91av

SANDWICHED between the brackish waters of Takahoko lake and Obuchi lake in northern Japan lies a stretch of land that could change our planet’s future. All our worries about sky-high oil prices and damaging greenhouse gases could fade if the Japanese government decides to make this the home of a project that could lead to almost unlimited amounts of cheap, clean electricity within 50 years.

Scientists had originally earmarked the land at Rokkasho as one of two possible sites for a vast nuclear fusion experiment called ITER. The aim of ITER is to tame the same nuclear fusion process that powers the sun and produce 10 times as much energy as is it takes to run the machine. In June, after years of political wrangling, officials from six governments finally decided to build ITER in southern France. But despite losing out, Rokkasho may yet be home to another project that turns out to be just as important in realising the dream of a fusion power plant.

The basic idea of nuclear fusion is simple enough. Instead of burning fossil fuels and releasing the electromagnetic energy stored in the chemical bonds that hold molecules together, you unleash the binding energy that grips protons and neutrons in nuclei. Because the force inside nuclei is vastly stronger than electromagnetism, nuclear reactions free around 10 million times more energy than chemical ones.

Researchers have identified the most promising nuclear reaction for generating power: take the heavy isotopes of hydrogen – deuterium and tritium, which have one and two neutrons per atom, respectively – and heat them to 100 million °C in a doughnut-shaped chamber called a tokamak. Some of them will fuse together to form helium, releasing neutrons as well as colossal amounts of energy.

Much of the physics is well understood and researchers are on the way to handling almost all the technical challenges of building such a reactor (see “Fusion fundamentals”). Except for one. No one knows what will happen to the reactor’s structure when it is blasted for years by hundreds of megawatts’ worth of neutrons travelling at a fifth of the speed of light. These subatomic cannonballs will fly out in such large numbers that every atom in the walls of the reactor will be knocked out of position around 30 times a year.

Most of the atoms will simply spring back into place, but some won’t, slowly riddling the material with faults that could turn into cracks. And some of the neutrons are likely to react with nuclei in the reactor structure, leaving it radioactive for decades after the power station closes down. “No one has ever subjected materials to these conditions,” says Chris Llewellyn Smith, director of Culham Science Centre, a fusion facility in Oxfordshire, UK, and head of the scientific committee that advises European politicians on fusion. “There’s just nowhere to do this.”

Enter the patch of land at Rokkasho. Fusion researchers hope it will become home to a centre called the International Fusion Materials Irradiation Facility, which will test potential reactor materials. By blasting them for long periods with beams of neutrons, IFMIF should discover whether steel and other materials will be able to withstand conditions inside a fusion power plant or will simply crumble.

The best estimates are that it will take 10 years to plan and build IFMIF, followed by several years of testing to find the best materials. So running this research in parallel with the experiments at ITER – the so-called fast-track approach – could knock up to 15 years off the time it takes to get to commercial fusion power.

In February, fusion researcher Ian Cook and his colleagues at Culham published a report detailing the necessary steps on the way to commercial power generation. If IFMIF gets the go-ahead this year, a demonstration reactor could produce its first electricity in 2031 and the first commercial power plant could fire up in 2048 (see Diagram). They could even be switched on earlier given enough funding to do more of the groundwork in parallel, for example by building two ITER-type machines, two IFMIFs or additional prototype reactors.

“The fast-track approach could knock up to 15 years off the time it takes to get commercial fusion power”

Change of fortune

All this is a huge turnaround for the fortunes of fusion power. A few years ago its prospects looked bleak. The 1990s saw budgets for fusion research slashed across the world. And in 1998 the US government pulled the plug on its support for ITER, after computer simulations published two years previously by Bill Dorland and Michael Kotschenreuther, both then working at the University of Texas at Austin, made pessimistic predictions about ITER’s performance. They suggested that turbulence inside the plasma would ultimately prevent the ITER tokamak from achieving the temperatures and plasma confinement that its designers hoped for.

So what has changed? Fusion researchers can claim part of the credit. In the wake of the Dorland and Kotschenreuther bombshell, they pored over existing results and conducted new experiments. They found that although plasmas do experience the type of turbulence Dorland and Kotschenreuther predicted, its impact is not as bad as was feared. What’s more, theorists and experimentalists have made a series of advances in understanding plasmas and, crucially, in controlling them (see “Why ITER might work”).

But the main reason is the changing political climate. Worries over high oil prices, the security of oil supplies and climate change have made fusion more attractive to politicians. In particular, Europe and Japan have been battling to host ITER as if it were the Olympics, keen to reap the financial rewards and kudos that the project will bring.

This political wrangling led to an unexpected twist. In 2003, with the US back in the ITER programme after the project was redesigned to be smaller and cheaper, US energy secretary Spencer Abraham called a ministerial-level meeting of the six ITER partners – the US, EU, Japan, China, Russia and South Korea – in Washington DC. The aim was to hammer out a deal between the two remaining contenders for the ITER site: Cadarache in France and Rokkasho in Japan.

However, in the weeks before the meeting a bidding war erupted. Japan offered to increase its contribution to the cost of ITER from 30 to 48 per cent if it hosted the experiment. In response the EU made the same offer if ITER was built in France. Neither side seemed prepared to back down.

Llewellyn Smith and his colleague Frank Briscoe realised that this was a godsend: if both Europe and Japan put in the extra money they were offering, there was enough on the table for both ITER and IFMIF. Llewellyn Smith called the UK government’s chief scientific adviser, David King, on his cellphone as King waited to catch a plane to the Washington meeting. Llewellyn Smith suggested throwing IFMIF into the negotiations as a sort of consolation prize for whichever country failed to get ITER.

Runner-up

King liked the idea, and raised it at a critical point when the meeting appeared to be deadlocked. The idea took off and, after more meetings, formed the basis for a deal between Europe and Japan under which Cadarache gets ITER and Japan gets €680 million to spend on one or more options on a long shopping list that includes IFMIF. Among the other projects vying for money are a remote control centre that would allow researchers in Japan to operate ITER, an advanced computer simulation centre and a major upgrade to Japan’s JT60 tokamak in Naka.

But there are clouds on the horizon. Japan’s consolation prize money is not enough to fund all the options on the table, and some fusion researchers are worried that IFMIF will lose out altogether. Japan is expected to reach a decision in the next few months, and its choice could derail the whole fast-track schedule. “I wouldn’t want to bet on what the Japanese are going to choose,” says Ned Sauthoff of the Princeton Plasma Physics Laboratory in New Jersey and head of the US ITER project.

“The reactor’s materials will be blasted for years by hundreds of megawatts’ worth of neutrons travelling at a fifth of the speed of light”

Another worry is that US politicians have not yet backed the fast-track plan, even in principle. Arguments over how to fund its share of ITER could even result in the US pulling out yet again before the final agreement is signed next year. And even if all governments embrace the fast-track approach and stump up the necessary $1.2 billion a year, there is no guarantee that fusion power will arrive as scheduled, or even at all.

That’s partly because scaling up to ITER and its successors could reveal some unpleasant surprises. At the moment physicists rely largely on the results from small volumes of plasma to predict how large volumes will behave. But those extrapolations may not be valid, because the way the plasma behaves at its edges has a huge impact on the temperature and density that can be achieved deeper inside, and this is still poorly understood for large volumes of plasma.

“No one can tell you how the edge plasma will behave in a machine we haven’t built yet,” says Bill Dorland at the University of Maryland.

But in spite of these unknowns, almost all fusion researchers agree that fusion power can be made to work. With the physics looking so promising, it is now a game of politics.