Bob Bovey was a Burke Scholar on graduation from USNA and received his doctorate from The Johns Hopkins University. He commanded SAND LANCE.
In the Navy and Atomic Energy Commission (AEC) joint Naval Reactors (NR) office, the mission and management was a seamless web encompassing research and development (R&D), acquisition and construction, and plant operation and maintenance. NR’s vision of its reach was as broad. It saw itself as responsible for creating or providing materials, processes, and qualified people. The first two responsibilities required a great deal of fundamental research.
Submarine nuclear-power development lay on the intersection of the development of nuclear power over time and the world of submarine technology generally. The focus of this review is 1949-1959, this is not entirely arbitrary. In 1949 the Naval Reactors Branch of the AEC was established, headed by the same man who had earlier been appointed head of the Navy Bureau of Ships office, Code 390, for the same purpose. The name changed several times over the years, but the combined office was usually referred to as Naval Reactors, or NR. In 1959 the SKIPJACK (SSN585), a hull form optimized for submerged performance and powered by a standard SSW nuclear power plant, went to sea. Nuclear submarines had reached maturity. For th is and other reasons, 194 9-1 9 5 9 was the decade on the time continuum when nuclear power moved from a fuzzy idea to a mature industry.
Although developing nuclear power was crucial to creating a true submarine, it was only one part of the submarine technology continuum. Without nuclear power, earlier submarine hulls had to be designed in recognition that the ships spent most of their time on the surface. At the same time nuclear propulsion was being developed, however, the Navy was conducting parallel developments in several submarine-related areas, including designing a hull form optimized for high-speed submerged operation and testing it extensively at sea in ALBACORE, starting in 1953 Therefore, the program described here is only a partial picture of a much more complex reality.
In January 1939, in a conference in Washington, D.C., Niels Bohr and Enrico Fermi announced that Otto Hahn and Fritz Strass man had split the nucleus of a uranium (U) atom3• Ross Gunn of the Naval Research Laboratory (NRL) heard this presentation and “became immediately convinced of the importance of quickly initiating navy research .. .toward the goal of nuclear power plants for submarines . . . “~A few days later, Gunn asked Rear Admiral Harold G. Bowen to initiate work at NRL. Bowen allocated $1,500 to Gunn, “the first government money spent on the study of atomic fission.”
NRL began research into the technology of gaseous diffusion to enrich uranium in the fissionable isotope, U-235, for fueling such a submarine. The Manhattan Project adopted this gaseous diffusion technology in 1944 to produce the highly enriched uranium (HEU) for the Hiroshima atomic bomb.
On 2 December 1942, Fermi’s University of Chicago experimental group achieved the first controlled and sustained nuclear chain reaction, 10 years and 4 months before NR’s Mark I initial criticality. During World War II, three reactors were built for producing nuclear weapons materials. These and five small research reactors were operating in 1946. The technology that existed for developing a reactor that would produce usable power was scattered and buried in classified files, not at all readily available.
In June 1946, a group of Navy officers and civilians were assigned to Oak Ridge to learn about the state of nuclear technology. In August, General Leslie Groves of the Manhattan Project approved a contract with the General Electric Company for a paper study of a liquid-metal-cooled reactor fora destroyer. Earlier, General Electric had agreed to “operate the plutonium production plant at Hanford, Wash., in exchange for a promise that the government would provide a nuclear development laboratory for the company at Schenectady.
“This laboratory became the Knolls Atomic Power Laboratory and eventually, over the period 1950-1955, was subsumed under the NR program. In sum, a good deal of research began shortly after World War II.
The five officers and three civilians studying at Oak Ridge facility developed an initial pool of information and concepts. They then toured the country, visiting laboratories and experts to refine their ideas. The team leader, Hyman G. Rickover, developed the initial research agenda to fill gaps in scientific knowledge required to support what he saw to be essentially an engineering program.
