deleted] of the aircraft nuclear propulsion program. Despite these statements by [classified matter deleted] no fuel material has been developed (1957) which can operate without emission of fission products at the required temperature for the required time. Actual operation of the [classified matter deleted] fuel elements has resulted in continuous emission of fission products into the gas stream at a rate unsatisfactory for naval application. The [classified matter deleted] proposed reference design uses a fuel element similar to [classified matter deleted] but requires 5 times the heat flux and 10 times the power level and also assumes no release of fission products.

(b) There is no satisfactory choice of coolant gas.-Very little experience exists (1957) with power generation systems using a high-temperature gas other than air. Experience with large graphite reactors is of limited applicability; the proposed [classified matter deleted] design requires temperatures on the order of 1,000° F. higher and radiation intensities 10 to 100 times higher. Helium has the advantage of chemical inertness, but leakage (particularly through rotating seals) appears to be such a problem that sufficient quantities of makeup gas could probably not be stored aboard ship to insure continued operation. In addition, world resources of this critical material are being rapidly depleted. Other gases such as carbon dioxide or nitrogen present serious problems of radiochemical decomposition and of chemical reaction with component materials.

(c) There is no satisfactory neutron moderator.-The coolant water in pressurized water reactors also slows down or moderates the neutrons; there is no satisfactory moderator for doing this under the conditions assumed in the proposed reactor. Zirconium hydride is proposed by [classified matter deleted] for this purpose. This material, however, has a large number of feasibility questions associated with it and it is doubtful that it would prove satisfactory for the purpose. If another material such as graphite or beryllium oxide were used as a moderator, questions of feasibility will still have to be answered for the substitute material, and a bigger reactor and therefore a heavier plant will also result.

(d) The feasibility of major components has not been established.-The blowers, compressors, seals, high-speed instruments, large valves, and steam generators of the proposed system are very different from existing equipment. This is due to the combined requirements for size, pressure, temperature, leak-tightness, and in some cases fast response time. In addition, many of the basic facts from which these developments would have to evolve must themselves be developed. (e) There is almost no experience with large, closed-cycle, gas-turbine power systems of any kind.—There are serious questions concerning the control and reliability of such systems which can be answered only by experience with a large closed-cycle system, operating with the reference gas reasonably near the required temperatures and pressures. No such experience exists today (1957). Development of such technology could be done more easily without the additional requirements imposed by a nuclear application.

b. Space and weight

(1) From the technology already developed (1957) in existing programs it appears that gas-cooled reactor plants would not compare favorably with pressurized water plants for naval application. Present experimental evidence casts serious doubt that naval gas-cooled reactor plants would be lighter or smaller than pressurized waterplants.

(2) The [classified matter deleted] report states that "the major weight reduction made possible with gas-cooled reactors results from a decrease in the total amount of shielding required." The [classified matter deleted] reference design assumes that, since helium does not become radioactive, "shielding over the piping and machinery is not required." This assumption is not valid. Shielding would be required in a naval gas-cooled reactor plant to protect personnel from radioactivity in the coolant which could result from diffusion of fission products through fuel element cladding, erosion and corrosion, from contaminants introduced into the coolant system during operation and maintenance, and from possible fuel element defects or failure.

(3) The next largest weight and space reduction in the [classified matter deleted] reference design results from using only one reactor, whereas two reactors were used in the Bureau of Ships design which was used by [classified matter deleted] for comparison. This weight and space reduction has nothing to do with the difference between gas-cooled and pressurized-water reactors.

(4) The remaining differences in propulsion plant weight and space are small and are more than offset by weight and space additions which would be required

to compensate for other unsuitable features in the [classified matter deleted] designs, such as inadequate safety and casualty protection, a requirement for complex and high speed control equipment, and several inadequate plant reliability features. Examples are:

(a) Thermal aspects of the reactor design are inadequate; correcting this would require changes such as increased reactor core size, decreased gas temperature, or greater coolant flow rate, which would increase component and shield weights.

(b) Practical limitations on size of components and piping would probably require the use of more loops per reactor, or reduced gas temperature or pressure, which would increase overall plant weight and space.

