Science in Society Archive

Thorium the Answer to Clean Energy?

The Fukushima disaster has rekindled interest in thorium reactors, which may be both safer and more economical as their proponents claim, but do we need nuclear power at all? Prof. Peter Saunders

The US, the UK, and many countries in Europe are planning to build new fleets of nuclear reactors. These are to be pressurised water reactors (PWRs) like those already under construction at Olkiluoto in Finland and Flamanville in France. The fuel will be U-235, a fissile isotope of uranium. Over 99 per cent of naturally occurring uranium is the non-fissile isotope U-238, so before it can be used the uranium has to be enriched, i.e. processed to increase the proportion of U-235. This is usually done by making uranium hexafluoride, which is a gas, and separating molecules with the different isotopes using either filters or centrifuges [1] (see also [2] Energy Strategies in Global Warming: Is Nuclear Energy the Answer? SiS 27). The process is technically challenging and expensive, which is why the western powers were so concerned when they learned that Iran was developing its own capacity for enrichment. For light water reactors such as PWRs, the proportion of U-235 is generally increased from about 0.72 per cent of the uranium mass to at least 3 per cent.

The enriched uranium hexafluoride is then chemically processed into uranium dioxide powder which is compressed into pellets, sintered into ceramic form, and loaded into tubes made from a zirconium alloy. These are then formed into fuel rods. The depleted uranium, i.e. what is left behind after the enrichment process, contains enough U-235 that it is hazardous and requires special handling and disposal. Depleted uranium is sometimes used for armour-piercing weapons because of its high density, and while war zones are not well suited to research and cancers can take a long time to develop, there is strong evidence that the minute particles formed when a shell fragments are carcinogenic [3].

It has been known for a long time that thorium, the element two places below uranium in the periodic table, can also be used as fuel for nuclear reactors. Thorium has only one naturally occurring isotope, Th-232, so there is no separation to be done. What is more, Th-232 is not fissile, so the ore is not dangerous to mine and process.

If, however, an atom of Th-232 absorbs a neutron and becomes Th-233, it emits two electrons and becomes U-233, which is fissile. The conversion can be done off-site and the U-233 used in a conventional reactor, or Th-232 can be used together with U-235, but there are also designs in which only thorium is used as fuel and the conversion takes place within the reactor itself.


 A Liquid Fluoride Thorium Reactor (LIFTR, pronounced ‘lifter’) consists of two main parts, a ‘core’ surrounded by a ‘blanket’. The core contains U-233 tetrafluoride in a fluoride salt. The blanket contains Th-232 tetratfluoride, also in a fluoride salt. Both are kept molten by the heat of the reactor. The core produces the energy output and also neutrons that cross into the blanket and convert more thorium to U-233. The uranium is then chemically separated from the unconverted thorium and transferred to the core to produce more energy and more neutrons [4].

One advantage of a liquid fuel is that it is much easier to separate the waste products. In particular, xenon, which is a serious problem in solid fuel reactors, just bubbles to the top and can be removed. As a result, all the thorium can be used, whereas in a conventional reactor the fuel rods have to be removed when only about 5 per cent of the U-235 has been used up because by that time the fuel rods have been degraded. Because so much of the fuel is left, the spent rods are highly radioactive.

Adding thorium to the blanket, transferring uranium to the core, and removing waste can all be done on a continuous basis. There is no need to shut the LIFTR down. The waste contains the same products as from a uranium reactor but in very different amounts, and it is far less hazardous. This is largely because it takes more neutron-capture to produce the transuranic elements such as plutonium from thorium than from uranium.

A reactor should have a negative temperature coefficient of reactivity, i.e. if it becomes hotter the reaction should slow down. The Chernobyl reactor did not, at least not under all circumstances, which was a major factor in its explosion. Modern uranium reactors are designed to ensure that they always do, but it is especially easy to arrange in a LIFTR because as the fuel gets hotter it expands, reducing the effective area for neutron absorption.

In almost all commercial uranium reactors in use today, water is the coolant and moderator. The water is maintained at high pressure to raise its boiling point and make it more efficient at transporting heat. The coolant of a LIFTR is a fluoride salt that remains liquid up to 1 400 °C at atmospheric pressure. This is well above the operating temperature of the reactor, about 800 °C. As a result, a LIFTR does not need the same very strong piping, pressure vessel and containment building. If a failure does occur it will be a manageable leak rather than an explosive release of radioactive material.

