Tuesday, 25 June 2013

China’s faces technical challenges to develop thorium based nuclear power

By 2017 the Chinese are expecting to be running an experimental molten salt reactor, whilst other countries and commercial organisations, are being left behind. China’s first step towards liquid fuelled thorium reactors is, in essence, to reproduce the work done by Oak Ridge National Nuclear Laboratory (ORNL) in the sixties, when ORNL built and ran the Molten Salt Reactor Experiment MSRE. But what are the problems and challenges facing the Chinese as front runners in the drive to start up a liquid fuelled thorium reactor (LFTR)? 
Alvin Weinberg
Under Director Alvin Weinberg, back in the sixties and seventies a number of them were identified by the MSRE at Oak Ridge National Nuclear Laboratory.  In the test environment some of these challenges seemed to have answers, whilst others are still to be solved.  They are all written up in reports and published papers, now freely available on the internet, or at the ORNL archives.  You can be sure that they have been well studied by the many Chinese visitors to Oak Ridge.
Materials Hastelloy - N
The MSRE ran at around 670 deg C and the combination of high temperature and radiation made the selection of materials critically important.
The Molten Salt Reactor Experiment
Based on laboratory tests Hastelloy –N (originally known as INOR-8), a nickel alloy with molybdenum and chromium developed by ORNL, was chosen for the reactor containment vessel and for piping and heat exchangers. 
 Chemical Composition of Hastelloy N™
•     Carbon 0.04-0.08
•     Chromium 6.00-8.00
•     Molybdenum 15.00-18.00
•     Iron 5.00 max
•     Manganese 0.8 max
•     Aluminum + Ti 0.5 max
•     Boron 0.01 max
•     Sulfur 0.03 max
•     Nickel remainder

