Aircraft Nuclear Propulsion - Technology

Technology

The principles behind using atomic energy for the propulsion of aircraft were developed early in the atomic age. As early as 1942 Enrico Fermi and his associates involved with the Manhattan District Project discussed the use of atomic power to propel aircraft.[1] It was in 1946 that a study by John Hopkins University's Applied Physics Laboratory delineated the potentials and problems of using atomic power for aircraft propulsion. Chief amongst the problems at the time was the lack of data on the effects of radiation on materials which would be used in a design.[2] Some of the other basic problems were the possible release of radioactive fission products or isotopes during normal operation or due to any accident, shielding the crew and persons on the ground from radiation, and the selection of test sites and ranges. There was the potential for the release of radioactive materials to the atmosphere and the problems of direct radiation during operational use.[3] The requirements for an operational nuclear aircraft were that, even under the most adverse conditions, the aircraft did not add materially to the general background atmospheric radioactivity and that while in use the aircraft restricted all harmful radiation to within the craft or a predesignated exclusion area.[4]

In 1946 the interest in atomic aircraft developed into a long-lived project know as NEPA, for Nuclear energy for the Propulsion of Aircraft.[5] The NEPA project, which started in May, was controlled by the United States Air Force (USAF) and was therefore oriented towards developing both an atomic- powered long-range strategic bomber and high-performance aircraft. Nuclear power showed promise in both fields because of its dual nature of long-lasting fuel supply and the high temperatures theoretically possible using a reactor. However, in a paper in 1957 Kelly Johnson and F. A. Cleveland, both of Lockheed Aircraft Corporation, wrote, "It appears that the strategic bomber, by requiring both high speed and great endurance and because of the inherent low-altitude potential advantages over similar chemical airplanes, will be the first candidate for a nuclear power plant."[6]

The NEPA contract was with the Fairchild Engine & Airframe Co., and the work was conducted at Oak Ridge. By the end of 1948 the USAF had invested approximately ten million dollars in the program.[7] Extensive studies were conducted under NEPA from 1946 until 1951, at which time it was replaced by the joint Atomic Energy Commission (AEC) / USAF ANP program. The ANP program set forth the ambitious goal of full-scale development of aircraft reactor and engine systems. One of the factors that led to the creation of the ANP program was a study done at MIT by a group convened by the AEC in 1948 to look at the potential uses of atomic powered flight. "This study group, known as the Lexington Project, came to the conclusion that nuclear aircraft (manned) were likely less difficult than nuclear ramjets, which, in turn, would be less difficult than nuclear rockets to develop."[8] Ironically, this turned out to be the reverse of the proper order of difficulty, as later research and development would prove. Although nuclear ramjets, under Project Pluto, and nuclear rockets, under Project Rover, were successfully tested at the levels needed for operational use, an operational level atomic aircraft powerplant was never developed. In 1954, Raymond Clare Briant, who was then the director of the ANP Project stated that "manned nuclear aircraft pose the most difficult engineering development job yet attempted within this century."[9]

Unfortunately the ANP program wasn't very well organized. Instead of trying to develop one aspect of the technology to a working stage the effort was spread out over a number of areas. Part of the problem was that, under the conventional guidelines, the AEC was responsible for reactor development while the Air Force was responsible for development of the remainder of the system. Therefore the project was divided into two parts which needed to work closely together, but these two parts were managed by totally separate entities.

