WASH-1222 with Comments: Part 2

Introduction: I have, in three previous posts discussed the effects that Milton Shaw's beliefs, managerial style, and policies had both on nuclear research at some and probably all National Laboratories. Shaw's ruthless methods of imposing his views even lead to the firing of Alvin Weinberg over disagreements about nuclear safety. In my last post, on the introduction to WASH-1222, I suggested that not only did Shaw have mistaken beliefs about the maturity of reactor technology, but that these beliefs cost reactor owners tens of billions of dollars. I may elaborate on this at a further time. I also argued that Shaw's mistakes about technology extended to breeder reactors, He held the mistaken belief that the maturity of LMFBR's had reached a level of maturity similar to that of LWRs.

In my last post, I began a review of a WASH-1222, a document prepared under Shaw;s direction. In the introduction to the document i found evidence that Shaw's mistaken beliefs, coupled with the consequences of his own bureaucratic decisions for which he failed to acknowledge responsibility, and shocking errors in logic, led Shaw to discount a promising new reactor technology, the Molten Salt Reactor. At the same time, Shaw was implicitly pushing other technologies, even though many the shared many of the problems of Molten Salt Reactor technology, while for other problems with technology Shaw favored, were solved by using the molten salt approach. Shaw claimed that unanticipated costs might incurred during the corse of development. My intention at the moment is to to post the next three sections of WASH-1222. There are a few points that might require further comments, so I might add them to morrow.

AN EVALUATION OF THE MOLTEN SALT BREEDER REACTOR

II. SUMMARY

The MSBR concept is a thermal spectrum, fluid-fuel reactor which operates on the thorium-uranium fuel cycle and when coupled with on-line fuel processing, has the potential for breeding at a meaningful level. The marked differences in the concept as compared to solid-fueled reactors make the MSBR a distinctive alternate. Although the concept has attractive features, there are a number of difficult development problems that must be resolved; many of these are unique to the MSBR while others are pertinent to any complex reactor system.

The technical effort accomplished since the publication of WASH-1097 and WASH-1098 has identified and further defined the problem areas; however, this work has not advanced the program beyond the initial phase of research and development. Although progress has been made in several areas (e.g., reprocessing and improved graphite), new problems not addressed in WASH-1097 have arisen which could affect the practicality of designing and operating a MSBR. Examples of major uncertainties relate to materials of construction, methods for control of tritium, and the design of components and systems along with their special handling, inspection and maintenance equipment. Considerable research and development efforts are required in order to obtain the data necessary to resolve the uncertainties.

Assuming that practical solutions to these problems can be found, a further assessment would have to be made as to the advisability of proceeding to the next stage of the development program. In advancing to the next phase, it would be necessary to develop a greatly expanded industrial and utility participation and commitment along with a substantial increase in government support. Such broadened involvement would require an evaluation of the MSBR in terms of already existing commitments to other nuclear power and high priority energy development efforts.

III. RESOURCE UTILIZATION

It has long been recognized that the importance of nuclear fuels for power production depends initially on the utilization of the naturally occurring fissile 235U; but it is the more abundant fertile materials, 238U and 232Th, which will be the major source of nuclear power generated in the future. The basic physics characteristics of fissile plutonium produced from 238U offer the potential for high breeding gains in fast reactors, and the potential to expand greatly the utilization of uranium resources by making feasible the utilization of additional vast quantities of otherwise uneconomic low grade ore. In a similar manner, the basic physics characteristics of the thorium cycle will permit full utilization of the nation's thorium resources while at the same time offering the potential for breeding in thermal reactors.

The estimated thorium reserves are sufficient to supply the world's electric energy needs for many hundreds of years if the thorium is used in a high-gain breeder reactor. It is projected that if this quantity of thorium were used in a breeder reactor, approximately 1,000,000 quad (1 quad = 1 quadrillion Btu) would be realized from this fertile material. It is estimated that the uranium reserves would also supply 1,000,000 quads of energy if the uranium were used in LMFBRs. In contrast, only 20,000 quads would be available if thorium were used as the fertile material in an advanced converter reactor because the reactor would be dependent upon 235U availability for fissile inventory make-up. (Note: a conservative estimate is that between 20,000 and 30,000 quads will be used for electric power generation between now and the year 2100.)

