The Asan Institute just held its large North Korean conclave, and one of the more interesting things to come out of it was new work by Scott Kemp and Joshua Pollack on North Korea’s enrichment capabilities. Their paper is not out yet, but has already received wide coverage in Japan, South Korea and the US, including in the New York Times; Pollack was kind enough to share his notes.
Back in August, David Albright and Robert Avagyan at the Institute for Science and International Security noted the expansion of a building in the fuel fabrication complex that houses the gas centrifuge plant for uranium enrichment. They estimated that the area is now covered by an extended roof that is roughly twice the size of the previous one.
Kemp and Pollack go a step further by considering the technological capabilities that might make an indigenous enrichment program viable. North Korea was detected importing large amounts of centrifuge-related parts and materials for about four years from 1999 through 2003. This information was key to the onset of the second nuclear crisis. Recall at that point that the concerns did not center on the plutonium track—Yongbyon had been shut down under the Agreed Framework—but rather on a clandestine HEU program with blueprints and technology from Pakistan and other sources.
Kemp and Pollack focus on six technologies that would be crucial for such a program:
- Uranium hexafluoride (UF6);
- Vacuum pumps to remove air from centrifuges prior to feeding UF6 gas into them;
- Ring magnets used as top bearings for the centrifuges;
- Frequency inverters, allowing control of the speed of the centrifuges;
- Maraging steel;
- Computer numeric controlled flow-forming machines, specialized lathes required for making symmetrical shapes such as cylindrical maraging steel rotors. (This is an area where evidence seems particularly clear as such machines are often on view during leadership visits. Pollack has even posted numerous pictures at Arms Control Wonk.)
In the first four areas, Kemp and Pollack found various sources–scientific reports and even patent awards to North Koreans describing domestic production. Evidence in other areas, such as maraging steel, is more suggestive but nonetheless important enough to set off an alarm. They estimate that by 2009, research in these areas had been completed even if it is hard to say whether production is up and running.
The implications are substantial, and tie back to the Chinese sanctions list that we commented on earlier in the week. Kemp and Pollack draw two important conclusions about sanctions efforts:
- We will not be able to even glean what is going on by monitoring technology trade, no matter how good our forensics;
- Export controls, sanctions and interdiction won’t necessarily work to slow down or stop the program.
They also note that these activities could be going on anywhere in the country, since they use little electricity or heat and can easily be mistaken for routine economic activity.
Kemp and Pollack might be a little pessimistic about the value of limiting dual use transfers; the Chinese are clearly concerned and presumably believe that they can get some traction on the program by monitoring dual-use technologies more closely. But the problem is real, as we know from the fact that North Korea has deposits of uranium. It is harder to stop indigenous activities than those that rely on an international source of supply.