IAEA-TECDOC-1450 is a document that was issued by the International Atomic Energy Agency in July of 2005. It is a fairly comprehensive treatment of the use of thorium in many different types of nuclear reactors.

Most of the attention in the paper is focused on the use of solid thorium oxide fuel assemblies in light-water and heavy-water reactors, with some consideration of how to use thorium in liquid-metal fast breeders as well.

On pages 29 and 30 of the document there is a description of a “molten-salt” reactors that could utilize thorium as well.

The overall benefits of thorium are described in TECDOC-1450 as:

(1) Thorium is 3 to 4 times more abundant than uranium and widely dispersed worldwide.

(2) Unlike the abundant isotope of uranium (uranium-238), the fissile daughter of thorium formed by neutron absorption (uranium-233) produces enough neutrons from fission per neutron absorption to sustain the further conversion and consumption of thorium indefinitely in a thermal spectrum reactor. Uranium-238 can only accomplish this in a fast-spectrum reactor.

(3) Thorium can be consumed as a nuclear fuel with very little production of actinides heavier than uranium (transuranic elements), which are the primary sources of long-term radiotoxicity in existing uranium-based nuclear fuel.

(4) Thorium has only a single valence state (+4) while its fissile daughter uranium has two major valence states (+4 and +6) which facilitates easy chemical separation of bred uranium from thorium.

(5) Thorium–based fuels and fuel cycles have intrinsic proliferation-resistance due to the formation of 232U via (n,2n) reactions with 232Th, 233Pa and 233U. The decay of 232U includes two decay products (212Bi and 208Tl) that emit hard gamma radiation that make the detection of U232-contaminated fuels exceptionally straightforward.

The IAEA study also identified challenges related to the use of thorium as a nuclear fuel. These included:

(1) The high melting point of ThO2 (3500C) makes fabrication of solid thorium-oxide based fuels challenging.

(2) Thorium-oxide-based nuclear fuels are chemically inert and difficult to reprocess.

(3) The 232U contained in thorium-based fuels has high radiation levels which make solid fuel fabrication difficult using existing procedures.

(4) In the conversion chain of 232Th to 233U, 233Pa is formed as an intermediate, which has a relatively long half-life (~27 days) and a significant neutron-absorption cross section. If 233Pa absorbs a neutron before decaying to 233U, the consumption of thorium as a fuel is thwarted and 234U is formed instead. Ideally, 233Pa would be isolated from the neutron flux as soon as it is formed and then reintroduced to the reactor after it has decayed to 233U. This step is highly impractical in most solid-fueled reactors, and so overall neutron flux (and power) must be reduced to reduce neutron-absorption in 233Pa. This reduces the economic viability of the reactor design.

The challenges related to the use of thorium in a solid-fueled reactor were recognized early in the Atomic Age by early pioneers such as Nobel Laureate Eugene Wigner and Dr. Alvin Weinberg, the inventor of the light-water reactor. In order to realize the benefits of thorium as a nuclear fuel while circumventing the disadvantages, they proposed that thorium be utilized in reactors whose fuel was in a fluid form. This would enable the elimination of fuel fabrication that was difficult with thorium, as well as facilitating the rapid reprocessing of thorium fuels, taking advantage of the chemical differences between thorium, protactinium, and uranium to achieve simple chemical separation of the fuels.

Further research at the Oak Ridge National Laboratory in the early 1950s identified liquid-fluoride salt mixtures as the ideal medium in which to utilize thorium as a nuclear fuel. This is because only thorium in fluoride form (specifically, thorium tetrafluoride) can truly be in solution, rather than a suspension or a slurry. To reduce the melting temperature of the thorium tetrafluoride, they proposed that it be mixed with fluoride salt mixtures that were chemically compatible with high-nickel container materials and also possessed attractive neutronic properties, specifically lithium-7 fluoride and beryllium fluoride. In such a liquid-fluoride reactor, thorium would absorb neutrons from the fission reaction and be transmuted first into protactinium, and then by decay into uranium. The chemistry of uranium and thorium enabled very easy separation by fluorination of the uranium from a tetrafluoride to a gaseous hexafluoride. The chemistry of protactinium also permitted separation from thorium, potentially allowing a high-flux thorium reactor to operate while still minimizing fuel losses to protactinium absorption. The fluoride salt mixtures of the reactor (7LiF-BeF2-233UF4 in the core and 7LiF-BeF2-ThF4 in the blanket) were very chemically stable and thus would allow low pressure operation at high temperatures, which meant that high-temperature reactor operation was possible. This in turn implied excellent power conversion efficiencies if coupled to an appropriate power conversion system such as a closed-cycle gas turbine, as well as applications enabled by high temperature reactors such as sea water desalination from waste heat, thermo-chemical generation of hydrogen, and the synthesis of other transportation fuels for a variety of applications.

The low-pressure operation of the reactor also permitted the reactor to incorporate many impressive passive safety features, such as a freeze plug in the bottom of the reactor vessel that would melt if the reactor overheated and would drain the core fluids into a vessel specifically intended for the passive removal of decay heat, rather than one that needed to sustain nuclear operation. The capability of the fluoride reactor to productively utilize thorium as a nuclear fuel, while avoiding its principal disadvantages, indicate that the liquid-fluoride thorium reactor should be the baseline reactor design considered for the successful developm
ent of the thorium resource.