Under the ANP program the General Electric Co., at Evendale, Cincinnati was issued a contract to develop a direct-cycle turbojet, and Pratt & Whitney Aircraft Division of United Aircraft Corp. was authorized to study an indirect cycle and work was started at the Connecticut Aircraft Nuclear Engine Laboratory (CANEL).[10] In the direct air cycle air enters through the compressor stage of one or more turbojets. From there the air passes through a plenum an is directed through the reactor core. The air, acting as the reactor coolant, is rapidly heated as it travels through the core. After passing through the reactor the air passes through another plenum and is directed to the turbine section of the turbojet(s) and from there out through the tailpipe.[11] An indirect system is very similar, except that the air does not pass through the reactor itself. After passing through the compressor the air passes through a heat exchanger. The heat generated by the reactor is carried by a working fluid to this heat exchanger. The air then passes through the turbine and out the tailpipe as above. The working fluid in the indirect cycle is usually a dense fluid, such as a liquid metal, or highly pressurized water. This allows more heat energy to be transfer, thereby increasing the efficiency of the system. [12]

In an article in the SAE Journal, L.W. Credit wrote, "Of three alternatives for achieving flight reliability in nuclear aircraft through component or system redundancy, the single-reactor, all-nuclear aircraft seems to be the optimum design." [13] The other two alternatives were a dual-reactor system and a combination nuclear-chemical (combustion) system. Originally the ANP program was to develop an indirect cycle, single reactor propulsion system. However, a petition by General Electric to the government allowed them to develop the direct cycle system. GE claimed that the direct cycle was simpler and therefore would have a shorter development time. For the indirect cycle system, Pratt & Whitney developed the super-critical water reactor, in which the working fluid is water heated to 1,500 degrees fahrenheit, but kept in a liquid state by pressurizing to 5,000psi. This avoided the problems of using a liquid metal working fluid. The United States has never favored the operational use of liquid metal reactors. To date all military reactors in active service, with the exception of the one liquid sodium reactor on the attack submarine USS Seawolf, have been of the Pressurized Water Reactor (PWR) type. Even the USS Seawolf experienced enough problems that the liquid sodium reactor was replaced with one of a PWR design after a few years in service.

Part of the ANP program was the X-6 program. Beginning in 1952, the designated goal of the X-6 program was to produce two flying testbeds powered by atomic energy. The test program started by testing shielding problems. A B-36 was converted for this purpose. This aircraft was referred to as the Nuclear Test Aircraft (NTA). The NTA began its life as a Convair B-36H bomber, but after conversion it was redesignated as an NB-36H. It was modified to carry a small air cooled reactor in the aft bomb bay and to provide shielding for the crew. The NTA incorporated shielding around the reactor itself and a totally new nose section which housed a twelve ton lead and rubber shielded compartment for the crew. There were also water jackets in the fuselage and behind the crew compartment to absorb radiation. The reactor was made critical in flight on several occasions and the aircraft was used for many radiation and shielding experiments.

Convair's successful flight program with the B-36 carrying a flight test reactor (July 1955 - March 1957)" showed that the "aircraft normally would pose no threat, even if flying low. The principal concerns would be: (a) accidents which cause the release of fission products from the reactors, and (b) the dosage from exposure to leakage radioactivity (in the direct cycle concept).[14]

It was decided that the risks caused by radiation were no greater than the risks that had been incurred during the development of steam and electric power, the airplane, the automobile, or the rocket.[15]

The B-36 was also to provide the basis for the actual X-6 aircraft. At the time the B-36 was the only existing, time tested, airframe large and powerful enough to carry the expected engine and shield weight. The engine chosen was the J53 turbojet.[16] At the time the J53 was a conventional turbojet in the planning stage at General Electric. The J53 was a high- performance design and it was felt that conversion to nuclear power would present no more difficulty than any other design then in use. In the early stages of the program, before GE's petition, it was planned to connect the J53 to a liquid-metal reactor for use on the X-6. The original propulsion system was to have weighed 165,000 pounds. This was composed of a 10,000 pound reactor, 60,000 pounds of reactor shielding, 37,000 pounds of crew shielding, and a total engine weight of 18,000 pounds plus an additional 40,000 pounds for ducts and accessories.[17] After experiencing development problems with the J53, GE resorted to the J47 as the powerplant. J47s converted for nuclear testing were referred to as X-39s.

It should be noted that the United States was not the only country working on atomic aircraft in the early years. The Soviet Union had a few projects of their own. One aircraft, a flying boat, proposed in 1950 would have had a flying weight of 1,000 tons.

