The early universe was veiled with a fog of neutral hydrogen and helium. Even the brilliant ultraviolet (UV) light of stars in the first primordial galaxies could barely penetrate the all-absorbing curtains of gas, leading some astronomers to term that era the cosmic "Dark Ages." Only over hundreds of millions of years could those first stars gradually heat and ionize the surrounding gas, transforming an opaque ocean of space into the transparent void of today. Without that transformation, no telescope would be able to see the UV radiation emitted by sources beyond our own galaxy. All our knowledge of UV sources would be confined to the Milky Way.

Determining precisely when that transformation from darkness to light (known as re-ionization) took place, and what caused it, will yield important insights into the history of galaxy formation in the infant universe. New theoretical work by J. Stuart Wyithe (University of Melbourne) and Abraham Loeb (Harvard-Smithsonian Center for Astrophysics), reported in the February 26, 2004 issue of Nature, shows that the cosmic Dark Ages lasted for more than a billion years, because the billion-year-old universe was still mostly neutral.

The Big Bang created a universe filled with hot, ionized hydrogen and helium. After some 380,000 years, the universe expanded and cooled enough for electrons and atomic nuclei to recombine, forming UV-absorbing, neutral hydrogen and helium atoms. Transformation from the consequent darkness into now-pervasive transparency required that the gas filling the universe be re-ionized, or once again split into its constituent electrons and nuclei. The question is: When did that re-ionization occur?

"A simple model of instant and complete ionization, using data from the WMAP (Wilkinson Microwave Anisotropy Probe) satellite, implies that re-ionization took place about 200 million years after the Big Bang, around a redshift of 17. However, the actual re-ionization process was not instant and complete. It was more complicated," says Wyithe.

Redshift is an increase in the wavelength of light from a distant object, caused when the light is stretched due to the expansion of the universe. This increase in wavelength makes the object appear to be redder than it actually is. The larger the redshift, the more distant the object.

Quasars light the way
To draw their conclusion, Wyithe and Loeb studied quasars -- very distant and very bright objects that can be seen across billions of light-years of space. Their extreme brilliance means that quasars can be used as cosmic flashlights to illuminate not only their immediate environment, but also all of the intervening space between the quasar and the Earth.

At the heart of every quasar lies a supermassive black hole gulping down nearby matter. As matter spirals into the black hole, some of its gravitational energy is converted into light.

Ultraviolet radiation from a quasar will ionize gas in the surrounding intergalactic medium (IGM). While bright quasars were not numerous enough in the early universe to cause widespread re-ionization, each of them was powerful enough to form a local "bubble" of ionized gas.

The size of the bubble depends on the properties of the surrounding gas. If the nearby IGM is mostly neutral, the quasar has to work harder to ionize the gas, and can only create a small bubble. An IGM that is already mostly ionized means less work for the quasar and a larger bubble.

"A good wintertime analogy is shoveling snow. After a light snowstorm, you may be able to shovel your entire driveway in only 30 minutes because there isn't much material to move. After a heavy snowstorm, it may take 30 minutes just to clear your sidewalk because there's so much more material to move. Similarly, the higher the fraction of neutral material around a quasar, the less space it can 'clear' through ionization over the quasar's lifetime," says Loeb.

Still mostly neutral after 1 billion years
To calculate the fraction of neutral hydrogen in the early universe, Wyithe and Loeb used the two most distant quasars known, one at a redshift of z~6.3 and one at z~6.4, corresponding to distances of about 13 billion light-years. They predicted the expected size of the ionized "bubble" around each quasar, assuming reasonable values for the emission rate of ionizing photons and time elapsed (i.e. the quasar's lifetime), then compared their predictions to the observed sizes. The conclusion was clear.

"The observed bubble sizes were small-so small that the fraction of neutral hydrogen had to be large, in the range of tens of percent. So, even one billion years after the Big Bang, when re-ionization should have been well underway according to WMAP, most of the intergalactic medium was still neutral," says Wyithe.

Wyithe and Loeb's finding points to a complicated history of star formation and the associated re-ionization of the IGM. While the re-ionization process likely began with the formation of the first stars at z~30 (about 100 million years after the Big Bang), it took a considerable amount of time for complete re-ionization. The universe endured an extended period of partial re-ionization lasting for billions of years.

It is even possible that re-ionization occurred in two distinct phases powered by the first and second generations of stars.

"The first stars were hot because they were made of the pristine gas left over from the Big Bang. As soon as heavy elements were produced in their interiors and dispersed into their surroundings by supernova explosions, the additional stars forming out of the enriched gas were cooler and less efficient at ionizing the universe," says Loeb. "It is conceivable that the IGM became fully ionized, as we see today, only after enough of those second-generation stars formed."

Headquartered in Cambridge, Mass., the Harvard-Smithsonian Center for Astrophysics is a joint collaboration between the Smithsonian Astrophysical Observatory and the Harvard College Observatory. CfA scientists, organized into six research divisions, study the origin, evolution and ultimate fate of the universe.