Spencer Reiss, WIRED magazine, reports:

China is staring at the dark side of double-digit growth. Blackouts roll and factory lights flicker, the grid sucked dry by a decade of breakneck industrialization. Oil and natural gas are running low, and belching power plants are burning through coal faster than creaky old railroads can deliver it. Global warming? The most populous nation on earth ranks number two in the world - at least the Kyoto treaty isn’t binding in developing countries. Air pollution? The World Bank says the People’s Republic is home to 16 of the planet’s 20 worst cities. Wind, solar, biomass - the country is grasping at every energy alternative within reach, even flooding a million people out of their ancestral homes with the world’s biggest hydroelectric project. Meanwhile, the government’s plan for holding onto power boils down to a car for every bicycle and air-conditioning for a billion-odd potential dissidents.

What’s an energy-starved autocracy to do?

Go nuclear.

While the West frets about how to keep its sushi cool, hot tubs warm, and Hummers humming without poisoning the planet, the cold-eyed bureaucrats running the People’s Republic of China have launched a nuclear binge right out of That ’70s Show. Late last year, China announced plans to build 30 new reactors - enough to generate twice the capacity of the gargantuan Three Gorges Dam - by 2020. And even that won’t be enough. The Future of Nuclear Power, a 2003 study by a blue-ribbon commission headed by former CIA director John Deutch, concludes that by 2050 the PRC could require the equivalent of 200 full-scale nuke plants. A team of Chinese scientists advising the Beijing leadership puts the figure even higher: 300 gigawatts of nuclear output, not much less than the 350 gigawatts produced worldwide today.

To meet that growing demand, China’s leaders are pursuing two strategies. They’re turning to established nuke plant makers like AECL, Framatome, Mitsubishi, and Westinghouse, which supplied key technology for China’s nine existing atomic power facilities. But they’re also pursuing a second, more audacious course. Physicists and engineers at Beijing’s Tsinghua University have made the first great leap forward in a quarter century, building a new nuclear power facility that promises to be a better way to harness the atom: a pebble-bed reactor. A reactor small enough to be assembled from mass-produced parts and cheap enough for customers without billion-dollar bank accounts. A reactor whose safety is a matter of physics, not operator skill or reinforced concrete. And, for a bona fide fairy-tale ending, the pot of gold at the end of the rainbow is labeled hydrogen.

A soft-spoken scientist named Qian Jihui has no doubt about what the smaller, safer, hydrogen-friendly design means for the future of nuclear power, in China and elsewhere. Qian is a former deputy director general with the International Atomic Energy Agency and an honorary president of the Nuclear Power Institute of China. He’s a 67-year-old survivor of more than one revolution, which means he doesn’t take the notion of upheaval lightly.

“Nobody in the mainstream likes novel ideas,” Qian says. “But in the international nuclear community, a lot of people believe this is the future. Eventually, these new reactors will compete strategically, and in the end they will win. When that happens, it will leave traditional nuclear power in ruins.”

Now we’re talking revolution, comrade.

Known as China’s MIT, Tsinghua University sprawls across a Qing-dynasty imperial garden, just outside the rampart of mirrored Blade Runner towers that line Beijing’s North Fourth Ring Road. Wang Dazhong came here in the mid-1950s as a member of China’s first-ever class of homegrown nuclear engineers. Now he’s director emeritus of Tsinghua’s Institute of Nuclear and New Energy Technology, aka INET, and a key member of Beijing’s energy policy team. On a bright morning dimmed by Beijing’s ever-present photochemical haze, Wang sits in a spartan conference room lit by energy-efficient compact fluorescent bulbs.

“If you’re going to have 300 gigawatts of nuclear power in China - 50 times what we have today - you can’t afford a Three Mile Island or Chernobyl,” Wang says. “You need a new kind of reactor.”

That’s exactly what you can see 40 minutes away, behind a glass-enclosed guardhouse flanked by military police. Nestled against a brown mountainside stands a five-story white cube whose spare design screams, “Here be engineers!” Beneath its cavernous main room are the 100 tons of steel, graphite, and hydraulic gear known as HTR-10 (i.e., high-temperature reactor, 10 megawatt). The plant’s output is underwhelming; at full power - first achieved in January - it would barely fulfill the needs of a town of 4,000 people. But what’s inside HTR-10, which until now has never been visited by a Western journalist, makes it the most interesting reactor in the world.

