ON THE GROUND, we take toasters and hair dryers for granted, even though they need a thousand watts or more of electricity. There are no such luxuries on spacecraft because power must be generated on board, and there are limits to the size of the power system and the amount of fuel the craft can carry. The heavier the load, the more it costs to launch it. Today’s satellites are frugal with power, which is usually supplied by arrays of solar cells. A typical communications satellite needs between 1 and 1.5 kilowatts of electricity, and the space shuttle manages on 12.5 kilowatts.
These demands are modest, thanks largely to advances in microelectronics, which require very little power. But the space missions of the future will require much more power, and new technologies to supply it.
Lack of power could be a fatal flaw in the US’s plans for the Strategic Defense Initiative. In January, a panel at the National Academy of Sciences called space power systems ‘a pacing item for the successful development of the SDI’. The panel said: ‘Either major innovations in power systems and power system components will be required or SDI power requirements will have to be relaxed.’
Advertisement
Hardware for the SDI is the most power-hungry equipment on the drawing boards. Planners have divided the power needed into three broad levels. The lowest level is a steady output of up to several hundred kilowatts during the 10-year lifetime of a satellite. That power would drive sensors, laser and microwave radars, communications systems and monitoring equipment.
Higher levels of power – in the megwatt range – would be needed to put weapons on alert. In that state, weapons normally on standby would be primed to fire on command in a matter of seconds. The exact requirements for this are poorly defined, but the academy’s report says that alerts could last for up to a year. Firing the weapons would consume the highest power levels, perhaps 50 to 200 megawatts.
NASA, too, will need more power for its future missions. Earl VanLandingham, deputy director of its propulsion power and energy division, says that the first phase of the space station planned by the US will require 75 kilowatts, supplied by arrays of photovoltaic cells. Last month, the space station office of NASA, Washington DC, suggested a solar dynamic system, where concentrated solar energy is converted thermodynamically into electricity. The second phase of the space station (which NASA is still waiting for Congress to approve) would need 300 kilowatts, supplied by solar dynamics. Eventually, says VanLandingham, the space agency will have to turn to nuclear reactors if it is to supply the 1.5 megawatts that a base on the Moon would need.
Like NASA, military planners want more power, but they also want systems that will withstand the Soviet Union’s antisatellite weapons. Richard Verga of the SDI Organisation (SDIO) says that solar cells could be made less vulnerable to attack by placing them inside a funnel-shaped structure, so that light could reach them only from the direction the funnel was pointing. The main problem is that solar cells must be exposed to the environment somehow: if sunlight cannot reach them, they cannot do their job. A nuclear reactor, on the other hand, can be sealed inside a container.
The military’s need to make sure that its power systems can survive an attack is the main reason why NASA and the SDIO have quite different approaches to research in this area. NASA’s main concern is efficiency, measured as the power output per unit weight rather than as the fraction of sunlight converted to electricity. The Pentagon’s prime concern is ‘survivability’, even at the expense of weight and efficiency.
NASA spends about $5 million a year on photovoltaic technology. So far, no solar cell can convert sunlight to electricity with an efficiency greater than 31 per cent. The Sandia National Laboratories in Albuquerque, New Mexico, reached this figure last year with a cell made from a layer of gallium arsenide and a layer of silicon. Separately, each material has a peak efficiency of 28 per cent. The advantage of the two-layer structure is that even though gallium arsenide generates more energy per photon absorbed than silicon, it does not absorb as broad a band of wavelengths. So the silicon salvages energy from light that passes through gallium arsenide. These record efficiencies are possible only if sunlight is concentrated between 350 and 500 times its normal intensity, something that present satellites cannot do.
The US Air Force is also doing research on technology designed to exploit the Sun’s rays, and it is particularly interested in survivability. It has developed protected, or ‘hardened’, systems but at a cost in weight: unhardened solar cells promise to deliver about 35 watts per kilogram, against only 10 watts from the first generation of hardened cells.
Despite their interest in solar technology, both the SDIO and NASA are spending far more money on developing nuclear reactors for space. Up to now, American spacecraft have relied heavily on a marriage between nuclear technology and thermoelectric generators. RTGs, or radioisotope thermoelectric generators as the resulting systems are known, are powered by heat from the natural decay of between 1 and 10 kilograms of plutonium-238, a short-lived artificial isotope. The US launched its first RTG in 1961, and so far the systems have powered 21 of its spacecraft, including the Apollo missions to the Moon. In 1964, one radioisotope generator burned up in the atmosphere on re-entry, as planned. After that, NASA redesigned the modules to survive re-entry, and two have, including one on the ill-fated Apollo 13 mission, when an oxygen tank exploded and threatened the lives of astronauts en route to a third Moon landing. Now RTGs are mainly fitted in probes to the outer planets, where sunlight it too weak for solar power. The Galileo probe to Jupiter, scheduled for launch in October, and Ulysses, which will take measurements over the Sun’s poles, will be powered by RTGs.
