ASK any astronomer what a black hole is, and you’ll usually get the same answer: the shrunken remains of a star much more massive than the Sun, which can suck in matter like a cosmic vacuum cleaner. But one astronomer has found the first signs that black holes may have been born long before the stars. Mike Hawkins of the Royal Observatory in Edinburgh claims that space is teeming with tiny black holes—almost small enough to get your arms around—that were manufactured just moments after the big bang. The nearest might be only 30 light years away—a mere stone’s throw in astronomical terms.
Ironically, the clues do not come from studies of dark objects but from quasars, the brightest objects in the Universe. Quasars are typically as bright as hundreds of normal galaxies, yet they generate all their energy in a region about the size of the Solar System. Most astronomers believe that they are powered by supermassive black holes, typically as heavy as a billion Suns. The light would then come from an “accretion disc” of gas, which is heated to millions of degrees as it swirls down into the hole.
Since quasars were discovered in 1963, astronomers have found that they sometimes fluctuate in brightness. Some rare quasars vary their light output over a very short time—months or even hours. Most astronomers agree that these short-term fluctuations are probably due to “hot spots” in the accretion disc, perhaps caused when stars from a surrounding galaxy collide with the disc. But quasars also fluctuate on longer timescales of between 5 and 10 years, and no one knows why. “In the 30-odd years since quasars were discovered, we have amassed an enormous amount of data, but we still don’t really understand what they are,” says astronomer Bohdan Paczynski of Princeton University.
Advertisement
Inside story
Without any obvious mechanism for the long-term fluctuations, astronomers have tended to put them down to unknown changes within the quasar or overlying galaxy. “The prejudice of most astronomers was that such changes are intrinsic to quasars,” says Hawkins. “There was no plausible explanation for the cause but no one doubted that one would eventually turn up.”
However, over the past couple of years, Hawkins has followed up a controversial alternative. He suggests that the long-term variations do not arise within the quasars themselves, but that they are illusions caused by bodies passing between quasars and the Earth. If he is right, then the Universe may be chock-a-block full of black holes, only about a metre across, left over from the big bang.
Hawkins became interested in quasars entirely by accident. In 1975, he planned a project using the UK Schmidt telescope at Siding Spring Observatory in Australia together with the COSMOS automatic plate scanning machine at Edinburgh. He hoped to monitor a square of sky 4.5 degrees across in three separate colour bands. The goal was to complete a long-term study of objects such as variable stars that brighten and dim a few times each year.
However, by about 1980, it became clear to Hawkins that stars made up only a minority of the objects that vary in brightness. Far more common were mysterious objects whose brightness varied over a few years. Hawkins took a look at their spectra and found that they had bright lines at certain wavelengths. “What I saw were powerful emission lines—the unmistakable fingerprint of quasars,” he says. This led Hawkins to restructure his monitoring programme to look specifically for quasars. Several years later, he had pinpointed about 95 per cent of the 1500 quasars that astronomers would expect to find in that patch of the sky.
Hawkins found that every single quasar in his sample was varying in brightness on a timescale of between 5 and 10 years. The variation was like a smooth sine curve, and typically amounted to a 30 to 100 per cent change in brightness—much greater than the few per cent involved in the short-term variations. The strangest thing was, however, that the light variations did not seem to be coming from within the quasars themselves. Somehow, the light altered during its journey to Earth.
The first sign, says Hawkins, came from the periods of variation for quasars at different distances. Astronomers assume that if the brightness variations occur inside the quasars, the timescale of the changes should generally be longer for distant quasars than for nearer ones. This is because of the expansion of the Universe, which causes a “time dilation” effect. Light emitted by a very distant variable quasar at two different points in the cycle should be farther apart by the time it reaches Earth because the Universe has expanded during the immense journey. But Hawkins’s data showed no evidence of the effect.
Colour clues
A second hint that the long-term brightness variations do not come from the quasars themselves came from the colours of quasars as their brightness changed. Brightness variations in astronomical objects nearly always come hand in hand with changes in colour. For instance, if a star becomes hotter and brighter, it also becomes more blue. But having studied the quasars in blue, red and ultraviolet light, Hawkins found that their colours did not change significantly.
