Power and Purpose in Space
Without nuclear-powered spacecraft, we'll never get anywhere
by Robert Zubrin, Mars Society
Wood, wind, water, coal, oil, gas, and nuclear energy: of those major power sources, only one—nuclear—can work in space. Like it or not, humanity is going nowhere, astronautically speaking, without the power of the atom.
Because of technical and political factors, nuclear technology has been confined to applications that produce rather little power, mostly in deep-space probes. Lacking nuclear energy as a robust and diverse option, for both propelling spacecraft and powering their on-board systems and instruments, we have been forced to make extensive use of solar power for systems and instruments and to rely exclusively on chemical rockets for propulsion.
Neither one is adequate. To power spacecraft instruments, the sun's thin rain of energy is of little use much beyond Mars's orbit. For propulsion, chemical rockets run up against the basic burden of space travel: weight. The cost of a space mission scales more or less proportionally to the mass of the spacecraft involved. Up to 90 percent of the weight of a typical chemical rocket at launch is propellant. So to keep this load to a minimum, propulsion experts choose chemical propellants whose reactions are very energetic. The more energy per unit mass of propellant, the less propellant needs to be carried, and the cheaper the mission will be.
Rocket engines fueled with hydrogen and oxygen have already come quite close to the practical performance limit of chemical propulsion. Spend all the money you want; you won't do much better. Nuclear reactors, on the other hand, pack a million times as much energy per unit mass as the best chemical fuels. The challenge, of course, is getting the power out usefully and safely.
Kettles in the sky
The simplest way, which was demonstrated extensively in ground tests in the United States in the 1960s, is to use the reactor to heat a fluid and then eject that fluid out a rocket nozzle. This "flying steam kettle" technique, with hydrogen as the working fluid, yields about four times as much energy per unit mass as a chemical rocket based on hydrogen and oxygen fuel. Much higher energies are possible in theory, but in practice, the exhaust gas would become too hot to handle with any known materials.
To compare the propulsive efficiency of different rocket technologies, engineers use a characteristic called specific impulse. It indicates the amount of time, in seconds, that the technology can put out a pound (4.45 newtons) of thrust while expending a pound (0.454 kg) of propellant. According to the laws of physics, if you quadruple the energy density of a rocket's propellant by, for example, using the nuclear steam kettle rather than a chemical engine, you get a doubling of the rocket's specific impulse.
The actual figures are 450 seconds for the chemical engine and 900 seconds for the steam kettle (more formally known as a nuclear thermal rocket). If you run the numbers, you find you need half as much propellant, at most, with the nuclear rocket. (You'd probably get away with less than half because the rocket pushes not only the payload but also the propellant—and the nuclear rocket has much less propellant.)
Doubling specific impulse is not bad, but it's just the beginning. If we convert the nuclear power into electricity and then use it to accelerate an ionized propellant through an electrostatic grid, we can boost specific impulse by a factor of 10, to around 5000 seconds. Such a nuclear-electric propulsion system would be perfect for some applications, such as propelling uninhabited probes to the outer planets.
Power is money
Alas, it would be tricky to apply this kind of scheme to a large spacecraft, for example, one with a number of human occupants. For a nuclear-electric rocket with a specific impulse of 5000 seconds to eject mass (propellant) at a rate that is high enough to push such a big ship, you'd need a huge nuclear reactor capable of generating thousands of megawatts. It might be possible to use a smaller reactor of about 100 kW, but the power would have to be dribbled out, resulting in electric propulsion burntimes lasting years. That kind of duration would probably be unacceptable for human missions.
One way around this difficulty would be to ship the reactor to a place like the moon or Mars; use the reactor's energy to convert local materials into chemical propellants; and then use those propellants in ordinary chemical engines, with their high thrust-to-weight ratios, to return to Earth. While the specific impulse of these engines would be no higher than for any other chemical rockets, the effective specific impulse of such a system, from the point of view of the overall mission, is multiplied many times. That's because only a small fraction of the propellants used would need to be launched from Earth.
To understand the economics of this concept, start with the conclusion of a decade-old report on the subject by the U.S. Department of Energy. The department estimated that it would cost about US $500 million to produce a working first-generation space nuclear power reactor, with each additional reactor of the same design costing about $100 million.
That's not cheap, but consider the possibilities and alternatives. On Mars, such a reactor could power an unattended chemical plant that would combine atmospheric carbon dioxide with water to produce the rocket propellants methane and oxygen. And with abundant propellants available on the Martian surface, a mission to Mars becomes much simpler to envision.
Moreover, the mission could be accomplished with today's technology. It would begin with two direct trips from Earth to Mars, each lofted by a heavy-lift vehicle in the Saturn-V class. The first would deliver an empty return spacecraft, a reactor, and the automated chemical unit that would make the propellant for the return trip. The second would deliver a habitation module containing the crew.
Cost analyses at NASA's Johnson Space Center have shown that such a live-off-the-land approach to human Mars exploration should reduce program costs nearly tenfold compared to traditional Battlestar Galactica plans based on transporting all mission consumables, including the vast load of propellant needed for a roundtrip.
Bottom line: the ability to manufacture propellants in space would pay enormous dividends. But you would have to have the power to do it.
