Introduction and Basic Principles

The generation and use of energy is central to the maintenance of organization. Life itself is a state of organization maintained by the continual use of sources of energy. Human civilization has reached the state it has by the widespread use of energy, and for the large fraction of the world that aspires to a higher standard of living, more energy will be required for them to achieve it.

Therefore, I embrace the idea that we need energy, and probably need much more of it than we currently have. We should never waste energy, and should always seek to use energy efficiently as possible and practical, but energy itself will always be needed.

This weblog is about the use of thorium as an energy source of sufficient magnitude for thousands of years of future energy needs. Thorium, if used efficiently, can be converted to energy far more easily and safely than any other energy source of comparable magnitude, including nuclear fusion and uranium fission.

Briefly, my basic principles are:

1. Nuclear reactions (changes in the binding energy of nuclei) release about a million times more energy than chemical reactions (changes in the binding energy of electrons), therefore, it is logical to pursue nuclear reactions as dense sources of energy.

2. Changing the binding energy of the nucleus with uncharged particles (neutrons inducing fission) is much easier than changing the nuclear state with charged particles (fusion), because fission does not contend with electrostatic repulsion as fusion does.

3. Naturally occuring fissile material (uranium-235) will not sustain us for millennia due to its scarcity. We must fission fertile isotopes (uranium-238, thorium-232) which are abundant in order to sustain energy production for millenia. Fertile isotopes such as U-238 and Th-232 basically require 2 neutrons to fission (one to convert, one to fission), and require fission reactions that generate more than 2 neutrons per absorption in a fissile nucleus.

4. For maximum safety, nuclear reactions should proceed in a thermal (slowed-down) neutron spectrum because only thermal reactors can be designed to be in their most critical configuration, where any alteration to the reactor configuration (whether through accident or intention) leads to less nuclear reactions, not more. Thermal reactors also afford more options for achieving negative temperature coefficients of reactivity (which are the basic measurement of the safety of a nuclear reactor). Reactors that require neutrons that have not been slowed significantly from their initial energy (fast-spectrum reactors) can always be altered in some fashion, either through accident or intention, into a more critical configuration that could be dangerously uncontrollable because of the increased reactivity of the fuel. Basically, any fast-spectrum reactor that is barely critical will be extremely supercritical if its neutrons are moderated in some way.

5. "Burning" uranium-238 produces a fissile isotope (plutonium-239) that "burns" inefficiently in a thermal (slowed-down) neutron spectrum and does not produce enough neutrons to sustain the consumption of uranium-238. "Burning" thorium-232 produces a fissile isotope (uranium-233) that burns efficiently in a thermal neutron spectrum and produces enough neutrons to sustain the consumption of thorium. Therefore, thorium is a preferable fuel, if used in a neutronically efficient reactor.

6. Achieving high neutronic efficiency in solid-fueled nuclear reactors is difficult because the fuel sustains radiation damage, the fuel retains gaseous xenon (which is a strong neutron poison), and solid fuel is difficult to reprocess because it must be converted to a liquid stream before it is reprocessed.

7. Fluid-fuel reactors can continuously strip xenon and adjust the concentration of fuel and fission products while operating. More importantly, they have an inherently strong negative temperature coefficient of reactivity which leads to inherent safety and vastly simplified control. Furthermore, decay heat from fission products can be passively removed (in case of an accident) by draining the core fluid into a passively cooled configuration.

8. Liquid-fluoride reactors have all the advantages of a fluid-fueled reactor plus they are chemically stable across a large temperature range, are impervious to radiation damage due to the ionic nature of their chemical bond. They can dissolve sufficient amounts of nuclear fuel (thorium, uranium) in the form of tetrafluorides in a neutronically inert carrier salt (lithium7 fluoride-beryllium fluoride). This leads to the capability for high-temperature, low-pressure operation, no fuel damage, and no danger of fuel precipitation and concentration.

9. The liquid-fluoride reactor is very neutronically efficient due to its lack of core internals and neutron absorbers; it does not need "burnable poisons" to control reactivity because reactivity can continuously be added. The reactor can achieve the conversion ratio (1.0) to "burn" thorium, and has superior operational, safety, and development characteristics.

10. Liquid-fluoride reactors can retain actinides while discharging only fission products, which will decay to background levels of radiation in ~300 years and do not require long duration (>10,000 year) geologic burial.

