A spacecraft must be powered, otherwise it can't accelerate or is useless. Electromagnetically driven propulsion systems require a reliable power supply. The ion rocket and VASIMR are likely to be employed over the direct nuclear rocket for reasons of radiation safety. Large spacecraft sent to deep space this will require a power source that is considerably superior to what currently exists. Current power systems are solar photovoltaic generators and weak decay reactors that are not true nuclear reactors. Photovoltaic systems work well for small craft that travel close enough to the sun. The RTG radioactive type of decay reactors have been employed on craft sent to the outer gas giant planets.
It is likely that some form of nuclear energy will have to be used to send spacecraft by propulsive means to speeds of 100 km/sec or higher to the outer planets. Solar energy may work well for the inner planets. However, the energy requirements for reaching Mercury are comparatively high and a high specific impulse propulsion system must be powered to reach the planet in a timely manner. Solar energy may work well enough for this domain of space exploration. However, solar radiation at the gas giant planets is a fraction of the irradiance here on Earth. Solar photovoltaic cells would have to be inordinately large.
The use of nuclear energy in space is controversial. Many express concern over the risk of contamination if there is an accident, such as if nuclear material is released in a crash during the launch phase. A similar concern exists over putting nuclear powered satellites in Earth orbit. In low Earth orbit small amounts of atmospheric drag can bring the craft back to Earth and release radioactive materials. This happened in 1978 to a Russian nuclear powered satellite Cosmos 954 which fell in Canada. These matters doubtless can't be ignored. The launch of such materials has to have a degree of safety assurance so that they are not released in an accident. It is further probably unwise to park nuclear reactors in Earth orbit. Yet if we are to explore the outer planets it is likely that nuclear reactors will be required to power their propulsion systems.
The most common method for powering spacecraft is with solar photovoltaic cells. These have their origin with Albert Einstein who recognized that electrons generated from photons interacting with a metal were an indication of the particle-like nature of photons. This result was important in the development of quantum mechanics. The photovoltaic cell employs two materials with different materials with different potential functions for electrons. An electron excited in one material is drawn to the other through a circuit. This is the basis for the p-n junction photovoltaic cell.
Back electrical contact
Fig. 4.1. The left is a schematic of how photons induce an electron-hole charge separation in an n-p junction and the operation of a solar cell. The right is photograph of an actual photovoltaic cell.
Fig. 4.1. The left is a schematic of how photons induce an electron-hole charge separation in an n-p junction and the operation of a solar cell. The right is photograph of an actual photovoltaic cell.
The most common photovoltaic cell is the silicon cell [4.1]. Silicon (Si) is a group 14 atom by virtue of its placement on the periodic table. A Si atom has 4 valence electrons in its outer shell, just as carbon. Silicon atoms may covalently bond with each other to form a crystalline solid with a long range ordering of atoms in three dimensions. Si may also form an amorphous solid with no long range order. The scale for this crystalline order defines terms for the crystalline structure of silicon; poly-crystalline, micro-crystalline, nano-crystalline depending upon the size of the crystal "grains" which make up the solid. Solar cells are constructed from each of these types of silicon, the most common being poly-crystalline. Silicon is a semiconductor, which has certain bands of energies electrons exist in, and electrons are forbidden to exist in an energy state between these bands. The two main bands are the conduction band for electrons that flow between the Si ions and the valence band of electrons bound to the Si ions. These forbidden energies are called the band gap, which requires quantum mechanics to understand fully. This will be avoided here. However, the existence of band gaps is central to the physics of semiconductors.
In a solid the electrons involved with the conduction of a current have energies in the Fermi level, which obeys strange physical properties due to quantum mechanics. However, for silicon this Fermi level is forbidden. This makes silicon a poor electrical conductor at ordinary temperatures. To improve the conductivity of silicon impurities are introduced, or "doped," with very small amounts of atoms from either group 13 or group 15 of the periodic table. These dopant atoms take the place of the silicon atoms in a few lattice slots in the crystal, and bond with their neighboring Si atoms with analogous electronic interactions the Si atoms exhibit. Group 13 atoms have only 3 valence electrons which results in an electron deficiency, and group 15 atoms have 5 valence electrons which results in an electron excess. Since the group 13 atom bonds to the crystal in the same way as silicon does this results in a positive hole. Similarly the group 15 atom gives an excess electron. The holes and electrons may then move freely around the solid and in so doing change the conduction properties of the solid. Silicon with an excess of holes is a p-type semiconductor and silicon with an excess of electrons is defined as n-type. Common group 13 atoms are aluminium and gallium, and common group 15 atoms are phosphorus and arsenic. Both n-type and p-type silicon are electrically neutral, for they have the same numbers of positive and negative charges, yet n-type silicon has negative charge carriers and p-type silicon has positive charge carriers.
