Plutonium is the nuclear nightmare. A by-product of conventional power-station reactors, it is the key ingredient in nuclear weapons. And even when not made into bombs, it is a million-year radioactive waste legacy that is already costing the world billions of dollars a year to contain.

And yet, some scientists say, we have the technology to burn plutonium in a new generation of “fast” reactors. That could dispose of the waste problem, reducing the threat of radiation and nuclear proliferation, and at the same time generate vast amounts of low-carbon energy. It sounds too good to be true. So are the techno-optimists right — or should the conventional environmental revulsion at all things nuclear still hold?

Fast-breeder technology is almost as old as nuclear power. But after almost two decades in the wilderness, it could be poised to take off. The U.S. corporation GE Hitachi Nuclear Energy (GEH) is promoting a reactor design called the PRISM (for Power Reactor Innovative Small Modular) that its chief consulting engineer and fast-breeder guru, Eric Loewen, says is a safe and secure way to power the world using yesterday’s nuclear waste.

The company wants to try out the idea for the first time on the northwest coast of England, at the notorious nuclear dumping ground at Sellafield, which holds the world’s largest stock of civilian plutonium. At close to 120 tons, it stores more plutonium from reactors than the U.S. and Russia combined.

Britain’s huge plutonium stockpile makes it a vast energy resource.

While most of the world’s civilian plutonium waste is still trapped inside highly radioactive spent fuel, much of that British plutonium is in the form of plutonium dioxide powder. It has been extracted from spent fuel with the intention of using it to power an earlier generation of fast reactors that were never built. This makes it much more vulnerable to theft and use in nuclear weapons than plutonium still held inside spent fuel, as most of the U.S. stockpile is.

The Royal Society, Britain’s equivalent of the National Academy of Sciences, reported last year that the plutonium powder, which is stored in drums, “poses a serious security risk” and “undermines the UK’s credibility in non-proliferation debates.”

Spent fuel, while less of an immediate proliferation risk, remains a major radiological hazard for thousands of years. The plutonium — the most ubiquitous and troublesome radioactive material inside spent fuel from nuclear reactors — has a half-life of 24,100 years. A typical 1,000-megawatt reactor produces 27 tons of spent fuel a year.

None of it yet has a home. If not used as a fuel, it will need to be kept isolated for thousands of years to protect humans and wildlife. Burial deep underground seems the obvious solution, but nobody has yet built a geological repository. Public opposition is high — as successive U.S. governments have discovered whenever the burial ground at Yucca Mountain in Nevada is discussed — and the cost of construction will be huge. So the idea of building fast reactors to eat up this waste is attractive — especially in Britain, but also elsewhere.

Theoretically at least, fast reactors can keep recycling their own fuel until all the plutonium is gone, generating electricity all the while. Britain’s huge plutonium stockpile makes it a vast energy resource. David MacKay, chief scientist at the Department of Energy and Climate Change, recently said British plutonium contains enough energy to run the country’s electricity grid for 500 years.

Fast reactors can be run in different ways, either to destroy plutonium, to maximize energy production, or to produce new plutonium. Under the PRISM proposal now being considered at Sellafield, plutonium destruction would be the priority. “We could deal with the plutonium stockpile in Britain in five years,” says Loewen. But equally, he says, it could generate energy, too. The proposed plant has a theoretical generating capacity of 600 megawatts.

Fast reactors could do the same for the U.S. Under the presidency of George W. Bush, the U.S. launched a Global Nuclear Energy Partnership aimed at developing technologies to consume plutonium in spent fuel. But President Obama drastically cut the partnership’s funding, while also halting work on the planned Yucca Mountain geological repository. “We are left with a million-year problem,” says Loewen. “Right now there isn’t a policy framework in the U.S. for solving this issue.”

He thinks Britain’s unique problem with its stockpile of purified plutonium dioxide could break the logjam. “The UK is our best opportunity,” he told me. “We need someone with the technical confidence to do this.”

The PRISM fast reactor is attracting friends among environmentalists formerly opposed to nuclear power. They include leading thinkers such as Stewart Brand and British columnist George Monbiot. And, despite the cold shoulder from the Obama administration, some U.S. government officials seem quietly keen to help the British experiment get under way. They have approved the export of the PRISM technology to Britain and the release of secret technical information from the old research program. And the U.S. Export-Import Bank is reportedly ready to provide financing.

Britain has not made up its mind yet, however. Having decided to try and re-use its stockpile of plutonium dioxide, its Nuclear Decommissioning Authority has embarked on a study to determine which re-use option to support. There is no firm date, but the decision, which will require government approval, should be reached within two years. Apart from a fast-breeder reactor, the main alternative is to blend the plutonium with other fuel to create a mixed-oxide fuel (mox) that will burn in conventional nuclear power plants.

Britain has a history of embarrassing failures with mox, including the closure last year of a $2 billion blending plant that spent 10 years producing a scant amount of fuel. And critics say that, even if it works properly, mox fuel is an expensive way of generating not much energy, while leaving most of the plutonium intact, albeit in a less dangerous form.

Only fast reactors can consume the plutonium. Many think that will ultimately be the UK choice. If so, the PRISM plant would take five years to license, five years to build, and could destroy probably the world’s most dangerous stockpile of plutonium by the end of the 2020s. GEH has not publicly put a cost on building the plant, but it says it will foot the bill, with the British government only paying by results, as the plutonium is destroyed.

