Really dark matter: Is the universe made of holes?

23 August 2011 by Marcus Chown
Magazine issue 2826. Subscribe and save
The universe's missing mass may be locked up in legions of tiny black holes, says a controversial new theory

MOST of the matter in the universe gives out no light, or at least so little that it is currently undetectable. Yet we know it is out there because its gravity keeps stars and galaxies in their orbits. Pretty much everyone thinks that this so-called dark matter is made of hitherto undiscovered subatomic particles. Physicists are hopeful they will find a candidate in high-speed collisions at the Large Hadron Collider at CERN, near Geneva, Switzerland.

But could we have got dark matter all wrong? Mike Hawkins thinks so. He believes that rather than particles, what we call dark matter is actually legions of black holes created shortly after the big bang.

It is a controversial claim. Yet Hawkins, who is an astronomer at the Royal Observatory, Edinburgh, in the UK, believes he has persuasive evidence gathered over years of observations. If he is right, it radically changes our picture of the cosmos. "We live in a universe dominated by black holes," he says.

The story dates back to 1975, when Hawkins began monitoring a portion of the southern sky, night after night, using the UK Schmidt Telescope at Siding Springs in Australia. After five years, in addition to the stars he was looking for, he found something unexpected: thousands of objects that were brightening and dimming extremely slowly. They turned out to be quasars, the super-bright cores of newborn galaxies at the edge of the universe.

Yet Hawkins was puzzled. Quasars are powered by matter that shines fiercely as it swirls into a supermassive black hole typically billions of times the mass of the sun. Their light does vary, for example as stars are ripped apart by the black hole's enormous gravity. But these flare-ups make quasars brighten and fade within a few days, not years and decades as Hawkins found.

He wondered if the variability was down to a phenomenon called microlensing. If a massive body happens to pass between us and a quasar, its gravity will bend the quasar's light and magnify it. Perhaps that was responsible for the light variation, rather than it being something intrinsic to the quasar itself. Was it sufficient to explain the decade-long variation Hawkins was seeing?

His calculations showed that it could, provided that the bodies passing in front of the quasar had a mass roughly that of the sun. So what could these massive, microlensing bodies be?

The obvious answer is stars. However, they are ruled out for a compelling reason that dates back to the first few minutes after the big bang. The universe is the way it is now because ever since then it has consisted of a certain ratio of photons to protons and neutrons, which are collectively called baryons.

We can tell what this ratio is from the cosmic microwave background - radiation left over from the big bang. It turns out that there can be at most only twice as many baryons in the universe as we see tied up in stars and galaxies. That's a problem for microlensing of quasar light, because to see the effect with every quasar, which we do, means the bodies bending the light around must add up to far more than double the number of baryons we see in the universe. "There just isn't enough baryonic matter for the lenses to be ordinary stars," says Hawkins. "They must be dark matter."

The baryonic argument rules out many other possible candidates that belong to a family called MACHOs, or massive astrophysical compact halo objects. MACHO is a catch-all name for any dark body that might account for dark matter, and includes neutron stars, black holes and stars so small that they give out little or no light.

But if everything made of ordinary matter is ruled out, that leaves Hawkins with only one candidate: small black holes weighing about the mass of the sun and not much bigger than a village. Stephen Hawking at the University of Cambridge and others have said that these could have formed spontaneously when the universe was about 1/100,000th of a second old.

Star quality

At this time, the building blocks of matter known as quarks went from being free to condensing, like droplets forming in steam. The "droplets" were protons, neutrons and a plethora of other subatomic particles that can only be created in collisions between high-energy particles today.

This process would have resulted in isolated super-dense regions of particles that could have shrunk under their gravity to form primordial black holes. "The characteristic mass of such holes is about the mass of a star," says Hawkins. "Exactly what is required to be quasar microlensing candidates."

Hawkins has had tough time convincing other astronomers. Although no one knows what dark matter is made of, most researchers now favour particles called neutralinos, which are quite unlike ordinary baryonic matter. One reason is that our most promising theories of particle physics points to neutralinos as dark matter. Another is that a wide-ranging search for MACHOs has largely drawn a blank.

