(BCG, 2006), which was funded by Areva, found that the cost of limited recycle using MOX/PUREX in the United States could be comparable to the cost of the once-through fuel cycle, and an earlier study by OECD/NEA (1994) found that reprocessing was about 14 percent more expensive per kilowatt-hour of generated electricity than was the once-through fuel cycle.
Another example discussed earlier in this report is the case of a system using a fully closed fuel cycle (including fast reactors as well as fuel cycle plants). The LCOE for such a system remains speculative, but it is likely to be more expensive than the once-through approach, as a large number of fast reactors and reprocessing plants will be required. On the other hand, as discussed in the section on “Alternative Fuel Cycles,” such a closed cycle would produce a smaller volume of long-lived high-level waste than the once-through fuel cycle produces, and the long-term heat load could be reduced owing to the destruction of a large fraction of transuranics in the used fuel. If fission products were also removed from the fuel and handled separately, the short-term heat load could also be reduced, potentially allowing closer packing of waste in a repository. In this case—although a quantitative analysis has yet to be done—the increased expense for the reprocessing, the fuel fabrication, and the fast reactors might be counterbalanced by reduced cost for waste disposal if one or more future repositories become unnecessary.
The AEF Committee’s estimates of the potential supply from nuclear power in 2020, 2035, and 2050 are discussed in this section and tabulated in Table 8.1. The committee has estimated the maximum deployment of new nuclear plants that could be built under an accelerated deployment program, as described in Part 1 of this report; however, no attempt has been made to predict what will in fact be built. Any such prediction is intrinsically uncertain because it depends on many factors, including the economic conditions in the United States and around the world over the coming decade.
The contribution of new nuclear power plants to the U.S. electricity supply before 2020 is likely to be limited because new plant construction requires a long lead time: it can take some 4 years to obtain a construction and operating license and 4–7 years to build the plant (consistent with current world trends, as shown