A machine for testing nuclear weapons opens for business
WHAT DO YOU GET when you focus 192 lasers onto a pellet the size of a match head and press the “fire” button? The answer, hope physicists at the National Ignition Facility (NIF) in Livermore, California, is: the most powerful machine on the planet. The NIF, scheduled to go into operation on May 29th 2009, is designed to create conditions like those found in stars – and also in the explosions of hydrogen bombs. To do that requires, for the brief instants when it is operating at full tilt (a total of three thousandths of a second a year), that it has a power of 500 trillion watts, about 3,000 times the average electricity consumption of the whole of planet Earth.
The pellets at which this energy is directed are made of frozen hydrogen. The aim is to make those pellets undergo nuclear fusion – the process that causes stars to shine and hydrogen bombs to explode. Although the justification for building the NIF has changed over the years (originally there was talk of it being a prototype for fusion-based power stations), it is the resemblance to bombs which has saved the project from the budgetary chop. For the NIF provides America with a way to carry out nuclear-weapons tests without actually testing any weapons.
Had the NIF been a purely scientific project, it would almost certainly have been cancelled. It has cost $4 billion so far, almost four times the original estimate, and is running more than five years behind schedule. Construction started in May 1997 but the initial design proved impractical and was sent back to the drawing board. In 2000 the Department of Energy, which is responsible for the Lawrence Livermore National Laboratory, the NIF’s host, altered the design and revised its budget and deadlines. And in July 2005 Congress actually voted to suspend construction of the machine – relenting only when extra money was found to compensate for cost overruns that had threatened to penalise the work of two other energy-department laboratories that drew their cash from the same pot.
What ultimately saved the NIF from cancellation was that its backers persuaded politicians it was vital to the “stockpile stewardship” programme for America’s nuclear bombs. Although America has not ratified the Comprehensive Test-Ban Treaty, it suspended the testing of its nuclear weapons in 1992. Instead of weapons development, nuclear-weapons scientists are now engaged in a programme intended to ensure that the country’s existing warheads will continue to function predictably as they age. This work uses “subcritical” tests that do not involve full nuclear detonations, and computer simulations of how a weapon would explode.
Such simulations are all well and good, but they must, from time to time, be tested against the real world. That is where the NIF comes in. It will, if it works, create real nuclear explosions, not subcritical phuts. These explosions will be too small to count as nuclear tests within the meaning of the treaty (which America tries to abide by, even though it has not signed). They will, however, be big enough to yield information useful to nuclear-weapons scientists.
Each laser pulse will begin as a weak infra-red beam. This is split into 48 daughter beams that are then fed into preamplifiers which increase their power 20 billion times. Each of the daughters is split further, into four, and passed repeatedly through the main amplifiers. These increase the beams’ power 15,000 times and push their wavelengths into the ultraviolet.
The pellet itself contains a sphere of deuterium (a heavy form of hydrogen, with nuclei consisting of a proton and a neutron) and tritium (even heavier hydrogen, with a proton and two neutrons) that is chilled to just a degree or so above absolute zero. The beams should compress the sphere so rapidly that it implodes, squeezing deuterium and tritium nuclei together until they overcome their mutual repulsion and fuse to form helium (two protons and two neutrons) together with a surplus neutron and a lot of heat. If enough heat is generated it will sustain the process of fusion without laser input, until most of the nuclear fuel has been used up.
Physicists hope that in a year or so the NIF will become the first machine to achieve a nuclear-fusion reaction that produces more energy than it takes to ignite, albeit for only a fraction of a second. Sceptics reckon that the machine may not be capable of such a feat. Creating a sustained nuclear-fusion reaction that could generate power is the goal of another mammoth experiment, the International Experimental Thermonuclear Reactor, which is being built in Cadarache, France. Plenty of people are sceptical about the likely success of that project, too. Like the NIF, it appears to be slipping behind schedule. Full experiments to test nuclear fusion as a power source seem likely to be delayed until 2025.
If the NIF does work, the bomb-scientists will be ecstatic. Astrophysicists will be pretty pleased, too. Although they will get only about 200 of the annual budget of between 700 and 1,000 runs, they will be able to use their time on the machine to simulate the interiors of giant planets, stars and exploding supernovae, by varying the compositions of the pellets to match what they think those things are made of. Bombs or no bombs, astronomy will start to move from being an observational to an experimental science. At a mere $140m a year, then, the NIF is a snip.
This article was first published in The Economist in May 2009.
Preventing nuclear trafficking is easier than policing it
NO NUCLEAR MATERIAL, no bomb. It’s as simple as that. Hence the renewed, unanimous call by the United Nations Security Council for Iran to cease its suspect uranium-enriching and plutonium work. The same is true for terrorist groups such as al-Qaeda, known to be seeking nuclear materials and other weapons of mass destruction as a “religious duty”. The difference is that Iran can produce its own fissile material; terrorists have to steal theirs.