Power is knowledge
Nuclear power is also necessary for the robotic exploration of the outer solar system. Solar energy diminishes proportionally to the square of the distance from the sun. So at Jupiter it is only 3.7 percent as strong as it is on Earth. And out there, unfortunately, is precisely where you need power the most, both to heat the spacecraft in the frigid void and to transmit data over the longer distances back to Earth.
We have long used nuclear generators on our outer-solar-system probes, including Pioneer, Voyager, Galileo, and Cassini. But these units are puny: NASA's standard radioisotope thermoelectric generator (RTG) module puts out a mere 300 W. NASA also makes extensive use on its probes of 1-W radioisotope heating units, which keep spacecraft instruments warm enough to operate.
We are going to need a lot more power—tens or hundreds of kilowatts—if we are to explore our outer solar system in an efficient and systematic way. Why? While on Earth, it has been said, knowledge is power, in the outer solar system, power is knowledge.
Data transmission rates are proportional to transmitter power, all other things being equal. A probe equipped with a 30-kW nuclear reactor would return 100 times as much data from the outer solar system as a conventional mission equipped with a standard 300-W RTG. And returning data is what a science mission is all about. Equipped with such a power supply, an outer solar system probe could use multikilowatt transmitters, akin to those now employed by the U.S. military, instead of the 40-W traveling-wave-tube systems that are the current standard on NASA interplanetary missions.
By the way, with 30 kW of power, a spacecraft could also be outfitted with nuclear electric propulsion, which would probably double its payload. Wonderful as that would be, the 100-fold increase in data returned would be much more important. Such a huge increase would in itself easily justify the added expense of a nuclear system.
The higher data rates produced by space nuclear power would let us make very high-resolution images in a variety of spectral bands (visible, infrared, and so on). This is a tremendously powerful scientific tool that we have never been able to exploit in the outer solar system. Increase the data rate by a factor of 100 and you can increase the spatial resolution by a factor of 10. Instead of seeing things the size of cars, you'll be able to see things the size of cats.
The number of pictures you could return also grows in direct proportion to data rate. Instead of returning stills, you could return movies, real motion pictures of atmospheric phenomena globally and on a small scale—a meteorologist's dream. And having so many pictures greatly enhances the probability of capturing transient phenomena like lighting, avalanches, or volcanic activity. There's no telling what we'd find, because with the tools we've had available to date, we've hardly been able to look.
The table shows the data transmission rates possible for a probe in orbit around one of the other planets in our solar system. In one case, I assume a 300-W RTG and a fully functional X-band dish 5 meters in diameter; in the other, a 30-kW nuclear spacecraft transmitting with two smaller X-band dishes 3 meters in diameter. The receiver in either case is one of the big 70-meter Deep Space Network antennas.
With standard data compression, 1 kb/s translates into about three good photographs transmitted per hour. Even if the nuclear mission were twice the cost of the conventional one, it would still be a tremendous bargain if you figure in the overwhelmingly greater returns.
Getting beneath the surface
Most alluringly, higher power will let us do things we simply can't do otherwise, as I suggested earlier. Take active sensing, which involves probing a planet with electromagnetic waves and combines radar and radio occultation science. To use it to maximum effect will demand much more power.
Mapping rates show why. The linear resolution of radar imaging improves linearly with the power available, as does the rate at which planetary surface can be mapped at a given spatial resolution. For missions involving flybys, these rates make all the difference. For example, the 60 W-Cassini radar mapper, which will for several years fly repeatedly past Titan, will manage to map less than 10 percent of that Saturnian moon because of low power and slow imaging rates.
Another benefit of high power is that mapping can be done from farther away. This counts for a lot near a planet like Jupiter, where belts of intense radiation may rule out close fly-bys. Higher power also means that ground-penetrating radars can go much deeper. That kind of probing could prove revelatory for Jupiter's Galilean satellites, one of which appears completely covered with ice and liquid water and has been the subject of intense speculation about whether it has the biochemistry to support life.
It's hard to overstate the value of such a capability. Much of what's interesting about a planet is underground. Here on Earth, if we could stare at only the surface of the sea, we would know almost nothing about what is going on beneath the waves, in the vast, teeming worlds of the ocean's reefs, ridges, and trenches. Similarly, observing a planet passively from orbit gives us a very limited view. There could be underground rivers and oceans on Mars or Io, for all we know. Without the kind of power required for deep active sensing, we're almost blind.
Radio occultation science, which is the study of celestial bodies by grazing their atmospheres with radio transmissions, is similarly enhanced by high transmitter power. For example, we would need at least 10 kW to mount a meaningful investigation of the dynamic structure of Saturn's rings. Also, tunable lasers could sound the atmospheres of the major planets for optically active chemical compounds. There are lots of other possibilities in active sensing—too many to describe here.
Have power, will travel
By restarting the United States' languishing space nuclear power program, NASA and the Bush administration are making a critical contribution to science and the human future. Near-term robotic exploration of the outer solar system needs nuclear power units offering tens to hundreds of kilowatts, which is about the same size as the units we will someday use to support human activities on the moon and Mars. The development of space nuclear power sources in this class is fully within our capabilities, and the rewards for doing so will be immense, even unimaginable.
Copied with permission from Robert Zubrin.
Project Orion
Nuclear Space Organisation
Come and join us! We're going to the stars.
Nuclear Space Central
Nuclear Space Forums