Nuclear reactors offer much more power than RTGs, which are based on natural radioactive decay rather than nuclear fission. Both the US and the Soviet Union are working on new versions. Typically for the American space programmes, the SP-100, rated at 100 kilowatts, is encountering delays. The Soviet Union has already tested its second-generation reactor, called Topaz, twice in space. One operated for six months and the other for a year.
Since 1968, the Soviet Union has launched about 30 satellites powered by nuclear reactors, mostly reconnaissance satellites to track shipping. The reactors contain about 30 kilograms of uranium, and American experts believe that they generate between 5 and 10 kilowatts of electricity by nuclear fission of uranium-235. The first generation of Soviet reactors operated for about three months before being boosted towards a stable higher orbit. Two never reached it: in 1973, Cosmos 954 scattered radioactive debris over northwest Canada as it came back down, while Cosmos 1402 completely dispersed during re-entry. Heat from these reactors was converted into electricity thermoelectrically; in other words, a current was produced in two metals maintained at different temperatures. The prototype Topaz reactors also produce 10 kilowatts but Soviet officials say that the design can be modified to generate several hundred kilowatts.
The Topaz reactors generate electricity thermionically; that is, a heated material releases electrons, generating a current (‘Making heat work harder’, ¿ìè¶ÌÊÓƵ, 16 June 1988). Thermionic generators can be placed inside the core of the reactor, while thermoelectric generators are kept outside the nuclear container where temperatures are cooler and energy conversion is less efficient. According to reports in the Western military press, the Soviet Union has offered to sell Topaz reactors to the US.
The sole American nuclear reactor in space was launched in April 1965 on one of the Air Force’s satellites. The reactor produced 500 watts for 43 days until, as the official explanation ran, ‘spacecraft instrumentation not directly connected with the reactor operation malfunctioned, causing a sudden reactor shutdown’. The satellite remains in an orbit, at an altitude of 1287 kilometres, from where it should not return to Earth for several thousand years.
A question of weight
The US started the SP-100 programme in 1983. The original goal was to generate 300 kilowatts, but budget cuts in 1987 led planners to go for a more modest 100 kilowatts. The US Department of Energy, the SDIO and NASA joined forces to fund the SP-100, spending $75 million, in 1988. The energy department, which supplies about two-thirds of the money, plans to test on the ground a reactor from General Electric of the US in 1992. The aim is to have a proven reactor ready by the mid-1990s .
The design differs greatly from that of Earth-based reactors, but sticks to largely proven technologies. The power unit, including a reactor fuelled with highly enriched uranium, will be separated from the rest of the satellite by a long boom containing a power cable, keeping everything as far as possible from the reactor’s heat and radiation. Only the part of the reactor facing the rest of the satellite will be shielded.
Some problems are emerging with the design. Last summer, an independent review panel, convened by the government under the leadership of Glenn E. Cunningham of the Jet Propulsion Laboratory, Pasadena, expressed concern over the reactor’s weight: the prototype’s of 5422 kilograms was ‘substantially above’ the design goal of 3000 kilograms. The designers have taken the conservative-sounding approach of staying with thermoelectric conversion, a proven but inefficient technology. Congress’s General Accounting Office has expressed concern that there is a high risk that the programme will fail because of a lack of data on the type of thermoelectric converters in the SP-100 design. That process is only about 4 per cent efficient, so the reactors must produce a thermal energy of 2.5 megawatts in order to generate 100 kilowatts of electricity. The remaining heat has to be removed. It will be dissipated by a liquid-lithium cooling system that includes fixed and movable panels to radiate energy from the satellite.
NASA thinks that the SP-100 is powerful enough for its uses, but the Pentagon and the DOE are working on nuclear and other technologies to generate from a few megawatts to hundreds of megawatts to provide the power for star wars. In 1985, the two departments started the Multimegawatt Reactor programme, aiming to identify by 1992 at least one nuclear system that would meet the SDIO’s requirements. Six contractors are preparing paper studies of multimegawatt reactors, and the departments expect to select two or three concepts for further development this summer.