A solar cell consists of a layer of an n-type silicon on a p-type silicon layer. Photons will, as Einstein explained the photoelectric effect, cause the respective charge carriers, electrons and holes, to jump into p-type and n-type silicon respectively. This is an energetically excited state. A circuit between the two silicon layers permits the electrons and holes to travel back to their respective silicon types. The excess energy is then extracted by attaching a load to the circuit. More specifically, the electrons in the valence band, those electrons that bind the Si atoms together absorb a photon and are excited into the conduction band of the p-type silicon. The presence of holes in the conduction band decrease the energy gap between the valence and conduction bands. Similarly photon absorption in the n-type silicon produces holes in its conduction band. This results in the generation of electron-hole pairs that are quantum mechanically correlated with each other. This energy imbalance is restored by permitting a current to pass between the layers. This p-n junction is the basis for the diode as well.
The efficiency of solar cells is 10-15%. Thus for a solar irradiance of 1000 W/m2 a meter square of photovoltaic will light a 100 to 150 watt bulb. Efficiencies have improved some and the production costs have declined. It is likely that solar photovoltaics will be an increasing aspect of the electrical generating infrastructure around the world in the 21st century. The initial use of solar photovoltaic cells was with satellites, where early high costs could only be justified for space applications. Earth orbiting satellites and spacecraft sent within the inner solar system have almost exclusively employed solar panels. For spacecraft sent to the outer solar system radioisotope thermal generators must be used. Currently solar power most often generates electricity for systems on board the craft. However, solar power could also drive an ion thruster for a craft in the inner solar system. There is some interest in generating large amounts of power by solar photo-voltaics for use on Earth. This energy is to be converted to microwaves and beamed back to Earth for their collection and electrical power distribution.
Currently for deep space application radioisotope thermoelectric generators (RTG) are employed [4.2]. The concept employs a piece of radioactive metal in contact with thermocouplers which convert the heat of the material into electrical power. This is not a true nuclear reactor, but only relies upon the radioactive decay of a radionuclide. The heat difference between the radioactive material and the cold of space form the energy difference. This temperature difference is analogous to a waterfall with a paddle wheel, from which energy may be derived. The radioactive material form a core in a metal container that conducts heat. Thermocouplers on this heat conducting wall absorb this heat. The temperature difference with the other end of the thermocoupler, a heat sink, permits a heat flow through this device, where some of this heat energy may be converted to electrical energy. The thermocoupler is an infrared analogue of the photovoltaic cell. It is a p-n junction, where one junction is heated and the other attached to the heat sink.
The radioactive material must decay slowly to have a half-life long enough for the duration of the spacecraft mission. Further, the decay process must either be by a nuclear or weak interaction, which produce a (alpha) particles (helium nuclei due to nuclear decay) or ¡3 (beta) radiation (electrons or positrons due to weak decay) respectively. Some nuclei exhibit electromagnetic decay processes, due to a "reshufflling" of charges in the nucleus. This results in high energy photons called 7 (gamma) rays that are highly penetrating and thus less desirable. Even still the charged alpha and beta radiation scatters with charges that make up the nuclear material to produce secondary X-rays. This process is called bremsstrahlung radiation, which occurs when a charge is accelerated. Radioactive processes often have a small neutron emission level. Neutrons can damage materials over time. As such the container around the radioactive material must be shielded to protect the thermocouplers and other components from damaging exposure. This places constraints on the types of radioactive material that can be used. The most acceptable is 94Pu238 which has the longest half-life and the lowest shielding requirements. Only 1/10 of an inch of material is need for shielding of 94Pu238, where the container is often all that is required. 94Pu238 has a half-life of 87.7 years and is a low gamma and neutron emitter. This means that the RTG will have a power loss by a proportion 1-0.51/97'7 every year. The RTG employed on the Voyager craft were rated at 470 W, which means this has dropped to 80% since launch. This is further compounded by the degradation of the thermocouplers as well.
RTGs are only about 5-10% efficient. This means that over 90% of the radionucleide generates unused heat energy. This is a factor in launch
weight constraints. However, RTGs are relatively cheap to construct and plutonium is available from breeding reactors. RTGs do not have the controlled reactor issues that would arise with a sustained nuclear fission system on a spacecraft. They have then been the optimal compromise for spacecraft with moderate power requirements.
