A once-through cycle would require fissile materials as seed (e. g. U-235 or plutonium) each time the reactor is refuelled. To take advantage of the fertile thorium and its capability for generating fissile U-233 in a once-through cycle, extended cycles would be required to permit continued generation of U-233 and its fissioning to sustain reactor operation. In this regard, the excellent refractory properties of thorium oxide make it a good candidate for extended burn-up.

From the perspective of sustainability and resource utilization, however, there is still a dependency on U-235 with a once-through cycle (or plutonium if this is used as fissile material). In fact, if thorium were to be used in today’s thermal reactors, the conversion factor would be less than one but still greater than the conversion factor achieved in a standard uranium-plutonium cycle. For example, in a LWR using thorium-based fuel, the conversion factor is 0.7 (compared to 0.6 for uranium-plutonium fuels) and may reach easily 0.8 or even 0.9 in other types of reactors such as PHWRs or HTRs. A self-sufficient equilibrium thorium cycle, i. e. a conversion factor equal to or greater than 1, can even be reached in some thermal reactors. Examples are the Shippingport LWR reactor (see Section 8.2.1), CANDU-type reactors and, especially, molten salt reactors (MSR). In MSRs, U-233 breeding is promoted by keeping burn-up and specific power low (which entails an economical penalty) and by continuous removal of Pa-233 from the core by on-line reprocessing. Breeding in a thermal spectrum is not possible with the uranium-plutonium cycle so that this represents a real advantage of the thorium cycle. This is important because of the more favourable characteristics of thermal reactors compared to fast neutron reactors, e. g. lower fissile inventory in the core and, probably, a lower investment cost.

Various studies have investigated the use of thorium in thermal reactors, since many combinations of fuel cycles are possible with a mix of various types of reactor, operating as symbiotic systems. It transpires that thorium can be mixed with four types of fissile material:

• Highly enriched uranium (> 90% of U-235), called the ‘Th-HEU’ cycle. This was the reference fuel for HTR reactors in the 1970s (in the USA and in Germany). But today use ofthis fuel would raise serious proliferation concerns.

• Mid-enriched uranium (20%), which is called the ‘Th-MEU’ or sometimes ‘denatured’ cycle. The underlying idea is to disable direct use of uranium material for nuclear weapons fabrication.

• Plutonium, whatever its origin and isotopic composition, called the ‘Th-Pu’ cycle.

• Uranium-233, when available in large amounts after reprocessing of thorium — based spent fuel.

To summarize, the results of these studies show that thorium use in non-breeder thermal reactors would allow a global saving in uranium usage from a few tens of per cent to a maximum of roughly 80%, when equilibrium of the reactor fleet is reached. The precise figure depends on reactor types (and reactor type combinations) and recycling options.

With regard to the use of thorium in fast neutron reactors (FNRs), a number of studies (performed particularly in Russia for the BN-800 reactor, but also in France and elsewhere in Europe) demonstrated the possibility of achieving self­sufficiency in a Th-232/U-233 fuel cycle, that is to say achieving a conversion factor greater than one. However, Th-232/U-233 fuel performance regarding breeding in FNRs is not as good as uranium-plutonium fuel performance. For example, thorium-based FNRs need very large material inventories in the blankets to achieve negative feedback reactivity effects and a conversion factor greater than 1. The main reason is that plutonium has an eta factor (see Section 8.1.2) slightly better (1.33) than that of U-233 (1.27) for fission by fast (as opposed to thermal) neutrons. Another reason is that the fission cross section of thorium in the fast range is much lower than that of U-238 (one third or so). In summary, the use of a thorium cycle in FNRs is not very attractive, though there are claims that, for sodium-cooled fast neutron reactors, this leads to a reduced positive sodium void coefficient when compared to the standard uranium-plutonium core.5

Globally, if thorium were to be intensively used in non-breeder thermal reactors in closed cycle (i. e. U-233 recycling) the world’s fissile resources would be increased by around a factor 2 or maybe more in the very long term (provided that enough natural uranium is available to sustain such a cycle). If breeder fast reactors were intensively used with uranium-plutonium fuel (and with a conversion factor at least equal to 1), the energy potential of uranium natural resources would be multiplied by a factor 50 to 100. In that case, thorium breeding would multiply again this already huge energy potential by an additional factor 2 or so (depending on available thorium resources).