By Hui Zhang
Senior Research Associate, Project on Managing the Atom, Belfer Center for Science and International Affairs, Harvard Kennedy School
On February 12, 2013, North Korea conducted its third nuclear test, and a number of seismic stations around the world detected the event. Before and after the test, there has been much anticipation in the media that we might learn through off-site sampling analysis whether North Korea exploded plutonium bomb like it did in the in 2006 and 2009 tests, or a new device using highly enriched uranium (HEU).
Indeed, many experts have suggested that the test was an HEU explosion. North Korea has only a small supply of plutonium—material that it had stopped producing by 2008—and had more recently demonstrated an operational capability to enrich uranium, which would support a much larger arsenal of weapons given North Korea’s huge deposits of natural uranium. Without a doubt, confirming the type of nuclear weapon it tested is highly desirable. However, the seismic signals are useless in this regard. The question is, then, can the off-site environmental sampling analysis distinguish a plutonium explosion from a HEU explosion?
An underground nuclear test produces noble gases as fission products; in particular, radioactive xenon isotopes are often vented above ground and detected in air samples. However, if the underground containment is effective, it is possible that not enough gases would leak out to be detected. Indeed, there are no reports documenting any releases from the 2009 North Korean test, while in the 2006 test, there was a release of gas detected.
The International Monitoring System (IMS) established by the Comprehensive Nuclear-Test-Ban Treaty has been mandated to establish a worldwide network of detector systems capable of detecting the four radioxenons: Xe-131m, Xe-133m, Xe-133, and Xe-135. An air sampling of xenon would be dependent on a number of factors including, e.g. how much xenon isotopes would be released into atmosphere (related to total inventory, leak rate, etc); meteorology conditions and dilutions (atmospheric dispersion), timely access to the plume (e.g. decay factor and plume location), the background radionuclide concentration, and the minimum detectable concentration.
Since the amount of the xenon that is released by an underground test is very uncertain, any clue to the nature of the fissile material would have to come from looking at isotope ratios. These ratios are different for uranium-235 and plutonium-239 fissions. Based on an analysis I did with colleagues in 2007, we concluded that if one were able to detect and analyze the radioactive xenon within the first few hours after an explosion, it would be possible to distinguish between the xenon from plutonium and uranium explosions. If the air samples were taken two days after the test, however, such determinations would be very difficult. Beyond two days after the explosion, there is no way to detect from sampling of radioactive isotopes whether the test used plutonium or HEU.
While the off-site sampling would be very difficult to confirm a plutonium or uranium explosion(s), the radioactive xenon isotope ratios would be able to confirm whether the seismic event was caused by a nuclear explosion. For instance, analysts would use the activity ratio of xe-135 to xe133 to identify a nuclear test, although they might need to make sure that there was no leakage from nuclear reactors in the vicinity of the test, or if there was a recent leakage, there was no contribution to the samples. (See this paper).
It should be noted that if the unfissioned materials (e.g. plutonium or HEU debris) are vented and collected, it would be easy to determine whether the explosion was from plutonium or HEU device. However, unlike the noble gases, the unfissioned materials would not easily escape into the atmosphere in a normal underground test. Even if some plutonium or HEU debris is vented, in North Korea’s case, it would be hard for that material to be transported over long distances and across borders for off-site air sampling.
Finally, another interesting question is how big an explosion occurred. What was the likely yield of the weapon that was tested? Unfortunately, using off-site air sampling would not help to be able to estimate the yield. In the case of North Korea, we will need to depend on the records of seismic detection.