The Molten Salt Reactor Family: Fuel

I intend to offer a series of posts designed to explain the sometimes bewildering complexity of Molten Salt Reactor Technology. This first post explains two nuclear fuel breeding cycles.

Rather than offering a single potential reactor design, the Molten Salt Reactor (MSR) idea offers a large number of design options, each of which would require a significant amount of research, before a prototype reactor could be built. The Molten Salt Reactor designer is faced with a bewildering number of elective choices, each offering a set of advantages and disadvantages. Each choice that the designer makes will dictate a number of design features some of which require further choices.

Lets start with nuclear fuel. My father first demonstrated that not only U-235 but also Pu-239 could be used as a reactor fuel in MSRs. During the ORNL Molten Salt Reactor experiment Oak Ridge scientists tested the use of the three fissionable materials that can be used as nuclear fuels, Plutonium-239 (Pu-239), Uranium-235 (U-235) and Uranium-233 (U-233). Once during the operation of the Molten Salt Reactor Experiment (MSRE) they used all three potential fuels in the reactor at the same time.

Of the three potential fuels, U-233 had some significant advantages. Neither U-235 nor Pu-239 produced enough neutrons per neutron hit, to support breeding more nuclear fuel at a slow (thermal) neutron speed range. U-233, produced by breeding thorium did produce enough neutrons to breed thorium at a slow temperature range. We will see that this offers a very large advantage. U-235 is not efficiently produced by breeding, while Pu-239 can only be produced in the breeding range with fast neutrons.

Breeding means that for every fuel atom used in the nuclear process, at least one new fuel atom is produced. Thus in a plutonium fast breeder, if a neutron strikes a plutonium atom, it is very likely to fission into two smaller atoms, almost always with three neutrons left over. Those neutrons will be moving fast and will contain a lot of energy. Fast neutrons are more likely to produce fission in plutonium atoms than slow neutrons. Neither U-235 nor Pu-239 produce enough neutrons to maintain breeding if they encounter a slow (also called thermal) neutron. Thus Plutonium can only be produced as a nuclear fuel in so called fast reactors. There are, as we shall see, some major disadvantages to fast reactors.

Fast reactors are often thought of as having liquid sodium as their coolant, although liquid lead, and a liquid lead-bismuth mixture have also been used as a coolant in fast reactors. In addition it is possible to build fast Molten Salt Reactors. The stability of Molten Salt Reactor operations in enhanced by Xenon-135 removal. Xenon-135 is a radioactive gas that is a byproduct of nuclear fission and has a very large neutron cross section. Because it is very likely to capture neutrons, Xenon-135 can adversely effect a chain reactor in a reactor. Thus it would be highly desirable to get Xenon-135 out of a reactor core quickly after it is produced. That is impossible in a solid core reactor, but it is not difficult to do in a Molten Salt Reactor. The presence of Xenon-135 adversely effects to the ability of reactors to breed nuclear fuel, so any MSR that is designed as a thorium breeder would have a system for moving Xenon-135 out of its core.

There are decided advantages for fuel reprocessing with MSRs. Compare the fuel reprocessing technique for a Molten Salt Reactor with the fuel reprocessing technique proposed for the Integral Fast Reactor (IFR) a LMFBR. In two fluid MSR, the blanket salt flows out of the blanket, and protactinium and U-233 are withdrawn from it by chemical processes. Once they are processed out of the carrier salt, the U-233 is re-fluoridated and returned to the core. The protactinium is set aside until it undergoes a nuclear transformation to U-233, and then that U-233 is returned to the core. In a IFR, the spent fuel is fished out of the reactor core, and once recovered, dumped into a molten salt bath, in which it dissolves. Then by use of electroplating, various material from the old fuel, for example plutonium, are separated out of the bath, and deposited on electrodes. Eventually the separated metal, is recovered, melted and mixed into an alloy, which is then cooled enough to serve as fuel elements, and then returned into the reactor. The MSR fuel reprocessing technology is much simpler than the fuel reprocessing technology designed for the IFR.

In addition fast reactors require 10 times as much nuclear fuel to produce a chain reaction as thermal breeder reactors. It does not really matter if the fast reactor is cooled by liquid metal of liquid salts, a fast breeder reactor just needs a who lot more fuel in order to operate than a thermal breeder reactor does. This makes fast reactors poor candidates to replace fossil fuels like coal with nuclear power, because many reactors will have to be built quickly, and fueling enough fast reactors quickly will be a big challenge.