R&D within NR broadly followed three parallel tracks-pressurized water reactor (PWR), liquid metal (sodium) reactor, and gas-cooled reactor. Gas-cooled reactors were abandoned early ( 1949) by NR for naval use,’ although the issue was revisited from time to time. For example, in a 12 April 1957 hearing of the Subcommittee on Research and Development of the Joint Committee on Atomic Energy, Rickover was being pressed by several members who clearly were enthusiastic about gas-cooled reactors. In the context of civil reactors, he responded to Representative Chet Holifield’s question, “If you had the privilege of naming the reactor you would like to go into, which one would you select?” with “Gas cooled.”8 Indeed, gas-cooled reactors have been pursued subsequently for land-based applications.
Liquid-sodium-cooled reactor development preceded the formation of NR. In 1946, under AEC contract, General Electric had begun designing a sodium-cooled breeder-a reactor that created more fissionable material than it burned during operation. The sodium-cooled reactor was pursued through a full-scale operating, land-based prototype and into operation in USS SEA WOLF(SSN 575), which went to sea in 1957.
In 1959, SEA WOLF was converted to a PWR after a series of debilitating maintenance problems directly related to the sodium coolant.9 However, a liquid-metal coolant R&D was continued under AEC, the Energy Research and Development Administration (ERDA), and Department of Energy (DOE) sponsorship in the liquid-metal fast-breeder reactor (LMFBR) program until President Carter terminated it in 1977 on the grounds that production of fissionable material was inconsistent with efforts to stop the proliferation of nuclear weapons. The LMFBR program is of some interest in the current context because the NR R&D management approach was applied with more formality, and hence more visibility, than it had been in the early NR program itself.
The PWR turned out to be the dominant technology to emerge from the NR program. The specific examples below are therefore from PWR-related research. NR’s success in naval propulsion led the AEC to task in NR_led team to design and construct the PWR at Shippingport, Pa. This PWR became the world’s first purely commercial nuclear power plant in December 1957, when its generators transmitted electricity to the Duquesne Light company grid. The Shippingport reactor was not only larger than NAUTILUS one, it also employed a seed-and-blanket design in which a central cylinder of HEU was surrounded by an annulus of natural uranium.11 The PWR remains the dominant nuclear power technology in the world. R&D has continued worldwide on gas, liquid-metal, and water-cooled nuclear power plants to the present.
Overall Organization
The NR organization evolved from a loose network of interested individuals in 1947, which largely ignored an existing Navy office, to a formal organization in January 1949. This formal organization was unusual because the director was dual-hatted (in the Navy and the AEC). While there were many changes over the years, for this discussion, a simplified organization chart (see figure) will do.
For S&T management, the left leg was more important because almost 90 percent of NR R&D funding, in the neighborhood of$ I 00 million in FYI 958, flowed through it.
The NR HQ in Washington grew to about 90 scientists and engineers, both officers and civilians, by 1957. These people worked interchangeably for the AEC and Navy. In addition, 150-180 people in the Navy Bureau of Ships worked with NR almost exclusively.14 By 1959, the NR HQ had grown to about 120 scientists and engineers.
At the beginning of the decade, the main sources of science support to NR were first Oak Ridge and later Argonne Laboratory in the AEC system. The importance of this support declined by the early 1950s because NR built its own laboratory system. Two main AEC laboratories were established during this decade: Bettis Atomic Power Laboratory near Pittsburgh, Pa. (established in 1949 and operated by Westinghouse Corporation), and Knolls Atomic Power Laboratory near Schenectady, N. Y. (assigned to work for NR on 12 April 1950 and operated by General Electric Company).’6 Most of the R&D work done on naval reactors was performed in these two facilities, and no other work was done for other government or private programs. At the end of the decade, a third, smaller laboratory was established at Windsor, Conn. (owned and operated by Combustion Engineering). Together, these facilities employed about 2,000 scientists and engineers, plus supporting people.17 Bettis alone employed about 5,300 people, of whom 1,300 were scientists and engineers. 18 Reactor prototypes operated at the Schenectady and Windsor sites, but most were at the AEC facility in Idaho. These prototypes were used for conducting engineering tests, training submarine crews, and conducting physics and materials research.’9 At the same time, many other scientists and engineers who worked for the subcontractors were designing equipment for the naval nuclear program.
