Research Programs

Advanced Fuel Cycle Technology and Economics

Actinide Management in LWRs

The once through nuclear fuel cycle adopted by the majority of countries with operating commercial power reactors raises concerns about the radioactive waste, which has to be isolated from the environment for thousands of years. Moreover, plutonium and other actinides, after the decay of fission products, could become targets for weapons proliferators. These concerns can be addressed if a closed fuel cycle strategy, where all Trans-uranic actinides (TRU) are recycled back into reactors and destroyed by fissioning. This can be accomplished in dedicated fast reactor systems, however, the development and deployment of such innovative systems is technically and economically challenging. Therefore, the possibility of constraining the generation of long lived radioactive waste through multi-recycling of TRU in existing Light Water Reactors (LWR), which could be implemented on a faster time scale, is explored. The objective of the study is to identify the extent to which LWRs can consume the actinides they produced without negatively impacting their safety and the cost of such a closed fuel cycle.

Fertile free fuels (FFF) were selected as the most attractive candidates for TRU burning in LWRs because they do not generate additional TRUs. Burning of TRU in dedicated fully fertile free LWR cores were explored. Although the FFF exhibited excellent TRU destruction performance in a typical LWR fuel lattice geometry, significantly reduced Doppler effect, impaired void reactivity feedback and small effective delayed neutron fraction challenged reactor control and safety. Therefore, the Combined Non-Fertile and UO₂ (CONFU) assembly concept was proposed for multi-recycling of TRU in existing PWRs. The assembly assumes a heterogeneous structure where about 20% of the UO₂ fuel pins on the assembly periphery are replaced with FFF pins hosting TRU generated in the previous cycle. The possibility of achieving zero TRU net was demonstrated. A comprehensive neutronic and thermal hydraulic analysis as well as numerical simulation of reactivity initiated accidents demonstrated the feasibility of TRU containing LWR core designs of various heterogeneous geometries. The power peaking and reactivity coefficients for the TRU containing heterogeneous cores were estimated to be comparable to those of conventional UO₂ cores.

A typical concern when recycling the actinides in a thermal spectrum is the buildup of higher actinides, which emit spontaneous fission neutrons and hard gammas. The majority of TRU nuclides reach their equilibrium concentration levels in less than 20 recycles with the exception of Cm246, Cm248, and Cf252, which complicate TRU fuel reprocessing and fabrication. However, allowing longer cooling times of the spent fuel before reprocessing can drastically reduce the SF neutron radiation problem due to decay of Cm244 and Cf252 isotopes with particularly high SF source.

The environmental impact of the sustainable CONFU assembly based fuel cycle is limited by the efficiency of TRU recovery in spent fuel reprocessing. TRU losses of 0.1% from the CONFU fuel reprocessing ensure that the CONFU fuel cycle radiotoxicity reduction to a level comparable to that of the original natural uranium ore within 1000 years. The cost of the sustainable CONFU based fuel cycle is found to be about 60% higher than that of the once through UO₂ fuel cycle, but the difference in total cost of electricity between the two cycles is only 8%. The higher fuel cycle cost is a result of the cost of reprocessing and secondarily of the increased level of needed UO₂ enrichment. The CONFU closed fuel cycle is comparable to that of a closed cycle using an advanced fast actinide burning reactor (ABR), but the main advantage of CONFU is the possibility of near-term deployment and higher certainty of the costs involved. The cost of the CONFU fuel cycle is projected to be considerably lower than that of a cycle with an accelerator driven fast burner system.

The enhanced CONFU concept is under development, which would allow one to achieve net transuranic destruction. First results indicate that a CONFU assembly can achieve a net TRU destruction of actinides using a limit of 5 ^w/_o U-235 driver pins and a maximum of 20 ^v/_o TRU in inert zirconia-magnesia matrix pins. Further, it has been shown that this net TRU destruction can be accomplished in each recycle with multi-recycling of actinides for up to 10 recycles. This assembly has sufficient reactivity to support a 3-batch, 18-month cycle for a total of 4.5-years total in-core irradiation. The thermal margins for operation are reasonable, if somewhat degraded compared to a typical all UO₂ assembly. Alternative recycling schemes are also studied including recycling of plutonium together with neptunium and americium with curium in a different stream allowing for different cooling times to facilitate fuel handling. Finally, a proliferation resistance, repository impact, and economic assessment are also conducted.

In parallel to the reactor physics feasibility studies, inert matrix materials suitable to host actinides are being explored. These include metals nitrides, carbides and oxides. For LWR applications, ceramic oxides appear more promising than other materials, although no consensus has been reached as to the best material. Zirconia (stabilized by yttria or calcia) and cerium oxide are currently considered most promising as hosts for the actinide oxides. Spinel (MgAl₂O₃) has been investigated as a host for zirconia fuels to enhance fuel thermal conductivity. The sensitivity of spinel to fission product damage necessitates the use of dispersion of actinide-host particles of a large particle size, 150-200 m m, in a continuous matrix of spinel. Because large particles have significant fission gas release due to high local burnup, the selection of the right size of the particle still remains an open issue. A performance model of a ceramic dispersion fuel, which has been developed by modification of the NRC developed FRAPCON-3 code, shows that dispersion of particles into an inert matrix can be potentially advantageous over UO₂, but the issue of high local burnup of the particle and associated high fission gas release remains to be resolved.