There are two breeding cycles, the Uranium 238 breeding cucle, and the thorium 232 breeding cycle. Both cycles have some advantage. Plutonium-239 produces more neutrons per fission event than thorium, but fewer fission events per neutron in the thermal spectrum. In fact Pu-239 produces so many fewer fission events in the thermal spectrum than in the fast spectrum, that it is impossible to achieve a positive breeding ratio for the U-238/Pu-239 breeding cycle in a thermal reactor. On the other nand U-233 produces about as many neutrons per fission event in the thermal range as in the fast range, and about as many fission events. That means that the Th-232/U-233 breeding cycle is as effective in the thermal range asin the fast range, and because thorium breeding only requires about 10% of the nuclear fuel in the thermal range as U-238 breeding requires in the fast range, thorium breeding cycle reactors can be deployed far faster.

In addition Liquid fuel reactors have advantages over solid fuel reactors. Once a sollid fuel is inserted into a reactor it almost always stayes there for a year or more, while fission products build up in the fuel. We have already seen that Xenon-135 becomes a reactor control problem, although Xenon-135 eventually reaches an equalibrium because of its short half-life. The presence of Xenon-135 in a nuclear core, can interfear with a reactor's capacity to bread, especially in the thermal breeding range. Thus Thorium fuel cycle breeder reactor are better candidates for rapid deployment than U-238 fuel cycle breeders, and liquid fuel thorium breeding reactors have advantages as solid fuel thorium breeder. Liquid fueled thorium breeders, as we have already noted, have advantages over solid fuel U-238 breeders. Thus the Thorium fuel cycle Molten Salt Reactor (often called the LFTR) would seem to offer several advantages over U-238 fuel cycle liquid metal fast reactor.

In the next post of this series I intend to explain the difference between single fluid and twoi fluid Molten Salt Reactors.

Understanding Molten Salt Reactors: 1. How are MSRs Different from LFTRs?

The name Molten Salt Reactor (MSR) is more inclusive than the name Liquid Fluoride Thorium Reactor (LFTR). The LFTR is a type of Molten Salt Reactor that features the use fluoride salts and a thorium fuel cycle. Strictly speaking a LFTR need not be designed to produce fuel in a breeding range, but breeding is an important justification for the use of the thorium rather than the uranium fuel cycle.

In chemistry a salt is an ionic compound that is produced when acid undergoes a neutralizing reaction with a base. Some salts are compounds of metallic and non-metallic elements. Salts become molten (or liquid) when heated. Different salts have very different melting temperatures. Some salts may melt at relatively low temperatures, while other salts require several hundred degrees centigrade of heat before they melt. Mixed combinations of salts may melt at a lower temperature than individual salts will melt.

Some molten salts are very good conductors of heat, and some molten salts may not boil until they reach 1400 degrees centigrade, or even higher. These qualities make molten salts potentially excellent coolants for high temperature reactors. Research on high temperature fluoride salt cooled reactors continues at Oak Ridge National Laboratory (ORNL).

Two families of salts have been identified as potentially excellent reactor coolants. They are Fluoride and Chloride salts. Of these two salt families, ORNL scientists quickly chose the former, as presenting fewer developmental challenges while offering greater opportunities for commercial reactor use. Liquid chloride salts were identified as presenting more problems for reactor developers. Chlorine was, in particular, less useful as a neutron moderator than fluorine. However, Chloride salts were suitable for fast reactors, and a Molten Chloride Fast Breeder Reactor was both possible, and probably technologically less challenging than the Liquid Metal Fast Breeder Reactor, as a uranium fuel cycle breeder. In addition, the MCFBR would have offered far fewer safety problems than the LMFBR, while offering a route to technologically superior and lower cost fuel reprocessing.

ORNL Scientists however, also noted that liquid fluoride salt offered a superior performance precisely because of their neutron moderating performance. Moderation removes energy from neutrons. Heavily moderated neutrons are called thermal neutrons, and reactors that are built with large amounts of neutrons moderating materials such as graphite or heavy water are called thermal reactors. Neutron moderation decreases the amount of fissionable material required to sustain a chain reaction, and graphite and heavy water moderated reactors can produce chain reactions with natural uranium. While graphite and heavy water moderated reactors are useful tools for the production of plutonium, they are not particularly efficient tools for burning Pu-239 or even U-235 as nuclear fuels. However, it is possible to breed thorium in thermal reactors. Thermal reactor breeding of thorium offers significant advantages over fast reactor breeding of U-238. Thermal Reactors can operate with 10% perhaps as little as 3% of the fissionable material required to operate fast reactors. Thus thermal reactors can potentially be started at a far more rapid rate than fast reactors.

