Thorium: Reduce, Reuse, Recycle

One of the basic principles of the modern environmental movement is the simple mantra to “reduce, reuse, and recycle”. It is my intention to show in this essay that the technology of the liquid-fluoride reactor, coupled with the energy source thorium, make it possible to achieve these goals to a far greater degree than other nuclear energy technologies.


Liquid-fluoride reactors are based upon the use of dissolved actinide fluoride salts in a carrier medium of low-absorption fluoride salt solvents. The most common formulations that have been considered and demonstrated for this mission are solvents based around low-melting point mixtures of beryllium fluoride (BeF2) and lithium fluoride (LiF) isotopically enhanced in the more-abundant component lithium-7. The actinide fluorides most commonly employed are thorium tetrafluoride (ThF4) and uranium tetrafluoride (UF4). LiF-BeF2 salt mixtures have very low neutron absorption properties, excellent heat capacity, stability under intense radiation, and the ability to dissolve appreciable amounts of thorium or uranium tetrafluoride.

Liquid and Solid Fluoride Salt Mixtures

Despite providing some degree of neutron moderation, LiF-BeF2 mixtures are not terribly good neutron moderators, thus liquid-fluoride reactors generally employ solid moderating materials in order to moderate neutrons to thermal energies. Graphite is most commonly employed, being abundant, relatively inexpensive, and chemically compatible with the salt. Graphite is not “wetted” by the fluoride salt and can be sealed in ways that limit the intrusion of fission product gases (especially xenon) into the structure of the graphite. Thorium as a nuclear fuel is not as well-known as uranium, but has properties that have special merit for nuclear use. Thorium also has a number of drawbacks for its use as a common nuclear fuel, but fortunately, by using thorium in fluoride form, nearly all of these drawbacks can be eliminated or strongly mitigated.

Thorium is common in the Earth’s crust, consisting of about 10 parts per million of common continental crust, approximately three to four times more common than uranium. Thorium is not fissile and consists of a single natural isotope (232) but thorium can be converted to a fissile fuel by the absorption of a neutron followed by a short period of beta decay. After absorbing a neutron, thorium-232 is transmuted into thorium-233, which then beta-decays with a half-life of 22 minutes into protactinium-233, which is chemically distinct from the parent thorium. Protactinium-233 has a half-life of about 27 days, after which is beta-decays to uranium-233, which is fissile and has impressive properties. Uranium-233 produces enough neutrons from fission by a thermal neutron to sustain the continued conversion of thorium to energy, even accounting for normal losses, provided that the reactor is neutronically efficient.

Reducing the Production of Transuranic Nuclear Waste

One of the biggest concerns about today’s approach to nuclear power generation concerns our use of low-enrichment uranium (LEU) in solid-uranium-oxide-fueled light-water reactors. In these reactors, LEU fuel is irradiated by thermal neutrons and a significant amount of plutonium is produced from the uranium-238 that makes up 95-97% of the original fuel. Some of this plutonium is consumed as the solid-oxide fuel rod is further irradiated, but from the plutonium other isotopes of plutonium are formed by neutron capture, and then higher actinides like americium and curium are produced. From the small fraction of U-235 present in the fuel even some long-lived neptunium-237 is produced. After an irradiation period of 3-4 years, the fuel rod can no longer sustain addition irradiation and is removed and placed in a spent fuel pool for cooling as high-heating decay products move inevitably towards stability.

