ONE DAY soon, a courier will knock gingerly on the door of a physics lab and
hand over a box labelled 鈥淏lack Hole Kit: Handle With Care鈥�. Then he鈥檒l probably
run. So perhaps the box should also bear the reassuring words of the device鈥檚
inventor, Ulf Leonhardt: 鈥淭he concerned reader should note that optical black
holes are safe.鈥� No one, repeat, no one, is going to be sucked into
oblivion.
Leonhardt and his colleague Paul Piwnicki at the Royal Institute of
Technology in Sweden have devised a way to create black holes in the lab. Inside
their machines, light will be sucked in, never to return.
These strange devices may uncover some even stranger natural laws. They might
prove that the only thing able to escape from black holes, Hawking radiation,
really exists. They might also provide an experimental stage for quantum
gravity, the attempt to unite quantum mechanics with Einstein鈥檚 general theory
of relativity. This synthesis has evaded physicists for decades, and it could
lead to still grander theories that unite all the forces of nature.
Advertisement
Real black holes suck in light using their ultrapowerful gravity. Any object
with mass distorts space-time, creating a kind of depression in the cosmos, and
something passing through this depression feels a pull towards the centre,
rather like a golf ball passing over the edge of a hole. If it is going slowly,
the ball falls into the hole; if it is going fast enough, it is only deflected.
But black holes are so massive, and the depressions they create are so deep,
that nothing goes fast enough to escape them. Even if you are moving at the
speed of light, to stray beyond a frontier called the event horizon is a bad
idea. You will be sucked into the hole.
Being the fastest thing in the Universe, light is the hardest thing for even
a black hole to grab. So how can it be sucked in by a machine in the lab?
Creating a real black hole would not only be incredibly difficult, it would be
foolhardy鈥攜ou don鈥檛 want to be sucked into a device of your own making.
Leonhardt and Piwnicki have come up with a simple answer: instead of trying to
capture light when it鈥檚 going full tilt, they plan to slow it down. A lot.
The speed of light is constant in a vacuum, but it changes when the light
travels through another medium. In water, for instance, light goes at roughly
three-quarters of its speed in empty space, slowing to 220 million kilometres
per second. That鈥檚 still pretty quick, but in the past year researchers have
done much better: 8 metres per second in a vapour of rubidium atoms (New
Scientist, 20 February 1999, p 10), and as slow as 50 centimetres per
second鈥攍ess than walking pace鈥攊n a special, ultracold state of
matter known as a Bose-Einstein condensate.
And as Leonhardt and Piwnicki have shown (Physical Review Letters,
vol 84, p 822), when a light-slowing medium moves, it can pull the light along
with it. Light facing a fast enough headwind will go backwards. 鈥淚f the velocity
of light is low compared with the velocity of the medium, then the motion of the
medium is overwhelming,鈥� Leonhardt says.
Blowing light around isn鈥檛 quite the same as sucking it in. But Leonhardt and
Piwnicki believe that if they created a swirling vortex in a medium like this,
it would actually swallow light.
On Earth, the nearest things we have to black holes are vortices. Tornadoes,
for example, can suck up trees, roofs and trucks. Their power comes from the low
pressure in the centre of the vortex. And according to Leonhardt and Piwnicki鈥檚
calculations, a vortex should apply the same sort of inward force to light.
If the vortex rotates much faster than the light can move, any ray that
strays too close to its centre will get caught and dragged inexorably inwards
(see Diagram).
The light will eventually be absorbed by the gas. So just like a
real black hole, the vortex has an event horizon beyond which escape is
impossible.
鈥淭his is a really exciting idea,鈥� says Lene Hau of the Rowland Institute for
Science in Massachusetts and of Harvard University. Hau led the team that first
slowed light down to a pedestrian pace in a Bose-Einstein condensate. Within a
couple of months, she hopes to slow light down to just a centimetre per
second.
But condensates have a few drawbacks. To prevent light escaping, the material
in a vortex must move much faster than the light within it. Even when light
speed is just a centimetre per second, the vortex would only begin to suck in
light when it moves at 2 metres per second. However, quantum vortices generally
try to minimise their angular momentum by splitting into several slower
vortices. So it would be difficult to create a single vortex that spins fast
enough. 鈥淚鈥檓 already raising my eyebrows at two metres per second鈥攖hat鈥檚
quite a bit,鈥� Hau says.
Setting the trap
What鈥檚 more, if you spin a condensate rapidly, all the gas is squeezed out
towards the edges, leaving a hollow core. With most gases, this 鈥渆ye of the
hurricane鈥� would be wider than the event horizon, leaving no condensate at the
centre to trap the light. Making black holes out of condensates begins to look
difficult.
An ordinary gas might work better. 鈥淚f you use a classical gas, you can
accelerate it to large velocities and create classical vortices,鈥� Leonhardt
says. A spinning bath of rubidium atoms kept at about 100 掳C might just do
the trick, he believes. Researchers have already managed to slow light to 8
metres per second in this kind of system. To trap light travelling at this
speed, you need a vortex spinning at more than 300 metres per second. That鈥檚
pretty fast, but not impossible to achieve.