Proponents of fast reactors see them as the nuclear application of one of the totems of environmentalism: recycling.

The idea of fast breeders as the ultimate goal of nuclear power engineering goes back to the 1950s, when experts predicted that fast-breeders would generate all Britain’s electricity by the 1970s. But the Clinton administration eventually shut down the U.S.’s research program in 1994. Britain followed soon after, shutting its Dounreay fast-breeder reactor on the north coast of Scotland in 1995. Other countries have continued with fast-breeder research programs, including France, China, Japan, India, South Korea, and Russia, which has been running a plant at Sverdlovsk for 32 years.

But now climate change, with its urgency to reduce fossil fuel use, and growing plutonium stockpiles have changed perspectives once again. The researchers’ blueprints are being dusted off. The PRISM design is based on the Experimental Breeder Reactor No 2, which was switched on at the Argonne National Laboratory in Illinois in 1965 and ran for three decades.

Here is how conventional and fast reactors differ. Conventional nuclear reactors bombard atoms of uranium fuel with neutrons. Under this bombardment, the atoms split, creating more neutrons and energy. The neutrons head off to split more atoms, creating a chain reaction. Meanwhile, the energy heats a coolant passing through the reactor, such as water, which then generates electricity in conventional turbines.

The problem is that in this process only around 1 percent of the potential energy in the uranium fuel is turned into electricity. The rest remains locked up in the fuel, much of it in the form of plutonium, the chief by-product of the once-through cycle. The idea of fast reactors is to grab more of this energy from the spent fuel of the conventional reactor. And it can do this by repeatedly recycling the fuel through the reactor.

The second difference is that in a conventional reactor, the speed of the neutrons has to be slowed down to ensure the chain reactions occur. In a typical pressurized-water reactor, the water itself acts as this moderator. But in a fast reactor, as the name suggests, the best results for generating energy from the plutonium fuel are achieved by bombarding the neutrons much faster. This is done by substituting the water moderator with a liquid metal such as sodium.

Proponents of fast reactors see them as the nuclear application of one of the totems of environmentalism: recycling. But many technologists, and most environmentalists, are more skeptical.

The skeptics include Adrian Simper, the strategy director of the UK’s Nuclear Decommissioning Authority, which will be among those organizations deciding whether to back the PRISM plan. Simper warned last November in an internal memorandum that fast reactors were “not credible” as a solution to Britain’s plutonium problem because they had “still to be demonstrated commercially” and could not be deployed within 25 years.

Critics argue that plutonium being prepared for recycling ‘would be dangerously vulnerable to theft or misuse.’

The technical challenges include the fact that it would require converting the plutonium powder into a metal alloy, with uranium and zirconium. This would be a large-scale industrial activity on its own that would create “a likely large amount of plutonium-contaminated salt waste,” Simper said.

Simper is also concerned that the plutonium metal, once prepared for the reactor, would be even more vulnerable to theft for making bombs than the powdered oxide. This view is shared by the Union of Concerned Scientists in the U.S., which argues that plutonium liberated from spent fuel in preparation for recycling “would be dangerously vulnerable to theft or misuse.”

GEH says Simper is mistaken and that the technology is largely proven. That view seems to be shared by MacKay, who oversees the activities of the decommissioning authority.

The argument about proliferation risk boils down to timescales. In the long term, burning up the plutonium obviously eliminates the risk. But in the short term, there would probably be greater security risks. Another criticism is the more general one that the nuclear industry has a track record of delivering late and wildly over budget — and often not delivering at all.

John Sauven, director of Greenpeace UK, and Paul Dorfman, British nuclear policy analyst at the University of Warwick, England, argued recently that this made all nuclear options a poor alternative to renewables in delivering low-carbon energy. “Even if these latest plans could be made to work, PRISM reactors do nothing to solve the main problems with nuclear: the industry’s repeated failure to build reactors on time and to budget,” they wrote in a letter to the Guardian newspaper. “We are being asked to wait while an industry that has a track record for very costly failures researches yet another much-hyped but still theoretical new technology.”

But this approach has two problems. First, climate change. Besides hydroelectricity, which has its own serious environmental problems, nuclear power is the only source of truly large-scale concentrated low-carbon energy currently available. However good renewables turn out to be, can we really afford to give up on nukes?

Second, we are where we are with nuclear power. The plutonium stockpiles have to be dealt with. The only viable alternative to re-use is burial, which carries its own risks, and continued storage, with vast expense and unknowable security hazards to present and countless future generations.

For me, whatever my qualms about the nuclear industry, the case for nuclear power as a component of a drive toward a low-carbon, climate-friendly economy is compelling. [A few months ago, I signed a letter with Monbiot and others to British Prime Minister David Cameron, arguing that environmentalists were dressing up their doctrinaire technophobic opposition to all things nuclear behind scaremongering and often threadbare arguments about cost. I stand by that view.]

Those who continue to oppose nuclear power have to explain how they would deal with those dangerous stockpiles of plutonium, whether in spent fuel or drums of plutonium dioxide. They have half-lives measured in tens of thousands of years. Ignoring them is not an option.