(c) Inability of the plant to sustain without reactor meltdown a rapid decrease in gas flow rate; several complex solutions to this problem were discussed in the [classified matter deleted] report, but no weight or space allowance for this was made in the [classified matter deleted] designs. c. Inherent disadvantage

(1) Regardless of weight and space considerations, any closed-cycle gascooled reactor for naval propulsion would have a number of inherent disadvantages which would seriously reduce plant safety and reliability and increase plant complexity as compared to a pressurized water reactor plant. These disadvantages are:

(a) Rupture of the reactor coolant system could lead to meltdown of the reactor core far more easily in a gas-cooled plant than in a pressurizedwater plant. Meltdown can be prevented in a water-cooled reactor, but not in a gas-cooled reactor, by keeping the core covered with coolant, even unpressurized.

(b) Rapid reduction of coolant flow could lead to meltdown of the core in a gas-cooled reactor. However, loss of coolant pumping power in a pressurized-water plant will not result in core meltdown or other damage to the plant, since the heat capacity of the water coolant is sufficient to remove decay heat.

(c) Flooding a gas-cooled reactor with water would introduce a large increase in reactivity and possibly lead to a major safety hazard; increasing reactivity in this way is not possible in a pressurized-water reactor.

(d) A gas-cooled reactor plant would not have the self-regulation features inherent in pressurized water reactors.

(e) Plant control systems and their instruments for gas-cooled plants would require faster response and higher accuracy than those needed in pressurized water plants. The greater sensitivity of gas-cooled reactor plant control systems would lead to operational and maintenance difficulties. (2) The above disadvantages result from the following inherent characteristic of gas-cooled reactors for naval application:

(a) Low heat capacity of the coolant.

(b) High temperatures and large temperature gradients.

(c) Small negative or possible positive temperature coefficient of reactivity.

(d) Coolant passages in the reactor not normally filled with a good moderating material such as water.

d. Cost

Naval gas-cooled reactor plants would not be cheaper to construct or to operate than pressurized water plants; actually, they would probably be more expensive. At temperatures required for naval application, gas-cooled reactors require more special materials, more complicated components, more complex and sensitive control systems, and more provisions to insure acceptable safety. The cost of operation would be increased because of the higher uranium content and shorter life inherent in gas-cooled reactor cores.

C. Conclusions

In a letter to [classified matter delected] dated November [classified matter deleted], 1956, the Chief of the Bureau of Ships rejected the [classified matter deleted] proposal, stating:

"The Bureau has reviewed the above [classified matter deleted] reports and your development program proposal of [classified matter deleted] October 1957. As a result, the Bureau has concluded that sufficient technology has not yet been developed in key areas such as fuel elements, moderators, control materials, physics, safety, and the properties of gaseous coolants to warrant the

undertaking of a naval gas-cooled reactor development program at this time. A summary of the Bureau's review is enclosed for your information.

"As you know, the Atomic Energy Commission has underway an extensive effort to develop high-temperature gas-cooled reactor technology in its aircraft, maritime, Army, and civilian power progress. While the requirements of those AEC projects are different than those for naval application, the Bureau considers that on the basis of the technology already developed gas-cooled reactor plants would not compare favorably with pressurized-water plants for naval application."

APPENDIX 3

Memorandum for the Secretary of Defense.
Subject: Nuclear power for surface warships.

JANUARY 23, 1963.

Dr. Seaborg's letter to you of January 7 reviewed progress the Atomic Energy Commission is making with its program of developing reactors for surface ships, particularly aircraft carriers. It also suggested an overall appraisal of the future of nuclear propulsion for surface warships to permit the Commission to determine the level and scale of effort in this field in the next few years. Additionally, it raised the question of whether it is too late to reconsider the decision to build CVA-67, the CVA in the 1963 shipbuilding program, as a conventionally fueled ship.

The Commission has been most generous in its support of naval nuclear propulsion since the inception of the program and is understandably concerned that the scale of application of nuclear power to surface warships has been limited. It is certainly appropriate that we reappraise our position and plans in the field.

With respect to CVA-67, you will recall the Navy asked the fiscal year 1963 carrier be built with nuclear propulsion. This nuclear carrier employed four reactors and the estimated end cost was $410 million. Subsequent to the decision to build CVA-67 as a conventionally fueled ship, we conducted a detailed review of the Navy's nuclear propulsion program which I forwarded to you last March. This review indicated that the advantages of nuclear propulsion were well recognized and that the Navy policy calls positively for movement in the direction of nuclear propulsion. The review concluded that cost is the major factor in determining whether nuclear-powered surface ships will be built by the Navy as compared with conventional propulsion. Further, the best means at our disposal to reduce cost of these nuclear-powered ships is to expand the nuclear shipbuilding effort. These conclusions are similar to those reached previously by Chiefs of Naval Operations, Secretaries of the Navy, Secretaries of Defense, and many Members of the Congress who have discussed the nuclear-powered Navy over recent years.

