In addition to its original purpose, of recovering uranium and plutonium, reprocessing also offers the possibility of:

(i) Controlling proliferation risks. The production and potential isolation of fissile materials which can be diverted from the civil fuel cycle for military purposes is an intrinsic risk in nuclear power. Although the fissile content of plutonium derived from high burnup fuel is not opti­mal, power reactor plutonium can nevertheless be used in a weapon,4 so the creation of plutonium is, by definition, a proliferation risk. While irradiated fuel is lethally radioactive and specialised facilities would be

needed to recover fissile material, the fission products will decay over a few hundred years to the point where plutonium could be easily recovered. This raises complex ethical issues and the recovery of plu­tonium for separate treatment, either as a waste or as a fuel, within a few years of production can be attractive for these reasons.

(ii) Reducing waste volumes. Well over 90% by mass of spent fuel is uranium. If a ‘‘once through’’ or ‘‘open’’ fuel cycle is adopted, the irradiated fuel will be packaged and disposed as waste. As a result, the volume of waste for disposal will be very substantial, and the associated costs will be high. For example, in some disposal concepts being considered for the UK, fuel elements containing 2-4 tonnes heavy metal (masses of nuclear materials are often expressed in tonnes heavy metal (tHM), i. e. equivalent mass of uranium or plutonium in the material) depending on fuel type and heat production could be packaged in a cast iron insert, then in a copper container between 2 and 5 m long and 0.9 m diameter, with 5 cm thick walls.5 By contrast, reprocessing spent fuel and conversion of the high level waste to glass will produce less than 100 kg (0.04 to 0.05 m3) of glass per tonne of uranium reprocessed, reducing the volume of highly radioactive material for disposal. Since high level waste and spent fuel are heat gen­erating wastes, they need to be widely spaced in a disposal facility to limit the heat load, so the volume of waste disposed has a large effect on the facility footprint and consequent cost. If there is only a limited volume of host rock, reducing waste volume may be very helpful. Finally, a disposal facility, for example the currently suspended Yucca Mountain facility in the USA, may be legally limited to a specific volume of waste, in which case volume reduction by removal of uranium may well be attractive.

(iii) Controlling high level waste radiotoxicity. The majority of the fission products in spent fuel have relatively short half lives, so that, at timescales longer than a few hundred years, the activity is dominated by relatively radiotoxic actinide elements (see Figure 2). If, in addition to conventional reprocessing to remove uranium and plutonium, a further separation of minor actinides (e. g. neptunium, americium and curium) from high level waste is carried out, the radiotoxicity of the waste can be reduced by several orders of magnitude beyond a thousand years or so. Of course, the concentrated minor actinide stream has to be managed separately, which prompts much of the current interest in transmutation processes.