Another possible novel used-fuel recycling process for LWR fuel would remove uranium by making use of the volatility of uranium hexafluoride. Some fission products would also be removed leaving plutonium, minor actinides and some residual fission products. This so-called ‘plutonium ash’ could be used as CANDU fuel (Dyck et al., 2005).

Detailed reactor studies and MOX fuel fabrication and irradiation tests have confirmed the feasibility of MOX fuel use, including ex-weapons plutonium, in the CANDU reactor (Chan et al, 1997; Dimayuga, 2003; Dimayuga et al., 2005) as well as in the ACR-1000 (Ovanes et al, 2009).

Some countries are looking at alleviating public concerns about the back-end of the fuel cycle by partitioning and transmuting the used fuel. In this concept, some of the long-lived actinides from used LWR fuel would be fissioned in fast reactors designed for that purpose. By utilizing the actinides in the used LWR fuel first as fuel in a CANDU reactor, the total number of fast reactors needed for ultimate destruction of the LWR actinide waste can be significantly reduced. In this role, the CANDU reactor would be an effective intermediate burner between LWRs and fast reactors, by reducing both the decay heat and the radiotoxicity of the used LWR fuel (Hyland and Dyck, 2007; Hyland et al. , 2009a; Hyland and Gihm, 2010).

Thorium fuel

A large variety of thorium fuel cycles could be utilized in the CANDU reactor, the ACR-1000 or the PT-SCWR. The fissile component could be provided by plutonium from reprocessed LWR fuel, LEU or from recycled U-233. The use of thorium in the CANDU reactor can be tailored to meet national considerations, such as availability of fissile and fertile material, the availability of recycling and fuel fabrication facilities and strategic objectives. The use of thorium would provide protection against the increasing cost of uranium as resources dwindle, and would help assure long-term resource supply and diversity.

The CANDU reactor and its evolutionary variants offer the potential of a staged approach to thorium fuel cycles. The simplest, near-term fuel cycle option is the ‘once-through’ thorium fuel cycle, in which economic and resource benefit is derived from the use of thorium without the need for recycling (although the fissile U-233 created is available for future recycling when and if needed). In the near term, the fissile material required to initiate the thorium cycle could be provided by LEU fuel. Alternatively, if plutonium were available from conventional reprocessing or advanced recycling technologies, it could provide the external source of neutrons needed to initiate the cycle. In this case, plutonium would be consumed and fissile U-233 would be created.

Regardless of the source of fissile material, the CANDU reactor provides several options for configuring the arrangement of fissile driver material and fertile Th-232 (Boczar et al, 2002a, 2002b):

• a homogeneous fuel in which the fissile and fertile components are co-mixed in the same fuel pellet and bundle (Hyland et al, 2009b)

• heterogeneous ‘mixed bundle’ designs in which the fissile and fertile components are in separate elements in the same bundle

• heterogeneous ‘mixed channel’ designs, in which the bundles containing thorium and the bundles containing the driver fuel are irradiated in separate channels. This option allows a different irradiation time for the thorium and driver fuel bundles/channels

The full benefit of the thorium fuel cycle would require recycling of the U-233 produced during irradiation, which could be done in existing CANDU reactor designs. Looking to future developments, the Self-Sufficient Equilibrium Thorium (SSET) cycle offers the potential of a near-breeder thermal reactor, which is self­sustaining in fissile material (recycled U-233). To achieve the SSET cycle would require even greater improvements in neutron economy. This could be achieved by further optimizing the lattice design, removing the adjuster rods, increasing heavy-water purity, reducing the flux level (which would reduce parasitic absorption in Pa-233), removing the isotope Zr-91 from the zirconium structural materials in the fuel and core, which is a stronger absorber of neutrons, and ultimately by using a gas coolant (which would add around 15 mk reactivity and eliminate the positive coolant void reactivity). To illustrate the potential of this reactor and fuel cycle, in Canada, if all the plutonium from used natural uranium fuel from the existing CANDU reactors operating over their lifetime were to be used to initiate the SSET cycle, two to three times the existing nuclear capacity in Canada (currently about 15 GWe) could be sustained indefinitely (Boczar et al, 2010).

The CANDU reactor offers a phased approach to the use of thorium, starting with the simple technology existing today and progressing through stages to a closed thorium cycle. For example, China is a country having both LWR and CANDU reactors and has indicated interest in pursuing the thorium fuel cycle in their CANDU reactors (WNN, 2009). A phased approach has been proposed towards achievement of energy sustainability through the use of the thorium fuel cycle in CANDU reactors in China (Boczar et al, 2010). The main elements of that strategy are more widely applicable:

• In the short term, obtain benefit from and experience in the use of thorium in existing CANDU 6 reactors using a mixed-bundle approach (a CANFLEX bundle containing thorium in the larger central eight elements with enriched uranium in the outer 35 elements). Pure thorium bundles could also be irradiated for long periods of time in the peripheral channels.

• In the medium term, maximize the benefit from reprocessing used LWR fuel by using the recycled plutonium in homogeneous plutonium/thorium fuel in CANDU 6 or EC6 reactors (Mao et al. , 2009) or in ACR-1000 reactors (Ovanes et al, 2009). The RU could be used in existing CANDU 6 plants or in new EC6 reactors, either as-is or blended with DU to form NUE fuel. If actinide destruction is desired, this could also be pursued in CANDU reactors.

• In the long term, the thorium cycle would be closed by recycling the U-233 in EC6 reactors, or in CANDU reactors designed to have even higher neutron efficiency and a conversion ratio approaching unity (the SSET fuel cycle). A thorium-fuelled CANDU reactor would also be synergistic with fast reactors.