Our most recent review has been stimulated by recent experiences with our nuclear-powered surface ships, recent OSD actions on the Navy shipbuilding program, and the fact that the nuclear propulsion plant now recommended by Chairman Seaborg represents substantial advances in power and core life over those originally considered for the fiscal year 1963 carrier. This most recent review has made us increasingly aware of the improved operational effectiveness of nuclear-powered ships and of the need for positive action to prevent losing our recognized leadership in this extremely important and unique technological area.

During the last 18 months the Long Beach, Enterprise, and Bainbridge have joined the fleet. They have logged a total of more than 115,000 miles; their propulsion plants have demonstrated their reliability and capabilities. In each case, the nuclear propulsion plant has met or exceeded its specifications.

Enterprise has now been through a rigorous year of fleet operations which has proved her to be the best carrier in the fleet. In her first year she has logged 65,000 miles and 12,000 aircraft landings. In addition to the advantages of sustained high speed and consequent reduced vulnerability to submarine attack, Enterprise has reported that we can expect fewer aircraft losses and longer plane life on a nuclear carrier due to the elimination of stack gases.

We are aware that is is difficult to assign a dollar value to the military advantages of a nuclear carrier. The tremendous number of significant operational advantages, offensive and defensive, for both task force and individual ship operation that result from the virtually unlimited cruising range, long endurance, sustained speed, structural improvements, superior electronic performance, 1 See app. 2, pp. 201-229.

tactical flexibility, and freedom of movement of the nuclear carrier are certainly worth considerable premium. Resources stockpiled by providing nuclear fuel in peacetime can be a vital element in wartime operations.

Our studies, which have taken into account both experiences with present nuclear ships and the predicted characteristics of the improved four-reactor plant indicate that this cost premium will be about 20 percent in investment, operating, and maintenance costs, as compared to a conventionally powered carrier, over a 20-year period. Specifically, the increased cost to provide nuclear power in the CVA-67 is now estimated at about $115 million including the cost of the initial fuel which will last for at least 7 years of operation. This represents a significant reduction from the differential cost of about $200 million to build a repeat Enterprise. The operating costs of such a ship would also be less than for an eight-reactor Enterprise design. This change to nuclear propulsion, as well as other changes to the hull design to achieve significant (over 50 percent) increases in payload, aviation fuel, and ammunition, would result in an end cost estimate of about $403 million for the nuclear CVA-67. Although 6 to 9 months would be needed for the machinery and hull redesign effort, I would expect no significant delay in the delivery of the ship to the fleet; it would be completed in calendar year 1967.

It should be noted that the cost picture of future nuclear-powered warships is much influenced by the procurement policy we follow today and our limited capabilities to predict costs, logistics, and operational patterns into the operating period of 1968-88. Although the initial cost of installation has decreased with the development of improved technology and experience in production methods, the increase in initial cost for each nuclear ship has been a significant factor in peacetime budgeting reviews and has made it difficult to give these nuclear ships the full consideration required by their important potential contributions to our major offensive and defensive capabilities. It is apparent that a cost premium in the short term is inescapable in order to keep alive technical and manufacturing progress, and AEC interest, toward the goal of cheaper nuclear ships in the long term.

As a result of my review of all pertinent considerations, as highlighted by Dr. Seaborg's letter, I conclude that the substitution of an improved fourreactor nuclear propulsion plant for the oil-fired plant now programed for CVA-67 is both feasible and desirable. It will provide us with a superior fighting unit and permit us to exploit the long-range potential military advantages of nuclear power in our major combatant surface ships. It will let us utilize the most advanced proven technology available in building our fleet of the future and will fill a notable gap in the nuclear propulsion program created by the recent deferral of the fiscal year 1963 DLG (N). It will give impetus and support for nuclear propulsion technology, moving us further along the road to the nuclear Navy we envisage for the future. In view of the extreme importance which is attached to this matter, you may desire to establish a special task force to review this subject before you act on my recommendation and forward a reply to Dr. Seaborg. To this end I would be pleased to nominate Vice Admirals Schoech and Rickover to sit with your most senior and qualified representatives. It seems clear that the subject of nuclear propulsion for surface ships is a very active one, and of considerable concern to many Members of Congress. I suggest therefore your consideration as a matter of urgency.

