¿ìè¶ÌÊÓƵ

As surfers ride the waves along the coastline of southern Californian,
scientists nearby are creating waves of a different sort. Their waves are
formed in high temperature plasmas and the charged particles that surf
on them approach the speed of light. The technique, researchers say, could
lead to a new generation of smaller and more powerful particle accelerators
than the giant machines running today.

The largest accelerators are tens of kilometres long. The world’s longest,
at the particle physics laboratory at CERN near Geneva, is circular and
measures 27 kilometres round. And the Superconducting Supercollider would
have been 87 kilometres in circumference, if it had been built in the US.
Why so big? Accelerators work on the principle that charged particles speed
up in an electric field. The stronger and longer the field, the more energy
it can pass to a particle. But the fields in today’s machines cannot be
made any stronger without ripping electrons from the fabric of the apparatus,
causing sparks which prevent the accelerator from working. So, using conventional
technologies, the only way to accelerate particles to the high energies
physicists want is to make the electric fields, and hence the machines,
longer.

Plasma wave accelerators work differently. Inside them a powerful laser
beam heats a thin pencil of gas in a larger chamber to form a plasma. At
temperatures of about 100 000 kelvin, the atoms in the gas dissociate into
positive ions and a sea of free electrons. The plasma waves are created
in this sea by laser light. The alternating electric and magnetic fields
that are part of the laser beam’s electromagnetic wave together force electrons
out of their path when the intensity is at a maximum. But they allow the
electrostatic repulsion between the electrons to force them back when the
intensity falls to zero.

The crests and troughs of these waves correspond to regions of high
and low electron density. This variation creates powerful electric fields
which sweep through the plasma approaching the speed of light. The fields
exist over distances of a few ten millionths of a metre. Confined to such
a scale, the electric fields can be extremely powerful without affecting
the apparatus and can accelerate the particles surfing on these waves to
very high energies.

The bigger the plasma waves, the stronger the electric fields. To make
them as powerful as possible, physicists rely on the phenomena of resonance.
There is a certain frequency at which the electrons in a plasma move back
and forth naturally. When the frequency of the light matches this, the plasma
waves are at their largest.

FIND THE BEAT

Unfortunately, a plasma reflects light that exactly matches its own
natural frequency. So, in 1979, John Dawson, a physicist at the University
of California in Los Angeles, and Toshiki Tajima at the University of Texas
in Austin, proposed generating plasma waves using two laser beams of higher
but slightly different frequencies that could pass through the plasma.
When these overlap they interfere to form a ‘beat wave’ of a lower frequency
which matches the natural frequency of the plasma.

Initially, physicists found it extremely difficult to accelerate particles
using the beat wave technique. The intensity of laser light must be extremely
uniform to create plasma waves and this can only be achieved when it is
precisely focused. One of the main problems facing physicists is creating
long enough waves: the waves last only for the very short distance over
which the beam is precisely focused. ¿ìè¶ÌÊÓƵs have found, however, that
the plasma itself can focus light, extending the region of uniform intensity
by seventy times. But there is a long way to go. The best experiments have
achieved a uniform intensity over distances of no more than a centimetre
or so.

SHORT BUT STRONG

Physicists need to increase this to a few metres to get the best performance
out of their machines. This is the maximum length for a plasma accelerator.
Although they are extremely powerful, the electric fields are very short,
stretching from the crest of the plasma wave to the trough. When an electron
has surfed into the trough it cannot gain any more energy. In practice,
this happens after the wave has travelled a few metres. To reach very high
energies physicists expect to build future machines in short stages, accelerating
the electrons more at each stage.

Turbulence is also a problem. The ions in a plasma are heavier than
electrons and so take longer to react to light passing through. At the head
of the plasma-wave train, the light beam is constantly plunging into fresh
plasma so any movement of ions here is insignificant. But the waves behind
begin to interact with these ions causing the plasma to become turbulent
and the waves to decay. The solution is to fire the lasers in pulses with
enough time between each to allow any turbulence to settle down.

The lasers that heat the plasma cause their own set of problems. In
the early 1980s, their optics were not good enough produce a uniform plasma.
This lead to plasmas in which the density of electrons varied from place
to place. As the density changes, so does the resonance frequency making
plasma waves difficult to sustain for more than a fraction of second. And
because of imperfections in laser optics, the intensity of the beam along
its cross section could vary – also producing variations within the plasma.
By the end of the decade, scientists were able to solve both these problems
as more powerful lasers with better optics became available.

The first breakthrough in plasma wave acceleration came in 1992 when
a team at UCLA led by the physicist Chandrashekar Joshi accelerated electrons
to 9 million electronvolts (MeV) over a distance of only 1 centimetre. They
found that the waves were unable to pick up slow moving electrons. In order
to catch the waves in the way surfers have to paddle to catch ocean waves,
Joshi had to inject an electron beam into the plasma that had already been
accelerated to 2.8 MeV. In April this year the team announced that they
had reached 30 MeV – using fields that were thirty times larger than those
possible in conventional accelerators. If they can make their design work
over ten centimetres, they could reach 300 Mev.

Robert Bingham at the Rutherford Appleton Laboratory in Oxfordshire
is confident of further success. ‘There is no reason why physicists cannot
accelerate electrons to energies greater than 1000 MeV over about ten centimetres,’
he says. To produce these energies with current techniques requires accelerators
about a kilometre long. Bingham says a similarly powerful plasma-wave accelerator
would fit on a benchtop although a rather large one.