A LIFTR is also easily made safe against the possibility of a complete power failure, as happened at Fukushima. This is done by the simple device of having a freeze plug at the bottom of the core. Salt is kept below its melting point by an electric fan. If the fan stops for any reason, including of course a total failure of electric power, the salt melts and the contents of the core flow into a basin whose geometry ensures that the reactions will stop.

 In the 1960s the US Oak Ridge National Laboratory built an experimental molten salt reactor. This did not have the blanket that a LIFTR would use to produce U-233 but it allowed many of the other features to be tested during the 5 years it was operated [5]. The results were encouraging, but by this time the US had decided to concentrate on reactors using uranium dioxide. The Director of Naval Reactors, Admiral Rickover, had chosen this type to install in nuclear submarines, and the civil programme was able to share the research and the momentum. In the context of the cold war, the fact that uranium reactors produce a considerable amount of plutonium was seen as an advantage; today, of course, it is a drawback because of the hazard and the danger of proliferation of nuclear weapons.

In January this year, the Chinese Academy of Sciences announced a programme to develop what they call thorium molten salt reactors [6]. The aim is to have them in operation within 20 years. In the west, several private companies are planning LIFTRs.

Accelerator driven systems

Carlo Rubbia, a Nobel Prize winner and former director of the European nuclear physics establishment CERN, has proposed an alternative method for converting Th-232 to U-233 [7]. The proton beam from a particle accelerator impacts on a target of heavy metal and this produces the required neutrons. The energy output is proportional to the strength of the proton beam; in particular if the beam is turned off, the reactor stops.

Because they use thorium, accelerator driven systems have many of the same advantages as LIFTRs. They can be configured so as to use depleted uranium from conventional reactors, thus helping reduce the amount of radioactive waste that has to be disposed of.

Rubbia is now working with a private company AKER to develop the concept, which they call the Accelerator Driven Thorium Reactor (ADTR) [8]. They predict the time to market will be around 2030 and claim that the cost per unit of electricity will be “extremely competitive” with conventional nuclear reactors and other energy sources.

A contrary view

A year ago, the UK National Nuclear Laboratory (NNL) published a short assessment of thorium as a source of nuclear energy [9]. In their view, the advantages of thorium are very much overstated. Compared with uranium reactors they see no advantage in cost. They point out that a country like the UK, which has reserves of neither thorium nor uranium, would be no less dependent on imports.

The NNL argues that U-233 should be regarded as posing a high proliferation risk, claiming that any U-238 that was added could be separated out by centrifugation. They argue that there are only modest gains to be made in radiotoxicity, though they concede that a full thorium recycle does provide an incentive “in the long term.”

The NNL critique is addressed to the use of thorium in conventional reactors because they consider that the construction of new types such as high temperature reactors and accelerator driven systems would be viable only in the long term, forty years or more. The assessment briefly mentions the accelerator driven system and says nothing at all about the LIFTR.

Ironically, the NNL is highlighting a weakness of all nuclear power. The lead times are so long and the costs of construction so great, that once we have decided on a design we are locked into it for a very long time, while circumstances and alternative sources of energy can change out of all recognition.

To conclude

On the face of it, thorium looks a much better prospect than uranium. It certainly seems to be safer, and we are told it should produce electricity at about the same cost as a PWR, bearing in mind that it is very difficult to work out the true cost of existing nuclear power plants, let alone those that haven’t been built yet. To that extent, it is welcome news that China and India, both committed to nuclear power, are planning to use thorium, though the Indian reactors will not use thorium alone. And we might wish that our own governments were looking into thorium rather than being so firmly committed to the conventional PWR.

The real question is whether we need nuclear power at all. The nuclear lobby is forever warning us that if we do not start at once to build a new fleet of nuclear reactors, the lights will go out all over Europe. It is certainly true that we cannot go on relying heavily on fossil fuels and being profligate in our use of energy. But the vast amounts of time, effort and resources that we are being asked to spend on nuclear energy would be better spent on energy efficiency and renewables (see [10]  Green Energies - 100% Renewable by 2050, I-SIS publication).