A common application of RTGs is as power sources on a spacecraft sent far into the outer solar system. Spaceprobes that travel far enough from the Sun, the inverse square law gives a ~ (ro/r)2 factor in irradiance decrease, where solar panels are no longer viable. The deep space probes Pioneer 10, Pioneer 11, Voyager 1, Voyager 2, Galileo, Ulysses and Cassini employed RTGs [4.3]. RTGs were used to power the two Viking spacecraft to Mars and powered scientific experiments left on the Moon by the crews of the Apollo 12, 14-17 lunar missions. Apollo 13 carried an RTG, but the lunar lander was used to carry the crew back to Earth after the service module failed. The RTG, or pieces of it, ended up in the ocean Tonga trench.
The Galileo and Cassini missions generated unpopular attention by some over the launching of nuclear materials into space. Considerable protest was made over this. The protests were in some ways high on drama and low on technical reality, but these protests can't be ignored. The concern at large is over the use of nuclear power in space, issues of launch safety and the obvious connection between space nuclear power and military applications.
The SNAP-10A was the only true nuclear reactor used in space by NASA. The Russians employed nuclear reactors with their RORSAT program. The SNAP-10A reactor powered a satellite in low Earth orbit in 1965. However, after 43 days a voltage regulator failed, which caused the reactor to be jettisoned into high Earth orbit. Nuclear power in space has become a flashpoint for those on the anti-nuclear side of the nuclear debate.
It is well known that a nuclear reactor maintains a self-sustained nuclear chain reaction. This approach was pioneered by Enrico Fermi and Leo Szilard in 1942 at the University of Chicago. This first nuclear reactor, sometimes called a pile, was a step in the development of the atomic bomb during World War II. It demonstrated a sustained nuclear fission process with uranium was possible. Neutrons which fissioned the uranium were moderated with the use of graphite. The trick with a sustained nuclear reaction is to prevent the exponential runaway process of a nuclear chain reaction from resulting in a sudden release of energy. A nuclear bomb conversely exploits this. Fuel rods of uranium or plutonium are suspended in some moderating material, graphite or water, to slow neutrons. Control rods are inserted between the fuel rods to absorb neutrons and are inserted into the pile or withdrawn to maintain the nuclear reaction at a controlled rate. Of course for a nuclear reactor that produces large amounts of energy complex cooling systems are required, which must still operate if the reactor is shut down for there is still heating from nuclear decay processes.
The most elementary nuclear reaction is the absorption of a neutron by 92U235 to form an transient unstable isotope 92U236 [4.4]. This isotope
fissions to produce up to three neutrons and fission daughter products. Some y rays are produced by the rearrangement of charged species in the reaction, or by a secondary electromagnetic interaction. The fission daughter products have a high velocity and interact electromagnetically with other nuclei. The relative accelerations between these charged nuclei results in bremmstrahlung radiation in the form of y ray and X-ray photons. Typically the fission daughter nuclei have atomic weights of 100 AMU and 132 AMU. The binding potential per nucleon, a proton or neutron is 7.5 MeV, MeV = million electron volts, for 92U235. An electron volt eV is 1.6x 10-19 j of energy. If the nucleus is split into two nuclei of equal mass, from the average binding energy per nucleon the energy difference in binding energy between 92U235 and its fission products is (8.4-7.5) 235 MeV or 211 MeV. The energy is mostly carried by the nuclear fission products ~ 170 MeV, about 5 MeV is carried of by neutrons and the rest is in y rays and the radioactive energy of the daughter products. An average chemical reaction between two molecular species is around 10 eV, so the nuclear fission produces around 10 million times the energy per elementary interaction. To convert this energy to its heat equivalents, a single gram of 92U235 produces 2 x 1010 cal, while a gram of carbon, the main constituent of coal, produces 7827 cal. A nuclear reactor that consumes a ton of uranium in a year will in principle generate as much energy as a coal fired plant which consumes over a million tons of coal. The ratio between the size of a bomb and its yield for chemical and nuclear explosives scales similarly.
