There is rampant speculation that the recent seismic event in North Korea was a nuke test that produced much less energy than planned – in other words, a “dud”. Without rehashing commentary both official and unofficial, suffice that it is probably a safe bet that North Korea actually did test a nuclear device and that the yield ranged from roughly 0.5 to 2.0 kilotons (only the Russians have offered a higher estimate, which we can take as an outlier).
North Korea has repeatedly shown itself to be technologically backward – for example, after eight years of trying, it remains unable to launch an ICBM. Arguably, 60-year-old nuclear technology is nearly as challenging as 50-year-old rocket technology. Consequently, it is extremely unlikely that North Korea would attempt anything more advanced than a 1st generation plutonium bomb along the lines of the “Fat Man” that was first tested by the U.S. in July 1945. A 1st generation device should have a nominal yield of around 20 kilotons – hence, the “dud” speculation.
In a previous article, I stated that a highly enriched uranium bomb of the kind being pursued by Iran was so simple to make that a “trained orangutan” could build it. Only trained homo sapiens, however, can construct a plutonium bomb because there will always be some amount of Pu-240 present along with the Pu-239. This is because a major decay route for Pu-240 is spontaneous fission, which is attended by the release of neutrons. The higher the neutron flux, the faster it is necessary to assemble the critical mass in order to avoid what is called a “pre-initiation”. There is a good discussion here. As the article notes:
[I]f pre-initiation occurs at the worst possible moment (when the material first becomes compressed enough to sustain a chain reaction) the explosive yield of even a relatively simple first-generation nuclear device would be of the order of one or a few kilotons.
A U-235 bomb can assemble its critical mass more slowly than a Pu-239 bomb – this is why the former can utilize a simple gun design while the latter requires the more sophisticated implosion design.
One way a plutonium bomb could “fizzle” would be for some failure in the timing mechanism or the shaped charges to result in a non-uniform compression of the core. But, even if these elements worked flawlessly, a fizzle could still take place if the core wasn’t compressed rapidly enough to compensate for the Pu-240 concentration.
Which is the more likely problem? Making shaped charges is simple. Whole categories of military weaponry (such as anti-tank grenades and shells) rely on shaped charges and the North Koreans presumably make such weapons domestically. Properly timing the firing of the charges is also simple. The North Koreans could have had a quality control failure with their bomb. However, it would have been easy to conduct multiple tests of the design and the components before they were ever wrapped around a plutonium core; therefore, this avenue of failure seems remote. On the other hand, if there were a problem with the production reactor or with how it was operated, the plutonium could have had a greater percentage of Pu-240 than the firing system could accommodate without running the risk of pre-initiation. Determining the proper reactor geometry and the necessary fuel “burnup” for a particular percentage of Pu-239 is a non-trivial undertaking though every other nuclear power has managed it without apparent difficulty. Given their limited resources, the North Koreans may have tried to maximize their plutonium output, which would have necessarily meant a higher burnup. This, in turn, would have increased Pu-240. The closer the Pu-240 concentration was to the limits of their bomb design, the more likely it was for a miscalculation resulting in a fizzle.
If North Korea’s bomb was a fizzle and excessive Pu-240 was the culprit then fixing the problem may be much more difficult than has heretofore been assumed. It would mean that the existing stocks of plutonium might be unsuitable for a reliable weapon without doing a complete redesign of the firing mechanism. Running a new batch of plutonium at much lower burnup in order to produce a higher percentage of Pu-239 would be the alternative. In either case, instead of a month or two before North Korea achieves a successful bomb it could easily be a year or two – maybe even more.
Update 14-Oct: Some corroboration that North Korea detonated a nuke has now been reported.