Rethinking what the first stars were like
The Big Bang wasn’t all it has been cracked up to be. Sure, it created the universe. But after the heat of the primordial fireball faded, the cosmos plunged into darkness. The universe was cold and black — a sea of hydrogen and helium atoms mixed with a mysterious dark form of matter making its presence known only by its gravity. No stars.
It took a series of violent events — starting about 100 million years after the Big Bang—to end the cosmic Dark Ages. First, the evenly spread dark matter gathered into clumps, pulling in hydrogen gas that coalesced into clouds. Then pressure inside the clouds grew strong enough to fuse atoms, triggering nuclear reactions. The first stars created this way looked like roses with diaphanous petals, unfolding against a sea of darkness. The universe was finally in bloom.
The first stars marked a milestone in the history of the universe, bringing light and warmth back to the cosmos. Later, those primeval stars met their end in spectacular explosions known as supernovas, which seeded the universe with its first dollops of oxygen, carbon and silicon. Those elements made it possible for a second generation of stars to form.
The second-gen stars eventually burned through the opaque fog of hydrogen atoms and set the skies twinkling. These stars gathered into the first recognizable galaxies — dwarf galaxies of a few million stars. Dwarf galaxies merged, and after billions of years life emerged in one of the bigger galaxies, on a smallish backwater planet called Earth.
On that much, astronomers agree. But new simulations that track the star-formation process further than ever before are casting doubt on earlier ideas about the properties of the first stars. They’ve been cast as loners and extremely massive, for instance. But now the massive-loner theory is in dispute. And that has profound consequences for nearly everything that happened next, because the mass of the first stars may have determined the size of the first galaxies and how quickly the second generation of stars could assemble to form them.
“There is widespread confusion and disagreement,” says astronomer Jason Tumlinson of the Space Telescope Science Institute in Baltimore. “I can no longer say with any confidence what the first stars were like.” But, he adds, “that’s what makes the field so exciting.”
New simulations, new ideas
Retracing the steps of star formation is a tricky business. Less than a decade ago, computer simulations by Tom Abel of Stanford’s Kavli Institute for Particle Astrophysics and Cosmology and his colleagues indicated that the first stars were whoppers — between 30 and 300 times as heavy as the sun — and that each formed in solitary confinement within separate clouds of gas (SN: 6/8/02, p. 362). The gas showed no sign of fragmenting into several stars; instead, it appeared that the condensing object would keep growing to become one behemoth. And because massive stars die out in just a few million years, none of these first stars could still exist in the universe today.
Although the researchers could follow the steps toward star formation during the first 100 million years or so of cosmic history, they could not track the additional 100,000 years it takes for an infant star to grow to its final size. The team had to stop because supercomputers couldn’t — and still can’t — precisely track the rapid changes in density a cloud core undergoes as it becomes a star.
Using a mathematical trick, however, other teams have now gone slightly further, simulating about 1,000 years more of the star-formation process. Rather than attempting to track the rapid changes in the dense cloud core, these teams in effect ignore the core, treating it as a sink or black hole, with material falling onto the central region simply disappearing from sight.
Adopting that approach, the researchers have found evidence that a disk of material that forms around each of the embryonic stars can fragment into several fledgling stars, much the way the disk of material around the infant sun broke into clumps that formed the planets (SN: 2/26/11, p. 18).
The net result, as these astrophysicists now see it, is that stars could have been born in pairs or even threesomes. Since they coalesce from the same cloud, each partner would be lighter than if it had formed in solitary confinement.
“Whether at the end of this process one, two or a few massive stars will remain is currently unknown,” says Abel. Some studies even suggest that very small fragments, weighing no more than the mass of the sun, might form. Because low-mass stars take billions of years to burn out, some of the first stars could have survived to the present day, some researchers suggest.
To find out what the first stars were like, researchers are now looking to the scars those stars left behind — the extent to which they broke apart nearby atoms of hydrogen gas.
For instance, if most of the first stars were single and massive, they would have transformed the early universe into a giant hunk of Swiss cheese. That’s because big stars emit copious amounts of ultraviolet light, which ionizes surrounding gases — stripping electrons from the neutral hydrogen and helium atoms that veiled the cosmos during the Dark Ages. The birth of each individual star would create an ionized bubble, or hole, in the gases around it. Over time, the universe would be riddled with these holes. Once the holes grew large enough to overlap, the universe would be almost completely ionized — as evidence suggests it has been ever since the cosmos was a few hundred million years old.
But if the very first stars were extremely massive, they could have prevented other stars from forming. The energy from their ultraviolet emissions would break molecules of hydrogen into atoms. Without hydrogen molecules, which provide a clump-promoting cooling effect, the dark matter at the heart of star formation would not have enough gravity to pull gas into a star.
If the new simulations showing that primeval stars were born with partners are correct, the universe might never have gone through a Swiss cheese phase, Zoltán Haiman of Columbia University thinks. If the partnerships were close enough, one star would be more likely to collapse to become a black hole and draw matter from the other, emitting X-rays in the process. Far more penetrating than ultraviolet light, the X-rays would rapidly strip electrons from hydrogen and helium atoms throughout the cosmos, leaving a uniformly ionized universe instead of holes, Haiman suggested in the April 7 Nature.
The stellar-partnership scenario could explain an enduring puzzle in the universe today, suggests a team led by I. Félix Mirabel of the French Atomic and Alternative Energies Commission in Gif-sur-Yvette, France and the Institute for Astronomy and Space Physics in Buenos Aires. The leading theory of dark matter predicts that the Milky Way should be surrounded by hundreds of dwarf galaxies, but observers have found only about 25. Mirabel’s team suggests in the April Astronomy & Astrophysics that the other dwarf galaxies exist but can’t be seen because they’re starless — shadowy leftovers from the early universe, when such galaxies were too small to either forge or hold onto the first stars.
