On 3 March 1997 I visited the new Sellafield MOX Plant (SMP) with Martin Forwood of Cumbrians Opposed to a Radioactive Environment and Nigel Chamberlain of the Campaign for Nuclear Disarmament.

The construction of the SMP is almost complete and parts of it are undergoing testing for start-up. Uranium start-up is due this year and MOX fuel fabrication is due to start in 1998. British Nuclear Fuels Limited (BNFL) has made many statements implying that it can fabricate weapons-grade plutonium from the United States and Russia into MOX fuel for the Lead Test Assemblies (LTAs). These LTAs would be used to test reactor operation with MOX made from weapons grade plutonium.

You can imagine my surprise when the Assistant Director of MOX for BNFL, Arthur Roberts, said the following in response to a question I posed to him about the capability of the SMP to make weapon-grade plutonium into MOX:

I wrote down his statement while on the tour and checked it with him several times to make sure that there was no misunderstanding. He amended it slightly during the review and the quote above is what was approved by him. I released the statement to radio and television in Cumbria the next day, 4 March. Either Roberts or Simon Marshall, the BNFL Works Manager for MOX Fuels, reaffirmed it on television. (I saw the clip on the run and do not remember which one of them spoke for BNFL.)

The main problem appears to be that weapon-grade plutonium would not meet the criticality criteria that have been set forth in the "safety case' for which the plant has been designed. I understand informally from a source outside BNFL that the problem might be with the uranium-plutonium mixing equipment. This equipment (and/or other equipment) would have to be modified, or a new line built to mix weapon-grade plutonium with uranium. Evidently, this would take time. Thus, the best argument for using BNFL to make a substantial number of LTAs -- that the testing of MOX made from weapon-grade plutonium could begin relatively soon if existing MOX facilities are used to make LTAs -- does not appear to be valid for the SMP. Roberts and Marshall (who was also present during my visit) did not want to state which modifications would be required or how long it would take to make them. There are currently two MOX fuel fabrication lines in the SMP, one for plutonium dioxide from PWRs and one for that from BWRs.

Another surprising finding that came up during the tour related to materials accounting. SMP is not set up to routinely check the weight or isotopic composition of the plutonium that it receives as feed material. In case there is reason to be suspicious, the canister could be sent to facilities associated with THORP, which is next door. But BNFL does not plan to routinely verify the shipper's claims about plutonium content. Apparently, the rationale underlying this set up is that BNFL would be fabricating MOX from plutonium dioxide produced at THORP. Since this Plutonium would already be assayed at THORP, there would be no need to check its composition and weight again prior to feeding it into one of the two the MOX fuel fabrication lines.

The lack of materials accounting facilities to routinely verify the shipper's claims of plutonium content is a serious design defect from a proliferation standpoint. If BNFL is to fabricate weapon-grade plutonium into MOX, it must be able to verify the shipper's claims about the context of the cans. What if the isotopic composition of plutonium shipped from abroad was not that claimed by the shipper and some reactor-grade or fuel-grade plutonium was surreptitiously put in the place of weapon-grade plutonium?

It is unclear whether any of the other facilities in Europe would face problems similar to the SMP or whether the issues regarding criticality or materials accounting are specific to the particular MOX plant design used in SMP. Further checking needs to be done as to the capabilities of the other MOX fabrication facilities in Europe.

BNFL has not make its safety or non-proliferation analysis public. It is imperative that BNFL, Cogema, or Belgonucleaire make the safety analyses for these plants public so that all relevant issues, including criticality and materials accounting issues, can be adequately and publicly scrutinized. This must be done before these companies are given contracts to fabricate weapon-grade plutonium into MOX.