Nuclear fusion gets quadruple boost
19:00 11 July 01
Eugenie Samuel, Boston
The decades-long effort to build a nuclear fusion reactor has
received a major boost. In experiments at the US National
Fusion Facility in San Diego, researchers have quadrupled
the rate of fusion in superhot deuterium gas.
Fusion reactors aim to
reproduce the Sun's power
source, but the problem is
containing the hot plasma. The
San Diego team achieved more
stable containment and higher
pressure by carefully
manipulating the magnetic fields that control the spinning
This brings us a step closer to a commercial reactor that
could provide enormous amounts of energy with hardly any
pollution or waste.
The team, which includes researchers from Columbia and
Princeton universities, as well as General Atomics of San
Diego and others, is using DIII-D, a tokamak reactor whose
heart is a doughnut-shaped cavity 4.5 metres in diameter.
Inside the cavity a plasma of deuterium is heated to 100
million kelvin and held in place with powerful magnetic fields.
Deuterium is a heavy isotope of hydrogen, and when its nuclei
collide under this intense pressure, some of them fuse to form
helium, releasing large amounts of energy.
The goal of fusion research is a reactor that produces much
more energy than the large amounts needed to run it. The
experimental tokamaks that exist around the world, such as
the Joint European Torus (JET) reactor at Culham near
Oxford, have to date not progressed far beyond the
Early theoretical and experimental
results suggested that there is a limit to
how pressurised the plasma can be
before it begins to bulge unpredictably.
But then other work in the early 1990s
suggested that you could solve the
problem by spinning the plasma around
the cavity as if it were a racetrack.
Experiments at Columbia and DIII-D
had shown it was easy to set the
plasma spinning, but it tended to slow down and become
unstable again. Now the DIII-D team has found out why: the
plasma was magnifying tiny imperfections in the magnetic field
that contained it.
So they fitted sensors to detect the imperfections - some as
weak as the Earth's magnetic field - and then corrected them
with arrays of magnets in the cavity controlled via feedback
"It takes very little power because the errors are about one
part in a thousand," says Ronald Stambaugh of General
Atomics. The team found that the spinning plasma did not
slow down and they could ramp up the pressure to twice the
previous limit, quadrupling the rate of fusion.
Rob Goldston, who worked on Princeton's tokamak until it
was closed in 1997, says he is very excited by the result.
"This is a very deep insight into the behaviour of stable
DIII-D is only about one-eighth the size you'd need for a
commercial reactor, and such a reactor would have to run on
a mixture of deuterium and its heavier sibling tritium.
Despite these differences, Stambaugh believes the principle
will work in a commercial model. "We're doing this research
with the belief that the physics will transfer," he says.
Researchers in Europe, Japan, Russia and Canada are now
lobbying governments to fund a prototype called ITER that
would produce power (New Scientist, 14 October 2000, p
4). Michael Watkins of JET says that the work done at
Culham, combined with the DIII-D team's method, should
have real benefits for the ITER project. "Tokamak research is
in a very strong position now," he says.