The laboratories reported administratively through an AEC field office, in which NR representatives were posted, and a program field office was located at each site to carry out functions such as budgeting, contracting, administrative control, etc. The relation between NR headquarters and the laboratories was usually direct on technical matters of most interest to an examination of S&T. Communications with the Navy and the AEC were conducted through NR headquarters.
Early Research
Early in the Navy team’s stay at Oak Ridge (June 1946-June 1947), it concluded that the necessary technology base for designing propulsion reactors did not exist. Each team member took a subject area and set out to read, listen to, and question Manhattan Project personnel about it. Each member also wrote a series of papers, which were reviewed by his colleagues. These initial papers were the first step in creating the necessary database. Adding to this database systematically became a primary function of NR.
The striking feature of the research initiated by NR in the late 1940s and early 1950s was its elementary nature, its attention to the sorts of basic measurements and analyses that physics and engineering students perform in class. It was exactly the kind of work that many scientists and graduate engineers disdain; yet, it was precisely the kind of information needed before the reactors could be de-signed. For example, in reviewing existing data on water, NR was surprised to discover how little was known about the properties of water itself or its effects on materials. 21 Over the years, NR coordinated a variety of laboratory studies on corrosion and wear in water systems. Throughout the 1950s, NR sponsored a series of reactor engineering handbooks that were the foundation of the nuclear industry as a whole. The series included the Liquid-Metals Hand-book ( 1950). The Metallurgy of Zirconium ( 1955), A Bibliography of Reactor Computer Codes ( 1955), The Metal Beryllium ( 1955), Reactor Shielding Design Manual (1956), Corrosion and Wear Handbook for Water-Cooled Reactors (1957), The Metallurgy of Hafnium (not dated, post-1957), and the three-volume Physics Handbook ( 1959-1964 ).
Support of Engineering Development-Zirconium as a Structural Material
One development within the PWR materials track serves to illustrate two points about the interplay between scientific research and engineering development that was commonplace within the program. First, research was often done to understand the properties of materials that seemed attractive based on preliminary knowledge. Second, research sometimes unexpectedly uncovered possibilities that demanded further R&D to exploit.
By December 194 7, Oak Ridge had completed a very preliminary design of a PWR.23 One of the problems in building a PWR was to find a material that would be strong enough and workable to support and clad the uranium fuel elements, had little tendency to absorb neutrons, and resisted corrosion by hot water under high radiation. Many materials, including stainless steel, aluminum, and beryllium, were studied. An Oak Ridge engineer, Samuel Untenneyer, had suggested zirconium (Zr) because of its mechanical, metallurgical, and corrosion characteristics; however, it had two big disadvantages Zr had never been produced in quantity, and it seemed to have a large neutron capture cross section. However, in late 1947, Herbert Pomerance, an Oak Ridge physicist, had discovered that the large cross section recorded in earlier tests was mostly the result of a hafnium (Ht) impurity in the Zr test material. Therefore, removal of the Hf would make the Zr neutron capture cross section quite low. However, the removal of this previously undetected alloying material might also degrade Zr’s mechanical, metallurgical, and corrosion properties.
Based on the evidence accumulated by the end of 1947, [Rickover] committed to Zr as the metal for fuel-element structural material and fuel-plate cladding. This decision set in motion four parallel tracks of materials scientific research and engineering work. One path was to verify the properties of pure Zr and perhaps discover alloying materials to improve them. The second was to mass-produce Zr. These two tracks converged onto the third, which was to design, test, and manufacture hundreds of fuel elements. The fourth track concerned Hf and is addressed in the following section. Each of these tracks involved iterative but overlapping scientific research and engineering problem solving.