Other ways of producing plasma waves may be even more promising. Instead
of matching the natural frequency of the plasma to a beat wave, it is possible
to create waves using one intense pulse of light from a single laser. Such
a pulse pushes electrons out of the way as it travels through the plasma.
When it has passed, the electrons spring back. To produce the waves at
the natural frequency of the plasma, the pulse must be half as long as
those waves. The accelerated electrons surf in electric fields created
in the wake of these pulses – hence the name wakefield accelerators.

The most impressive new work involving wakefield accelerators has been
carried out in Japan. A 19-strong team of physicists led by Kazuhisa Nakajima
based at the KEK particle physics laboratory in Tsukuba, has accelerated
electrons to 18 MeV over a distance of only 0.6 millimetres. Although this
may not be as high as the energies achieved with the beat wave technique,
the electric fields were ten times stronger. Bingham is cautious about
future applications, however. He says that the laser pulse can be so intense
that it can push the electrons out of the plasma thus destroying the waves.

Andy Sessler, a physicist at the Lawrence Berkeley Laboratory in California,
is more upbeat however. He points out that if this technique worked over
a distance of a metre it would reach energies of 30 GeV – about a third
of the energy possible with the 27-kilometre Large Electron Positron collider
at CERN.

One problem with both beat wave and wakefield accelerators is that
they accelerate very few electrons at a time. The main reason for this is
that the lasers work only in short bursts at a time, releasing so much energy
that they must cool down for twenty minutes between pulses. So wakefield
techniques, for example, involve electrons surfing behind a single pulse
of light. To make the most of the energy, scien-tists hope to use the same
pulse more than once by piping it back to the beginning or by reflecting
it into another accelerating chamber.

One new accelerating method does not rely on plasmas at all. Instead,
it exploits the fact that highly energetic electrons can travel faster than
light in certain gases and liquids. At these speeds the particles emit
bluish light, known as Cerenkov radiation, in a cone about their path, rather
like the shock waves created by a supersonic jet. In doing this, the particles
lose energy and slow down.

Physicists can accelerate electrons by reversing the process. A beam
of particles gains energy if light hits them at the Cerenkov angle – the
angle subtended by the cone of Cerenkov radiation. The technique is known
as inverse Cerenkov acceleration.

NO COMPARISON

Using this method, scientists at the Brookhaven National Laboratory
in New York have accelerated electrons to 3.7 MeV over a distance of 12
centimetres. This may not be as much as the beat wave or wakefield techniques
generate, but Wayne Kimura, the physicist in charge of the experiment, says
the comparison is unfair. Kimura achieved his result with a 700 megawatt
laser while Joshi used two lasers, both about a thousand times more powerful.

Back in Oxfordshire, however, Bing-ham believes that inverse Cerenkov
techniques do not offer the potential of plasma wave accelerators. Although
it is inherently less complex because there is no plasma to sustain, the
technique is less efficient because the laser has to illuminate a larger
region and is therefore less intense. Kimura is continuing his experiments.
In the next part of his project he hopes to reach 100 MeV.

While 1000 MeV energies are theoretically possible with benchtop-sized
accelerators, physicists want teraelectronvolt (TeV) accelerators which
are at least a thousand times more powerful, to carry out their experiments.
Such machines will be built in stages with each section accepting electrons
from the previous one and accelerating them to higher energies. The sections
may have to contain progressively denser plasmas. The greater the density,
the larger the plasma waves they can support but these have to be driven
by more powerful lasers. To reach such energies, scientists believe that
plasmas might have to exist in solid form and the plasma waves be driven
by X-ray lasers. TeV accelerators may be less than a kilometre long although
Bingham cautions that the technology is still many years away – if it is
possible at all.

The research shows that lasers can accelerate electrons but producing
a well-focused beam that physicists can use is difficult. The plasma itself
can help. The waves create magnetic as well as electric waves and these
act to pinch the electrons into a tightly packed beam. Nevertheless, Robert
Cahn, a physicist at the Lawrence Berkeley Laboratory, believes these
methods are unlikely to lead to accelerators that will be useful for probing
the heart of matter in the next twenty years. ‘When high temperature superconductors
were discovered, some physicists wanted to redesign the SSC to include them.
But fundamental advances cannot always be applied immediately in the way
you want.’

Others agree. A recent report on the future of high-energy physics commissioned
for the American government found that although some ingenious ideas had
been put forward, none of these were yet suitable for a high-energy physics
accelerator.

SILICON SCULPTING

But physicists are not the only people to use accelerators. The high-energy
electrons from plasma-wave accelerators could be fired into a target to
produce X-rays. This would have applications in the semiconductor industry,
for example, where the small wavelength of X-rays makes them ideal for etching
fine detail onto silicon chips.

An important feature of plasma-wave accelerators is that they produce
small bunches of electrons separated by the wavelength of the plasma. In
turn, these produce small, rapid bursts of X-rays which could be extremely
useful in examining structures such as chemical bonds and muscle tissue
that would be destroyed or damaged by a continuous beam. According to Bingham,
these accelerators could be ready in ten years and would be much smaller
than existing X-ray sources which fill a small room.

Much depends on the success of the next generation of experiments. At
the Rutherford Appleton Laboratory, Bingham hopes to collaborate with Joshi
and a team from Imperial College in London to accelerate electrons to 100
MeV in about a centimetre using plasma-beat wave techniques. If all goes
to plan, they hope to start producing results within a year. ‘Once the experiments
have been done, it’s amazing how quickly industry can pounce on technology
like this, improve it and turn it into a viable product.’ Chip manufacturers,
chemists and biologists may have to wait only ten years to see the fruits
of plasma-wave accelerators. High-energy physicists, however, are destined
to wait much longer.

Keay Davidson is a science writer for the San Francisco Examiner.