We spent a large portion of Friday, September 9th at the Department of Energy and Climate Change, where we were privileged to discuss details of LFTR technology, including the details of the power conversion system with top DECC officials. Many DECC staff members then attended the open presentation on LFTR.
My public presentation began with an introduction of our mission statement at Flibe Energy: “to supply the world with affordable and sustainable energy, water and fuel.” I talked about the significance of Alvin Weinberg’s work and noted that Weinberg’s son Richard was present in the audience.
Weinberg had a vision of farming using water desalinated by the energies of thorium and the molten-salt reactor, but that vision was at odds with the stated policy of the United States under Richard Nixon to develop a plutonium fast-breeder reactor, so Weinberg’s research was cancelled and he was fired.
We formed Flibe Energy to fulfill that vision of a world set free by the energies of thorium. Energy independence is possible and affordable. A lifetime’s supply of energy can be held in the palm of your hand if LFTR technology is utilized.
Nature gave us three options for nuclear fission energy. There’s not enough U-235. The traditional thought has been to use plutonium-239 bred from uranium-238. But thorium was the third option–thorium bred into uranium-233.
There was a significant difference between the three options. Two were suitable for nuclear weapons, the third was not. This was because of the presence of uranium-232 which produces gamma radiation in its decay process, leading thorium to be set aside during the Manhattan Project. This had significant implications for the future development of thorium.
Thorium is made to release energy by absorbing a neutron, forming first protactinium-233 and then uranium-233. Uranium-233 releases enough neutrons when absorbing a thermal neutron to sustain the conversion of thorium to uranium-233 and its energy release.
The secret to using thorium effectively is to use “flibe”–a mixture of lithium fluoride (LiF) and beryllium fluoride (BeF2) that is the basis for thorium-burning molten-salt reactors. Flibe Energy derives its name from this material.
The LFTR is very safe in an accident scenario due to the use of liquid fuel. Decay heat after shutdown poses a real challenge for solid-fueled reactors because coolant must be introduced to the fuel. In fluid-fueled reactors, a “freeze plug” contains the fluid in the reactor, preventing the core fluid from draining. If power or cooling is lost, the fluid fuel melts the plug and drains into a drain tank where heat rejection to the environment is maximized rather than minimized. In simpler terms, the reactor can “downshift” into a new configuration in the event of a shutdown.
Solid-fueled reactors, and especially pressurized solid-fueled reactors have real challenges removing decay heat. You have to have a variety of emergency core coolant injection systems at different pressures, redundant systems, emergency power systems. All of it can be done, but it’s not cheap. In contrast, the simple, low-pressure approach used by a liquid-fluoride reactor is affordable and attractive.
Examining the large containment building of a typical pressurized-water reactor like Watts Bar in the US shows that the building is much bigger than the reactor due to the need to contain the steam that would evolve in the event of a pressure release. By contrast, a liquid-fluoride reactor doesn’t operate at high pressure and doesn’t have a coolant that undergoes a phase change like water, so the containment building can be much smaller, close-fitting, and less expensive.
The theoretical thorium fuel cycle is transformed into hardware in the liquid-fluoride reactor design. A “blanket” of flibe salt containing thorium jackets the core of the reactor, which has flibe salt containing uranium-233. As new uranium-233 forms in the blanket, it is chemically removed and introduced into the core, continuously refuelling the reactor and compensating for the fuel that was consumed. The fission of the fuel, in turn, generates the neutrons that convert thorium into uranium-233 in the blanket. Fluoride volatility is the secret to this chemical transfer. This works due to the simple chemical fact that uranium has two valence states (+4 and +6) while thorium has only one (+4). Only in fluoride chemistry is this simple and proven chemical approach possible.
The pumping action of the core salt continuously re-homogenizes the core composition. This is a huge advantage over solid fuel, where each location in the fuel element sees a different flux from the reactor and has different burnup and composition.
Each of the chemical steps anticipated for the LFTR (fluorination, reduction, and distillation) was actually demonstrated in the operation of the Molten-Salt Reactor Experiment at Oak Ridge National Lab from 1965 to 1969.