In our current approach to civilian nuclear power, these irradiated uranium oxide fuel rods are not reprocessed to separate and partition their different chemical components, but are instead bound for disposal in a deep geological repository in Nevada. There after several hundred years the transuranic actinides still present in the spent nuclear fuel will generate the bulk of the heating that dictates their spacing in the repository and its ultimate capacity. Furthermore, the transuranic actinides carry the vast majority of the radiotoxicity that repository licensers must deal with as they plan for the performance of the repository over the next ten thousand years. Reducing the amount of transuranic waste that will be sent to any future repository would therefore be an important goal of a future approach to civilian nuclear power generation, and this is eminently doable by using thorium in a liquid-fluoride reactor. Transuranic waste production can be drastically reduced by a clever combination of the inherent properties of the thorium fuel approach and by the flexibility of the liquid-fluoride fuel form. Thorium, with an atomic mass of 232, begins the nuclear energy generation process at least five neutron absorptions removed from the first transuranic isotope that could be generated. As previously mentioned, thorium-232 absorbs a neutron, transmuting to protactinium-233 and then uranium-233, which is fissile. In a thermal neutron spectrum, uranium-233 tends to fission 90% of the time it absorbs a thermal neutron. The other 10% of the time is converts to uranium-234. Another neutron absorption in uranium-234 leads to conversion to uranium-235, which is also fissile and represents another opportunity for destruction through fission. Uranium-235 fissions in a thermal neutron spectrum approximately 85% of the time, and the other 15% of the time is converted to uranium-236. Uranium-236 has a rather low neutron absorption cross-section, and only after absorbing a neutron is the first transuranic isotope of this approach produced: neptunium-237. Neptunium can be removed from the fluoride salt mixture readily by fluorination from NpF4, which is in solution to NpF6 which is gaseous. Thus, unlike our current approach to nuclear power where the majority of the fuel (97% U-238) is a single neutron absorption away from the production of the first transuranic isotope (Pu-239), in the thorium-based approach, the fuel is five neutron absorptions away from the production of a transuranic isotope, and in the course of those absorptions roughly 98.5% of the original fuel is removed by fission. Thus, by using thorium in the fluoride reactor rather than uranium in the solid-oxide reactor, it is possible to REDUCE the amount of transuranic material generated by a very large factor.

Reusing Nuclear Fuel

As previously mentioned, today’s approach to nuclear fuel employs low-enrichment uranium is solid-oxide form in zirconium cladding, cooling and moderated by ordinary water. As an oxide, uranium is quite chemically stable and able to achieve high temperatures without melting down. Unfortunately, as an oxide, uranium is also subject to the low thermal conductivities common to most all oxides, and therefore high temperatures at the centerline of the solid fuel element become an inevitable consequence of heat transfer out the surface of the fuel element. In fact, the centerline fuel temperature of a uranium oxide fuel element, relative to the melting temperature of uranium oxide, is one of the key geometrical constraints.

As uranium oxide fuel is irradiated, fission products and transuranics accumulate in the ceramic oxide matrix. Intense radiation from the fission process and the decay of fission products also damages the fuel structure, causing dislocations and swelling in the crystalline matrix. Especially damaging to the fuel element are in the in-growth of gaseous fission products such as xenon and krypton, which further distend and crack the fuel structure. One of the isotopes of xenon (135) has a huge appetite for thermal neutrons and causes control transients during the changing of power settings within the reactor. After a period of time the uranium oxide fuel element has been depleted of fuel, swollen, cracked, distended, inflated, and compromised by the fission process and must be removed before cladding failure leads to the loss of fission products and other radioactive isotopes to the water loop of the reactor system. Spent solid-oxide fuel rods must be replaced by new fuel rods and are sent to a cooling pond where decay heat can be removed. Although the spent fuel still contains large amounts of unused fuel in the form of both uranium and other actinides, that fuel cannot be accessed until a reprocessing program takes place that involves chemically changing the solid uranium oxide into a liquid uranium nitrate fuel form through the application of strong nitric acid. Then a combination of chemical processes in aqueous and hydrocarbon solvents takes place to separate gaseous fission products, other fission products, transuranics, and uranium from one another. The resulting waste streams from these processes can be utilized productively, but the cost is significant due to the aggressive chemical steps involved and the chemical intensiveness of the new forms. Many, many recycles of the fuel would be needed to “burn-down” the uranium-238 present in the original spent fuel to energy (through fission) and the costs involved in reprocessing dictate that spent nuclear fuel is rarely subjected to more than one or two recycles before it is disposed. Thorium and the fluoride reactor present an entirely different approach to fuel management that makes repeated recycling not only easy but economically advantageous. That is because nuclear fuel in the liquid fluoride form rather than in the solid oxide form has distinct advantages. It is already in a chemically stable form as a fluoride. There is no reagent to treat the fuel that will be favored over its current state. Thus it is protected from chemical attack, combustion, burning, or corrosion. But more importantly, as a fluid is it in a form where chemical processes can be employed directly to remove fission products or to add new fuel to compensate for burnup. Additionally, the ionic nature of liquid-fluoride salt renders the fuel essentially impervious to radiation damage. Despite the passage of large amounts of gamma radiation, neutron radiation, alpha radiation, etc. the fuel remains chemically unaltered and with a complete retention of its physical properties. Gaseous fission products, including the important fission product poison xenon-135, are effortlessly easy to remove from liquid-fluoride salt. They simply come out of solution in the pump bowl during the pumping of the fluid through the loop. This has the additional benefit of keeping pressures low and allowing the reactor to change power states rapidly without concern for the effect of xenon on power changes.