The 92U235 nucleus is readily fissionable by slow, or thermal neutrons. Thermal neutrons refer to neutrons slowed by reactor moderators. This was the first isotope of uranium considered for nuclear energy and the atomic bomb. However, this isotope composes only .7% of uranium isotopes. The remaining percentage is largely 92U238, which is not fissionable by slow neutrons. Yet it was found that 92U238 produced two new atomic elements not previously identified. This reaction is tt238 , TT239 AT 239 i - r>„239
The two new atomic elements are neptunium (Np) and plutonium (Pu), which do not occur naturally on Earth. 94Pu239 fissions with slow or thermal neutrons, which made it the optimal nuclear fuel for reactors and for bomb making material. For reactors plutonium will fission in a stable manner. However, for bombs it was found that a quick exponentially exploding chain reaction could not be achieved unless the material was rapidly compressed. Hence high explosives are used to implode the material as it receives neutrons from a source trigger.
For space power applications it is obvious the energetic advantages are clear [4.5]. As a simple model a spacecraft with 1000 kg of chemical pro-pellant is compared to a spacecraft with the same 1000 kg of reaction mass propelled by an ion thruster or VASIMR thruster powered by a nuclear reactor. For a small nuclear reactor there might be about a 10% energy conversion from nuclear energy to the kinetic energy of the reaction mass and spacecraft. This is due to thermodynamic losses of heat during energy conversion steps. If the reactor has 1 kg of fissionable material, this has as much energy as about 106 kg of chemical propellant. Thermodynamic losses reduce the usable portion of this by a factor of 10. Kinetic energy of a body with a mass to moving at a velocity v is K = ^mv2. The nuclear powered spacecraft then has about 105 times as much kinetic energy available. So if the chemical rocket may change the velocity of the spacecraft by Av, often called "delta vee" the nuclear powered craft has a Av as much as 316 times as large. Thus a nuclear powered VASIMR craft could reach velocities of ~ 3,000 km/sec, compared to a chemically propelled craft of comparable mass and reaction mass with a Av ~ 10 km/sec.
For interstellar travel this is still hopelessly slow. This example is about one percent the speed of light, where this might at best be improved by a factor of 10 for any fusion powered spacecraft. This is why nuclear energy simply can't send a spacecraft to a nearby star within any reasonable time frame. Yet, nuclear power provides clear advantages for interplanetary spacecraft are apparent.
Concerns over the safety of launching actinide materials from the Earth exist. It raises the question of whether nuclear fusion might be used instead. As yet controlled nuclear fusion that offers a net energy output has not been achieved. Yet it is possible that in the 21si century this may change. If nuclear fusion is obtained and employed in spacepower or propulsion the advantages could improve by a factor of 10 or more.
Nuclear fusion is the opposite of nuclear fission, in that light nuclei are forced together into heavier elements and energy is released. Fission is the splitting of heavier nuclei with the release of energy. The nuclear process exhibits an energy output for fusion for very light elements and conversely energy produced by fission requires heavy elements. Iron defines the middle point, where for elements lighter than iron more energy is required to fission them than the energy liberated, and the fusion of nuclei to form elements heavier than iron is also an energy absorbing process. Such processes are endothermic, while nuclear processes of fusion and fission at the opposite ends of the periodic table for light and heavy nuclei respectively are exothermic. This binding curve of energy is mostly an empirically known aspect of nuclear physics. Yet it is known that light nuclei under tremendous heat and pressure will fuse into heavier nuclei and release energy. This is the process which powers the sun and other stars.
The first application of nuclear fusion was the hydrogen bomb first proposed by Edward Teller. The bomb was developed by Teller and Stanslaw Ulam, where Ulam solved some of the most important problems, in the late 1940's and detonated in 1950. This development was during the ramp up of the cold war between the United States and the Soviet Union. Isotopes of hydrogen are driven together to counter their mutual electrostatic repulsion by heat and pressure to initiate the nuclear fusion reaction. The hydrogen bomb requires a small fission explosive, called a plutonium trigger, to initiate this. This leads to the idea that if a plasma of hydrogen isotopes of deuterium and tritium could be sustained at high enough temperatures and pressures that they might fuse into helium to produce energy. This program ran right away into difficulties of maintaining a plasma in a stable configuration. The standard fusion reaction is between a deuterium nucleon, an isotope of hydrogen consisting of a proton and a neutron D = 1H2, with tritium T = 1H3 according to
This process is used in nuclear explosives, and is the favored one by controlled fusion researchers. The energy released is large, and energy of the a particle, or 2He4 which consists of two protons and two neutrons, is relatively large. These two species exist in the lowest energy level with opposite spin states. An additional nucleon must exist in a high nuclear energy level, which for helium is the unstable 2He5 isotope. The Pauli exclusion principle dictates that only one particle of half integer spin can exist in a state. This is stated here as a "matter of fact" as this is a result of quantum mechanics. The protons and neutrons in the 2He4 may then lay in the bottom of the nuclear energy potential in a minimal quantum level with opposite spins. That 2He4 is at low energy means a large energy is released. However, it has to be pointed out that this suffers from one difficulty. Most of this energy is carried away in a neutron. For a nuclear explosive this is not a problem, for this neutron will scatter about with other nuclei and thermal-ize the environment. In nuclear fusion this neutron can't be manipulated readily by electromagnetic means. So much energy is lost in this process, and neutrons damage materials.
