The linkage between nuclear energy and rocket technology long antedates Los Alamos and Peenemunde. The earliest pioneers of the Space Age soon realized the attractiveness of atomic propulsion for space flight.
Soon after then end of the Second World War, analysis began of the prospects for turning this long-held dream into reality. These early analyses quickly established that engine specific impulse increases with the square root of the exhaust temperature, and increases with the square root of the decrease in the molecular weight of the exhaust gas. The quest for improved performance thus led to a search for ever higher reactor operating temperatures, and molecular (or monatomic) hydrogen was soon identified as the optimal propellant. In addition, it was soon realized that attractive system designs required minimizing the mass of radiation shielding, which in turn required minimizing the size of the reactor core itself, since reflector and shield mass grows with the cube of the radius of the core. Reactors considered included so-called "fast" unmoderated reactors, as well as moderated thermal reactors, which require more shielding but use less fuel than fast reactors.
A - EARLY SPECULATION
As early as 1907, American rocket pioneer Robert Goddard had concluded that nuclear propulsion would be essential for interplanetary exploration, if only a means could be found to liberate the energy of the atom. Other early pioneers either quickly despaired of the practicality of nuclear propulsion, like Tsiolkovsky, or ignored nuclear propulsion entirely, like Oberth.(1)
Goddard was perhaps the first to contemplate the application of atomic energy to rocketry in 1906 and 1907, though he noted that the rate of decay of radium was far too slow to provide a practical means of interplanetary travel.(2) While he failed in his initial attempt to have this speculation published, it is characteristic of Goddard's penchant for secrecy that these calculations remained in his private notebooks, with Esnault-Pelterie being the first to publicly suggest the use of atomic energy for rocketry. However, Goddard remained convinced of the ultimate utility of solar and atomic propulsion for interplanetary travel:(3)
"It is evident, from the calculations made regarding the use of solar energy in space, that the most extreme speeds will be produced by solar, rather than by chemical energy.... If it is possible to obtain infra-atomic energy, the matter of transportation would be comparatively simple, and a large body could be sent from the solar system... Further, atomic disintegration may open the way for the creation of what might be called artificial atoms, in which energy might be stored by many high speed particles. This tremendous amount of energy could be liberated when these artificial atoms were broken up, or the particles were removed gradually."
Although his contributions have been obscured by subsequent developments, Robert Esnault-Pelterie continued in Verne's tradition of French leadership in the interplanetary project. Esnault-Pelterie was one of the pioneers of French aviation, whose contributions include the first all-metal monoplane, which he built in 1907.(4) REP's work culminated in 1930, with the publication of his Astronautics, which constituted a landmark review of the problems and prospects of space travel. A subsequent edition in 1934 gave considerable attention to interplanetary travel, including the applications of nuclear power.(5)
On 15 November 1912, Esnault-Pelterie presented a paper to the Physics Society of France. In one of the first scientific discussions of the problems of space travel, he suggested that atomic energy would hold the key to solving the problem of reaching the Moon and other celestial bodies. Although long a proponent of nuclear propulsion, by the early 1930s the work of others on the potential of chemical propulsion had convinced him that nuclear propulsion would not be required to accomplish lunar missions.(6)
" ... I must agree with the opinion of the Germans (who have considered the question): even with an oxygen-hydrogen mix, a trip to the moon and back is at the limit of our possibilities... This formidable realization can be only precarious and subject to the terrible hazard of a prolonged trip outside the field of gravity."
B - BRITISH INTERPLANETARY SOCIETY
Members of the British Interplanetary Society were skeptical of the long-term promise of chemical propellants, and from the outset were proponents of nuclear propulsion:(7)
"Any chemically-fuelled spaceships will be unwieldy, fabulously expensive beasts with fuel consumptions measured by the thousands of tons for a single voyage... The giant chemically-powered spaceship may never be developed -- just as the steam-driven aeroplane was never built... Atomic power is hardly likely to advance the conquest of space by more than ten years, but it may make it a really practical proposition almost from the beginning, which otherwise would certainly not have been the case... the "cheapest" journey to Mars -- as far as fuel is concerned -- lasts 258 days. With an atomic ship, travelling by a more direct route at quite a moderate speed, it need take only two or three weeks."
Thus the Society produced the first design for a nuclear powered interplanetary spacecraft (Figure II-1).(8) Eric Burgess, a Fellow of the Society, also concluded that nuclear propulsion would be essential for interplanetary travel, and produced a surprisingly modern looking design for such a vehicle.(9)
C - REACTOR CONCEPTS
A nuclear rocket engine works by transfering heat from the reactor core to a fluid propellant, and the performance of the engine increases with higher operating temperatures.(10) The most effective propellants, which produce the highest specific impulse, are those with the lowest molecular weight. Thus the ideal propellant for a nuclear rocket is mon-atomic hydrogen. But since molecular hydrogen only disassociates at very high temperatures (above 2300o K at pressures of 10 atmospheres), liquid molecular hydrogen is typically used as a propellant. Gains in specific impulse can result from reactor operations which produce mon-atomic hydrogen in their exhaust.