In the air-conditioned chill of the visitors’ area, a grad student runs through the basics. Instead of the white-hot fuel rods that fire the heart of a conventional reactor, HTR-10 is powered by 27,000 billiards-sized graphite balls packed with tiny flecks of uranium. Instead of superhot water - intensely corrosive and highly radioactive - the core is bathed in inert helium. The gas can reach much higher temperatures without bursting pipes, which means a third more energy pushing the turbine. No water means no nasty steam, and no billion-dollar pressure dome to contain it in the event of a leak. And with the fuel sealed inside layers of graphite and impermeable silicon carbide - designed to last 1 million years - there’s no steaming pool for spent fuel rods. Depleted balls can go straight into lead-lined steel bins in the basement.

Wearing disposable blue paper gowns and booties, the grad student leads the way to a windowless control room that houses three industry-standard PC workstations and the inevitable electronic schematic, all valves, pressure lines, and color-coded readouts. In a conventional reactor’s control room, there would be far more to look at - control panels for emergency core cooling, containment-area sprinklers, pressurized water tanks. None of that is here. The usual layers of what the industry calls engineered safety are superfluous. Suppose a coolant pipe blows, a pressure valve sticks, terrorists knock the top off the reactor vessel, an operator goes postal and yanks the control rods that regulate the nuclear chain reaction - no radioactive nightmare. This reactor is meltdown-proof.

Zhang Zuoyi, the project’s 42-year-old director, explains why. The key trick is a phenomenon known as Doppler broadening - the hotter atoms get, the more they spread apart, making it harder for an incoming neutron to strike a nucleus. In the dense core of a conventional reactor, the effect is marginal. But HTR-10’s carefully designed geometry, low fuel density, and small size make for a very different story. In the event of a catastrophic cooling-system failure, instead of skyrocketing into a bad movie plot, the core temperature climbs to only about 1,600 degrees Celsius - comfortably below the balls’ 2,000-plus-degree melting point - and then falls. This temperature ceiling makes HTR-10 what engineers privately call walk-away safe. As in, you can walk away from any situation and go have a pizza.

“In a conventional reactor emergency, you have only seconds to make the right decision,” Zhang notes. “With HTR-10, it’s days, even weeks - as much time as we could ever need to fix a problem.”

This unusual margin of safety isn’t merely theoretical. INET’s engineers have already done what would be unthinkable in a conventional reactor: switched off HTR-10’s helium coolant and let the reactor cool down all by itself. Indeed, Zhang plans a show-stopping repeat performance at an international conference of reactor physicists in Beijing in September. “We think our kind of test may be required in the market someday,” he adds.

Today’s nuclear power plants are the fruits of a decision tree rooted in the earliest days of the atomic age. In 1943, a Manhattan Project team led by Enrico Fermi sustained the first man-made nuclear chain reaction in a pile of uranium blocks at the University of Chicago’s Metallurgical Lab. A chemist named Farrington Daniels joined the effort a short time later. But Daniels wasn’t interested in bombs. His focus was on a notion that had been circulating among physicists since the late 1930s: harnessing atomic power for cheap, clean electricity. He proposed a reactor containing enriched uranium “pebbles” - a term borrowed from chemistry - and using gaseous helium to transfer energy to a generator.

The Daniels pile, as the concept was called, was taken seriously enough that Oak Ridge National Laboratory commissioned Monsanto to design a working version in 1945. Before it could be built, though, a bright Annapolis graduate named Hyman Rickover “sailed in with the Navy,” as Daniels later put it, and the competing idea of building a rod-fueled, water-cooled reactor to power submarines. With US Navy money backing the new design, the pebble bed fell by the wayside, and Daniels returned to the University of Wisconsin. By the time of his death in 1972, he was known as a pioneer of - irony alert - solar power. Indeed, the International Solar Energy Society’s biennial award bears his name.