Keeping nuclear weapons out of the hands of terrorists therefore involves a lot of gumshoe work to clamp down on traffickers. But it would surely be better to plug the security holes that allow the stuff to leak out in the first place. Changing the nuclear industry’s security culture is the immodest aim of the newly launched World Institute for Nuclear Security (WINS).
One measure of the scale of the plugging needed comes from the nuclear trafficking database of the International Atomic Energy Agency (IAEA), the UN’s nuclear guardian. From 1993, when the IAEA first started counting, to the end of 2007, there had been 1,340 recorded incidents of the misuse, theft or loss of nuclear materials; 18 of these involved highly enriched uranium (HEU) or plutonium, some in quantities as large as kilograms (it takes up to 25kg of HEU or 8kg of plutonium to make a bomb).
Most such cases have involved material filched from ill-guarded sites in Russia after the collapse of the Soviet Union. The number of serious incidents recorded has been falling. That may not be as comforting as it sounds. Traffickers are probably getting cleverer; by its nature, illicit trade goes largely undetected. But it helps that America has been assisting Russia and others with security upgrades, including better fences, surveillance equipment and radiation monitors, as well as security training, at military and civilian sites where nuclear materials are used or stored.
Now two things are about to change. Security upgrades in Russia are to be completed by the end of 2008. It remains to be seen whether improvements will last once the dollars and the chivvying stop. Meanwhile an industry using deadly materials spread across more than 40 countries may be about to expand farther and wider, as from Venezuela to Vietnam governments contemplate nuclear power as a source of cleaner energy.
The international legal apparatus to deal with this looks robust. The UN’s resolution 1540 obliges all governments to stop nuclear (and chemical and biological) bomb-making materials falling into terrorist hands. Some 75 countries have banded together in a supporting Global Initiative to Combat Nuclear Terrorism. The IAEA, meanwhile, can help governments with advice: protecting the Olympics, for example, against radiological attack, disposing of radioactive materials found in factories and hospitals, and fixing security breaches. It is also boosting the academic study of nuclear security.
But the weakest links will always be sites where materials are kept. WINS is a place where for the first time those with the practical responsibility for looking after nuclear materials – governments, power plant operators, laboratories, universities – can meet to swap ideas and develop best practice.
Start-up cash comes from America’s Department of Energy, the Nuclear Threat Initiative, a private Washington-based group that has long promoted nuclear clean-up activities around the world, and the Norwegian government. Eventually WINS will have to live on support from those who find its services useful.
It took the Chernobyl nuclear disaster in Ukraine for the nuclear industry to focus collectively on reactor operating safety. Preventing security lapses that would mean an even bigger catastrophe ought to be a winning cause.
This article was first published in The Economist in October 2008.
Nuclear power: Combining several small reactors based on simple, proven designs could be a better approach than building big ones
WHEN THE TWO BIG NUCLEAR REACTORS under construction at Flamanville in France and Olkiluoto in Finland come on stream, each will boast enough electricity-generating capacity to light up a city of 1.5m. But despite the best efforts of EDF and Areva, which are building the reactors, both are behind schedule and, at over $5 billion apiece, well over budget. With results like these, it is little wonder that the vaunted “nuclear renaissance” has failed to materialise. In fact, the number of operating reactors is in decline, spurring the nuclear-power industry to look for new approaches. Rather than relying on huge, traditional reactors costing billions, it is turning to small, inexpensive ones, many of which are based on proven designs from nuclear submarines or warships.
A global race is under way to develop small-reactor designs, says Paul Genoa of the Nuclear Energy Institute, an industry body in Washington, DC. He estimates that more than 20 countries have expressed serious interest in buying mini-reactors.
At least eight different approaches are being developed, mainly in America and Asia, by an army of 3,000 nuclear engineers, according to Ron Moleschi of SNC-Lavalin Nuclear, an engineering firm based in Montreal. Regulatory and licensing procedures are lengthy, so little will be built until around 2017, he says. But after that the industry is expected to take off. The International Atomic Energy Agency (IAEA) estimates that by 2030 at least 40 (and possibly more than 90) small reactors will be in operation. It reckons that more than half of the countries that will build nuclear plants in coming years will plump for these smaller, simpler designs.