In a modern incarnation of the liquid-fluoride reactor, there are two separate fluoride salts in action in the reactor core: the “fuel salt” and the “blanket salt”. The fuel salt is a mixture of uranium tetrafluoride in the lithium-beryllium fluoride carrier solvent. The uranium consists predominantly of uranium-233 but also contains U-234 and U-236 at equilibrium levels of concentration. Depending on the reprocessing approach it also contains fission products in the form of fluorides. The blanket salt is a mixture of thorium tetrafluoride in the lithium-beryllium fluoride carrier solvent. The blanket salt geometrically surrounds the fuel salt with a graphite barrier between them. Fission in the fuel salt produces neutrons, roughly half of which end up in the blanket salt, transmuting thorium to uranium by neutron absorption followed by beta decay. The uranium formed in the blanket is removed by the simple process of fluorination, whereby uranium as a tetrafluoride in solution is converted to a hexafluoride that is gaseous. Since thorium has no gaseous hexafluoride, it is left behind while uranium is removed in this simple, one-step process. Then the fuel salt is “refueled” by this same stream of fresh uranium hexafluoride by converting it from hexafluoride back into tetrafluoride through contact with hydrogen gas. Thus freshly generated uranium is continuously removed from the blanket salt and added to the core salt, where it subsequently undergoes fission that continues the process all over again. By keeping fissile materials out of the blanket by continuous reprocessing, the blanket fluid can be kept relatively free of fission products. The fuel salt, on the other hand, will accumulate fission products as uranium fission continues. The most troublesome fission product, xenon, is effortlessly removed by pumping action, but other fission products will become of increasing concern. Samarium, neodymium, and other lanthanides are fission products whose neutron absorption cross-sections are significant enough to merit attention. In order to purify the fuel salt, the first step is to remove the uranium fuel by fluorination. Then the carrier salt (LiF-BeF2) can be distilled from fission product fluorides in a high-temperature still. The remaining fission product fluorides constitute the equivalent of “high-level waste” from fluoride reactor reprocessing. The extracted LiF-BeF2 is recombined with the uranium and reinserted into the reactor core for another cycle of power generation.

The fluid nature of the reactor fluids allow them to be used over and over again, removing only the products that have been generated during operation (uranium in the blanket, fission products in the fuel salt). This ability to continually REUSE the reactor nuclear fuels represents a profound advantage over the solid-fueled uranium approach.

