The Road Not Taken

The idea of a fluid fueled thorium breeder was first proposed by Nobel Laureate Eugene Wigner, together with Wigner's protégé Alvin Weinberg, and highly regarded engineer Gale Young in 1945. Between 1945 and 1958 Wigner and Weinberg who rose to be director of Oak Ridge National Laboratory had focused on a heavy water fluid fuel reactor the aqueous homogeneous reactor. But in 1948, an young Oak Ridge engineer, Ed Bettis, invented a second type of fluid fueled reactor, the Molten Salt Reactor, which was to demonstrate far greater potential as a thorium breeder and power production reactor.

Between 1950 and 1976 Oak Ridge National Laboratory developed the revolutionary Molten Salt Reactor concept. R. C. Briant and Alvin Weinberg explained in 1958 that there were

Two very different schools of reactor design have emerged since the first reactors were built. One approach, exemplified by solid fuel reactors, holds that a reactor is basically a mechanical plant; the ultimate rationalization is to be sought in simplifying the heat transfer machinery. The other approach, exemplified by liquid fuel reactors, holds that a reactor is basically a chemical plant; the ultimate rationalization is to be sought in simplifying the handling and reprocessing of fuel.

Briant and Weinberg added:

At the Oak Ridge National Laboratory we have chosen to explore the second approach to reactor development. . . . it has long been recognized that a liquid fuel which did not require high pressure, in which thorium or its compounds could dissolve, and which did not decompose under radiation would indeed be a major invention for the reactor art. . . .

we have been investigating another class of fluids which satisfies all three of the requirements for a desirable fluid fuel: large range of uranium and thorium solubility, low pressure, and no radiolytic gas production. These fluids, first suggested by R. C. Briant, are molten mixtures of UF4 and ThF4 with fluorides of the alkali metals, . . .

Many scientists are still convinced that

intrinsic [Molten Salt] reactor characteristics address current concerns about waste management, fissile resources, safety, and economics.

Scientists have pointed to the ability of the MSR to not only largely eliminate the problem of nuclear waste from its on spent fuel, but to make the nuclear waste of other reactors, largely harmless. In addition the use of MSRs to destroyed plutonium extracted from dismantled nuclear weapons has been proposed by scientists in Russia and the United States.

The Molten Salt Reactor was developed

During the 1960's, scientists and engineers built and tested an experimental Molten Salt Reactor that served as a proof of concept. They also worked on the design and development of a 1000 MWe MSR, and more briefly on a 250 MWe MSR design. At that time the estimated cost of a MSR was roughly equal to the cost of a light water reactor. Beginning about 1970 the cost of Light water reactors began a dramatic price rise that far exceeded the rate of inflation. Among the factors driving the price increase were increases in reactor size and complexity, as well as added safety features. The design of the MSR actually shifted toward greater simplicity in the late 1960's and remained relatively fixed in the 1970's. The safety issues that plagued the Light Water Reactor in the 1970's were not problems with the MSR, because many of the LWR safety problems were simply not present in the MSR design. Although an AEC document WASH-1222 complained that the MSR was a less mature technology than either the Light Water Reactor or the Liquid Metal Fast Breeder Reactor. In fact the LWR suffered from such serious design instabilities that last minute design alterations cost LWR purchasers tens of billions of dollars. Oak Ridge MSR designers reported that highly detailed $700 million development program - $2.4 Billion in 2009 dollars - would produce a viable commercial reactor - while the supposedly mature Liquid Metal Fast Breeder Reactor ended up costing over the United States government over $20 billion in 2009 dollars without ever producing a viable commercial prototype.

Thus in 2009 the 1970's to 1980's 1 GWe MSR would have cost about half the current cost of LWRs, while offering superior technology, and decreased operating expenses. The MSR would have cost less, because it was simpler, required less materials and fewer labor hours to build. The MSR had many inherent safety features that were absent from LWRs. Thus money does not have to be spent compensating for inherent safety defects in MSR design.

In addition to the savings from shifting from LWRs to MSRs, shifting from large reactors, to small, modular, factory built reactors offered an opportunity for significant construction savings. Researchers found that work disorganization was a significant cause of conventional reactor costs. Over 25 percent of workers time in reactor construction projects was wasted by work disorganization. Shifting labor from a construction site to a factory would help to solve the work flow problem. in addition building a large numbers of of small reactors in a factory, increases the rational for the use of labor savings devices on assembly lines. A rapid construction cycle, means less money would be spent on accrued interest. The small rapidly manufactured, low cost MSR is called a LFTR, Liquid Fluoride Thorium Reactor, In addition to the cost saving options already mentioned, other options are possible. It is at least conceivable that LFTR costs as low as 1 Billion Dollars per GW are possible. This is a very preliminary conclusion, but I believe that much more research should be undertaken. However, It is safe to say that some tentative evidence suggests that LFTR capital costs may run as low as $1 billion per GW, and that is a fair likelihood that LFTR costs will run below $2 Billion per GWe. Furthermore, LFTR research that would be preliminary to building pre-production prototype could run as low as $2.4 billion, and we could say that $5 billion is a not unreasonable estimate of the required research investment. Again further research would be desirable and would probably add to our certainty about cost estimates.