Over a period of seven years, an international team called the Macho Collaboration monitored 3 million stars in the Milky Way's satellite galaxies, the Large and Small Magellanic Clouds, to see if there were any signs of MACHOs. The logic was simple. The dark matter in our galaxy is believed to be in the form of a vast spherical halo. If the halo is made of MACHOs, then occasionally one will pass in front of a star in one of the satellite galaxies and microlense its light.

The results were not promising for black holes. The team saw just 17 microlensing events and concluded that MACHOs - whatever they might be - make up less than 20 per cent of the mass of the halo. Later surveys have reported an even smaller percentage. Hawkins, though, believes the results were mistakenly interpreted and that the microlensing events are consistent with a halo made entirely of MACHOs. "There are many uncertainties with these projects, including the structure of the halo and detection efficiency," he says.

He points to several strands of evidence in favour of primordial black holes. For a start, everyone agrees that quasar light definitely varies over periods of many years. Sometimes, the light from a quasar is lensed by an intervening massive galaxy, and this can bend the light along several paths to form multiple images of the quasar. If the light variation is intrinsic to the quasar, any change in one image should be matched by an identical change in the others.

Although this is occasionally seen, the images more often vary independently. "This can only be explained by microlensing," says Hawkins, describing how a primordial black hole can drift into the path of light that goes on to form one image but not the others (see diagram).

A second strand of evidence comes from the timescale with which the light brightens and dims. If this is a feature of the quasar itself, then quasars farther away should vary over a longer period than closer ones. This is because things happen more slowly to distant objects, an effect called time dilation.

But according to Hawkins's analysis, there is no difference in variability between distant and more nearby quasars. "This is compatible with the variation being caused by microlensing," he says, "and not compatible with any form of intrinsic variation."

Bernard Carr of Queen Mary, University of London shares Hawkins's enthusiasm for primordial black holes. "I think his results deserve serious consideration," he says.

Still, no matter how much evidence Hawkins puts forward, he faces a problem. Quasars are complex beasts and our lack of understanding of them leaves the door wide open for alternative theories. For example, some have suggested that their variability could be due to changes in the rate at which matter flows into the black hole. "It's difficult to know how to counter such an unfalsifiable argument," Hawkins says.

Hawkins's hypothesis, on the other hand, is falsifiable. If one of the many experiments searching for particles of cosmic dark matter finds some, then it will be curtains for primordial black holes. So far there are some intriguing claims. The oldest comes from researchers working on the DAMA experiment, at the Gran Sasso underground laboratory in Italy. They claimed in 2000 to have found evidence for dark matter particles raining down on Earth. An improved version of their experiment continues to show such evidence. A second experiment called CoGeNT, deep in the Soudan mine in Minnesota, also claims to have seen the effect (New Scientist, 7 May, p 16).

Both claims are controversial, however, and several other experiments have failed to see anything. "If this holds up, then dark matter is indeed made of particles and I am wrong," Hawkins acknowledges.

What he really needs is a killer observation, but that won't be easy. If the halo is made of primordial black holes, there will trillions of them, greatly outnumbering stars in the galaxy. But they will be at least 40 light years apart - ten times the distance between the sun and its nearest neighbour, Alpha Centauri. Add to that their smallness - less than about 3 kilometres across - and finding a halo of black holes is worse than looking for a needle in the proverbial haystack.

The best way to find black holes, if indirectly, is in another microlensing survey. "The tragedy of the Macho Collaboration's experiment is that it stopped too soon," says Hawkins. "We really need to start the search again."

If he is right, and stellar-mass black holes do account for the universe's dark matter, they need not necessarily have been created in the very early universe. Carr and his colleague Alan Coley at Dalhousie University in Halifax, Nova Scotia, Canada, have been examining models which have the universe going through a series of big bangs and big crunches. In these, black holes created at an earlier time might survive these repeated bounces and still be with us today ( "Macroscopic black holes surviving from the last cycle of the universe would be observationally indistinguishable from ones forming in this cycle," says Carr. "So they might indeed induce quasar lensing."

We have long suspected that dark matter is quite unlike anything we come across here on Earth. Perhaps much of it wasn't even made in our universe.

"The case for primordial black holes as dark matter" by Michael Hawkins (Monthly Notices of the Royal Astronomical Society, vol 415, p 2744)
Marcus Chown is author of the award-winning app "Solar System for iPad" (iTunes)