Environmental Impact of the Nuclear Fuel Cycle

Nuclear power provides approx 17% of the world’s electricity, which is equivalent to a reduction in carbon emissions of 0.5 Gt of C/year. This is a modest contribution to the reduction of global carbon emissions, 6.5 Gt C/year. Most analyses suggest that in order to have a significant and timely impact on carbon emissions, carbon free sources, such as nuclear power, would have to expand total energy production by a factor of three to ten by 2050. A three fold increase in nuclear power capacity would result in a projected reduction in carbon emissions of 1 to 2 Gt C/year, depending on the type of carbon based energy source that is displaced. The impact of an expansion of this scale on the generation of nuclear waste and fissile material that might be diverted to the production of nuclear weapons is examined. There are three types of nuclear fuel cycles that might be utilized for the increased production of energy: open, closed or a symbiotic combination of different reactor types, such as thermal and fast neutron reactors. Within each cycle, the volume and composition of the nuclear waste and fissile material depend on the type of nuclear fuel, the amount of burn up, the extent of radionuclide separation during reprocessing, and the types of material used to immobilize different radionuclide. The relation between the different types of fuel cycles and their environmental impact has been the subject of recent discussions.


Nuclear Power and Carbon Emission

In 1997, the third Conference of the Parties COP 3 produced the Kyoto Protocol. Although signed by then vice president Al Gore, it has not been ratified by the Senate of the USA. To enter into force, the Protocol must be ratified by 55 parties representing at least 55% of the world’s emissions of greenhouse gases GHG in 1990. As part of the protocol, the developed countries must commit themselves to reducing their collective emissions of six GHGs to at least 5% below 1990 levels. The most prominent of these GHGs is Co2, which accounts for nearly 65% of the warming effect (Houghton et al 2001). The USA presently accounts for approximately 25% of the global emissions of Co2 with 5% of the world’s population. The Kyoto Protocol would require the USA to reduce present emissions by 22%, an annual reduction of 1.1 x 10 (sq9) t co2, equivalent to removing all the gasoline powered vehicles from Us roads (Loewen & Leon 2001). The USA produces nearly 20% of its electricity using nuclear power, and this is equivalent to avoiding the release of 6 x 10 (sq 8) tco2, if this electricity had been produced from carbon based fuels (Loewen & Leon 2001).

There is a pressing need for developing a timely strategy to reduce GHG emissions. Thus, a number of analyses are based on a goal of limiting the increase in CO2 emissions to twice 550ppm the pre-industrial levels 275 ppm by the year 2050(Fetter 2000, Sailor t al 2000). Present CO2 levels are just over 360 ppm, increasing at an average rate of 1.5 ppm/y. this adds 3.3 GtC/y to the atmospheric reservoir, which is 750 GtC (Houghton et al 2001). Models of Co2 emissions suggest that strategies for reduction must be initiated in developed countries by 2010 in order to meet the goal of only doubling of the Co2 concentration above the preindustrial level (Wigley 1997). Of all the technologies presently capable of contributing to a major reduction in carbon emissions, nuclear power is one of the most promising, simply because the technology is already operating on a substantial scale, and in principle, it could be deployed more rapidly on a global scale. The Nuclear Energy Institute NEI maintains that the US nuclear power generating capacity can be increased to 10 GW by 2012, equivalent to 0.022 GtC/y. the NEI supports a goal of adding 50 GW capacity equal to approx 50 new NPP, by 2020 equivalent to a reduction of 0.1 GtC/y.