Electricity is generated from the LFTR by pumping the core salt through the reactor, where heat is deposited in the salt by fission reactions. The core salt then leaves the core and gives up its heat to a coolant salt, probably just flibe without any other additions, which is turn gives up its heat in a heat exchanger with a gaseous working fluid for a gas-turbine engine. The gas turbines that would be used in a LFTR are analogous to the gas turbines used in airplanes, but instead of heating gas through the combustion of jet fuel, the gas is heated from the hot salt. The turbine in the gas turbine drives both a generator and compressor. The generator produces electrical power that is then distributed across the grid.
The remaining heat in the gaseous working fluid is rejected to the environment, and if the reactor is located near seawater, the waste heat alone could be used to drive a desalination process, producing another valuable commodity from the operation of a LFTR. That is a remarkable additional benefit from operating a LFTR that isn’t possible with reactors that use steam turbines for power conversion.
The temperatures that can be achieved in a LFTR are much higher than can be achieved in a reactor using pressurized water. This allows a reactor to convert heat to electricity at efficiencies from 45-50%, unlike the 30-35% possible in a reactor cooled by pressurized water.
The possible options for reactor coolants could be visualized in a 2×2 matrix, with temperature on one axis and pressure on the other. Water is a coolant that operates at moderate temperatures but at high pressure. Gas is another coolant that operates at high temperatures but at high pressures. Liquid metals cool reactors at moderate temperatures but have the advantage of operating at low pressures. But only liquid-salts can cool reactors at high temperatures AND at low pressures, the most desirable possible situation.
Another matrix of fuel options could be considered, with two different neutron spectra considered (fast and thermal) and the three different fuel options (U-235, U-238/Pu-239, and Th-232/U-233) yielding six possible combinations. A moment of consideration leads one to realize that only three of these six combinations are both possible and attractive.
Yet another matrix where the four coolant options are mapped against the three feasible and attractive fuel cycle options yields twelve different possibilities for reactor designs. Another round of consideration shows that only five or maybe six of these twelve possibilities are feasible and attractive. In these five or six configurations we find essentially all of the world’s reactor designs.
A waste comparison between a light-water reactor and a LFTR highlights the differences in fuel efficiency. Most of the uranium is rejected as waste from the LWR in the enrichment facility. Of the uranium that remains and is irradiated in the reactor, most emerges unconsumed and mixed with transuranics and fission products. By contrast, the fuel cycle and fuel form of the LFTR permits almost complete consumption of the thorium fuel added to the reactor.
If an attempt was made to power the world by thorium, only about 7000 tonnes of thorium would be needed in a LFTR to match today’s world energy output. By contrast, about 65,000 tonnes of uranium is mined each year to generate only a small fraction of world energy output.
Nuclear reactors based on liquid-fluoride technology were successfully built and operated at the Oak Ridge National Laboratory, with the Molten-Salt Reactor Experiment being the most important example. The MSRE ran from 1965 to 1969 and was the first and only reactor to run on all three fissile fuels. It also demonstrated most of the key safety aspects of a LFTR as well as demonstrated on a small scale the chemical processing techniques needed for a larger reactor.
Transuranics accumulate almost immediately in a uranium-based reactor, even more so in a thermal-spectrum uranium reactor, which is what nearly all reactors are today. By starting with thorium, there are two options for fission before forming the first transuranic nuclide, thus limiting the formation of transuranics drastically over uranium-fueled reactors. Plutonium-238 could be produced in small quantities in the thorium fuel cycle from the “unburned” results of the thorium fuel cycle, that will help NASA explore deep space.
Spent nuclear fuel is a real public concern. Even if Yucca Mountain had been built, the accumulation of spent nuclear fuel would have caused it to reach its statutory capacity rather quickly. An expansion of conventional nuclear power would simply accelerate the need for more geologic repositories like Yucca Mountain. But we’re not even planning to build the first Yucca Mountain.