11. A liquid-fluoride reactor operating only on thorium and using a "start charge" of pure U-233 will produce almost no transuranic isotopes. This is because neutron capture in U-233 (which occurs about 10% of the time) will produce U-234, which will further absorb another neutron to produce U-235, which is fissile. U-235 will fission about 85% of the time in a thermal-neutron spectrum, and when it doesn't it will produce U-236. U-236 will further absorb another neutron to produce Np-237, which will be removed by the fluorination system. But the production rate of Np-237 will be exceedingly low because of all the fission "off-ramps" in its production.

12. We must build thousands of thorium reactors to displace coal, oil, natural gas, and uranium as energy sources. This would be impractical if liquid-fluoride reactors were as difficult to build as pressurized water reactors. But they will be much simpler and smaller for several reasons. They will operate at a higher power density (leading to a smaller core), they will not need refueling shutdowns (eliminating the complicated refueling equipment), they will operate at ambient pressure and have no pressurized water in the core (shrinking the containment vessel dramatically), they will not require the complicated emergency core cooling systems and their backups that solid-core reactors require (because of their passive approach to decay heat removal), and their power conversion system will be much smaller and power-dense (since in a closed-cycle gas turbine you can vary both initial cycle pressure and overall pressure ratio). In short, these plants will be much smaller, much simpler, much, much safer, and more secure.

That said, I am not an apologist for the nuclear industry. I think that a fundamental mistake was made when thorium was overlooked as the prime nuclear fuel in favor of uranium, and this blog is an attempt to explain my position on that topic. In such a position, I think I stand in some good company. Dr. Alvin Weinberg, former director of the Oak Ridge National Laboratory and inventor of the pressurized-water reactor (he holds the patent) said in 1970:

The achievement of a cheap, reliable, and safe breeder remains the primary task of the nuclear energy community. (In expressing this view, I suppose I betray a continuing frustration at the slow progress of fusion research, even though the Russian success with the tokamak has quickened the pace.) Actually not much has changed in this regard in 25 years. Even during World War II, many people realized that the breeder was central. It is only now, with burner reactors doing so well, that the world generally has mobilized around the great aim of the breeder.

As all readers of Nuclear Applications & Technology know, the prevailing view holds that the LMFBR is the proper path to ubiquitous, permanent energy. It is no secret that I, as well as many of my colleagues at ORNL, have always felt differently. When the idea of the breeder was first suggested in 1943, the rapid and efficient recycle of the partially spent core was regarded as the main problem. Nothing that has happened in the ensuing quarter-century has fundamentally changed this.

The successful breeder will be the one that can deal with the spent core most rationally—either by achieving extremely long burnup, or by greatly simplifying the entire recycle step. We at Oak Ridge have always been intrigued by this latter possibility. It explains our long commitment to liquid-fueled reactors-first, the aqueous homogeneous and now, the molten salt.

The molten-salt system has been worked on, mainly at Oak Ridge, for about 22 years. For the first 10 years, our work was aimed at building a nuclear aircraft power plant. The first molten-salt reactor, the Aircraft Reactor Experiment, was described in a series of papers from Oak Ridge that appeared in the November 1957 issue of Nuclear Science and Engineering.

The present series of papers reports the status of molten-salt systems, and particularly the experience we have had with the Molten-Salt Reactor Experiment (MSRE). The tone of optimism that pervades these papers is hard to suppress. And indeed, the enthusiasm displayed here is no longer confined to Oak Ridge. There are now several groups working vigorously on molten salts outside Oak Ridge. The enthusiasm of these groups is not confined to MSRE, nor even to the molten-salt breeder. For we now realize that molten-salt reactors comprise an entire spectrum of embodiments that parallels the more conventional solid-fueled systems. Thus molten-salt reactors can be converters as well as breeders; and they can be fueled with either 239Pu or 233U or 235U.

However, we are aware that many difficulties remain, especially before the most advanced embodiment, the Molten-Salt Breeder, becomes a reality. Not all of these difficulties are technical. I have faith that with continued enlightened support of the US Atomic Energy Commission, and with the open-minded, sympathetic attention of the nuclear community that these papers should encourage, molten-salt reactors will find an important niche in the unfolding nuclear energy enterprise.

Weinberg's faith in the AEC was unjustified, for just a few years later they moved to kill the liquid-fluoride reactor in favor of the liquid-metal fast breeder. I think this was (and is) a mistake, for only in the liquid-fluoride reactor can we find the safety, economy, and efficiency needed to unlock the potential of thorium energy for tens of thousands of years.