Uranium Enrichment Plant Turns into a Big Waste Dump Nuke Info Tokyo No. 140

A report released in November by Professor Siegfried S. Hecker of Stanford University revealing the existence of a uranium enrichment facility in North Korea caused an international uproar. Though not a major focus of media attention, the report contained one line of particular relevance to Japan. According to Hecker, the chief process engineer at the North Korean facility claimed that the components were “modeled after the centrifuges at Almelo [in The Netherlands] and Rokkasho-mura.” If this were true it would have serious implications for Japan’s non-proliferation credibility. JNFL immediately denied that its technology had been leaked to North Korea. Perhaps the North Korean engineer simply meant that the enrichment technology used by Japan and North Korea was in both cases based on the same URENCO technology. However, even if the North Koreans did somehow manage to get their hands on blueprints from the Rokkasho Uranium Enrichment Plan, considering the plant’s unfortunate history, they might come to regret their choice of target for technology theft.

The Rokkasho Uranium Enrichment Plant was the first of the nuclear fuel cycle facilities built in Rokkasho. It was supposed to eventually have a capacity of 1,500 ton-SWU per year¹. The first cascade (RE-1A) began operating in December 1992 with a capacity of 150 ton-SWU per year. Gradually more 150 ton-SWU cascades were added. The seventh and last cascade (RE-2C) began operating in October 1998, bringing the total capacity to 1050 ton-SWU per year. Thereafter, no further cascades were added, because one after the other malfunctioning centrifuges had to be stopped.

The current status of the plant is shown in the table below.

The amount of the fissile uranium-235 isotope in natural uranium is only about 0.7%. Most of the uranium in natural uranium is uranium-238, which does not readily fission. The basic principle behind centrifuges used for uranium enrichment is that when uranium hexafluoride (UF6) gas is fed into a cylindrical rotor rotating at immense speed the heavier U-238 isotope moves to the periphery, while the lighter U-235 isotope concentrates in the center of the stream. Enriched uranium for use in nuclear power plants is produced by repeating this process in cascades of many centrifuges connected in sequence to raise the U-235 concentration to between three and five percent (“low enriched uranium”). In the Rokkasho Uranium Enrichment Plant uranium adhered to the rotors. This caused them to vibrate until the centrifuges eventually broke down.

Centrifuges are dual-purpose machines capable of producing low enriched uranium to fuel nuclear reactors to generate electricity, but also capable of producing highly enriched uranium (over 90% U-235) for nuclear weapons. The concern about North Korea’s development of uranium enrichment technology stems from this fact. This is also the reason why details of the centrifuges in the Rokkasho Uranium Enrichment Plant, including their size, the mechanisms involved and the number of centrifuges in the plant, are not publicly available. On nuclear security grounds this is all classified as sensitive information. Photos of cascades of cylinders in uranium enrichment facilities are publicly available, but they only show the outer containers. The inner workings of the centrifuges are not shown.

JNFL only ever published the number of centrifuges that had stopped operating. When the number of centrifuges that had ceased operating in a cascade reached the point where the cascade was barely able to continue enriching uranium, the whole cascade was closed down. As can be seen from the table, in April 2000 the RE-1A cascade was first to shut down. At the time, 4,244 centrifuges in this cascade had ceased to operate. Each of the seven cascades was forced to shut down after around ten years of operation. RE-2B was the last remaining operational cascade, but by 2007 part of this cascade was already out of operation and by 2008 3,000 centrifuges had stopped rotating. However, if the whole cascade were shut down this would in effect have meant that the whole plant was out of service. To save face, JNFL kept this last cascade going, but it was operating in form only. The RE-2B cascade was finally shut down in December 2010, so now there are no centrifuges rotating at the Rokkasho Uranium Enrichment Plant.

The physical reason why the plant had to be shut down altogether was the breakdown of large numbers of centrifuges, but the underlying cause was the failure of technological development carried out within the context of a government, bureaucracy and industry structure where no one took responsibility. Former Japan Nuclear Fuel Service (predecessor of JNFL) President Masatoshi Toyoda effectively admitted this in a memo entitled “Uranium Enrichment – the Void of 20 Lost Years”. Toyoda, who in 1989 was a member of the Japan Atomic Energy Commission’s “Uranium Enrichment Technology Expert Commission”, said, “The demonstration testing of the centrifuge developed by the Power Reactor and Nuclear Fuel Development Corporation (PNC) was insufficient. On the grounds that the product was defective and uneconomic, I opposed transfer of development to the private sector, but was over-ruled by the Atomic Energy Commission and the nuclear plant makers.”

Although JNFL refutes this claim, it plans to start again from scratch by replacing the existing centrifuges with a new design. The existing centrifuges have a metal rotor. They were designed by PNC (now Japan Atomic Energy Agency) along with Japan’s three nuclear plant makers, Hitachi, Toshiba and Mitsubishi Heavy Industries. However major problems, including accretion of uranium and cracking of the base, were identified with this and a revised design. Now JNFL plans to replace the old centrifuges with a new type of centrifuge. The new centrifuges were developed by Sumitomo Electric Industries and IHI Corporation and have a composite carbon fiber rotor. JNFL intends to install them in stages between 2011 and 2020 and bring the capacity of the plant up to the originally planned capacity of 1500 ton-SWU per year.

However, the work to remove the uranium that has adhered to the existing centrifuges is proving to be time consuming, so it is very likely that the schedule for introducing the new centrifuges will be delayed. Even if the plant achieves its design capacity of 1,500 ton-SWU per year without problems, it will still only be able to supply a quarter of Japan’s enriched uranium demand. Furthermore, the cost will be much higher than procuring enrichment services abroad, so it will never be an economic enterprise.

By Masako Sawai and Philip White

1. SWU = Separative Work Unit.
“Separative work” represents the amount of separation done by an enrichment process. It is a function of the concentrations of the feedstock, the enriched output, and the depleted tailings. It is expressed in units which are so calculated as to be proportional to the total input (energy / machine operation time) and to the mass processed. (Wikipedia)