By the mid-1950s, with President Eisenhower preaching “atoms for peace” before the United Nations, civilian nuclear power was squarely on the table. The newly created General Atomics division of General Dynamics assembled 40 top nuclear scientists to spend the summer of 1956 brainstorming reactor designs. The leading light was Edward Teller, godfather of the H-bomb, and his message to the group was prophetic. For people to accept nuclear power, he argued, reactors must be “inherently safe.” He even proposed a practical test: If you couldn’t pull out every control rod without causing a meltdown, the design was inadequate.

But Teller’s advice was ignored in the rush to beat the Russians to meter-free electricity. Instead of pursuing inherent safety, the nascent civilian nuclear industry followed Rickover into fuel rods, water cooling, and ever more layers of protection against the hazards of radioactive steam emissions and runaway chain reaction. To try to amortize the cost of all that backup, plants ballooned, tripling in average size in less than a decade and contributing to a crippling financial crunch in the mid-’70s. Finally, partial meltdowns at Three Mile Island in 1979 and Chernobyl in 1986 pulled the plug on reactor construction in most of the world.

Even where the pebble-bed concept took root, the industry’s woes conspired against it. In Germany, a charismatic physicist named Rudolf Schulten picked up the idea and by 1985 a full-scale prototype was online - too large, in fact, to meet Teller’s inherent safety test. Barely a year later, with Chernobyl’s fallout raining over Europe, a minor malfunction at the German reactor set off nightmare headlines. Before long, the plant was mothballed.

The twin disasters in Pennsylvania and Ukraine proved Teller’s point and inverted his hopeful formulation: The Union of Concerned Scientists pronounced nuclear power “inherently dangerous.” The industry, already staggered by overbuilding and runaway budgets, ground to a halt. The newest of the 104 reactors operating in the US today was greenlighted in 1979. And there our story might have ended, except

Even as the nuclear establishment was putting all its efforts into avoiding the klieg lights, scientists in two faraway places were carrying the torch for a better reactor. One was South Africa, where in the mid-1990s the national utility company quietly licensed Germany’s cast-off pebble-bed design and set about trying to raise the necessary funds. The other was China, where the Tsinghua team pursued a Nike strategy: Just do it.

Frank Wu’s glass-walled ninth-floor office at Innovation Plaza offers a commanding view of Tsinghua University’s leafy campus. That’s no accident: The university co-owns this complex of gleaming silver towers, designed as a magnet for high tech startups. Likewise Wu’s company, Chinergy, is a 50-50 joint venture between Tsinghua’s Institute for Nuclear and New Energy Technology and the state-owned China Nuclear Engineering Group.

“I just had a call from a mayor in one of the provinces,” says Wu, who came on board as CEO after a decade spent running financial services companies in the US (where he adopted the English first name). “He asked me, ‘How much do we have to pay to get one of those things here?’”

If Wu’s pebble-bed “thing” is, well, hot, it’s because Chinergy’s product is tailor-made for the world’s fastest-growing energy market: a modular design that snaps together like Legos. Despite some attempts at standardization, the latest generation of big nukes are still custom-built onsite. By contrast, production versions of INET’s reactor will be barely a fifth their size and power, and built from standardized components that can be mass-produced, shipped by road or rail, and assembled quickly. Moreover, multiple reactors can be daisy-chained around one or more turbines, all monitored from a single control room. In other words, Tsinghua’s power plants can do the two things that matter most amid China’s explosive growth: get where they’re needed and get big, fast.

Wu and his backers aim to have a full-scale 200-megawatt version of HTR-10 by the end of the decade. They’ve already persuaded Huaneng Power International - one of China’s five big privatized utilities, listed on the NYSE and chaired by the son of former premier Li Peng - to pick up half of the estimated $300 million tab. Concrete is scheduled to be poured in spring 2007.

By the usual glacial standards, that timeline is nuts for a reactor still on the drawing board. South Africa’s pebble-bed group has been working on plans for a demonstration unit near Cape Town since 1993. But with an estimated $1 billion budget and local environmentalists on the warpath, the project remains stuck where it’s been for nearly a decade: five to 10 years from completion.