Recycling the “Wastes” of Fission

Fission processes inevitably generate a variety of fission product elements and a large number of isotopes, most of which are neutron-rich and radioactive. The familiar double-humped distribution of fission products reflects the physical reality that each fission event results in two fission products, a “heavy” one and a “light” one. As each of these fission products tends to have many more neutrons than is needed for nuclear stability at its new “station” in life, rapid beta decay generally follows fission and most fission products assume a stable form quite quickly. When all of the isotopes of an element reach stability it can logically be asked whether or not they are worth chemical extraction and recycling to other, non-nuclear uses. Consider the case of xenon. Xenon is a noble gas and fission product that accounts for a fair fraction of the mass of fission products from uranium fission. Xenon has a variety of isotopes but the longest lived one (133) has only a half-life of 5.2 days. Therefore, proceeding on the rule-of-thumb that “ten half-lives and you’re gone” after 50 days of storage the xenon remaining from fission would be essentially non-radioactive. In a conventional solid-core reactor the xenon is bound up in the solid-oxide fuel rod and can only be extracted by chopping up and dissolving the fuel element, but in a fluoride reactor it is very easy to extract xenon. In fact, it will come out of solution with essentially no effort at all. Since xenon is a valuable gas, rather than vent the xenon to the atmosphere it can be separated from the krypton by cryogenic distillation and sold. NASA and commercial satellite operators, for instance, use xenon for ion engines for spacecraft. Future NASA missions to Mars that have considered using xenon have had to seriously consider whether the world supply of xenon was sufficient to make such missions possible. Xenon recovered from fission might increase xenon supply. Another valuable material from fission is neodymium. Within the last 20 years, the discovery of a neodymium-iron-boron alloy that can be used to make super-strong, super-light magnets has caused neodymium demand to increase tremendously. Ironically, one of the markets that is in greatest demand for neodymium is the wind turbine market. They need large electrical generators due to the diffuse nature of the wind energy source, and they need these electrical generators to be as lightweight as possible so that they can be mounted on top of large towers. Neodymium magnets are particularly suited to this demanding application. Neodymium is the third-most-common element generated from fission (by mass) and also achieves nuclear stability relatively quickly; its longest-lived isotope (147) has a half-life of 10.9 days. By aging the high-level waste from the distillation process in fluoride reactors appropriately, one could extract the neodymium trifluoride from the other fluorides and convert it to a metallic form through electrolysis or metallic reduction. The neodymium would then be available to sell to the burgeoning market. Xenon and neodymium represent two recycling opportunities where a period of “aging” is needed before the isotopes stabilize and partitioning and marketing is possible. But there are other isotopes in the “waste” stream of a fluoride reactor where the radioactive form of the isotope is the desirable and economic product. An example of this case is the life-saving medical isotope molybdenum-99. Currently, molybdenum-99 is generated in specially-designed medical isotope production reactors in Canada and rushed to medical facilities across North America. Mo-99 decays to technetium-99m, which is then extracted and introduced into human patients in order to facilitate diagnostic procedures. The market for Mo-99 is quite large, but in solid-fueled reactors, the Mo-99 produced by fission is not accessible until the fuel is reprocessed. Since that is an infrequent event in solid-fueled reactors, the overwhelming majority of the Mo-99 produced in such reactors is never productively utilized; rather it simply follows its decay chain to Tc-99. In a fluoride reactor, on the other hand, the fluid nature of the reactor makes it possible to continuously extract Mo-99 along with the other isotopes of molybdenum. Molybdenum forms a volatile hexafluoride much like uranium does, and when the fuel salt is fluorinated, U, Mo, and several other elements come out of solution as gaseous hexafluorides. These can then be separated on from another by distillation at different temperatures, much like crude oil is refined. The molybdenum could then be shipped to medical facilities, where the Mo-99 would decay to Tc-99m that could be chemically extracted and given to patients who need it. Xenon, molybdenum, and neodymium are three of the most common fission products but many others have value too. The fluid nature of the fluoride reactor makes RECYCLING of the so-called waste quite likely to be economically attractive in many circumstances.

Summary: Reduce, Reuse, Recycle

The environmental dictum of “reduce, reuse, recycle” has been considered in terms of the thorium-fueled, liquid-fluoride reactor and found to be a simple and unifying theme for the options that this technology makes available. Relative to a conventional, solid-fueled uranium reactor, one can drastically REDUCE the generation of transuranic actinides, REUSE the thorium and uranium fuel is a way that allows for complete consumption of the energy resources, and RECYCLE three of the most common fission products into economically useful and even life-saving applications. The thorium-fueled liquid fluoride reactor is worthy of significant further attention, investigation, and funding based on these and many other merits.

64 thoughts on “Thorium: Reduce, Reuse, Recycle

  1. I agree with Michael Burns.
    I was born in Scotland, and chose to become an American.
    I was raised a Presbyterian, and am now an atheist.
    But it boggles my mind that so many of my now fellow-Americans imagine that Christianity is compatible with outrageous wealth and the destruction of the environment, a.k.a. the rest of God's Creation!
    But I disagree with Robert Carroll if he thinks that the first priority should be shutting down the nuclear industry.
    We 'd be far better shutting down the entire fossil carbon industry, AND those damned ugly wind turbines, with nuclear power, preferably owned by the people's Government, like the BPA, the TVA, the BBC, and until the EU forced them to privatize, France's EDF .

  2. The really neat thing about the REAL cures for global warming, like either kind of breeder reactor (i.e. I'm including the IFR) is that they DO NOT "create jobs".

    Technology, if good, doesn't take huge manpower to maintain it. So if we solve these problems, we must also solve the problem of redistributing incomes and leisure fairly, and curtail the horrific power of the Employer to ruin the Employee, whether by malice or the stupidity of people like Lehman Bros.

  3. Dr. Sorensen,

    I appreciate the work you are doing to make the
    public aware of the great possibilities inherent
    in the use of Thorium as a nuclear fuel. I have
    a minor comment and a follow-on question:

    Under essay3rs (presumably index.htm), you write:

    Many, many recycles of the fuel would be needed to “burn-down” the uranium-238 present in the original spent fuel …

    Don't you mean uranium-235 here?