IV. HISTORICAL DEVELOPMENT OF MOLTEN SALT REACTORS

The investigation of molten salt reactors began in the late 1940's as part of the U.S. Aircraft Nuclear Propulsion (ANP) Program. Subsequently, the Aircraft Reactor Experiment (ARE) was built at Oak Ridge and in 1954 it was operated successfully for nine days at power levels up to 2.5 MWt and fuel outlet temperatures up to 1580ºF (1133 K). The ARE fuel was a mixture of NaF, ZrF4, and UF4. The moderator was beryllium oxide and the piping and vessel were constructed of Inconel.

In 1956, ORNL began to study molten salt reactors for application as central station converters and breeders. These studies concluded that graphite moderated, thermal spectrum reactors operating on a thorium-uranium cycle were most attractive for economic power production. Based on the technology at that time, it was thought that a two-fluid reactor in which the fertile and fissile salts were kept separate was required in order to have a breeder system. The single-fluid reactor, while not a breeder, appeared simpler in design and also seemed to have the potential for low power costs.

Over the next few years, ORNL continued to study both the two-fluid and single-fluid concepts, and in 1960 the design of the single-fluid 8 MWt Molten Salt Reactor Experiment (MSRE) was begun. The MSRE was completed in 1965 and operated successfully during the period 1965-1969. The MSRE experience is treated in more detail in a later section.

Concurrent with the construction of the MSRE, ORNL performed research and development on means for processing molten salt fuels. In 1967 new discoveries were made which suggested that a single-fluid reactor could be combined with continuous on-line fuel processing to become a breeder system. Because of the mechanical design problems of the two-fluid concept and the laboratory-scale development of processes which would permit on-line reprocessing, it was determined that a shift in emphasis to the single-fluid breeder concept should be made; this system is being studied at the present.

I have, in three previous posts discussed the effects that Milton Shaw's beliefs, managerial style, and policies had both on nuclear research at some and probably all National Laboratories. Shaw's ruthless methods of imposing his views even lead to the firing of Alvin Weinberg over disagreements about nuclear safety. In my last post, on the introduction to WASH-1222, I suggested that not only did Shaw have mistaken beliefs about the maturity of reactor technology, but that these beliefs cost reactor owners tens of billions of dollars. I may elaborate on this at a further time. I also argued that Shaw's mistakes about technology extended to breeder reactors, He held the mistaken belief that the maturity of LMFBR's had reached a level of maturity similar to that of LWRs.

In my last post, I began a review of a WASH-1222, a document prepared under Shaw;s direction. In the introduction to the document i found evidence that Shaw's mistaken beliefs, coupled with the consequences of his own bureaucratic decisions for which he failed to acknowledge responsibility, and shocking errors in logic, led Shaw to discount a promising new reactor technology, the Molten Salt Reactor. At the same time, Shaw was implicitly pushing other technologies, even though many the shared many of the problems of Molten Salt Reactor technology, while for other problems with technology Shaw favored, were solved by using the molten salt approach. Shaw claimed that unanticipated costs might incurred during the corse of development.

AN EVALUATION OF THE MOLTEN SALT BREEDER REACTOR

I. SUMMARY

The MSBR concept is a thermal spectrum, fluid-fuel reactor which operates on the thorium-uranium fuel cycle and when coupled with on-line fuel processing, has the potential for breeding at a meaningful level. The marked differences in the concept as compared to solid-fueled reactors make the MSBR a distinctive alternate. Although the concept has attractive features, there are a number of difficult development problems that must be resolved; many of these are unique to the MSBR while others are pertinent to any complex reactor system.

