A reality check on nuclear fusion at MIT

Despite skepticism, research on nuclear fusion--long considered the ultimate clean-energy source--quietly continues in labs around the world, including one at MIT.

Martin LaMonica Former Staff writer, CNET News
Martin LaMonica is a senior writer covering green tech and cutting-edge technologies. He joined CNET in 2002 to cover enterprise IT and Web development and was previously executive editor of IT publication InfoWorld.
Martin LaMonica
6 min read

CAMBRIDGE, Mass.--Nuclear fusion suffers from an image of being too good to be true. But researchers here say they are already doing nuclear fusion on a small scale and it's just a matter of time--decades, realistically--before it becomes practical.

The Massachusetts Institute of Technology houses an active research lab for nuclear fusion, which many consider the answer to weaning the world from its costly dependence on fossil fuels. A tour of the Plasma Science and Fusion Center last week, organized by the TEDx Boston conference, offered a crash course on nuclear fusion, a field of research which continues largely outside daily discussions of energy.

Inside MIT's plasma and nuclear fusion labs (photos)

See all photos

Nuclear fusion--where two atoms fuse to form a new element--is the energy source for the blazing hot sun and other stars. Here on Earth, people may associate nuclear fusion with what was called cold fusion, an electrochemical process introduced as a breakthrough in 1989 with much fanfare. It was later discredited but the basic approach of room-temperature fusion still retains a small group of followers, according to a "60 Minutes" report two years ago.

At MIT and other centers, such as the Lawrence Livermore National Laboratory's National Ignition Facility, the route to fusion is most definitely through very high heat in giant reactors, not tabletop chemical reactions. And therein lies one of the biggest challenges with nuclear fusion: just performing fusion tests is extremely expensive and takes years.

Even what could be called mainstream fusion techniques still draw skepticism, but it's easy to see why people find it so compelling. The magnetic nuclear fusion studied at MIT uses an abundant energy source (a form of hydrogen found in seawater), power plants would pack a lot of energy in a much smaller footprint than solar or wind, and any radioactive material could be handled relatively easily. The waste from today's nuclear power plants, which split atoms (nuclear fission) to get usable energy, should have storage designed for tens of thousands of years while fusion would need 50-year repositories.

But the punchline for nuclear fusion is that it has one big down side: it doesn't work.

For Dennis Whyte, professor of nuclear science and engineering at MIT, that's not quite right. Fusion actually occurs at MIT's lab but nobody has been able to do it on a continuous basis.

"I would state that we have the scientific and technical readiness to produce electricity on a very short time scale," Whyte said during a presentation. "It would not be economically efficient mostly because we couldn't demonstrate that it would be on all the time, which is essentially a technical argument. And it would cost much more than a (nuclear) fission power plant--it's hard to say exactly but at least a factor of 10 larger."

A worldwide research collaboration called ITER with ties to MIT has begun construction of a giant experimental magnetic fusion reactor in the south of France which has a number of technical goals, including sustained fusion and demonstrating the ability to produce more heat than the amount of energy that is put in. The reactor, expected to be about 20 stories high, will be one of the biggest science projects ever and cost an estimated $18 billion over 10 years.

Hot, magnetized torus
Despite the long time scales for development, nuclear fusion has started to attract the kind of funding start-ups normally get. British Columbia-based General Fusion earlier this year received an investment from by Amazon CEO Jeff Bezos and venture capitalists to demonstrate "magnetized target fusion" where pneumatic pistons larger than people compress plasma--a state of matter similar to a gas--to cause fusion to occur.

The end goal of civilian fusion efforts is a power plant that uses the heat from nuclear fusion to make steam which is converted into electricity in a turbine, just as today's power plants do. But rather than burn fossil fuels or split uranium atoms, magnetic-fusion researchers intend to produce energy by fusing two isotopes of hydrogen--deuterium, which has one neutron, and tritium, which has two neutrons. The result is a heavier substance--helium--and a huge release of energy.

The ITER tokamak nuclear fusion reactor is planned to begin operation by the end of the decade. Click on the image to go to an interactive schematic. ITER

MIT's research reactor, called Alcator C-Mod, is a doughnut-, or torus-, shaped reactor called a tokamak, which is a Russian acronym for a configuration designed in the Soviet Union. Fusion happens when materials reach the plasma state, where electrons break off from an atom's nucleus, which happens at very high temperatures over 10,000 degrees, Whyte explained.

A tokamak uses powerful magnetic fields to confine plasma so that a reaction can take place without damaging the reactor vessel. Applying high temperatures through microwaves delivers the heat to precipitate a reaction. "Eventually it will get hot enough that it will ignite and that's what drives the fusion," Whyte said.

Right now, the Plasma Science and Research Center can do this high-temperature fusion at millions of degrees Celsius for a couple of seconds at a time. Tests at Alcator C-Mod, which is just a few feet across at the core, are done continually to collect data on how plasma affects material in the reactor, the temperature and strength of the magnetic field, and properties of the plasma inside.

Making even this experimental process operate requires megawatts of power for short bursts, buffered by an on-site alternator, and producing magnetic fields 10,000 times larger than the Earth's to maintain the naturally unstable plasma, according to researchers. Since fusion requires high temperatures, large reactors are needed to minimize heat loss and get to the point where the reaction is self-sustaining, Whyte explained. The ITER tokamak alone, for example, will weigh 23,000 metric tons, or three times the weight of the Eiffel Tower, and have about 1 million components.

In our lifetime?
Mastering nuclear fusion requires deep understanding of plasma and materials science, but a functioning magnetic-fusion power plant would also require nuclear-power engineering to harness the energy reliably.

When deuterium and tritium are fused together, an atom of helium is produced and a neutron is fired off. The kinetic energy from that neutron--"imagine a Ping-Pong ball bouncing around," Whyte said--is the energy source to both generate usable heat and produce tritium.

The basic idea of nuclear fusion: two atoms--in this case hydrogen isotopes--fuse to form a new atom, helium, and excess energy. The energy from helium production would be used to heat the fusion process and the neutron carries energy that would be converted into electricity.

In a reactor, the plasma would be surrounded by a "blanket" of liquid metal that contains lithium. The neutron from the fusion reaction would blast into the lithium and undergo a nuclear reaction to produce more tritium. The tritium from that reaction would be captured and then "recycled" by feeding it back into the nuclear reactor.

At the same time, the large amounts of kinetic energy from liberated neutrons heat the liquid metal blanket. A heat exchanger would circulate water to remove the heat and make steam, which would be used to make electricity. The other energy source for the reaction--deuterium--is found in large amounts in seawater.

That, of course, is the theory. Getting to the point where the science is fully under control and engineering ready for working power plants is years away. How far away really depends on the pace of development, but it appears the prevailing view is that it's on the order of decades. The Plasma Science and Fusion Center at MIT, one of many around the world, has an annual budget of about $30 million a year. The multibillion dollar ITER project, meanwhile, has come under fire for escalating costs, according to an NPR report last year.

The years of research in the past have convinced more people of the potential of tapping fusion--essentially making a sun on Earth--for an abundant and environmentally benign energy source, Whyte said. In the late 1980s and early 1990s, some of the last theoretical barriers to magnetic fusion were raised.

"For people who believe in fusion, we don't believe in it because we have some religious faith in it. It's because we've done the research, we've gone through the numbers and there's nothing fundamental in nature that's stopping us," he said. "This wasn't clear 60 years ago when research started. There could have been a showstopper."