A range of options are available for rocket engine reactor configurations. One choice concerns whether the reactor operates using the fast neutrons produced by fission reactions (a fast reactor), or whether these neutrons are moderated to increase the likelihood of producing subsequent fission reactions (a thermal reactor). Between fast and thermal reactors are intermediate and epithermal reactors. Another design choice for thermal reactors relates to the internal distribution of fuel and other reactor elements -- which may be heterogenous or homogenous.
Fast and thermal reactors impose constraints on the choice of reactor core materials. If thermal reactors are to avoid the high fuel loadings required by fast reactors, core materials must have low absorption cross sections. Thus Niobium Carbide and Zirconium Carbide are suitable materials for coatings in thermal reactors. The use of Hafnium Carbide and Tantalum Carbide, with absorption cross sections one to three orders of magnitude greater, are only suitable for use in fast reactors. Tungsten is an attractive material, given its low vaporization rate, which is at least an order of magnitude less than other high temperature materials. But various tungsten isotopes have different absorption cross sections, with Tungsten-184 having a cross section only twice that of Niobium Carbide, while the absorption cross section of of natural Tungsten is ten times that of Niobium Carbide.
Fast reactors are normally preferred for long-life operations, while thermal reactors may be more suitable for burst-mode power or propulsion applications.(11)
"Missions requiring high level alert power equivalent to years of integrated operations can use a fast reactor. The high fissile loading of such reactors allows high total energy (megawatt years) opeation... missions with high level burst power, but small integrated total energy can use a thermal reactor. Such applications benefit from the relatively long neutron lifetime and large delayed neutron fraction characteristic of thermal reactors. This facilitates more precise and rapid control of the burst power, and the ability to use hydrogen coolant with minimal effect on reactivity."
1 - Fast Reactors
The fast reactor is based on chain reactions propagated by the high energy neutrons produced by fission reactions. In the absence of moderator elements, the fast reactor is essentially a homogenous reactor. Typically fuel will occupy from 30% to 50% of the core material volume. To compensate for the low probability of fast neutrons producing fission reactions, a fast reactor will require a large amount of fuel, with representative fuel loadings ranging from 200 kg to 700 kg. However, high fuel concentrations permit construction of fairly compact fast reactors, which means that other system components, can be of reduced size and mass relative to thermal reactors. As a result, fast reactors are of particular interest for low-power applications in which the mass of the reactor may be a significant percentage of the overall mass of the vehicle.
However, there are a number of challenges in fast reactor design. The high power densities that are characteristic of these reactors impose significant thermal stress on core materials. Thermal deformation can lead to altered propellant flow patterns, which can produce local hot spots leading to further theral damage. Thermal expansion of core elements must be accomdated while avoiding designs that are vulnerable to vibrations during reactor operation. In addition, the intense fast neutron flux from fast reactors can lead to signficant radiation damage to reactor and system components, and as a result requires more compentent and more massive shielding.
2 - Thermal Reactors
Fast neutrons produced by a fission reaction can be slowed by elastic collisions with the nuclei of non-absorbing materials of relatively low mass, such as berylium, lithium or carbon, which are called moderators. If moderators slow neutrons to the point that they are in thermal equilibrium with reactor materials, the neutrons are termed thermal neutrons.
The probability of a neutron being captured and producing a fission reaction is much higher for low energy thermal neutrons than for higher energy fast neutrons. Thus a reactor using thermal neutrons requires a much smaller critical mass than does a fast reactor. Thermal reactor fuel loadings are typically several tens of kilograms of highly enriched uranium, compared to the hundreds of kilograms required for fast reactors. This also permits less complicated fuel element and core material designs.
Good moderator materials should have low atomic weight, low neutron absorption cross-section, and high thermal conductivity. They must be chemically compatible with the propellant operating environment, or they may require additional coatings for protection. Beryllium is an excellent moderator, it has a relatively low melting temperature, which restricts its use in reactors to the role of an external reflector. Although graphite carbon is a less effective moderator than beryllium, it has outstanding high temperature operating characteristics. However, carbon requires coatings to protect it from erosion by hydrogen propellant.
3 - Heterogenous and Homogenous Reactors
Thermal reactor designs are characterized as either heterogenous or homogenous depending on whether the moderator material is largely external to the fueled core (heterogenous), or distributed within the core (homogenous).
