The most energetic laser system in the world, designed to produce nuclear fusion–the same reaction that powers the sun–is up and running. Within two to three years, scientists expect to be creating fusion reactions that release more energy than it takes to produce them. If they’re successful, it will be the first time this has been done in a controlled way–in a lab rather than a nuclear bomb, that is–and could eventually lead to fusion power plants.

The National Ignition Facility (NIF), at the U.S. Department of Energy’s Lawrence Livermore National Laboratory (LLNL), comprises 192 lasers that fire simultaneously at precisely the same point in space: a sphere of fuel two millimeters in diameter. They are designed to deliver 1.8 megajoules of energy in a few billionths of a second. That’s enough to compress the fuel to a speck 50 micrometers across and heat it up to three million degrees Celsius. The lasers, which were fired together for the first time last month, have so far produced pulses of 1.1 megajoules.

“Depending on how you count it, it’s between 60 and 100 times more energetic than any laser system that’s ever been built,” says Edward Moses, the principle associate director for NIF and Photon Science at LLNL. Eventually, the fusion reactions produced by each pulse are expected to generate at least 10 times the energy delivered by the lasers, a significant net gain that could be useful for generating power.

The $3.5 billion facility, which has been in development for 15 years, was funded primarily as a way to better understand nuclear weapons, after a ban on testing in the 1990s. NIF will produce tiny thermonuclear explosions that give scientists insight into what happens when a nuclear bomb goes off. That data can, in turn, be used to verify computer simulations that help determine whether the United States’ nuclear stockpile will continue to work as the weapons age. The data could also provide insight into the processes that power the sun and other stars, and answer other scientific questions. Finally, NIF could serve as a proof-of-concept design for a fusion power plant.

To generate fusion, 192 laser beams are generated, amplified, converted from infrared to ultraviolet light, and then aimed at a small gold canister the size of a pencil eraser. Inside that canister is a sphere containing the fuel: two isotopes of hydrogen called deuterium and tritium. The lasers are positioned all around the sphere to create the temperatures and pressures needed to ignite a fusion reaction. If all goes as planned, some of the hydrogen atoms should fuse, producing helium and releasing energy. This should, in turn, cause more fusion reactions until the fuel runs out. The whole process will take just a few billionths of a second.

Researchers have created fusion in the lab before, but their experiments required more energy than they produced. For example, a system at Department of Energy’s Sandia National Laboratories, called the Z machine, uses electricity instead of lasers to compress hydrogen isotopes and produce fusion. A significantly larger version of the Z machine would be needed to generate more energy than it uses. Moses says that the NIF could reach fusion “gain” in just two to three years, well ahead of the more famous ITER fusion project in Cadarache, France, which likely won’t be operational until 2018. “This has been a grand challenge for a long time, so hubris is the worst thing,” Moses says. “But we think we see our way through it. When we get a [fusion] burn in 2010 or 2011, we’ll be in a very exciting place. I think the world will wake up to the possibilities.”

Moses is referring chiefly to the possibilities offered by a fusion power plant. Fusion poses no danger of nuclear proliferation, produces little waste, and uses abundant sources of fuel, so it could provide plenty of clean power for many thousands of years. Some say the fuel–hydrogen–is virtually unlimited, although proposed reactors will use tritium, a hydrogen isotope made from lithium, which is scarcer.

The current facility isn’t built to generate electricity. But Moses says that with the right funding, a power plant using fusion from a system like the one at NIF could be running in a decade. In contrast, power plants based on the Z machine at Sandia or the ITER system in France are decades away.

Other experts, however, are more skeptical. “If NIF is successfully, they’ll still be a very long way from turning this into a practical energy source,” says Ian Hutchinson, professor and head of nuclear science and engineering at MIT. For example, he says, a power plant would require the lasers to fire much more frequently than the NIF lasers–5 to 10 times a second, rather than once every couple of days, as is possible now. (Each burst would release energy equivalent to about five kilowatt-hours of electricity: by comparison, an average nuclear power plant generates 12.4 billion kilowatt hours a year, while an average house requires about 1,000 kilowatt-hours per month.)

In contrast, ITER will use magnetic confinement of hot plasma to produce fusion, a system that produces a continuous stream of energy that could be more suited to generating electricity than the very short bursts of energy produced by NIF, he says.

Whether or not it leads to fusion power plants, NIF is significant, says Stewart Prager, the director of the Department of Energy’s Plasma Physics Laboratory at Princeton University. The science it will make possible “cannot be done elsewhere,” he says.