Analyses for the prospects of nuclear power have been presented by many, but two of the most detailed are by Fetter 2000, and Sailor et al. 2000. These analyses necessarily make many assumptions about future energy needs. Assuming a stabilization of Co2 concentrations to approx twice pre-industrial levels by 2050, and projecting a growth in world population to 9 billion (a 50% increase) and an increase in per capita energy consumption of 50%, the global energy demand in 2050 will be approx 900 exajoules (10 sq18 joules) per year (Sailor et al 2000). If nuclear power provides one third of the projected energy requirement (300 EJ/y), and the balance is divided equally between conventional fossil fuels and ’decarbonized’ fossil fuels, the 300 EJ from nuclear are roughly equivalent to 3300 GW-year (one GW y is the average annual energy output from a single large power plant) of capacity per year (present capacities are about 260 GW y/y). With this scenario, the projected 900 EJ/y of global energy use would still result in Co2 emissions that would equal 5.5 GtC/y (present levels are approximately 6.6 GtC/y). still, this would mean a more than a tenfold increase in nuclear power generating capacity requiring the cosntruction of over 3000 NPP before 2050 (at present there are 439 operating nuclear generating units). The impact of this expansion in nuclear power generation capacity is difficult to anticipate because it depends critically on the types of reactors and fuel cycles that are used, as previously discussed. The figures from the MIT study are based on an increase by a factor of three of nuclear generating capacity by 2050 (1000 GW e) (Ansolabehere et al 2003). Still, one must expect that the most immediate deployment of new reactors will bne of the Generation III+ type, not too different from the present water reactor technology, but with higher burn up of the nuclear fuel. Thus, one may use the present technology as a basis for extrapolating the environmental impact and use the factors of 3 to 10 as the range of what has been considered for the increase in nuclear power production. On this basis, the annual increase in spent fuel production would be between 27000 and 89000 tHM. The higher number is more than the presently planned capacity (70,000 tHM equivalent) for the proposed repository at Yucca Mountain. One approach to reducing the impact of the increased nuclear waste production would be to use reprocessing to minimize the volumes of waste produced andto utilize the fissile content of the SNF; however, this raises major issues related to the proliferation of nuclear weapons. A I GW y light water reactor produces 200 kg /y of Pu (enough for 20 nuclear weapons). If the nuclear energy capacity is increased to 3000 GW, then the annual production of Pu would be over 500 000 kg (Williams & Feiveson 1990). If one foresees a nuclear industry based on Pu-breeder reactors, the 3000 GW nuclear system would produce five million kil of plutonium per year. Alvin Weinberg 2000 has related the reduction , avoided increase , in Co2 content in the atmosphere to the amount of U consumed, that is, the percentage of U fissioned in nuclear power plants. A typical LWR without reprocessing has an efficiency (% of U fissioned) of only 0.5%, while a perfect breeder reactor cycle with reprocessing has an efficiency of 70%. Even if the presently estimated reserves for uranium are completely utilized (Weinberg 2000), the low efficiency system that we now use, LWR followed by direct disposal, will lower the Co2 increase by only 38 ppm. Either there will have to be a shift to breeder reactors and reprocessing, or alternative sources of U must be found. All of these figures are speculative, but they do emphasize that an increase in the role of nuclear power in reducing carbon emissions must be substantial and go hand in hand with the development of advanced fuel cycles and waste management technologies that do not presently exist on an industrial scale.

Just as important as evaluating the performance of the nuclear fuel cycle, one must also consider the size of the fluxes and reservoirs of the carbon cycle. Present CO2 emissions from fossil fuels and the production of cement are estimated to be 6.3 +- 0.4 GtC/y; emissions related to changes in land use (eg deforestation) are 1.6 +- 0.8 GtC/y (Schimel et al 2001). At present, the reduction of Co2 emissions that can be attributed to the use of nuclear power is o.5 GtC/y. thus, the uncertainties in the major fluxes in the carbon cycle are approximately the same as the present impact of nuclear power on Co2 emissions (Sarmiento & Gruber 2002). To quote from Falkowski et al 2000: ‘our knowledge is insufficient to describe the interactions between the components of the Earth system and the relationship between the carbon cycle and other biogeochemical and climatologically processes.’ we will need much more refined understanding of the carbon cycle and a more explicit description of the nuclear fuel cycle before one can quantify the impact of the fuel cycle on the carbon cycle. Lack of final conclusion is to be expected, as the uncertainties in the analysis of the environmental impact of the nuclear fuel cycle remain large and even unconstrained in the absence of an explicit description of t he type of fuel cycle to be used. This is made all the more evident by the failure of previous, although thorough, analyses (eg Pigford 1976) to anticipate the state of the nuclear power industry even 25 years into the future. It is also difficult to estimate the impact of the nuclear fuel cycle on the carbon cycle, because the uncertainties in our knowledge of the global carbon cycle remain large (Falkowski et al 2000). Still, even in the face of the uncertainties for both of these cycles, nuclear and carbon, one must acknowledge the tremendous amounts of energy available from the non carbon producing nuclear option. However, the scale of the increase in nuclear power production that will be required, up to a factor of 10, will lead to reprocessing of spent nuclear fuel, and this raises two issues: 1) the successful management and disposal of the radioactive waste, and 2) the potential proliferation of nuclear weapons. There is optimism about reaching a successful solution to the first problem.

By Rodney C. Ewing, Dept Geosciences, Univ of Michigan,

Source:
Energy, waste and the Environment; a geochemical perspective
(Edited by) R Giere, Univ Freiburg, Germany; and P. Stille, Dept of Earth and atmospheric Sceince, Purdue Univ, West Lafayetter, USA; ULP Ecole et Observatoire des Sciences de la Terre-CNRS, Strasbourg, France, p 19-21