Conventional aqueous reprocessing (PUREX) is complicated and messy. It was designed to extract very pure plutonium, not to have simple and safe reprocessing for a fuel cycle. By contrast, the techniques needed to process fluoride fuel are straightforward and simple–fluorination, reduction, distillation, and purification–all of which are rad-hard. Simplifying reprocessing was one of the basic motivations that drove Eugene Wigner to consider fluid fuel and one of the reasons he encouraged Alvin Weinberg to pursue this path. Fluorination techniques will not only benefit fluoride fuel, but they could be used to transform existing spent uranium oxide fuel into fluorides, which would simplify their further processing. Fluorination of uranium oxide is done every day as part of the preparation of conventional nuclear fuel–its extension to spent nuclear fuel is not a radical departure.
I showed an example of zirconium alloy cladding used in conventional nuclear fuel.
If we could exclude actinides from the waste stream, radiotoxicity would fall dramatically over time and within three hundred years the radiotoxicity of fission products would be less that the radiotoxicity of uranium ore. This is an achievable goal with the LFTR approach to fuel management.
I showed the Java simulation that I developed to model the evolution of spent nuclear fuel over time that I originally showed in my Google TechTalk “Is Nuclear Waste Really Waste?” Strontium-90 and cesium-137 dominate the consideration of radiotoxicity for the first few hundred years but after that it is dominated by plutonium and americium. If you can use the thorium fuel cycle to avoid the production of plutonium and americium and then retain the nuclear fuels (thorium and U-233) through effective recycling, you can reduce the radiotoxicity of waste drastically.
The existing fuel cycle wastes almost all of the energy inherent in the uranium ore, leaving a variety of waste streams. Thorium in a LFTR, by contrast, doesn’t even need to mined since it is already anticipated as a byproduct of rare-earth mining. It will be consumed efficiently in a LFTR, and even the fission products resulting from the reaction will likely see useful application in a variety of applications.
The technical advantages of LFTR include the inherent, passive safety possible in the reactor through the fluid fuel form and low-pressure operation. The high temperature operation of the LFTR offers high conversion efficiencies and air cooling offers siting flexibility away from water bodies. The reactor will have extremely low fuel preparation costs and no fuel fabrication costs. Fluoride salt chemistry is an excellent match with the thorium/uranium fuel cycle. The fluoride salts are chemically stable and impervious to radiation damage, enabling unlimited fuel burnup and continuous solvent recycling. Uranium-233 fissile fuel is highly unsuitable for weapons diversion and is readily downblended with depleted uranium in an emergency.
The technical challenges that lie before LFTR developers like Flibe Energy include the simple fact that salts can be aggressive to metals and other structural materials and that high-temperature operation naturally presents design challenges. But more importantly, the technology base has largely stagnated for 40 years. If we look for help from the conventionla nuclear industry we must remember that LFTR technology is very different from the water-cooled, uranium-fueled reactors that are the basis for current nuclear power generation, and is not yet fully understood by regulatory agencies and officials.
I skipped most of the rest of the slides and concluded the presentation to take questions.
The first question concerned the size of the reactors that we intend to develop. I responded that we are focused on powering military bases and these will likely be in the 20-50 MW power range. But success in that arena would lead to utility-class reactors with power levels above 100 MW.
The second question concerned the sustainability of thorium as an energy resource. I pointed to Weinberg’s 1959 paper where he estimated that thorium would sustain a population of 7 billion at Western standards of living for about 30 billion years.
The third question concerned the chemistry techniques of the reactor, and particularly the regulatory aspects. I mentioned the hope I have for progress in the regulatory arena, particularly moving beyond the light-water reactor as the reference system.
The fourth question concerned materials selection in the reactor and what was the state of materials science for the reactor. I went through the residual issues that remained after the operation of the MSRE, particularly tellurium cracking, tritium retention, and radiation-induced embrittlement in the Hastelloy alloy.
The fifth question concerned advocacy and investment, to which I responded how we are working on our business case. Our goal is for Flibe Energy to be a privately-funded venture.
The final question was about beryllium toxicity and whether it was an issue. I pointed out the advantages to the system that beryllium fluoride conferred despite its toxicity and that we planned to work around it.
After dinner at an Italian ristorante just off Trafalgar Square, I joined John Durham, Laurence O’Hagan and Richard Weinberg for an evening of reflection to the strains of a violin concerto at the church of St. Martin-in-the-Fields.