Article first published 14/09/11


  1. Enriched Uranium, Wikipedia. 2/09/11
  2. Bunyard P. Energy strategies in global warming; Is nuclear energy the answer? Science in Society 27, 12-15, 2005.
  3. Busby C, Hamdan M and Ariabi E (2010). Cancer, Infant Mortality and Birth Sex-Ratio in Fallujah, Iraq 2005–2009. International. Journal of Environmental Research and. Public Health 7, 2828-2837; doi:10.3390/ijerph7072828
  4. Hargraves R and Moir R (2010), Liquid fluoride thorium reactors. American Scientist 98, 304-313.
  5. Molten salt reactor. Wikipedia. 2/09/11
  6. “Future nuclear reactors will be safe and will make use of all their fuel”, Xu Qimin, Wen Hui News, 28 January, 2011.
  7. Rubbia C (no date). Sub-critical thorium reactors, accessed 2 September 2011,
  8. Aker Solutions (2010). Accelerator Driven Thorium Reactor Power Station, accessed 2 September 2011,
  9. UK National Nuclear Laboratory (2010). The Thorium Fuel Cycle. accessed 2 September 2011, Accessed 2/09/11.
  10. Ho MW, Cherry B, Burcher S and Saunders PT. Green Energies: 100% Renewable by 2050. ISIS/TWN, London/Penang, 2009.

Got something to say about this page? Comment

Comment on this article

Comments may be published. All comments are moderated. Name and email details are required.

Email address:
Your comments:
Anti spam question:
How many legs on a spider?

There are 8 comments on this article so far. Add your comment above.

David Russell Comment left 15th September 2011 14:02:01
The above article neglects to mention the availability of Thorium vs Uranium. The nuclear weapons stockpile of the Soviet Union has been de-commissioned and broken down by the US over the last few years. The nuclear material has been recycled and used in nuclear reactors. This stockpile is almost gone, there being perhaps another year's worth left. The present World-wide uranium production being mined is not enough to supply all of the reactors in the world. Thorium is evidently more common than uranium and will therefore be cheaper. Watch for a huge spike in the cost of uranium in the near future. Some countries may not be able to afford the fuel causing many reactors to be shut down. It's going to be very interesting, coming on the heels of the economic mess in Europe now. Britain may be opening up those old fireplaces and burning coal sooner than you think.

tblakeslee Comment left 15th September 2011 14:02:37
Nuclear can be clean and safe but nothing that exists now is really clean or safe. There are two exciting developments that can qualify: One is the E-cat, developed by Andrea Rossi with his own money. He hopes to ship a 1 MW, container sized water heater in late October. Another exciting development by Lawrenceville Plasma Physics is about ready to try fusion of boron and hydrogen. If it works as planned it will produce no neutrons or radiation but will directly produce electricity. An exciting prospect. Here is an article I wrote about them:

Rory Short Comment left 15th September 2011 23:11:50
It is the long lead times for nuclear development mentioned in this article that really struck me. These lead times linked to the huge capital investment needed should surely make governments think about pursuing the development of the different technologies needed to harvest different sources of renewable energy. It is not surprising to me that the hazardous PWR technology that has generally been used for civil nuclear reactors up until now should have its origins in the US military. It is not surprising because the unspoken priority of most Western governments is military power not the long term health of their citizens.

Todd Millions Comment left 16th September 2011 18:06:57
Actual overall averages are recently sloping to total renewable in 20-25 year range,and this could be accelerated.Costings(Capital-simple payback) appear to be closer to a thousandths than hunderds of times cheaper than nuke. Remembre too that-this has being happening under constant oil/nuke mafia sabotage and co opting. Both time and costings could be greatly reduced,probably to limits of 10 years and no more than present fuel expendetures for one year in an industrial soceity(averaged). This would be particularly true if economic slowdown curtails non essential energy use. So the amount of oil extractable from greasy quantitve eased bankers rendered carcasses should be explored.

David Shields Comment left 19th September 2011 01:01:52
Will you be analysing the Rossi Reactor?

Mae-Wan Ho Comment left 19th September 2011 01:01:40
David Shields, we have already reported on the Rossi reactor some time ago, based on information available. See here:

Philip Ward Comment left 18th February 2012 08:08:43
It seems to me that most objective observers reckon that Andrea Rossi's "cold fusion" apparatus doesn't work. I fail to see why ISIS is so uncritical of these maverick scientists. One of the issues not addressed in this article on Thorium reactors is the materials science angle. I suspect that finding a durable container for molten fluorides is going to be a pretty tall order. A short-term demonstration plant is one thing, but getting a reactor to last for, say, 40 years under such conditions, is another. Anyone who has seen salt baths, used for heat-treating metals, will know that molten halides are very corrosive.

Mark Campey Comment left 19th October 2013 06:06:48
The MSR would be a fully proven technology today if it wasn't for President Nixon shelving it's further development in the 1970's in favour of the plutonium bomb producing reactors of today. It's encouraging to see the 4th. Gen. LFTR and other variants being developed by China and India but if the UK were ever to rid itself of the stockpile legacy of medium/high level radioactive waste then MSR is the way to go about it.