The technical effort accomplished since the publication of WASH-1097 and WASH-1098 has identified and further defined the problem areas; however, this work has not advanced the program beyond the initial phase of research and development. Although progress has been made in several areas (e.g., reprocessing and improved graphite), new problems not addressed in WASH-1097 have arisen which could affect the practicality of designing and operating a MSBR. Examples of major uncertainties relate to materials of construction, methods for control of tritium, and the design of components and systems along with their special handling, inspection and maintenance equipment. Considerable research and development efforts are required in order to obtain the data necessary to resolve the uncertainties.

Assuming that practical solutions to these problems can be found, a further assessment would have to be made as to the advisability of proceeding to the next stage of the development program. In advancing to the next phase, it would be necessary to develop a greatly expanded industrial and utility participation and commitment along with a substantial increase in government support. Such broadened involvement would require an evaluation of the MSBR in terms of already existing commitments to other nuclear power and high priority energy development efforts.

III. RESOURCE UTILIZATION

It has long been recognized that the importance of nuclear fuels for power production depends initially on the utilization of the naturally occurring fissile 235U; but it is the more abundant fertile materials, 238U and 232Th, which will be the major source of nuclear power generated in the future. The basic physics characteristics of fissile plutonium produced from 238U offer the potential for high breeding gains in fast reactors, and the potential to expand greatly the utilization of uranium resources by making feasible the utilization of additional vast quantities of otherwise uneconomic low grade ore. In a similar manner, the basic physics characteristics of the thorium cycle will permit full utilization of the nation's thorium resources while at the same time offering the potential for breeding in thermal reactors.

The estimated thorium reserves are sufficient to supply the world's electric energy needs for many hundreds of years if the thorium is used in a high-gain breeder reactor. It is projected that if this quantity of thorium were used in a breeder reactor, approximately 1,000,000 quad (1 quad = 1 quadrillion Btu) would be realized from this fertile material. It is estimated that the uranium reserves would also supply 1,000,000 quads of energy if the uranium were used in LMFBRs. In contrast, only 20,000 quads would be available if thorium were used as the fertile material in an advanced converter reactor because the reactor would be dependent upon 235U availability for fissile inventory make-up. (Note: a conservative estimate is that between 20,000 and 30,000 quads will be used for electric power generation between now and the year 2100.)

IV. HISTORICAL DEVELOPMENT OF MOLTEN SALT REACTORS

The investigation of molten salt reactors began in the late 1940's as part of the U.S. Aircraft Nuclear Propulsion (ANP) Program. Subsequently, the Aircraft Reactor Experiment (ARE) was built at Oak Ridge and in 1954 it was operated successfully for nine days at power levels up to 2.5 MWt and fuel outlet temperatures up to 1580ºF (1133 K). The ARE fuel was a mixture of NaF, ZrF4, and UF4. The moderator was beryllium oxide and the piping and vessel were constructed of Inconel.

In 1956, ORNL began to study molten salt reactors for application as central station converters and breeders. These studies concluded that graphite moderated, thermal spectrum reactors operating on a thorium-uranium cycle were most attractive for economic power production. Based on the technology at that time, it was thought that a two-fluid reactor in which the fertile and fissile salts were kept separate was required in order to have a breeder system. The single-fluid reactor, while not a breeder, appeared simpler in design and also seemed to have the potential for low power costs.

Over the next few years, ORNL continued to study both the two-fluid and single-fluid concepts, and in 1960 the design of the single-fluid 8 MWt Molten Salt Reactor Experiment (MSRE) was begun. The MSRE was completed in 1965 and operated successfully during the period 1965-1969. The MSRE experience is treated in more detail in a later section.

Concurrent with the construction of the MSRE, ORNL performed research and development on means for processing molten salt fuels. In 1967 new discoveries were made which suggested that a single-fluid reactor could be combined with continuous on-line fuel processing to become a breeder system. Because of the mechanical design problems of the two-fluid concept and the laboratory-scale development of processes which would permit on-line reprocessing, it was determined that a shift in emphasis to the single-fluid breeder concept should be made; this system is being studied at the present.