Researchers, however, don’t agree on how these X-ray–emitting partnerships would affect the universe. According to Haiman, the partners would emit so much more heat than a lone star that they would delay the formation of the first galaxies.
The extra heat from the stellar partners could boost the temperature and pressure of surrounding gases and prevent any clump of matter weighing less than a billion suns from corralling the gas to make new stars. Waiting around until dark matter clumps were that heavy may have delayed the onset of galaxy formation by 100,000 years.
But other astronomers disagree. Some theorists argue that rather than delaying the first galaxies, X-ray–emitting binaries would promote cooling that would hasten star formation. Tumlinson notes that through a chain of chemical reactions, X-rays would promote the formation of the HD molecule, in which one hydrogen atom is replaced by its heavier isotope, deuterium. That molecule might act as a new coolant.
“People argue about this for hours at meetings and still there’s no consensus,” notes Tumlinson.
As the theorists continue to debate their models, observations to test their ideas are about to begin.
New arrays of radio telescopes will look for imprints that the first stars left behind on the clouds of hydrogen atoms surrounding them. Radio astronomers can tune in to radio waves from hydrogen atoms that existed at different epochs of the Dark Ages — before, during and after the first stars formed — thanks to shifts in wavelength caused by the expansion of the universe.
In particular, astronomers will look for radio emissions with wavelengths of 21 centimeters, which neutral hydrogen emits but ionized hydrogen cannot. If the Swiss cheese model is correct and the first stars were massive loners, observers should see the holes created when the stars broke apart the neutral hydrogen atoms.
By using 21-centimeter radiation to pinpoint if and when holes formed and merged, low-frequency radio telescopes such as LOFAR, a set of radio dishes spread across the Netherlands and other parts of Europe, will map out the history of the first stars, says Avi Loeb of the Harvard-Smithsonian Center for Astrophysics in Cambridge, Mass. Such maps should indicate whether the first stars were massive loners after all.
Last year in Physical Review D, Loeb and his Harvard-Smithsonian colleague Jonathan Pritchard calculated that even a relatively inexpensive single radio dish that would record the intensity of the 21-centimeter radio emission averaged over the entire sky could indicate when the first stars were born and how quickly they ionized helium and hydrogen atoms by emitting ultraviolet light or X-rays.
Other researchers are attempting to read a fossil record of the elements cast into space by the very first generation of stars. Theorist John Wise of Princeton University and his colleagues are trying to simulate the second generation of stars, dubbed Pop II, which are the first stars that got incorporated into galaxies. Because Pop II stars are small enough to be relatively long-lived, researchers can examine them to see what they inherited from their parents’ generation.
“Astronomers are actually able to see Pop II stars in galaxies” and learn about their predecessors, says Wise. In addition to giant, 30-meter ground-based telescopes that astronomers are now planning to build, the James Webb Space Telescope, which researchers hope will launch late this decade, will closely examine Pop II stars from the first galaxies.
But researchers aren’t just waiting for Webb to be launched. Astronomers using the European Southern Observatory’s Very Large Telescope in Chile are getting a head start by re-examining the surfaces of eight elderly Milky Way stars. The stars are at least 12 billion years old and are probably members of the Pop II generation, Cristina Chiappini of the Leibniz Institute for Astrophysics Potsdam in Germany and her colleagues report in the April 28 Nature.
The team found high abundances of two rare, heavy elements — strontium and yttrium — relative to iron. To explain the composition of those second-generation stars, the researchers propose that the first stars were massive and rotated rapidly, spinning about 250 times faster than the sun. By mixing different layers of nuclear-burning gases, these whirling dervishes could trigger a chain of nuclear reactions that could have produced the high levels of strontium and yttrium.
If the first stars were fast rotators, they would be more likely to end their lives as gamma-ray bursts, Tumlinson notes in a commentary accompanying the Nature article. Such bursts are the most powerful explosions in the universe and would serve as cosmic fireworks that would brilliantly signal the first stars’ demise.
The bursts would be the ultimate messengers — death throes that traveled billions of light-years through space to reach Earth. For Loeb, recording those signals would be the thrill of a lifetime. “This is our roots, our origins,” he says. The bursts would put humans face to face “with our earliest ancestors, one star at a time.”
Ron Cowen is a freelance science writer in Maryland.
C. Chiappini et al. Imprints of fast-rotating massive stars in the galactic bulge. Nature, Vol. 472, April 28, 2011, p. 454. doi:10.1038/nature10000.
I.F. Mirabel et al. Stellar black holes at the dawn of the universe. Astronomy & Astrophysics, Vol. 528, Apirl 2010, p. A149. DOI: 10.1051/0004-6361/201016357. Available online: [Go to].
J.P. Pritchard and A. Loeb. Constraining the unexplored period between reionization and the dark ages with observations of the global 21 cm signal. Physical Review D, Vol. 82, July 23, 2010. Available online: [Go to]
J. Tumlinson. A new spin on the first stars. Nature, Vol. 472, April 28, 2011, p. 426.
R. Cowen, First stars may still shine. Science News, Vol. 179, February 26, 2011, p. 18. Available online: [Go to]
R. Cowen. How a star is born. Science News online, July 31, 2008. [Go to]
R. Cowen. Cosmic Dawn. Science News, Vol. 161, June 8, 2002, p. 362. Available online: [Go to]
A. Loeb. How did the first stars and galaxies form? Princeton University Press, 2010.
See simulations of the first stars by Tom Abel and colleagues: www.slac.stanford.edu/~tabel/GB/index.html