It was planned to equip the giant airplane with four atomic turbo-prop engines. The wing span was more than 130 meters, and the total power of the engines exceeded one-half million horsepower. This airplane was supposed to carry 1,000 passengers and 100 tons of load at a speed of 1,000 kilometers per hour.[18]

It was planned to surround the reactor with five layers of shielding. The layers were supposed to be as follows: first layer - beryllium oxide reflector; second layer - liquid sodium for removing heat from the walls; third layer - cadmium, for absorbing slow neutrons; forth layer - paraffin wax, for slowing down fast neutrons; fifth layer - a steel shell, for absorbing slow neutrons and gamma-rays. Such multilayer 'armor' permits decreasing the weight and size of the necessary shielding. The coolant was liquid lead.[19]

The Soviets studied many of the same options the United States considered; both direct and indirect cycles, turbo-props, shadow shielding, and the special ground handling needed. One fact that is striking is that in the Soviet design the total weight of the atomic power plant was to be 80 tons.[20] 80 tons is equal to 160,000 pounds, which compared to the original figures for the X-6 propulsion system, which was 165,000 pounds, was practically identical.

The reference to 'shadow shielding' above is to the practice of dividing the shields between the reactor and the crew, the crew being in the 'shadow' created by the shields. This is also referred to as the divided shield concept.

If it were possible to put as much shielding on the reactor as is done on ground reactors, we could reduce the radiation therefrom to a negligible amount. But the total weight of shielding required to do this would be prohibitive; in fact, we are forced to the so-called 'divided shield' concept in order to reduce total shield weight to an acceptable amount. Divided shielding is, of course, simply a division of the shielding between the reactor and the crew compartment in such a fashion as to result in near- minimum total shielding weight.[21]

Distributing the shields lessens the total shield weight, but it also means that the majority of the aircraft would have been exposed to higher levels of radiation. And once on the ground more radiation would penetrate the surrounding area. These problems were to be overcome by newer materials and by designing the aircraft's servicing equipment with the higher radiation levels in mind. Divided the shields also had some other benefits;

The directional nature of the radiation leads also to the fact that aircraft structure and components are useful as shielding material, and judicious use of such things as the wing box, landing gear, pay load, and fuel for landing go-arounds can reduce the thickness of shielding required on the crew compartment rear face.[22]

The problem with shield weight was one of two major problems which surfaced during the program. The other was increasing reactor performance. The ANP program focused a great deal of effort on developing the divided shield concept, decreasing the required shield size by decreasing reactor size via increasing reactor power density, increasing the operating temperature of the reactor to boost efficiency and therefore aircraft performance, and utilizing the reduced shield mass in aircraft design.[23] Although work on an actual airframe never got very far, a great deal of work was accomplished on the power plants.

General Electric ran a series of very successful experiments using the direct cycle concept. These were referred to as the Heat Transfer Reactor Experiment (HTRE) series. The series involved three reactors, HTRE-1 through HTRE-3. HTRE-1 became HTRE-2 at the conclusion of its test program. HTRE-1 (and therefore HTRE-2) successfully ran one X-39 (modified J-47) solely under nuclear power. HTRE-3 was the closest to a flight article the program came. It was solid moderated, as opposed to the earlier reactors which were water moderated, and it powered two X-39s at higher power levels. HTRE-3 was limited by the two turbojets, but it could have powered larger jets at even higher power levels. HTRE-1 was principally a proof of concept reactor. "HTRE-1 achieved a number of full-power runs that demonstrated conclusively the feasibility of operating a jet engine on nuclear power."[24] HTRE-2 was simply HTRE-1 modified to test advanced reactor sections in a central hexagonal chamber. In this way new reactor designs could be tested without the need to build a totally new reactor from scratch. The experience gained from HTRE-1 and HTRE-2 was used in the construction of HTRE-3. HTRE-3 was the final test item designed to prove the feasibility of producing an actual aircraft powerplant. "The design and testing of HTRE-3 has advanced the direct-cycle program beyond the question of feasibility to the problems of engineering optimization."[25]