Currently the two main competing ideas are magnetic confinement and inertial confinement. In the first case a plasma is confined by magnetic means and in the second by the implosive effect of converging laser beams or particle beams. Magnetic confinement, called Tokamak reactors, confines a plasma in a toroidal chamber heated by electromagnetic means. A solenoid of arbitrary length will produce a constant magnetic field within it. However one that is wrapped around into a torus has a weaker magnetic field on its outer side compared to the inner side. This field misalignment is corrected for with poloidal magnetic fields. Currently the ITER (International Thermonuclear Experimental Reactor) is the leading experimental Tokamak proposed. Inertial confinement uses large lasers or electron beams directed onto pellets containing deuterium and tritium. Neither of these systems has produced a net energy output sustained in a workable manner.
A fusion powered craft will give the same performance as the fission powered craft without the issues of radioactive safety and launch concerns. This of course assumes that a fusion system may be made small and light enough for spacecraft power. A fusion powered ion or VASIMR drive has an expected Av one to ten times that of the nuclear fission case.
The Daedalus project takes nuclear fusion to a different level [4.6]. The propulsion system is an inertial confinement system. A rapid series of mini nuclear bursts propels the craft forward. In this case the reaction mass is the fusion products. The proposed craft will employ the fusion reaction
as the main process. This is far superior to the D-T process for the daughter products are two charged particles and their is no energy loss by neutrons. The mass of each a particles is ~ 3750 MeV, which is the reaction mass. Hence the ratio of the energy output in MeV to the reaction mass in MeV is .003. This is compared to the same ratio for a chemical rocket ~ 10~10. Hence a "ballpark" estimate of the energy increase over that expected for a chemical rocket is ~ 3 x 106. The Av for this fusion craft is then up to 1250 times that of a comparable chemical rocket, and the specific impulse would be on the order of 5 x 105 sec. For such a craft that initially starts out as mostly its fuel mass it is possible to reach velocities ~ 10% the speed of light. Such a craft could get to some of the nearest stars, such as a Centuri, within a century.
As a final note on nuclear power systems systems that exist today, whether commercial, and military systems or research designs, have a certain cultural history to them. Fermi demonstrated a nuclear chain reaction by configuring an atomic pile. Nuclear energy has rested upon this design basis ever since. The whole philosophy since 1942 has rested upon the idea of a controlled self sustained nuclear chain reaction. Similarly the idea of nuclear fusion has been that high temperatures and pressures are required. However, this need not be the only way to generate nuclear energy. Clearly a beam of neutrons could be directed at a thin target of uranium or plutonium to achieve much the same results. Indeed a fast neutron source could fission 92U238 directly. The fission products are then magnetically separated or trapped and their energy extracted. A beam of protons could accomplish much the same, but the beam would have to be at a higher energy in order to overcome the electric repulsion between it and the nucleus. Similarly for fusion a deuteron beam on a target of 3Li6 might achieve the same result. Further, there are none of the complexities with a self sustained chain reaction in the fission case. To turn off the reactor the beam is turned off. Similarly for the fusion case it is clear that there are none of the confinement issues with a solid piece of lithium.
This completes this chapter on power systems that are either current or within the foreseeable future. It is clear that interplanetary space exploration may be greatly expanded beyond current capabilities. However, only one power source as applied directly to propulsion is marginally capable of interstellar exploration, and the rest are simply not acceptable. It appears that radically new approaches will be required to realistically explore another star system, particularly if it is found to posses a life bearing planet. Extrasolar systems and possible astro-biology will be discussed later. Yet at this point we must explore issues of space navigation and special relativity. In order to talk about travelling to another star at some significant percentage of the speed of light we need to discuss special relativity and understand it well. Just as the above discussion on space propulsion required Newton's laws an understanding of special relativity is needed to illustrate propulsion systems required for interstellar exploration.
Was this article helpful?