But if the quasars themselves were not the culprits, what could be causing the long-term variability? The possibility that occurred to Hawkins was that small, massive objects of some kind were passing between the quasars and the Earth. Each object’s gravity would act like a converging lens, temporarily boosting the quasar’s light. Such gravitational lensing would affect all wavelengths of light equally, which would explain why the quasars brightened without changing colour.
Lensing could also explain the lack of any time dilation effect. For light variations caused by lensing, the time dilation depends on the distance of the lensing object, not the quasar. So providing the lensing object is relatively near, the time dilation will be small. As it turns out, any lensing objects should lie relatively close to us, an effect due to the curvature of space. At very large distances equivalent to looking back to the early Universe, the volume of space available to lensing bodies decreases. A decade ago, Jeremiah Ostriker and his colleagues at Princeton University showed that this means most bodies that could cause a lensing effect with quasars must be concentrated relatively close to Earth, within just a quarter of the average quasar distance. So if lensing objects were at work, the time dilation effect would be very difficult to detect.
From the timescale of the light variations and the typical distances involved, Hawkins was able to work out a rough mass for the lensing bodies. “What I discovered was that they had to be the mass of Jupiter,” says Hawkins.
This was an astonishing conclusion. For at least one object to be lensing every quasar, the Universe would have to contain a truly enormous population of Jupiter-mass objects. In fact, they would have to make up the vast majority of the mass in the Universe. James Gunn and William Press of the California Institute of Technology in Pasadena had made this crystal clear back in 1973 when they studied how much matter would have to be put in compact bodies spread throughout the Universe to ensure that every quasar was lensed (Astrophysical Journal, vol 185, p 397). It turned out that the combined mass of all the lenses would have to be several hundred times the mass we see in visible galaxies.
If Hawkins’s idea was true, this would be revolutionary. All these Jupiter-mass objects would give the Universe more than a “critical mass”, enough to one day halt its expansion and force it to collapse in a big crunch. The lensing objects could also solve one of the central mysteries of cosmology—the identity of dark matter. The Universe is dominated by nonluminous matter, which reveals itself only by its gravitational effect on visible stars and galaxies. “Most of the dark matter is made of these Jupiter-mass objects,” says Hawkins. “The nearest body could be only 30 light years from Earth.”
But what exactly could these Jupiter-mass objects be? If they are indeed the shadowy source of dark matter, there are constraints on what they might be. Most of the dark matter cannot be made of baryons, such as the protons and neutrons that compose normal atomic matter. This is because all the stars, gas clouds and galaxies we see already use up practically all the baryonic matter that was present in the Universe after the fireball of the big bang cooled off. Astronomers can calculate this by looking at the relative amounts of each of the light elements in the Universe.
Mystery matter
So if not made of baryons, what could the lensing objects be? Some of the prime candidates for nonbaryonic dark matter are hypothetical subatomic particles such as neutralinos, which are predicted by so-called supersymmetric theories which attempt to unify gravity with the other forces of nature. These suggest that all the particles we see in nature have supersymmetric partners—as yet undetectable—that would have been created in the big bang.
However, a collection of these particles probably could not collapse under gravity to make objects compact enough to act as lenses. The gravitational collapse would generate heat, which could not escape from the clump because neutralinos do not emit radiation. The pressure from this heat would prevent further shrinkage.
Hawkins could think of only one other possibility for his Jupiter-mass objects —black holes formed in the first moments of creation. The idea sounds crazy, as black holes are usually thought to form when the gravity crushing a very massive star overwhelms the outward pressure of its hot matter. This only happens for stars many times as massive as the Sun—it is impossible to make a Jupiter-mass hole in this way through the pull of gravity alone.
Yet according to some theorists, the fury of the big bang fireball may well have been enough to crush matter together to form these mini black holes. In the 1980s, David Schramm and Matt Crawford of the University of Chicago suggested that the right conditions occurred a mere millionth of a second after the big bang. At this time, the Universe consisted of a hot “soup” of free quarks. A moment later, the strong nuclear force switched on, pulling groups of free quarks together to form baryons such as protons and neutrons. During this transition, the density of certain regions of the fireball could have become gigantic—perhaps big enough to spontaneously form black holes.
In fact, the density would have been just right, according to the results of numerical calculations of the way quarks behave. The densities would have depended on exactly how much turbulence was around during the transition. Hawkins says that in 1995, a Japanese team led by Y. Iwasaka of the University of Tukuba in Osaka published results indicating that turbulence was high enough for black hole formation (Nuclear Physics B, vol PS42, p 499). What’s more, the calculations suggest that the black holes would have masses similar to Jupiter’s, just as Hawkins suggests.