Russia is an early adopter. Rosatom, the state nuclear-energy giant, is building a floating, towable power station in a St Petersburg shipyard. The Akademik Lomonosov, due to set sail in 2012 for waters near Russia’s far-east town of Vilyuchinsk, will be followed by at least four other floating nuclear plants for the country’s Arctic regions. Such power stations are less prone to earthquakes and avoid the difficulties of erecting nuclear facilities on frozen land, which can melt, jeopardising foundations, says Vladimir Kuznetsov of the IAEA. And at a mere $550m a pop they cost a fraction of what a traditional reactor does (though they also provide less power).
Rosatom hopes its plants will appeal to energy-hungry coastal or river cities all over the world. By manufacturing in Russia, the firm sidesteps some of the regulatory controls a client country would impose on a plant built and installed on its own soil. Another selling point is Russia’s willingness to bring home the nuclear waste. Demand for floating plants may also help Russia broaden, or at least retain, its nuclear expertise, which has suffered as engineers have gone abroad or retired.
“Engineers of small reactors stress their similarity to proven, existing designs such as those found in nuclear-powered ships and submarines.” Similar concerns are driving efforts to develop small reactors elsewhere. No new nuclear plant has come on stream in America since 1996. The industry was dealt a blow in October 2010, when Constellation Energy, a utility, dropped a joint plan with EDF to build a large nuclear plant in Maryland. In spite of strong political support and a reported $7.5 billion in government loan guarantees, Constellation balked at the initial capital outlay. Steven Chu, America’s energy secretary, sees miniaturisation as a way to revive the country’s once-mighty nuclear industry.
One advantage of small reactors is their modularity. Extra units can be added to a plant over the years, incrementally boosting output as capital becomes available and electricity demand rises. NuScale, of Corvallis, Oregon, offers “scalable” nuclear plants with reactors delivered by truck. A plant with 12 reactors, each with its own electricity-generating turbine, would cost about $2.2 billion and produce roughly a third as much power as a big facility. Since large plants can cost roughly three times as much, the cost of electricity would be about the same. Moreover, a modular facility would generate revenue as soon as the first reactor is fired up, after a few years of construction. A big reactor traditionally takes a decade to erect.
Hyperion Power Generation, a firm based in Santa Fe, New Mexico, is building components for what it calls a “nuclear battery”. The refrigerator-sized Hyperion Power Module (HPM) reactor will shift much of the building from field to factory, where a controlled environment reduces costs. Also, fewer workers and families must be moved, at great expense, to distant building sites. HPMs would be delivered by truck with enough uranium to run for about ten years. They would be constructed in batches with interchangeable parts and cost about $100m each. And they need little human oversight to operate. “Forget huge – let’s make a hand-held version of a power plant,” says John Deal, the firm’s boss. Five companies, located in America, Britain, Canada, China and India, have put down deposits for an HPM.
Engineers of small reactors stress their similarity to proven, existing designs such as those found in nuclear-powered ships and submarines, or, in Rosatom’s case, icebreakers. And some small-reactor designs have an important advantage over bigger reactors. Because less heat is generated, small water-cooled reactors can use simpler designs relying not on pumps, but on natural convection. And eliminating moving parts should make the new small reactors both safer and cheaper. For instance, Hyperion’s HPM dispenses with elaborate valve systems by using a molten metal as a coolant because, unlike water, it doesn’t need to be kept under pressure to absorb large amounts of heat.
Christofer Mowry, who heads civilian power at Babcock & Wilcox, a maker of nuclear-propulsion systems for the US Navy, says the company’s small reactor offers another source of savings. Because it can use existing power-transmission lines without overloading them, the mPower can act as a “drop-in replacement” for ageing coal furnaces without the need for costly refurbishment. The Tennessee Valley Authority, America’s biggest public utility, hopes to put two of the firm’s reactors into an old coal plant. Five other American utilities are also considering replacing coal furnaces with nuclear reactors, according to Philip Moor of the American Nuclear Society, an industry group. He estimates that in America alone perhaps 100 old coal plants could be converted to nuclear within a decade – a trice by the industry’s standards.
Not all nuclear nations have entered the fray. France has studied micro-reactors’ potential in spaceship propulsion, but for generating power on Earth, big reactors are best, says Christophe Béhar, in charge of nuclear energy at the country’s Atomic Energy Commission. New markets for large plants are opening up as developing countries strengthen their grids to cope with the huge amounts of power they produce. China is building two small helium-cooled reactors, but the electricity they produce will never be as cheap as that from big reactors, according to Mr Béhar. Just in case, the Chinese have also commissioned French firms to build two large nuclear plants.
In Japan, too, utilities’ interest in small reactors appears scant for now. Tatsujiro Suzuki, the vice-chairman of the Atomic Energy Commission in Tokyo, hopes it will grow. Today’s broad trend to loosen government controls on electricity prices may do the trick. Utilities are more willing to make massive investments if they can accurately predict future income. As prices are allowed to fluctuate more widely, shorter-term investments for smaller reactors will become more attractive. At least one Japanese engineering giant sees promise in the market for such devices. Toshiba says its 4S (“super-safe, small and simple”) reactor is capable of running for three decades without refuelling.
