It was probably the worst first nuclear test in history, say weapons experts who have carried out preliminary analyses of the seismological data. But the North Koreans will have learned a great deal from the nuclear blast they set off on 9 October. And although experts' estimates of the country's technological expertise aren't much altered, North Korea's intent is now unmistakable.

South Koreans express their outrage at North Korea's nuclear test (see seismogram, inset). Credit: AHN YOUNG-JOON/AP

Seismographs around the world registered shock waves from the event, which the North Korean press agency claimed originated from an underground nuclear test (see Map). The event, measured at magnitude 4.2 by the US Geological Survey, was probably too large to be caused by conventional explosives, but too small to be a completely successful nuclear test — unless, as some experts believe, North Korea was testing a more sophisticated device than previously thought.

As Nature went to press, there was disagreement over the size of the explosion (see 'The nuclear detectives'), although most governments and experts agree that the seismograph data suggest a device equivalent to around 500 tonnes of TNT. If that is the case, the test was probably what nuclear experts term a 'fizzle' — a device that failed to create a full chain reaction in the plutonium and so produced a smaller than planned nuclear explosion. Weapons experts assumed that North Korea had one or two nuclear devices, but the fizzle suggests that their programme might not be as sophisticated as thought. “It looks like it was the least successful first test ever,” says James Acton, nuclear-weapons expert at London think-tank VERTIC.

“They're an embarrassment to the nuclear club,” adds Jeffrey Lewis, executive director of the Managing the Atom project at Harvard's Belfer Center for Science and International Affairs. North Korea's bomb is thought to have been a relatively simple atomic device, working entirely on nuclear fission, in which the margin for failure can be reduced by adding extra mat-erial and conventional explosives. Lewis says he is unaware of a first test by any other country leading to such a failure. “Frankly I'm baffled,” he says. “This is supposed to be really easy.”

Experts guess that the North Korean device consisted of a series of conventional explosives that were detonated to compress a sphere of plutonium until it reached critical mass, at which point a runaway fission reaction causes a massive release of energy. But even with such a simple device, there can be problems. If more than a few percent of the fissile material consists of plutonium isotopes other than plutonium-239, the chain reaction can begin before the sphere reaches maximum compression. The neutron generator, which triggers the chain reaction, might have been triggered at slightly the wrong time. It's also possible that in an effort to conserve their limited supply of plutonium, the North Koreans put too little in their device. They might also have used too few conventional explosives or these did not work properly, leading to an asymmetrical implosion. “It's not that difficult, but a lot of things can go wrong,” says William Press, a senior fellow at Los Alamos National Laboratory (LANL) in New Mexico, and a member of the JASONs, a group of scientists that provides advice to the US government on nuclear issues.

The much more worrying possibility is that North Korea was testing a more sophisticated weapon. John Pike, director of GlobalSecurity.org, a Washington-based security website, believes it may have been trying to test the first stage, or 'primary', of a hydrogen bomb. In an H-bomb, the fission explosion is designed to trigger a much larger fusion reaction. Lewis discounts the idea: “I don't know of a country that's gone straight to the hydrogen bomb,” he says. Press adds that even a primary must have a yield higher than that detected on 9 October: “By far the most likely design is an unboosted, spherical-implosion, plutonium bomb,” he says.

But Pike points to evidence suggesting that North Korea might actually have carried out its first nuclear test in Pakistan in 1998, when the two states' nuclear programmes were closely entwined. Moving from a first nuclear test to a primary might not be that difficult, he says — for example, the Israelis started building a hydrogen bomb in the 1980s, after just one nuclear test in 1979.

Siegfried Hecker, former director of LANL, visited nuclear scientists in North Korea in 2004 and 2005. He too warns against underestimat-ing them. “I would have to assume they have the know-how to make a bomb,” he says. “We just don't know exactly what they were trying to do.” Hecker thinks it's unlikely the North Koreans were testing a primary. But he says the size of the explosion suggests that they may have been trying to miniaturize the bomb into something that could be carried on a ballistic missile. “That ups the difficulty significantly,” he says.

Even if the explosion was a partially failed test of a basic device, North Korean researchers will have learnt a great deal from it, says Acton. Comparing the results of the test with previous computer simulations will allow the North Koreans to assess the accuracy of their models. Even if the test reveals a seriously flawed design, Acton says that at an “absolute upper limit” the country is no more than two years away from developing a functioning weapon. “If it was a fizzle, then they could easily learn enough from it to be in business for their next test,” agrees Press.

According to most estimates, North Korea has enough material for between six and eight nuclear weapons. It may be hoping to mount those weapons on missiles such as the TaePodong-2, although the bomb tested on 9 October was probably far too big. The TaePodong-2 itself failed and fell into the Sea of Japan after just a 42-second flight when tested on 4 July.

Ultimately, the failure or otherwise of the North Korean weapon may have little effect on the response of nations such as the United States and China, according to Michael Levi, a physicist at the independent Council of Foreign Relations in New York: “From a political standpoint it makes essentially no difference whether it was a fizzle,” he says. “What matters is their willingness to test.”