During construction questions arose concerning the stress-rupture life and fracture strain, which were found to be drastically reduced by thermal neutron irradiation
An out-of-pile corrosion test program was carried out for Hastelloy-N[9] which indicated extremely low corrosion rates at MSRE conditions.  Capsules exposed in the Materials Testing Reactor showed that salt fission power densities of more than 200 W/cm3 had no adverse effects on compatibility of fuel salt, Hastelloy-N, and graphite.  With their chosen salt mixture (see below) and at the operating temperature, they found that Hastelloy N was adequately resistant against corrosion and embrittlement for its design life in the MSRE.  Since this was an experimental reactor, however, this is not the same as saying that it will last the 30 years necessary for a commercial reactor.
Later experience showed that the mechanical properties of Hastelloy – N do deteriorate as a result of exposure to thermal neutrons but that the addition of small amounts of titanium and hafnium significantly improve its performance. to select the most resistant alloy much more long term operational experience is needed.
MSRE core
Graphite Moderator
As in Light Water Reactors, the graphite moderator blocks were found to swell and crack in the high radiation environment.  As a minimum this creates a maintenance issue since the blocks have to be replaced.  More seriously, depending on the reactor design, it can block fluid pathways and lead to hot spots.
Research into graphite production methods and coatings was carried out by ORNL but it remained an unsolved problem in the early seventies.
There are many ways to design a nuclear reactor and other possibilities of moderating the neutron flux may need to be tried.  
Alternatively one could opt for an un-moderated molten salt reactor design working in the fast spectrum.  Such a reactor would be able to burn actinides and eliminate almost entirely the production of highly radioactive waste having a long half life.  This is the subject of research at Grenoble, France but it has never been attempted on the scale of an experimental reactor so far.
Tritium removal
Tritium is generated in small quantities within the molten salt as a breakdown product of lithium 6. Tritium, or hydrogen-3, is made by bombarding lithium-6 (6Li) with a neutron (n). This neutron bombardment will cause the lithium-6 nucleus to fission, producing helium-4 (4He) plus tritium (3T) and energy.
Tritium is highly radioactive and must be sequestered in a secure storage facility.  If, however, the lithium salt used is purified to 99.995% Lithium -7, then the quantities of tritium produced amount to only a few hundred grams per year from a 1GWe reactor. 
At the temperatures concerned, however, tritium can pass through the heat exchanger material and get into the secondary cooling loop from where it could escape into the environment.  ORNL developed a secondary loop coolant system that would chemically adsorb the few hundred grams of residual tritium to a less mobile form, so that it could be trapped and removed from the secondary coolant, rather than diffusing into the turbine power cycle.  ORNL calculated that this technique would, by itself, reduce tritium emissions to the environment to acceptable levels 
The tritium was then extracted and successfully oxidized over copper oxides and captured in a pair of water bubblers.  No tritium was found in the exhaust gases.  
The process of capturing the already low concentration of tritium from the molten salt was found to be not very straightforward and more work is need to develop this or other process options.
The Lithium Problem
ORNL chemists did extensive pioneering work to find the best mixture of salts to give the desired physical and chemical characteristics.  Many were tested but they opted for Flibe, a mixture of lithium and beryllium fluorides.
Present day research programmes may need to revisit the choice of lithium because it’s necessary to use only lithium 7 and to remove the 7.5 % of the naturally present lithium 6 isotope.  Otherwise, depending on the concentration, lithium 6 would either transmute in the reactor into tritium, as mentioned above, or prevent the reactor from starting up due to its high neutron absorption characteristics.
But this is only part of the lithium problem.   At present only two countries are carrying out lithium isotope separation, China and Russia.
Since lithium 6 can be used to make tritium its production is strictly monitored. 
Interestingly in the US Castle Bravo hydrogen bomb, tested at Bikini Atoll on 1st March 1954, it was assumed that only the 30% of lithium-6 present in the lithium charge would react in a hydrogen bomb, but the contribution from an overlooked reaction due to the presence of the remaining 70% of lithium-7  caused an unexpected increase in yield of 250%,  making the 15 Megaton explosion the largest thermonuclear device tested at that date.
The US stopped separation of these isotopes, using the mercury amalgam method, a number of years ago due to concerns about its toxicity and the fact that large quantities of mercury had been lost from the Oak Ridge National Nuclear Laboratory inventory in an unknown manner.
So the lithium problem presents a difficulty to any private sector group not working in co-operation with a government having a form of lithium isotope separation technology. 
Other salt mixtures not using lithium are possible and many were tested at ORNL in the laboratory but so far they are completely untested in reactors and may affect the choice of reactor materials.
Storage of Solidified Salts
At the end of the MSRE programme the molten salts were stored as a solidified material.  It was found that fluorine and uranium hexafluoride were continuously released by radiolysis.   As a temporary measure the solidified contents were periodically reheated to induce recombination, but eventually the uranium was removed from the salts in an expensive and challenging cleanup programme.  A solution to this problem would be to remove the uranium prior to storage of the solidified salt by sparging the molten salt with fluorine gas, to create uranium hexafluoride which can be re-introduced into the reactor. 
Plutonium must also be removed to prevent radiolysis, which can be done by the addition of sodium carbonate to create plutonium dioxide. 

This is by no means a full list of the issues which arose during the operation of the MSRE. (Section 7 of the Weinberg Foundation's recent Report on Thorium-Fuelled Molten Salt Reactors gives several more). So there are plenty of technical challenges to address as the Chinese firstly repeat the MSRE experiment, and then extend it into larger reactor designs.  Personally I’m extremely pleased that the Chinese have the vision, the money and the forward thinking to restart and extend the pioneering work that was done at ORNL in the sixties and seventies   The use of thorium reactors running at atmospheric pressure will be much safer and will produce much less radioactive wastes than the current 50 year old designs of the existing Light Water Reactor fleet.  For me safe and abundant nuclear power is the only way of avoiding the prospect of runaway global warming, because conservation of energy, and the intermittent nature of the main renewable sources, cannot provide more than a part of the answer to the rapidly increasing global demand for energy.

Under Weinberg's leadership ORNL had a world leading  combination of nuclear scientists, engineers, chemists and metallurgists all working under one organisational umbrella.  They were capable of taking any brief concerning nuclear power generation and turning it into reality quickly and efficiently.  The USA allowed much of this expertise to dissipate when the Nixon administration fired Alvin Weinberg because he disagreed with the administration’s chosen programme of liquid metal cooled fast breeder reactors.  By the time that programme was closed down the USA  had forgotten about the highly promising liquid fluoride technology that is once again coming to the fore. 