Homogenous thermal reactors are similar in construction to fast reactors, except that some fuel elements are replaced by moderator materials within the reactor core. The proximity of fuel and moderator requires the moderator to operate at temperatures similar to those of the fuel, which constrains the choice of moderator materials to those that can withstand high temperatures and the reducing environment produced by the hot hydrogen propellant. Although some moderators such as graphite carbon have excellent high temperature characteristics, they require refractory coatings to minimize hydrogen errosion.
The simplest heterogenous reactor consists of a fuel core surrounded by a moderator, which serves as a neutron reflector, reducing the number of neutrons that escape from the core, and thus reducing the amount of fuel required to sustain reactor operations. The separtion of the moderator from the fuel elements also reduces the temperature of the moderator, which increases the range of moderator materials that may be used to include berylium and water, which have thermal properties that are inconsistent with use in homogenous reactors.
4 - Shielding
Although most of the 200 Mev released in a single fission of a uranium atom is in the kinetic energy of the fission fragments, the reaction also produces fast neutrons and gamma rays that are hazardous to vehicle components and crew members. Thus some form of shielding must be interposed between the reactor and the rest of the space vehicle. Hydrogen is an effective neutron shield, and thus hydrogen propellants, or hydrogenous materials, such as lithium hydride, are effective in slowing fast neutrons. High cross-section absorbers, such as boron, are included in shields to capture the resulting thermal neutrons. Shielding against gamma rays results from various types of interactions between the gamma ray photon and electrons, and thus depends on the density of electrons in the shield material. Heavy elements, such as lead, bismuth and tungsten, as well as iron and nickel, are effective gamma ray shields.
D - EARLY DEVELOPMENTS
However, other rocket pioneers were skeptical of the prospects for nuclear propulsion. Wernher von Braun concluded:(12)
"... barring entirely new discoveries in the field of nuclear energy, it is not likely that chemical propulsion will be replaced by an atomic power plant. In our proposed lunar trip, the landing on the moon's surface and the subsequent take-off require several hundred tons of thrust, and this involves the transfer of energy on such a scale that chemical propellants will prove to be superior. Nuclear rocket drives may some day be successful for flights between heavenly bodies for surveying and photographing purposes, but not when landings must be made. The weight alone of an atomic plan as we now conceive it would make the landing and take-off of huge rocket ships not only uneconomical but virtually impossible."
In the wake of World War II there was little delay in studies to combine newly developed nuclear and rocket technologies. Douglas Aircraft first published a study of a fission powered rocket in July 1946, and H.S. Tsien analyzed the propulsion capabilities of a hydrogen cooled graphite reactor at MIT in May 1947.
1 - North American Aviation - 1947
2 - General Dynamics - 1950s
1. Reupke, William, "Nuclear Propulsion Before the Discovery of Fission," 7th Symposium on Space Nuclear Power Systems, 1990, (University of New Mexico, Albuquerque, NM, 7-10 January 1990).
2. Goddard, Robert, "On the Possibility of Navigating Interplanetary Space," unpublished manuscript, 3 October 1907, in The Papers of Robert H. Goddard, (New York, McGraw-Hill, 1970), volume 1, page 22.
3. Goddard, Robert, "The Ultimate in Jet Propulsion," unpublished manuscript, 1943, in The Papers of Robert H. Goddard, (New York, McGraw-Hill, 1970), volume III, page 1611-1612.
4. Blosset, Lise, "Robert Esnault-Pelterie: Space Pioneer," in Durant, Frederick, editor, First Steps Toward Space, (Washington, Smithsonian Institution Press, 1974), Smithsonian Annals of Flight, Number Ten, Chapter 2, pages 5-31.
5. Somewhat surprisingly, there is no English-language edition of these books, which precludes a more systematic treatment of Esnault-Pelterie's contributions. The excerpts quoted here were translated by Tiffany Tyler for this study.
6. ibid, page 202.
7. Clarke, Arthur C., "The Challenge of the Spaceship," Journal of the British Interplanetary Society, December 1946, page 66-79.
8. Gatland, Kenneth, "Orbital Rockets," Journal of the British Interplanetary Society, volume 10, number 3, May 1951, page 88-123.
9. Burgess, Eric, Rocket Propulsion, (London, Chapman & Hall, 1954), chapter eight.
10. Willaume, R.A., et al, Nuclear, Thermal and Electric Rocket Propulsion -- Fundamentals, Systems and Applications, AGARD/NATO First and Second Lecture Series, 12-21 November 1962 and 28 September - 2 October 1964, (New York, Gordon and Breach, 1967), is the primary source for the following discussion.
11. Brookhaven National Laboratory, Particle Bed Reactor Multimegawatt Concepts, BNL-39495, March 1987, page 6.
12. von Braun, Wernher, Whipple, Fred, and Ley, Willy, Conquest of the Moon, (New York, Viking, 1953), page 5.