts originate far beyond our galaxy. To appear as bright as they do, they had to be more energetic than anyone had imagined—far brighter than supernovas, Woosley's first love.

They also needed a source of energy far beyond what any ordinary star could provide. Perhaps the cataclysmic jolt of a collapsing star could somehow be harnessed to produce gamma rays. So Woosley set out to determine how a core-collapse supernova could generate a burst.

He and his collaborators, including Andrew MacFadyen of New York University, stage their explosions in computers. They start with a whopper of a star, about 40 times the mass of the sun, spinning so fast—several hundred miles a second at the equator—that it barely keeps from flying apart. Near the end of its life, unable to resist the pull of its own gravity, the core of the star collapses to make a black hole. But because the star has so much spin, some of the infalling material resists the tug of the newborn black hole. A swirling disk of material forms around the hole—a maelstrom deep within the doomed star.

"Rotation is the name of the game," says Woosley. Without spin, there would be no disk. And without a disk, there'd be no burst. Friction heats the disk, whipping around the black hole thousands of times a second, to 40 billion degrees (22 billion degrees Celsius), while new material keeps cascading in. Moments after the black hole forms, jets of superheated gas blowtorch outward.

Each jet may draw its energy directly from the friction in the disk, or from the newborn black hole, via the magnetic fields that link it to its surroundings. Like the original star, the black hole spins frenetically, which could cause the fields to stretch, twist, and snap like rubber bands, dumping vast amounts of energy into the disk.

Either way, the jet shoots outward, reaching the surface of the star in a mere ten seconds. If the star has retained its original, puffy envelope of hydrogen gas, the jet stops dead and the gamma-ray burst may fizzle. But if the powerful winds that blow from some massive stars have stripped away the hydrogen earlier in the star's life, the jet escapes, arrowing into space at more than 99 percent of the speed of light.

Now comes the burst: High-speed collisions between blobs of material in each jet produce a cascade of speedy electrons. The electrons whirl around the jet's magnetic fields, flinging out gamma rays. Over many days, as the jet plows into the thin gas between the stars, it generates an afterglow at visible, infrared, and radio wavelengths.

The February 2006 burst was dimmer than most, perhaps because the star was not massive enough to form a black hole. Woosley suggests that the same sequence of events—an implosion, a spinning disk, jets—can still happen when the stellar collapse ends with the formation of a fast-spinning neutron star rather than a black hole.

Even after the jets have erupted, the star has not yet exploded. "The jet gets to the surface of the star minutes beforehand," says Woosley. "The burst is a herald of the supernova."

It's not enough, however, to cause the explosion. "Just running a jet through a star won't make a very good supernova," says Woosley. "It will unbind some of the star, but most of it will fall back." To make a collapsing star explode, he says, "there needs to be something else."

In the stars that launch gamma-ray bursts, the spinning black hole and the disk may pump out enough energy to blow the star apart. But in most collapsing stars, the collapse ends when the Earth-size core crunches into a neutron star the size of a city, at a temperature of a hundred billion degrees (55 billion degrees Celsius). This is the point of maximum scrunch. The squeezed core rebounds like a squished sponge, launching a shock wave that races outward, ramming into the material that is still pouring down from the star's outer layers.

Astronomers once thought this shock would be enough to tear the star apart and generate the explosion, says Adam Burrows of the University of Arizona. Turns out it's not so simple.

Simulating a supernova gobbles enormous amounts of computer power, and even the largest supercomputers can't fully reproduce an exploding star in three dimensions. But over the years the models have improved, and the shock wave scenario has fallen apart.

Researchers found that less than a thousandth of a second after the shock wave is generated, a flood of tiny, nearly massless particles called neutrinos escapes from the center of the star. The neutrinos, born in the collapsing core, drain energy from the shock wave. The shock stalls, and—at least in the computer—the supernova is a dud.

Now Burrows and his colleagues are working with a computer model powerful enough to simulate how the core shakes and churns during the collapse, and they've finally seen how a collapsing star could turn around and explode. The turbulent infalling gas starts shaking the core, causing it to pulsate. Raining down from the star's outer layers, the gas wraps around the core, dancing over its surface and penetrating its depths.

"The core is oscillating, and the stuff falling onto the core is exciting it," says Burrows. In about eight-tenths of a second, the oscillations are so intense they send out sound waves. The waves exert a pressure that expels material, reinforcing the shock wave created by the star's collapse. They also amplify the core's vibrations in a runaway reaction, says Burrows, "until the star finally explodes."

For someone brave enough to come within hearing distance, the waves would be audible, roughly the F note above middle C.

Burrows acknowledges that sound waves may not be the full story. But his model tends to produce a lopsided explosion, and stars do indeed explode asymmetrically, with more punch in some directions than others. That was true for supernova 1987A, recorded 20 years ago, the closest and brightest supernova since 1604. Astronomers also have found that some of the neutron stars left behind by supernovas zip along at 500 miles a second (800 kilometers a second), as if the explosion had imparted an enormous kick in one direction.

Stronger evidence for the sound wave idea could come from two sprawling facilities, in Hanford, Washington, and Livingston, Louisiana, designed to detect gravitational waves—ripples in the fabric of space and time. Gravitational waves, predicted by Einstein's theory of general relativity but never directly observed, should be produced whenever immense masses shake and twist, as they do in the core of a supernova.

If sound waves really are at work inside a collapsing star, it should vibrate only at certain frequencies, generating matching gravitational waves. Burrows calculates that for a supernova in or near our galaxy, the existing detectors could pick up these signals—clues to a big, big noise.

Stars, it seems, really may go kaboom. Woosley, still in love with pyrotechnics, is delighted. "It's like God built the universe just for me."