It’s a pity that the US, like other western countries, has not yet found the courage or political will to overcome the inertia inherent in the nuclear establishment, which is committed to the Pressurised Light Water Reactor and the uranium–plutonium fuel cycle.   Coming from the UK, I’ve seen many examples over the years where pioneering research is done by underfunded organisations, only to be developed by other better placed countries.  It’s a symptom of the West’s debt-burdened economic decadence in the face of the challenge of the Far East, but in fifty years time, if all goes well, we’ll thank the Chinese for picking up and developing ORNL’s 50 year old Liquid Fluoride Reactor research, even if we have to pay the licence fees to China for our neighbourhood power station.

Although there are some signs of increasing openness, and their thorium programme is a good example, it’s also worrying that the Chinese, as communists, do not have a fully representative form of government.  I sincerely hope that the Chinese thorium programme doesn't get caught up in some sort of revolutionary “Chinese Spring” that could set their programme back decades.

Here are some other posts that might interest you if you have read this far.

An outline of the Chinese thorium program given by Kun Chen,Professor and Deputy Director, Department of Nuclear Safety and Engineering, Shanghai Institute of Applied Physics.

A detailed review of Robert Hargraves' book "Thorium- energy cheaper than coal".

China has a virtual monopoly on the production of rare earth elements of which process thorium is a byproduct.


  1. John,

    As a mechanical engineering doctoral candidate, I am working with heat storage for concentrated solar thermal power systems. One common approach to CSP is to use a central power tower heating a molten salt working fluid, which is used for many of the same reasons as in the MSR: high heat capacity, stability at high temperatures, atmospheric pressure capability.

    Given this similarity, I am aiming to push toward a hybridized CSP/MSR setup. In the long-run, such a system could stabilize the intermittent nature of solar power while reducing the fuel costs of nuclear (currently low, but will increase as the MSR becomes more prevalent).

    Aside from FLiBe and FLiNaK, what salt compositions did the MSRE team identify as being suitable for fission? Both CSP and MSR technologies could benefit from developing a joint market for a crossover salt, leading to an increase in production and drop in price.

    -Barry Osterman-Burgess

  2. Barry,
    in answer to your question, the best I can offer is to point you to this paper from the Oak Ridge archives http://www.energyfromthorium.com/pdf/FFR_chap12.pdf . It takes a detailed look at different salt mixtures and their chemistry. It's a very old piece of work though and there may well be more recent papers published elsewhere.
    I expect you are familiar with the Andasol plant in Andalucia, which uses technology for energy storage similar to that which you describe. Since you are working in the field you are probably aware of other schemes in the construction or planning stages.

  3. I've noticed that the temperatures of these molten salt systems are near the roasting temperature of lime for portland cement. Now that means introducing a lot of carbon into the atmosphere by burning a fuel. I wonder out loud if a roasting system can be incorporated in a diverted secondary loop (may be more practical in the solar case...)

  4. Hi Billy,
    Yes, it’s clear that molten salt systems can provide high temperature energy for process industries. Others have written about this as one of the advantages of Molten Salt Reactors such as Robert Hargraves in chapter 7 of his book ”Thorium – energy cheaper than coal”, where he also considers using energy from this source to synthesize ammonia for use as a vehicle fuel, for desalination and for generating hydrogen.

    I agree that all this is possible and I would like to see it happen but, just taking the case of cement plants, they are usually very near the quarries, they often have very little staff and minimal security. In fact they run on a shoestring in a very competitive market.
    As I comment in my review of Hargraves’ book http://johnpreedy.blogspot.fr/2012/11/thorium-energy-cheaper-than-coal.html it would seem to me to need a number of large energy intensive industrial sites grouped around a process heat generating plant which is run by a company specializing in nuclear power plant operation. Otherwise I doubt whether such process industries would want to consider taking on the regulatory burden, the extra cost and the expert staff necessary to operate such a facility. One could perhaps imagine a plant for synthesizing vehicle fuels run by one of the large oil companies, who could subcontract the construction and operation of a molten salt reactor to another company, but I think that it would have to be a very large scale investment in order to justify the decision NOT to burn oil as an energy source. It really would have to provide reliable very cheap energy 24/7!

    1. Thanks for the reply. I'm happy to hear others are thinking about best uses and multiple uses of that abundant energy source. it seems much broader planning particularly in the infrastructure that supports the power plant can improve synergies with other industries. For instance, planning a rail spur or hub at or near the plant could make transport of materials like lime practical. I recently heard of a plan to ship frozen chicken to China for processing then ship it back as finished product... I think some graph theory applied to existing transport routes and industries can result in optimal placement of plants, their new codependent industries and changes to transport paths.