Fission's Last Gasp: Thorium ADS
Is the glass half full?
Before starting research in depth for this series, I brushed off my notes from my undergraduate courses in nuclear physics and neutron activation analysis, and took an online course on nuclear energy. Based on the latter I thought that perhaps fission power based on the Thorium-U233 fuel cycle could overcome the major problems of the U235- or Uranium-Plutonium fuel cycle. Hence I left this post for last.
The possibility of using Thorium as the basis for fission power reactors has been around for a long time; an early - if not the first - promoter was Carlo Rubbia of CERN, who proposed a research program in 1993. [1] The initial focus was on building a reactor to utilize minor Actinide waste. A more recent IAEA document [2], issued in 2005, lists thirteen variations on the Thorium fueled reactor concept that existed at that time. One of those, the Accelerator Driven System (ADS), a sub-critical reactor that uses U233 produced via neutron activation of Th232 as fuel, is the subject of this post. All other Thorium reactor concepts utilize fuels from the existing U235/Plutonium fuel cycles, are critical, and therefore by definition are incapable of meeting our requirements for inherent stability, relative sustainability, and not being associated with current nuclear weapons production and maintenance.
The dominant chain of isotope production for the Th232-U233 ADS is depicted in the figure below, which is an enhanced, corrected version of Figure 25 from the above cited IAEA document. The basic chain leading from Th232 to U233 is highlighted in blue.
A few comments are in order before proceeding. As is evident from the figure, if one started with a given isotope, an entire family of others can be produced via neutron bombardment and subsequent alpha or beta decay (and there are other modes). In particular, it is possible to start with Th232 and produce U235 - and even U238 - via neutron activation. The process is probabilistic, and the number of atoms transmuted from one isotope into another via a single neutron capture is proportional to a quantity called the capture cross section, the number of neutrons of requisite energy per unit time, the number of target atoms, and the amount of time allowed. Also involved is the probability that the target will undergo some form of radioactive decay before it can be transmuted. It follows that performing two transmutations Target → Daughter1 → Daughter2 is the product of the probabilities P(T →D1) and P(I1→D2) which means that P(T→D2) is smaller than either P(T→D1) or P(I1→D2) since all the probabilities are numbers smaller than unity. It then follows that the production of long lived minor Actinides starting with Th232 is extremely inefficient, because it requires Th232 → U233 → U234 → U235 → U236 → U237 → U238 → U239 before the first atom of Pu239 is produced. So, when the claim is made that the Th232-U233 fuel cycle does not produce any long lived minor Actinides, the claimant is speaking practically, not in absolute terms.
An ADS based on the Th232-U233 fuel cycle uses a particle accelerator to produce high energy protons which produce neutrons as a result of high energy interactions with the atoms of a target, typically the molten lead that also serves as the reactor working fluid. The high energy neutrons thus produced interact with the Th232 to produce Th233, which then eventually becomes U233, while the thermal neutrons help sustain fission of the U233, producing heat that can then be used directly or converted to electricity.
In a practical ADS, energy generation can occur only when a quantity of U233 is present. One could start with nothing but Th232, using neutrons to transmute the Thorium into Pa233 via Th233, with the Pa233 undergoing beta decay to produce the Uranium, or one could start with a mix of Th232 and U233. In any event, it is necessary to reprocess the fuel mixture from time to time to remove the fission products. When this happens there is a risk of diversion of U233, a fissile material which is every bit as useful as U235 in bomb making. The only mitigating factor is the presence of highly radioactive sources of gamma radiation, which greatly complicates the separation process.
The principal advantages of an ADS employing the Th232-U233 fuel cycle, in comparison to other fission reactor designs are:
In common with other Thorium-based designs, a greater degree of sustainability than achievable with Uranium-based fuel cycles, owing to the estimated 3.3-3.6 fold abundance of Thorium as compared to Uranium on earth [3];
Inherent stability due to the non-critical design. As the reactor operates at a sub-critical level, runaway reactions are essentially impossible and the reactor shuts down when power is lost. The major safety issue involves dissipation of residual heat, which can be dealt with passively;
Reduced quantities of long-lived radioactive waste produced as a result of the inherent improbability of producing minor Actinides; and
Reduced possibilities for proliferation as a result of the characteristics of Thorium byproducts and U232.
On the other hand,
Though ADS employing the Th232-U233 fuel cycle does not facilitate the production of any weaponizable Plutonium isotopes, U233 itself is a bomb making material, so the nuclear weapons community could adapt to use of U233 in place of U235- or Plutonium pits in their weapons designs, and ostensibly peaceful reactors could be used to supply material for weapons. In fact, though advocates appear to fail to notice, the Th232-U233 fuel cycle has a huge advantage over using natural Uranium because the latter is mostly U238, a non-fissile isotope that cannot be chemically separated from U235. On the other hand, chemical means can be used to obtain nearly pure U233 from material produced using the Th232-U233 fuel cycle;
Though there is essentially no minor Actinide waste, fissioning U233 still produces long lived radioactive fission products on the order of 20%, comparable to other reactors. The radioactive waste problem is reduced but not solved by turning to ADS with Th232-U233;
Although Thorium is more abundant than Uranium, the known reserves of the element are insufficient to support widespread use of the material as a fuel for more than perhaps a few centuries; and
There is no reason to expect that energy production via ADS employing the Th232-U233 fuel cycle will be significantly less expensive than other forms of fission power. While it is true that Thorium is more plentiful and therefore potentially less expensive as a raw material, the reactor design is more complicated and thus potentially more expensive to build and operate.
My conclusion is that it is not possible to design a fission reactor that combines the virtues of safety, sustainability, environmental responsibility, nonproliferation and attractive cost. The world would be far better off were we to replace our nuclear power reactors with truly safe, truly sustainable and affordable sources such as wind, solar, geothermal, tidal and sustainably-generated Hydrogen.
But wait, you might say. What about fusion power? That's a different story addressed in another post. Check the archives!
Notes
[1] https://cds.cern.ch/record/256520/files/at-93-047.pdf?version=1
[2] https://www-pub.iaea.org/mtcd/publications/pdf/te_1450_web.pdf
[3] https://en.wikipedia.org/wiki/Abundance_of_elements_in_Earth%27s_crust
Reposted with minor formatting changes and a modified final paragraph 7 Jul 2025