Although Zr was selected in 194 7 as a reactor structural material for PWRs because of its favorable nuclear properties and corrosion resistance, it was not until March 1950 that Argonne and Bettis laboratories decided it would be feasible to assemble a fuel plate consisting of a U-Zr alloy fuel element clad with Zr.25 Research continued to improve the performance of Zr, and out of this, an alloy named zirca/oy was developed. Zircaloy was less expensive than pure Zr and had improved corrosion and mechanical properties. However, after deciding to use zircaloy as cladding for U02 fuel elements in the Shippingport reactor, in-pile and out-of-pile tests revealed unexpected Zr properties. Zr tended to absorb hydrogen (H) from high-temperature water systems. Irradiation affected this, and the behavior of H dissolved in Zr was not initially understood. Both in-pile and out-of-pile tests were used to study the redistribution of H in Zr under thermal and stress gradients. Together, they provided a basis for explaining and predicting the migrations. Further research revealed the role of nickel (Ni) contained in zircaloy in accelerating or increasing H absorption and pointed the way toward a class of Zr alloys free of this injurious feature.
In the meantime, R&D was carried out to produce Zr and zircaloy. For example, in 1948, U.S. Zr production was about 86 pounds at$ l 35-235per pound, all by the Foote Mineral Company.i~ In 1955, the AEC signed 5-year contracts with three producers to produce a total of 2.2 million pounds Zr per year at $4.80-8.00 per pound. In sum, research into some very fundamental physical phenomena continued in parallel with engineering design and even manufacturing. Research was the bootstrap that pulled the engineering development forward.
When one asks an NR alumnus for the important factors influencing the conduct of S&T by NR, the first answer is people. This was rooted in the NR emphasis on individuals rather than processes. Rickover required each staff member to have definite responsibilities and to be held personally accountable for every aspect of those responsibilities. To achieve a staff that could succeed in such an environment, NR devoted extraordinary attention and energy to selecting and training people.
The first NR people engaged in independent study and research for the June 1946-June 1947 year as a team at Oak Ridge. A second group trained at Argonne National Laboratory. Other additions followed a course of supervised independent study in the NR office. By June 1949, NR had negotiated with MIT to extend a longstanding naval architecture and marine engineering course to include a year of nuclear physics and engineering for Navy engineering duty officers sponsored by NR. In March 1950, NR and Oak Ridge began the Oak Ridge School of Reactor Technology, which had trained over I 00 NR, Navy and contractor employees by 1956. The school eventually provided hundreds of trained engineers for the nuclear power industry.
In the meantime, universities were graduating physicists and materials scientists. The major people thrust after the 1949-1959 decade was the selection and training of officers and enlisted men to operate nuclear-powered ships, although the renewal of the NR staff continued to receive great attention. From the early 1950s, the NR approach for the laboratories was different. It was up to the contrac-tor to select and educate its people, but NR evaluated these people and demanded replacement of those found deficient in capability or dedication.
In his later years, Rickover became a well-known critic of the American education system generally and scientific/engineering education in particular. However, in this decade and later, NR training programs focused on meeting its own needs for managers and operators.
In the beginning, Rickover insisted on focusing on specific projects that would lead to a practical nuclear power system. He was ruthless in eliminating research that did not contribute directly to these projects.29 Later, the focus was broadened somewhat, as discussed below. Still, NR wanted to be in control of R&D-to tell the researchers what was to be done. The NR director wanted advice, but in the end he wanted relevance and sensible work.
The general view was that when an HQ pushes a laboratory, the lab will say that the HQ is not competent to judge. However, that was not the point of NR’s philosophy.
It believed the laboratory is like a violinist in a symphony orchestra. HQ should not tell the R&D contractor what to do (how to play his violin), but the government office must be the conductor, telling all the instruments what to play, what aspects of research on which to focus, etc.
NR believed that is must not get into the dangerous situation that it regarded as usual for government, where the researcher does whatever he thinks is fun without knowledge of overall system issues. An example drawn from the LMFBR Program was also illustrative of NR experience. The program was having serious civil heat exchanger problems. The program director ended up in a fight with a talented academic who wanted to work on some esoteric aspect that probably would never have an application (but was frittering money away), to get him to work on the real problem. In general, the view from NR was that most government people overseeing science are not managerially oriented. They tend to be sympathetic to the laissezfaire approach of the labs and contractors. The NR view was that when they look at R&D, they need to ask “What is mission value?” In other words, R&D had to be mission oriented, and it had to be the government who judged. To do that, talent was needed. Hence, the focus on people for the HQ organization.
