As the essays published in this Roundtable so far have established, converting research reactors and fission radioisotope production facilities to the use of low-enriched uranium (LEU) can involve a number of difficult challenges. Some are purely technical, some are financial, and others, as is often the case with medical isotopes, are related to safety and medical regulations.
Experience has shown that technical challenges to conversion at facilities producing molybdenum 99 can usually be overcome. Facilities using targets based on highly enriched uranium (HEU) often have the independent expertise to develop new LEU-based processing methods or to adapt their existing HEU processes. Where sufficient expertise is lacking, projects like the Global Threat Reduction Initiative may provide assistance that enables conversion to proceed.
In some cases, financial challenges are larger than technical ones. These challenges are best overcome when every stakeholder in the fission radioisotope supply chain becomes convinced that conversion is highly important to global security. Convincing everyone of this can be difficult in itself, as the supply chain contains numerous links: manufacturers of uranium targets, research reactors themselves, processors of molybdenum 99, manufacturers and distributors of technetium 99 generators, and nuclear medicine centers. But gaining the cooperation of all these stakeholders — as well as that of local governments and international minimization initiatives — seems the most satisfactory way to finance the costs of conversion.
Alexandr Vurim's first Roundtable essay describes a situation in which regulatory issues are paramount to the success of a research reactor's conversion process; indeed, regulation plays a large role in many conversion scenarios. Gaining regulatory approval for LEU conversion can take a long time (and it can take even longer for new facilities). Also, partly for regulatory reasons, facilities undergoing conversion may need to maintain parallel production processes for a while — the existing HEU process to produce medical radioisotopes until conversion is complete, and an LEU process that, as it is being refined, provides a demonstration for regulatory authorities.
So far I have mostly discussed facilities that are already in operation. But I believe it is clearly desirable that newcomers to fission radioisotope production use LEU targets from the very beginning, even if HEU is produced locally. Some new producers will be able to develop LEU-based processing methods on their own; if not, they can draw on programs such as a project of the International Atomic Energy Agency known as Small-Scale Indigenous Production of [Molybdenum 99] Using LEU Targets or Neutron Activation, which has been useful for some countries embarking on small-scale production of molybdenum 99. Or, as Australia and Egypt have done, nations may import LEU-based technology directly from a country like Argentina.
Defining the cut-off. My colleague Charles Piani has asked why the cut-off between low-enriched and highly enriched uranium can't be increased to something greater than 20 percent uranium 235, and he suggests that 30 percent might be a more reasonable dividing line. Others have raised the same issue — after all, cut-offs are often somewhat arbitrary. And from the perspective of some research reactors, economic factors argue for a cut-off higher than 20 percent.
I believe, however, that the 20-percent limit makes sense. Alexander Glaser, a Princeton University professor and a member of the Bulletin's Science and Security Board, has argued, in effect, that the existing definition represents a good compromise between competing nonproliferation imperatives. That is, when research reactors use uranium enriched to higher levels, uranium-based proliferation risks grow, in the sense that diversion or theft represents a greater security concern; but when facilities use uranium enriched to lower levels, plutonium-based proliferation risks grow (larger amounts of plutonium are produced when uranium enriched to lower levels is irradiated). I tend to agree with Glaser that the existing cut-off "represents a reasonable and even optimum choice as a conversion goal for research reactors" — and, I would add, for fission radioisotope production.