Five to 10 years ago, a lot of today’s China was little more than blueprints. And Wu, who likes to tell visiting Americans how one of his previous companies beat Sun Microsystems for the contract to wire West Point, has distinct advantages. The INET team, some of whose members studied with Schulten in Germany, has been prototyping pebble-bed designs since the mid-1980s. Also courtesy of the Germans, they have the best equipment in the world for what is probably the stickiest technical problem: fabrication of fuel balls in quantities that could quickly grow to millions.

By the time Chinergy’s pilot plant is up and running, it’s likely that the 30 reactors the government has planned for 2020 will already be under way. By then, however, China’s grid is expected to be market-driven, and companies like Huaneng will have a free hand to put plants where they’re needed and charge whatever the market will bear. Chinergy’s strategy is tailored for this new environment. Power companies operating in regions making the transition from rural to industrial to urban will need to start small, but may suddenly find themselves struggling to meet unexpected demand. That’s where the modular concept comes into play: Wu plans to sell power modules - 200-megawatt reactors plus ancillary gear - one at a time, if necessary. Growing utilities will be able to add modules as needed, ultimately reaching the gigawatt range where conventional reactors now reign. Such installations will be affordable to start - and they’ll become cheaper to operate as they grow, thanks to economies of scale in everything from security and technicians to fuel supply.

Too good to be true? Not according to Andrew Kadak, who teaches nuclear engineering at MIT (including a course titled “Colossal Failures in Engineering”). Kadak is a big-nuke guy by background. From 1989 to 1997, he was CEO of Yankee Atomic Electric, which ran - and ultimately closed - the ’60s-vintage plant in Rowe, Massachusetts. Now he’s helping INET refine its fuel ball technology and working with the US Department of Energy to build a high-temperature gas-cooled reactor at the Idaho National Engineering and Environmental Research Lab.

“The industry has been focused on water-cooled reactors that require complicated safety systems,” Kadak says. “The Chinese aren’t constrained by that history. They’re showing that there’s another way that’s simpler and safer. The big question is whether the economics will pay off.”

In May, British eminence green James Lovelock, creator of the Gaia hypothesis that Earth is a single self-regulating organism, published an impassioned plea to phase out fossil fuels in London’s The Independent. Nuclear power, he argued, is the last, best hope for averting climatic catastrophe:

“Opposition to nuclear energy is based on irrational fear fed by Hollywood-style fiction, the Green lobbies, and the media. … Even if they were right about its dangers - and they are not - its worldwide use as our main source of energy would pose an insignificant threat compared with the dangers of intolerable and lethal heat waves and sea levels rising to drown every coastal city of the world. We have no time to experiment with visionary energy sources; civilization is in imminent danger and has to use nuclear, the one safe, available energy source, now, or suffer the pain soon to be inflicted by our outraged planet.”

Coming to terms with nuclear energy is only a first step. To power a billion cars, there’s no practical alternative to hydrogen. But it will take huge quantities of energy to extract hydrogen from water and hydrocarbons, and the best ways scientists have found to do that require high temperatures, up to 1,000 degrees Celsius. In other words, there’s another way of looking at INET’s high-temperature reactor and its potential offspring: They’re hydrogen machines.

For exactly that reason, the DOE, along with similar agencies in Japan and Europe, is looking intently at high-temperature reactor designs. Tsinghua’s researchers are in contact with the major players, but they’re also starting their own project, focused on what many believe is the most promising means of generating hydrogen: thermochemical water splitting. Researchers at Sandia National Laboratories believe efficiency could top 60 percent - twice that of low-temperature methods. INET plans to begin researching hydrogen production by 2006.

In that way, China’s nuclear renaissance could feed the hydrogen revolution, enabling the country to leapfrog the fossil-fueled West into a new age of clean energy. Why worry about foreign fuel supplies when you can have safe nukes rolling off your own assembly lines? Why invoke costly international antipollution protocols when you can have motor vehicles that spout only water vapor from their tail pipes? Why debate least-bad alternatives when you have the political and economic muscle to engineer the dream?

The scale is vast, but so are China’s ambitions. Gentlemen, start your reactors.

Contributing editor Spencer Reiss () interviewed Bjørn Lomborg in Wired 12.06.