Sceptics fear that these small, cheap reactors will not be enough to revive the nuclear industry. Mycle Schneider, a nuclear-energy expert who is an external lecturer at École des Mines, an engineering school in France, and also an adviser to Germany’s environment ministry, says licensing and building small plants will take far too long to be profitable. As the costs of solar, wind and biogas power continue to fall, investors will increasingly favour household energy-producing kit and transmission technologies that let consumers sell excess production to neighbours and utilities, he says. South Africa’s decision in September 2010 to abort construction of a small reactor, even though about $1.3 billion had been spent, illustrates the sort of financial risk the sector faces.
Others fret that lots of small reactors, rather than a few big ones, will be more vulnerable to a terrorist attack. Hyperion’s Mr Deal insists that neither a rocket-propelled grenade nor a tank round could smash a small reactor. Small reactors can be shielded by a heavy layer of concrete and buried, in effect making them safer than big ones, whose protective concrete domes can only be so thick, lest they collapse under their own weight.
What if a rogue government tries to take advantage of an affordable reactor to acquire nuclear expertise or materials for weapons work? Henry Sokolski, a former Pentagon official who heads the Nonproliferation Policy Education Centre, a think-tank near Washington, DC, says that Western intelligence agencies have overestimated their ability to monitor the spread of nuclear equipment and know-how. If new enrichment facilities are built to supply a slew of small nuclear reactors, materials and expertise useful in bomb-making may spread as a result.
TerraPower, an American firm backed by Bill Gates, thinks it has the solution. It is designing a small “travelling wave” reactor that, once kick-started with a tiny amount of enriched uranium, would run for decades on non-enriched, depleted uranium – a widely available material. This will be possible because the nuclear reaction, eating its way through the core at the rate of about one centimetre a year, would gradually convert the depleted uranium into fissionable plutonium – in effect “breeding” high-grade fuel and then consuming it.
Mr Gates points out that nuclear power has historically been dogged by five worries: safety, proliferation, waste, cost and fuel availability. “This thing is a miracle that solves all five,” he says. John Gilleland, TerraPower’s boss, says that a single enrichment plant would then suffice to produce all the enriched uranium needed to spark up the world’s mini-reactors.
The prospects for mini-reactors, like those for large reactors, depend on a combination of technical, commercial and regulatory factors. The stars do not seem to be aligning for large reactors. But they are no longer the only game in town.
This article was first published in The Economist in December 2010.
How to analyse smuggled uranium
BETWEEN 1992 AND 2007, according to Ian Hutcheon of the Lawrence Livermore National Laboratory, in California, 17kg of highly enriched uranium was seized from smugglers around the world, along with 400 grams of plutonium. In neither case is that enough for a proper atom bomb, but it is still worrying. Presumably, more is out there. Even if it is not, the material that has been found could have been used to make a “radiological” weapon, by blowing it up and scattering it around a city using conventional explosives. Dr Hutcheon is one of those charged with analysing this captured material, to discover how dangerous it really is and where it came from – and thus whether it has been stolen from legitimate nuclear projects or made on the sly. He showed off some of the tricks of his trade at a meeting organised by the American Association for the Advancement of Science in February 2010 in San Diego.
His main tool is a device called a secondary-ion mass spectrometer. This measures the flight path of ions (electrically charged atoms) through a magnetic field. The lighter an ion is, the more the field bends its trajectory. The spectrometer can thus distinguish between, say, 235U (the fissile sort, from which bombs are made) and 238U (which has three extra neutrons in its nucleus and is much less fissile). Natural uranium has only seven atoms per thousand of the former. Weapons-grade uranium is 95% 235U. The “depleted” uranium used in armour-penetrating shells, by contrast, is almost pure 238U.
Uranium that has been in a reactor, though, has other isotopes in it, 233U and 236U, for example. The quantities of these, plus isotopes of elements such as plutonium that are also created in reactors, vary from one reactor to another. The isotopic signature is changed, too, by the centrifuges used to separate 235U from 238U during the process of enrichment, and radioactive decay after processing creates yet further elements that can be detected this way. These give some idea of a sample’s age.
The result is a profile that is often characteristic of a particular type of reactor or centrifuge, and sometimes of an individual machine – and can also indicate how long ago the processing took place. That enables the good guys to improve security in the case where something has been pinched, the bad guys to be admonished if they have been up to something they should not have been doing, and everyone else to sleep more easily in their beds.
This article was first published in The Economist in February 2010.