    And, a clarifing question for me:
    I think your comments about U-233 being 5 neutron captures away from transuranics is really important.
    In one or more of your videos, you make the comment that transuranics created in the reactor are 'burned up', which I interpreted to be they would transmute up or down (mainly, down) the periodic table in response to their exposure to neutrons, alpha, beta or gamma rays.

    Was I hearing you correctly, or does the essay3rs paper imply that, after 5 or more neutron captures, there will be in fact a small residual of transuranics that will not be 'burned up' and will have to be stored for 250K years or so (a far smaller amount than normal reactors, but still, something which needs the be delt with)?

    Thanks for your response in advance!

  4. I live in Alberta Canada. Until the disaster in Japan, the Alberta's electricity suppliers were sending out a magazine about every month. This was to prepare the public to accept a uranium type nuclear reactor near Whitecourt Alberta. They explained that this was to supplement the power used by the oil industry. The oil industry uses an incredible amount of electricity here for pumping and primary refining (for ever barrel of oil produced a barrel is burned to produce it).

    I wrote a letter to the official opposition party in Alberta about Thorium reactors (the government in power only sees one line – oil), and about the possibility of creating a new industry in Alberta, because the oil is not going to last for ever. No one knows about Thorium and very few understand it.
    Alberta, because of our need to custom manufacture for the oil industry has every capability to build and mass produce Thorium reactors. We have more machine shops in Edmonton, for instance, than any other city in North America. We also have every trade including the best technologies in welding and fabricating. This is where Thorium reactors should be built.
    How do we sell the idea to government and the public?

    Tom Salken
    Edmonton, Alberta

  5. These reactors were demonstrated as working models as early as 1966. However, the demand for weapons grade plutonium put a stop to their widespread use. We no longer need the prodigious amount of plutonium and the waste has become a football. Thorium reactors aer much more inherently safe than uranium or mox. Time to start building them and getting rid of our spent fuel inventory in a useful way.

  6. @TofuNFiatRGood4U:
    the answer to your first question "Don’t you mean uranium-235 here?" is:

    No, the difficulty is not to burn U-235 (which burns well, but there is little of it left in the reactor at that point), but to "burn" U-238.
    The reason is that when U-238 is hit by a neutron, it becomes Pu-239, which is fissile; and we just need a new neutron to hit the Pu-239.

    But then, for the reaction to continue, each Pu-239 that fissions should emit at least 2 neutrons, one to hit a new U-238 (to produce new Pu-238) and the other to fission another Pu-238.
    This is not easy to obtain and it is the issue which sent the whole nuclear industry to try to build the "fast breeders" since Fermi time (and "burning" in the process immense… amounts of money since then); the other road was the MSR/LFTR.

    Regarding the second question ("there will be transuranics in the LFTR that will not be ‘burned up’ and will have to be stored for 250K years or so?"), the answer is:
    a) yes, there will be actinoid waste left by the LFTR too
    b) but after 300 years the radio-toxicity will be 10 000 times less than the one by the existing reactors.
    So that's the difference: the LFTR has transuranic waste, but it lasts 300 years, and that is manageable

  7. I have to correct in my previous post the answer to the the second question asked by TofuNFiatRGood4U.

    In the LFTR the plan is to burn the actinides and transuranics; only the fission products are sent to the waste stream.
    Of course this implies that the chemistry of the fission product removal is selective enough to capture just (or mostly) fission products (and not transuranics).

    I thank Robert Hargraves for the clarification.

  8. I have to correct in my previous post the answer to the the second question asked by TofuNFiatRGood4U.

    In the LFTR the plan is to burn the actinides and transuranics; only the fission products are sent to the waste stream.
    Of course this implies that the chemistry of the fission product removal should be selective enough to capture just (or mostly) fission products (and not transuranics).

    I thank Robert Hargraves for the clarification.

  9. Has anyone approched the Navy about using a LFTR in one of there smaller ships like a destroyer or crusser. Perhaps retrofit a mothball ship to demonstraite the fesability. If you think about it how many billons of dollor are spent just to fuel the fleet in a year. With the US budget being a polictical hot button I would be think they would be interested in atleasted expoloring this posibility.

  10. The Xenon from a LFTR is not useful at all. 135-Xe decays to 135-Cs which is a long lived radioactive material. In contrast a conventional reactor produces 126-Xe, which is essentially stable and possibly useful, though it may not be recovered today.

    Using a LFTR on a ship is a bad idea: The fuel is slightly water soluble.

Leave a Reply