Not surprisingly, Hawkins’s idea sparked lots of controversy when it was first proposed in Nature (vol 366, p 242) two years ago. Astronomer Martin Rees of Cambridge University was one of the first into the fray, arguing that 3C 273, the closest of all quasars, varies over the long term. If long-term brightness changes are due to lensing, the close quasars should be immune to the effect, since the chances of lensing bodies passing in front of them are tiny.
Hawkins agrees with that in principle, but he takes issue with Rees over the observations themselves. Hawkins maintains that, in fact, 3C 273 only shows the short-term variations that are easily explained by changes within the quasar.
Another criticism arises from the exact way in which the quasars brighten and then dim. Astronomers point out that lensing by compact objects causes the brightness of the lensed body to rise and fall in a symmetrical fashion. This characteristic led to the discovery of massive compact halo objects, or MACHOs, in our Galaxy (“The Galaxy’s dark secrets”, 9 April 1994, p 26).
So is there any evidence of symmetry in the long-term quasar variations? Hawkins went back to his data set to find out. He expected the effect to be difficult to spot because each quasar would be lensed, not by a single body moving across the line of sight (the event that produced the “clean” signal of a MACHO), but by several objects. Although each lens would produce a variation in the brightness of the quasar that was symmetrical in time, their combined effect would not be symmetrical.
However, as Hawkins points out, it should be possible to filter out this asymmetry by averaging over many brightening and dimming cycles. A large number of lenses would have passed in front of the quasar by then, and the different asymmetric combinations of lenses would cancel each other out. “This is indeed what I found,” says Hawkins (Monthly Notices of the Royal Astronomical Society, vol 278, p 787). “The rising part of each light curve is indistinguishable from the falling part, statistically.”
Peter Schneider of the Max Planck Institute for Physics and Astronomy in Garching, Germany, agrees. “This test is passed by Mike’s quasars with flying colours.” Nonetheless, like many physicists, Schneider says that statistical evidence is not enough. “The astronomical community is rarely persuaded by statistical evidence,” agrees Paczynski. He says that much more compelling proof would be one undeniably symmetrical light curve due to a single crossing of a lensing object. “What Hawkins needs to produce is a cleaner, stronger signal.”
There is also scepticism about the other evidence Hawkins has marshalled to support his lensing idea. His studies may not have been sensitive enough to pick up small changes of colour in the brightening quasars. And Schneider says that there are several ways of explaining the lack of time dilation without lensing. One explanation is that the variability timescale of quasars happens to change as the Universe grows older in such a way as to cancel the time dilation effect. “Quasars are known to evolve strongly with time,” says Schneider.
Marking time
Schneider points out a second possible reason for the lack of time dilation. The light from a quasar has been “red-shifted” by the expansion of the Universe before it reaches the Earth, so it would have had a shorter wavelength when it left the quasar. Shorter-wavelength light comes from the hotter regions that are nearer the quasar’s core and small regions are able to brighten in synchrony much faster than giant regions. “Hawkins sees smaller and more rapidly variable regions in his most distant quasars,” says Schneider. “This could compensate for the time dilation effect which acts to make any variability appear slower.”
Still, Schneider says that Hawkins’s idea is interesting, and worth testing further. And according to Rees, further clues could come from the lensing of quasars by foreground galaxies. This effect boosts the light of some distant quasars. Nobody knows whether this is due to lensing by the galaxy’s mass as a whole or to small-scale “microlensing” by individual objects within the galaxy. However, if microlensing is the dominant effect and it is also responsible for long-term quasar variability, then the variability should be greatest in quasars that appear to be close to galaxies.
Hawkins hopes to continue his quasar monitoring programme and get the better evidence his critics demand. “All my tests can be firmed up. I’ve only monitored my quasars for 20 years—which is only twice the period of their long-term variability. Many people say they want to see many more years of data.”
One thing is certain: the verdict will attract a lot of interest. “The big advantage of Mike’s idea is that it is testable,” says Andy Taylor, a colleague of Hawkins at Edinburgh. “It’s there to be shot down.” Schramm agrees: “It’s a fun idea. That’s what I like about it.”