All three of the HTRE reactors were of the standard direct cycle configuration, with the addition of a chemical combustor just upstream from the turbines. This combustor allowed the jets to be started on chemical power and then be switched over to atomic heat as the reactor was brought up to operating temperatures. The operational system may have also utilized a chemical combustor for use during takeoff and landing, and possibly target penetration, when the reactors relatively slow response time could be a disadvantage.[26]

The HTRE either met or exceeded their goals, but although all had reactor cores of roughly the size needed to fit into an aircraft, none of the HTREs were designed to be a prototype of a flight system[27]; the series showed that it then appeared "possible and practical with the technology in hand to build a flyable reactor of the same materials as HTRE-3 and similar in physical size."[28] Despite the fact that HTRE-3 didn't produce the power that would have been needed for flight, that was mainly because it was not an optimized design; it was designed simply as a research reactor, to prove the concepts needed for a flight article.

At the end of the HTRE run the probability of flying a reactor seemed high. The test runs showed that a reactor using the same materials as HTRE-3, and which could power a gas-turbine powerplant, could have been built at that time. Such a reactor would meet all of the requirements needed for a flight ready unit.[29] In their paper Kelly Johnson and F. A. Cleveland also stated that "when improved materials are available, we would expect the nuclear power plant to advance rapidly in its overall efficiency, with a consequent improvement in ability to install such power plants in airplanes of smaller size than those currently contemplated."[30]

While GE was working on the direct cycle, Pratt & Whitney (P&W) was working on the indirect cycle. However, progress went much slower that it did with the HTREs. P&W never ran a practical test system. In fact their work was limited to component testing. In addition to work on the super-critical water reactor P&W worked with liquid metal coolant designs. It was the latter that received the most attention. The two major designs were a solid core reactor, in which the liquid metal circulated through a solid reactor core, and a circulating-fuel design, in which fuel was mixed with the coolant and critical mass was achieved as the coolant circulated through a central core. After the circulating-fuel design showed promise, work on the super-critical reactor was halted. P&W did accomplish a great deal on the design of liquid metal cooling loops, corrosion prevention, and heat exchanger design. However, P&W work at CANEL never led to a test reactor, much less one which was flight ready. In the long run the indirect cycle showed more promise, but it also required a great deal more developmental work.

While these test programs were successful, there were other programs which weren't. A number of programs were begun at a great cost of time and money, only to be dropped when the program went through one of its many reorientations. The official U.S. government report on the ANP project lists such programs. A Flight Engine Test facility was built in Idaho for use to test the flight engine both on the ground and in the test aircraft. This facility cost over eight million dollars, yet it was never used during the ANP program, other than as a storage building, because the flight program was cancelled. A radiator laboratory was constructed at CANEL for use in studying liquid metal to air heat transfer. After spending over six million dollars the construction was halted with only a shell completed because the Air Force changed its mind. Another laboratory was built at CANEL to study vacuum conditions. This laboratory cost over a million dollars, and it entered use in March 1961, the same month that the ANP program was cancelled. These were only the largest of the wastes. There were numerous instances of wasted time and money, none of which can really be blamed on the technicians, since the leaders changed their minds and the equipment went unused.

Overall the technology seems to have been there, yet the ANP program died. GE's HTRE series proved that the direct cycle concept would work. P&W was making progress, slowly but surely, on the indirect cycle. The NTA reactor tests demonstrated that aircraft shielding could be done effectively. A myriad of smaller developments, new metals, synthetic lubricants, all worked out and were available to produce an aircraft. In 1960 Kenneth Gantz wrote, "The taming of the atom, coupled with the technological advances in aerodynamic and structural efficiencies achieved over the past several decades, now brings atomic-powered aircraft and missiles within our grasp."[31] But if it wasn't killed by technology, what were the reasons for the program's demise? The answer to that question follows in the next section.