Mission focus moved from a management precept to a crusade for Rickover. From 1974 through 1982, he embarked on a campaign against the system for contractor IR&D then in effect and for those who administered it. Rickover debated with senior political appoint-ees in the Navy and Office of the Secretary of Defense (OSD) and took his case to the General Accounting Office (GAO) and Con-gress. His fundamental issue was that much of the work being funded by the government in contractor organizations had no relation to military needs. He opened his argument at high levels on 21 June 1978 with a memorandum for the Secretary of Defense (SECDEF) via the Secretary of the Navy (SECNAV). He recommended that JR&D reviewers be guided by the technical evaluations of proposals, that only experts in the proposed work evaluate proposals, that proposals in which the benefits to the government did not warrant the cost be rejected, and finally that the entire system be changed so as to finance worthy R&D by direct contract so the government could supervise the work and retain appropriate rights to the resulting intellectual property. On 24 November 1978, the Under Secretary of Defense for Research and Engineering, William J. Perry, rejected these arguments.
Having said this, the focus was not entirely consistent. First, Rickover interpreted his nuclear-power charter broadly where research was involved. Speaking of the many technical publications of NR, he said.
By having these books available you get the people in the universities and in other places starting to think about the problem and making improvements… You will find that today these are the standard books in the United States on this subject. .. There are not any others with detailed scientific and engineering information in this field.
NR was also more relaxed with university research than with industrial research. The money involved was much less, and it was good Congressional politics to have research going on in many places. As a practical matter, NR found that it could get good results from universities because it was possible to press the faculty principal investigators to do good work without incurring Congressional ire, so long as the money kept flowing. University research, however, was undertaken with some reticence because of the folklore that just when the research reached the point that NR needed it, the professor would go on sabbatical.
One ofNR’s main features was that it internalized the matter of responsibility. For research and other work performed through contracts, NR distilled from this the concept of the demanding customer. The following description of this concept is extracted.
Direction and guidance provided by the customer for contractor activities can take different forms. In many instances, the customer will arrange with contractor organi-zations to perform specific functions like research and development, design, procurement, construction, testing, and quality assurance, but will retain management of the total effort. In other instances, the customer will enter into arrangements where managing the total effort will be assigned to a selected lead contractor. The latter may still perform functions like those cited or have them provided by other organizations. Depending on the organizational arrangements involved, there will be one feature common to all-the need for the customer to exercise management across a customer-contractor interface.
The key principle is that management and other capabili-ties of the customer’s organization should be used basically for one function: namely to require and otherwise bring about effective management by the contractor organization or organizations to assure performance in accordance with the contract. The decisive test for any action contemplated by the customer is whether it is conducive to this objective. The principal pitfall is that the customer will use its capabilities to compensate for continuing weaknesses of the contractor. Like other management principles, this one is logically compelling but difficult to apply.
A second principle is that the customer should set forth technical requirements in sufficient breadth and depth to assure that the product will meet customer objectives, but not in such degree as will stifle contractor management, initiative, and innovative capabilities. A corollary is that the customer needs to be able to adjust requirements, as practicable, to accommodate difficulties being encountered.
The prerequisite need in applying these principles is that the customer have in-house capability as measured by technical competence among its own employees to shape, guide, direct, and assess the activities and operations of its contractors … If the customer organization lacks technical strength, the contractor will not feel the same pressure to achieve excellence.
Having cited the need for strong customer technical capability, it is important to caution against its misuse. The general caution is that is should not be used to do work or perform functions for which the contractor is being paid … Many customer personnel would not perceive this as happen-ing: some would not find it objectionable if they did. Such individuals find professional satisfaction principally from making a contribution to the solution of problems … It takes a firm hand to keep them from subverting the larger interests of their own organization.
