A Perspective on the U.S. Nuclear Fuel Cycle

There has been a resurgence of interest in the possibility of processing the US spent nuclear fuel, instead of burying it in a geologic repository. Accordingly, key topical findings from three relevant EPRI evaluations made in the 1990–1995 timeframe are recapped and updated to accommodate a few developments over the subsequent ten years. Views recently expressed by other US entities are discussed. Processing aspects thereby addressed include effects on waste disposal and on geologic repository capacity, impacts on the economics of the nuclear fuel cycle and of the overall nuclear power scenario, alternative dispositions of the plutonium separated by the processing, impacts on the structure of the perceived weapons proliferation risk, and challenges for the immediate future and for the current half-century. Currently, there is a statutory limit of 70,000 metric tons on the amount of nuclear waste materials that can be accepted at Yucca Mountain. The Environmental Impact Statement (EIS) for the project analyzed emplacement of up to 120,000 metric tons of nuclear waste products in the repository. Additional scientific analyses suggest significantly higher capacity could be achieved with changes in the repository configuration that use only geology that has already been characterized and do not deviate from existing design parameters. Conservatively assuming the repository capacity postulated in the EIS, the need date for a second repository is essentially deferrable until that determined by a potential new nuclear plant deployment program. A further increase in technical capacity of the first repository (and further and extensive delay to the need date for a second repository) is potentially achievable by processing the spent fuel to remove the plutonium (and at least the americium too), provided the plutonium and the americium are then comprehensively burnt. The burning of some of the isotopes involved would need fast reactors (discounting for now a small possibility that one of several recently postulated alternatives will prove superior overall). However, adoption of processing would carry a substantial cost burden and reliability of the few demonstration fast reactors built to-date has been poor. Trends and developments could remove these obstacles to the processing scenario, possibly before major decisions on a second repository become necessary, which need not be until mid-century at the earliest. Pending the outcomes of these long-term trends and developments, economics and reliability encourage us to stay with non-processing for the near term at least. Besides completing the Yucca Mountain program, the two biggest and inter-related fuel-cycle needs today are for a nationwide consensus on which processing technology offers the optimum mix of economic competitiveness and proliferation resistance and for a sustained effort to negotiate greater international cooperation and safeguards. Equally likely to control the readiness schedule is development/demonstration of an acceptable, reliable and affordable fast reactor.