Fast reactor advocates claim that it is possible for fast reactors to breed at a much more rapid rate, but there appear to be safety problems involved in more rapid fast reactor breeding, and current documented advanced fast reactor designs appear to breed at a similar rate to thermal thorium breeding molten salt reactors.

The term breeding refers to the production more nuclear fuel than is used in a nuclear reactor. If a fertile material, either uranium 238 or thorium 232, is included with the fissionable material placed inside a reactor core along, the result will almost inevitably be the conversion of some of that fertile material into a fissionable material. In the case of thorium, a thorium 232 atom inside the reactor can absorb a neutron, and then becomes a thorium 233 atom. The nucleus, that is the center, of a thorium 233 atom is unstable. A neutron in the unstable Th 233 atom will eventually emit an electron, changing the neutron to a proton, the new proton in turn converts the atom to protactinium 233, and Pa 233 is also unstable. After a few days, a Pa neutron emits an electron, and as a consequence converts to a proton. The added proton makes the atom uranium 233. U 233 is fissionable, and can be used as reactor fuel.

The uranium fuel cycle is similar. If an U-238 atom absorbs a neutron a process that is similar to the process we find with thorium 233 occurs, and the U-239 atom is converted into plutonium 239.

Most reactors produce added nuclear fuel by converting U-238 into Pu-239. The amount of plutonium produced usually equals somewhere in the neighborhood of 60% of the amount of U-235 burned in a conventional reactor. Neither U-235 nor Pu-239 are ideal nuclear fuels at conventional neutron speed. WASH-1097 states

From a nuclear standpoint, the use of U-233 in a thermal reactor makes it possible to achieve higher fuel conversion ratios and longer fuel burnups than is practical with either U-235 or Pu-239. . .

The higher conversion ratios which can be obtained in thermal-spectrum reactors when using U- 233 instead of Pu-239 can result in a significantly better utilization of natural uranium fuel resources with thorium-fueled reactors than with the low-enrichment, light-water cooled uranium-fueled reactors . . .

WASH-1097 defines the fuel conversion ratio,

The fuel conversion ratio (CR) is the ratio of the amount of fissile fuel produced per unit of fissile fuel destroyed. . .

Breeding takes place when the conversion ration is greater than 1 to 1. While plutonium theoretically produces more neutrons and therefore faster breeding in fast reactors,

A higher breeding ratio can be obtained with Pu-239 than with U-233 in a very high-energy, fast- neutron spectrum reactor. On the other hand, in a degraded (10 to 100 keV) fast spectrum, U-233 would probably be as good as, or better than, Pu-239. Also, the variation of U-233 and Pu-239 cross sections with energy are such that improved reactivity coefficients would be obtained with the use of U-233 in a large sodium-cooled FBR. This leads to improved nuclear safety characteristics. . . .

The energy dependence of the fast-fission cross sections of Th-232 and U-238 is such that the use of Th-232 would produce an improved reactivity coefficient in a liquid-metal-cooled FBR. The fast fission cross-section of Th-232 is much lower than that of U-238 so that use of the latter leads to much larger conversion ratios in fast-spectrum reactors.

Thus not only is thorium breeding attractive in molten salt reactors, thorium breeding in fast reactors enhances their safety, and increases their conversion ratios. This point has not been lost on Indian reactor scientists who plan large scale production of thorium-uranium fast reactor breeding hybrids. Breeding thorium as a nuclear fuel in fast reactors would produce a large amount of fissionable U-233 that can be used as nuclear fuel in conventional reactors. This point has not been lost on Indian nuclear scientists, who plan to use the extra U-233 in Advanced Heavy Water Reactors. U-233 fueled non-breeder or converter Molten Salt Reactors offer attractive, safer and lower cost alternatives to conventional water cooled reactors, while reducing but not eliminating the nuclear waste problem. Thus, a modified Indian system could be developed, that would feature thorium breeding in IFRs, with U-233 burning MSRs. The rub for such a system would be that LFTRs would be cheaper to develop, cheaper to build, safer and cheaper to operate, and would virtually eliminate the nuclear waste problem.

The LFTR then is a thorium breeding MSR, that offers what may well be the simplest and best solution to the problem of producing sustainable nuclear power.