Unfortunately, these high velocities present a problem of their own. Hau
first has to prime her gas by firing in a laser beam of a precise frequency.
That puts the atoms into a special quantum state that allows the gas to slow
light of another frequency. But because of the Doppler shift caused by a rapidly
spinning vortex, atoms will see the laser beam鈥檚 frequency rise and fall as they
move back and forth. 鈥淚f you start to get large Doppler shifts, you might move
outside the right bandwidth,鈥� says Hau. The the priming wouldn鈥檛 work.
But Leonhardt isn鈥檛 unduly concerned. If this does turn out to be a problem,
he says, you could tune the laser so it worked just in the interesting region
near the event horizon.
Even if all these problems can be overcome, Matt Visser, a physicist at
Washington University, St Louis, thinks that Leonhardt and Piwnicki鈥檚 prototype
will need some tweaking before it is accepted by the relativity community as a
proper black hole. He says that they need to increase its sucking power by
making the medium flow towards the centre, as well as around it.鈥漈hey鈥檝e got the
basic structure right, and it鈥檚 relatively straightforward to modify this model
so you do get a proper black hole,鈥� Visser says.
Leonhardt disagrees with Visser鈥檚 calculations, but admits that inward gas
motion would help. In his simple spinning vortex, light approaching from one
side of the hole鈥攊nto the wind鈥攊s swallowed much more easily than
light approaching on the other side, which gets whipped around faster by the swirling gas
(see Diagram). It would be very hard to make such a vortex eat
light on both sides.FIG-mg22304101.JPG
The solution may be as simple as pulling the plug. If gas is pumped out of
the centre of the vortex, the rest of it will move towards the middle, dragging
the light with it. You would be sucking up light with a vacuum cleaner.
Whatever the eventual solution, Leonhardt believes that optical black holes
are about five years of experimenting away. All those years of work will prove
worthwhile if they help reveal the secrets of quantum gravity.
The two greatest theories of the 20th century are quantum mechanics, which
describes how particles interact with each other via electromagnetic and nuclear
forces, and Einstein鈥檚 general relativity, which describes how space is bent by
matter and energy, and how that produces the force of gravity. Blending the two
theories to create a quantum theory of gravity has proved a mathematical
nightmare.
Scientists need a theory of quantum gravity to describe the very beginning of
the Universe, when matter was incredibly dense. But with the tenuous matter
around us now, gravity is desperately weak on quantum scales. No one has devised
a way to measure its effects.
Leonhardt points out that in an optical black hole, light experiences a kind
of gravitational field鈥攁nd a very strong one. In an optical black hole,
the swirling gas tells space how to bend, at least as far as a beam of light is
concerned. 鈥淭his could be used for making predictions in quantum gravity,鈥�
Leonhardt suggests.
Visser is also hopeful. 鈥淚t gives you a realistic hope for experimentally
testing at least part of the final theory of quantum gravity,鈥� he says. 鈥淭hat
would be a vast improvement over the current situation.鈥�
There is already one famous prediction in this field just waiting to be
tested. Back in 1974, Stephen Hawking showed that black holes shine. According
to quantum field theory, pairs of particles constantly pop into existence,
recombine and disappear again. These 鈥渧irtual鈥� particles live on borrowed
energy, and they can鈥檛 exist for very long. But if the particle pair happens to
be born just above the event horizon of a black hole, gravity can rip the pair
apart. One of the particles falls into the hole while the other half gains some
energy, allowing it to zoom off into the cosmos. Leonhardt compares the black
hole to an amplifier, boosting vacuum noise into a real signal.
No one has been able to confirm Hawking鈥檚 prediction. The black holes we know
about are too big to produce a measurable effect. A big hole has gentler
gravitational forces at its event horizon than a small one, so it produces less
radiation. Microscopic black holes would be bright enough, and they might have
been produced in the early Universe, but none has yet been seen.
An optical black hole should produce Hawking radiation of its own, as pairs
of virtual photons are dragged apart by the flow. Leonhardt is now trying to
work out whether it will be detectable or not. He says he is convinced that
Hawking must be right, but it would be very satisfying to be able to prove
it.
鈥淚t would be very significant,鈥� agrees Visser, a specialist in relativity,
black holes and quantum gravity. 鈥淯nless we鈥檙e lucky and find a microscopic
black hole left over from the big bang, the only way we鈥檙e going to be able to
test Hawking radiation is through something like this.鈥�
This much could be done with an optical black hole made from classical gas.
And if physicists succeed in making holes using condensates, another aspect of
quantum gravity could be laid bare. In an optical black hole based on quantum
flows, the effective curvature of space would be quantised too.
No one knows exactly what this quantisation will do to gravity. One way to
measure it might be to see what Hawking radiation does to the vortices
themselves鈥攚hether they dissipate, as real black holes are supposed to do.
鈥淲e would be seeing something similar to the quantum structure of a
gravitational object,鈥� says Leonhardt.
Leonhardt is confident that his optical black holes could provide a new way
of tackling the most crucial questions in physics. His device might darken the
lab, but our view of the Universe could become a lot clearer.