A demanding customer will insist on developing clear, mutually agreed upon understandings about relationships with the contractor. True responsiveness by the latter always obliges the contractor to use his own good judgment in questioning suggestions made [by] the customer staff if the contractor believes them to be ill-advised. Responsiveness is to be measured, not by the extent to which the customer responds automatically to guidance from customer representatives, but rather by the degree of responsibility exhibited in analyzing such guidance and then in acting on it or recom-mending reconsideration as appropriate. It is also to be emphasized that differences in important matters are not to be held unduly long at lower levels, where they foster animosity and weaken cooperation.
Instead, they should be raised promptly to higher levels of management for resolution. The objective to be sought is open, constructive dialogue between the parties, giving the primacy to objective technical and other considerations and suppressing personal predilection and bias ….
The need for the demanding customer to have i11-ho11se capability emphatically should not be taken to imply that the numbers of personnel be large. A customer operating in a sound managerial relationship vis-a-vis a contractor should be able to provide the needed managerial oversight with far fewer numbers than the contractor is obliged to use . . . the objective should be to keep competence up and the numbers down.
In NR’s view, organizational funding was important to good S&T. An organization needed to have, as NR had, mission funding, which provided a steady diet. Organizations that did funding task-by-task ended up just feeding the tourists, those who came around evaluating projects for continued funding. Also, project officers were seen as risk averse. They would not support S&T.
Recalling that NR’s budget was nearly all R&D, most of it from the AEC and quite stable overall mission funding, controlling the dollars available then became an issue. In NR, the project officer had no money. He had to concur with plans of the technical branches. The technical area director had the money and covered the spectrum in his technical area. For example, reactor engineering covered current production, operations, and technology development, both to fix current problems and for the next generation. The project officers crosscut the technical directors. They were critics. Otherwise, inertia would be in control, and the technical branch would just keep working down a particular line. This implied that the advanced technology project officer was often in the position of arguing, “You guys are ‘ polishing the cannon ball’; it’s time to shift money to something else.” These money shifts could take place across technical branches.
Rfokover ” … held that it took years to train a man to be profi-cient in the peculiar kinds of technical and management problems faced in the nuclear project. .. “In particular, he viewed the idea of rotating officers after a 3-year tour, ” … as the height of folly. Virtually all his senior staff agreed that the navy’s rotation system.
…made adequate control of technological development [impossible ].39 Building and maintaining a management team for the long term was a major objective-one that was achieved to a large degree.
For example, a head count taken as of 1982 indicated that there were 21 section heads (technical groups, project offices, and support sections) at NR headquarters. Of these, 12 had joined NR in the 1949-1959 decade and the remaining 9 had joined in the 1960-1970 decade.40 Because of this continuity, NR had a stable of strong advocates in its technical directors. They knew they were responsible for the whole spectrum, including the next generation, which had to be better than the last one. Furthermore, they would still be in NR to take the responsibility. In NR, the technical director had a much longer life than the technical leader in a normal Navy organization.
While the issue of tenure in NR ended up being a positive with respect to S&T management, controversy continued throughout the life of the program about the negative impacts of Navy rotation policy (applied to officers outside the NR program) on the program generally. For example, in 1960 Congressman Price observed, “With the attitude of the Navy in regard to .. .it would indicate to me that perhaps they are considering nuclear-powered submarines and Polaris-type submarines as conventional a little too early … which might adversely affect you.” Rickover responded, “Nuclear power has brought many novel problems with it. The people in the Navy rotate very quickly. Nuclear power is hard to understand so they try to force it right back into the old system, which they do understand.
In the years 1949-1959, judging top-level government execu-tives’ support of naval nuclear propulsion R&D (as contrasted to their support for shipbuilding plans, personnel decisions, and other matters that were related to, but different from, R&D) is difficult because of the many and tangled threads that ran through the decade.