FRED KORTH.

APPENDIX 4

THE SECRETARY OF DEFENSE.
Washington, February 22, 1963.

Memorandum for the Secretary of the Navy.
Subject: Nuclear power for surface warships (U).

As suggested in your memorandum to me of January 23, 1963, members of my staff have conferred with Vice Admirals Schoech and Rickover on the subject of nuclear propulsion for CVA-67. However, I do not feel that the subject of nuclear propulsion for surface warships has yet been explored sufficiently to permit a rational decision.

The penultimate paragraph in your memorandum points out that it is important to proceed with the nuclear version of CVA-67 in order to "*** let us utilize the most advanced proven technology *** fill a notable gap in the nuclear propulsion program *** give impetus and support for nuclear pro

pulsion technology, moving us further along the road to the nuclear Navy we envisage for the future." These are forceful arguments, but I am sure you realize that they depend upon the assumption that the future Navy will, indeed, make full use of nuclear power. It is precisely this question which lies at the heart of the matter; far more so than the question of whether CVA-67 itself should or should not be nuclear powered.

Accordingly, I should like you to undertake a comprehensive, quantitative study of this matter. This study should consider the design of the future carrier striking force in the broadest possible context. You should consider the implications of nuclear power on the composition of the task force. How many escort vessels of what type should be included? Should ASW escort be provided in the conventional manner, or should it envision added emphasis on nuclear submarines? How is replenishment of aviation fuel and ordnance to be accomplished? Should the underway replenishment ships also be nuclear? How should the Navy be deployed around the world? Would nuclear power speak for a modification of the present concept of the 1st, 2d, 6th, and 7th Fleets? Realizing that we will have a large number of conventionally powered surface vessels in the inventory for some time to come, how should we approach the "ultimate" design? Is it feasible to take advantage of nuclear propulsion in the near future, or must we consider this as a long-range objective, whose benefits will not be available for some time? Should there be a priority among the various types of ships for application of nuclear propulsion, or would a uniform application be preferable? What are the implications on force size? Would nuclear propulsion allow us to reduce the total number of carriers and/or carrier task forces?

As a general guide, I am interested in achieving the most efficient possible naval forces, defining efficiency as achieving the most beneficial military results for a given expenditure. If nuclear propulsion permits an increase in this efficiency, then advantage should be taken of it. However, I do not feel that a proper evaluation of such possibilities can be made in the absence of a thorough and comprehensive study which goes beyond the narrow consideration of CVA-67 alone.

Accordingly, I am reserving decision on the question of nuclear propulsion for CVA-67. Please advise me when the study referred to above can be completed. ROBERT MCNAMARA.

APPENDIX 5

THE SECRETARY OF THE NAVY,
Washington, April 4, 1963.

Memorandum for the Secretary of Defense. Subject: Nuclear power for surface warships (U). Enclosures: (1) Advantages of nuclear propulsion in surface warships; (2) answers to specific questions raised by SecDef's memorandum of February 22, 1963, to SecNav; (3) policy on nuclear power for surface warships.

In my memorandum of January 23, 1963, to you, I discussed the results of our most recent review of nuclear power for the surface warships in the light of our recent operating experience with the prototype ships. I recommended that an improved four-reactor nuclear propulsion plant, as described in Dr. Seaborg's letter of January 7, 1963, to you, be substituted for the oil-fueled plant now programed for the fiscal year 1963 aircraft carrier CVA-67.

In your memorandum of February 22, 1963, to me, you stated that the assumption the future Navy will make full use of nuclear power lies at the heart of the matter of whether we should change the CVA-67 to nuclear propulsion. I concur. Your memorandum requested a comprehensive, quantitative study of this matter and posed a number of specific questions primarily related to the implications of nuclear power on the design and use of the future carrier strike force. You further stated that if nuclear propulsion permits an increase in the beneficial military results for a given expenditure, then advantage should be taken of it. As discussed below, I have concluded that nuclear propulsion does permit a significant increase in the beneficial military results for a given expenditure and that we must exploit and take maximum advantage of it.

Since I took office over a year ago, the Chief of Naval Operations and I have studied at great length the Navy's nuclear propulsion program and its application to surface warships. My memorandum to you of March 30, 1962, and Jan