In The Politics of Innovation: Patterns in Navy Cases, Vincent Davis took strong issue with account[s] in which [Rickover] is generally portrayed as the clear-cut hero, and all others in the plot are either his helpful accessories or his villainous opponents. . .which made it appear as if [Rickover] had been forced to wage a one-man campaign against a Navy high command generally unenthusiastic about developing nuclear powered submarines.43 Davis saw the decision to send the team to Oak Ridge in 1947 as, ” … representing the triumph of the nuclear power enthusiasts within the Navy with respect to a firm Navy commitment to press ahead into research and development on nuclear propulsion for submarines. All remaining problems were ultimately resolved, in large part because the highest officials in the Navy Department, including the Secretary and Chief of Naval Operations consistently gave this project their strong support. Others emphasize the difficulties in getting and keeping the highest officials engaged.
In January 1947 [the Chief of Naval Operations, Fleet Admiral] Nimitz himself had approved a recommendation supporting development of a nuclear submarine … Two years of planning and discussion had … all but stifled the idea that seemed so promising . . . No one in a responsible position in the Navy really opposed the idea of nuclear propulsion. . .In a larger sense the issue was. . .whether the potential impact of nuclear power on the Navy war-ranted more than routine development.
The judgment was made more difficult by the fact that two organizational superstructures stood over NR. Also, top managers in this management structure changed over the years as it coalesced and later evolved. Rickover was a masterful bureaucratic politician and played the two parts of the organizational superstructure over him to marshal support for the nuclear reactor program. Generally, the DoD superstructure was instrumental in overcoming early AEC reluctance and inertia to begin serious R&D into nuclear propulsion. Later, the AEC superstructure became far more important for NR R&D-most R&D funding flowed through it-while relations with the DoD superstructure were often acrimonious over matters other than R&D. However, by the end of the decade, Rickover could bypass both legs of the superstructure to a large degree, at will, and was empowered by Congressional connections, primarily with the Joint Committee on Atomic Energy, in R&D and many other matters.
The success ofNR from 1949 through 1959 was demonstrated by the performance of its product-the nuclear submarine-and speed with which it was developed and built. This success was even more impressive considering that the nuclear reactor technology and several supporting industries did not exit and had to be developed starting from almost zero. The reasons for such an astonishing achievement were many. This review has not attempted to account for all the factors that played a role. It has focused on NR’s S&T research, which was a major factor in the success achieved during the decade. What seem to be the key relevant considerations in NR’ s management of S&T research are summarized below.
- Based on its reliance on individual responsibility as a central management principle, NR regarded hiring highly qualified people as a central task. The training and educa-tion of its HQ personnel was given first priority. By June 1949, NR sponsored a course in nuclear engineering and physics at MIT for the Navy engineering duty officers. In March 1950, NR opened the Oak Ridge School of Reactor Technology, which provided basic fundamentals as well as reactor-specific training to hundreds of engineers for the nuclear power industry.
- NR, in its management of government-owned/contractor operated (GOCO) laboratories, universities, and contrac-tors performing research, was a demanding customer
- Clear definition of program performance goals and systematic, strict evaluation of the projects led to well-defined technology gaps, focusing research where it was most important to the overall goal. The NR program benefited immensely from having highly qualified person-nel set technical requirements in sufficient breath and depth to ensure that research products would meet its performance objectives.
- In addition, these highly qualified NR personnel were able to use sound technical judgment in evaluating project results and determining its progress. S&T project progress and results were scrutinized frequently and judged on technical grounds, after often tough, sometimes bruising debate.
- Clear program technical and schedule requirements were set early and, in turn, drove S&T project decisions on how much research was enough. Requiring research to support development schedules was instrumental in delivering working systems on time.
- NR, in its quest for solutions to an entirely new set of technical problems, maintained a strategy of pursuing several technologies simultaneously, thereby reducing long-term technical risk. The strategy was applied at several levels, from overall concepts to specific materials and from fundamental research through engineering development and operations at sea. Best known is the search for the best reactor cooling configuration, in which parallel efforts on PWRs, liquid metal (sodium), and gas-cooled reactors were conducted. Another example of this strategy is simultaneous work on Hafnium and Silver alloys for control rod material applications.
- NR R&D (including the S&T component) also benefited from stable budgets, most of which came from the AEC.