The main details in this programme remain to be defined, partly because the SDIO’s requirements remain vague. Planners seem to want a reactor that can generate low, steady outputs and still produce the 100 megawatts or more for 100 to 2000 seconds that it would need to fire a weapon, such as a barrage of small rockets or pulses from a high-energy laser. To meet these requirements, the reactor would probably have to operate at between 1100 and 1900 Degree C, far hotter than liquid-metal cooled reactors on the ground, which run at between 300 and 500 Degree C.
For bursts of power lasting 1000 seconds or less, chemistry may be more attractive. ‘You can do everything the SDI thinks it wants to do in the foreseeable future chemically,’ says Verga. One possibility is burning hydrogen and oxygen to drive a turbine, which converts the energy into electricity through an alternator or direct-current generator. According to Verga, the SDIO is spending about four times as much on chemical technology as on nuclear systems to find the multimegawatts it wants.
But chemical power has its problems, too. Engineers can choose between open cycles, which exhaust waste gas into space, or closed cycles, which seal wastes in the craft. Open cycles are attractive because they can carry away waste heat, a serious problem in space, and with no need for storage tanks they also reduce the mass of the system. The drawback is that effluents might damage the spacecraft. The National Academy of Sciences called this ‘the principal unknown in using chemical reactants to produce space power’.
Powerful weapons
Star wars itself may depend on success in developing open-cycle chemical power. The same panel says these open-cycle systems are the most efficient means of providing bursts of power in the megawatt range in terms of kilowatts per kilogram. If the systems cannot be developed, or if their effluents ruin the spacecraft, weapons or sensors, the panel warns, ‘the entire SDI concept will be severely penalised from the standpoints of cost and launch weights’.
Reactors are not the only way of using nuclear energy to generate high levels of power in space. Of the alternatives, one of the most controversial is the bomb-driven X-ray laser that the Lawrence Livermore National Laboratory in California is developing. The idea is that a nuclear explosion would excite a rod of material to emit a laser beam of X-rays. Normally, the energy from the nuclear blast would dissipate in all directions, but the X-ray laser rods would concentrate some of the energy into the comparatively narrow beams. If those beams could be aimed, advocates believe that they might destroy many military targets, including satellites, ballistic missiles and nuclear warheads. However, the 1967 Outer Space Treaty, endorsed by the US and the Soviet Union, bans the deployment of nuclear weapons in space, and the Partial Test Ban Treaty of 1963 prohibits testing the weapons in space. Unofficial reports also suggest there are serious technical problems with the bomb-driven, X-ray programme.
A less controversial approach is for thermal energy from a reactor to power a high-energy laser. This concept was originally demonstrated in 1974 at Sandia National Laboratories and at the Los Alamos National Laboratory in New Mexico. Researchers at the Department of Defense have now revived the concept. So far, the power levels are low, and anyway controversy surrounds plans to put nuclear reactors of any kind in orbit. Steven Aftergood, spokesman for the Committee to Bridge the Gap in Los Angeles, a group opposed to nuclear power in space, cites two main concerns – environmental safety and the potential uses of reactors.
As Cosmos 954 showed, a failure in a reactor can lead to radioactive debris being scattered over a wide area. The DOE acknowledges this, and its official description of the SP-100 proclaims: ‘Safety is of utmost importance in the US programme to develop nuclear power technology for use in space.’ Safety precautions include not operating a reactor until it reaches high-Earth orbit, where it should stay for hundreds of years. As a result, the dangerous products of nuclear fission would not be produced until the reactor is in orbit, and a failure at launch could release only uraniuum-235, which has a long half-life and is not highly radioactive. A backup coolant system would keep the SP-100 reactor intact even after losing its primary coolant of liquid metal. The control system is designed so that external damage or a component failure ‘would simply result in a loss of power through an orderly shutdown of the reactor’, according to an official DOE statement.
American planners are considering several alternatives once a reactor has finished its mission, including leaving it in its original orbit, raising it up even higher or sending it out of orbit altogether. A reactor unit that re-enters the atmosphere is designed to remain intact and ‘bury itself on impact in water, soil or pavement,’ according to a DOE report.
Aftergood concedes that uranium fuel is relatively safe to launch, but adds that plutonium-238 in an RTG is highly radioactive and extremely toxic, and ‘you don’t want that released into the atmosphere’. He also points out that the highly enriched uranium that will fuel the SP-100 could be used in the nuclear weapon, so safeguards are important to make sure it does not fall into the wrong hands. But Aftergood is more concerned about the fate of a burnt-out SP-100 reactor. The reactor will generate 25 times as much thermal power as the first-generation Soviet reactors, and will operate for seven years rather than three months, so it will contain far more products that are highly radioactive. Although Aftergood believes the Americans are concentrating on safety more than the Soviet Union is, he observes, ‘it’s a question of which will predominate, the increased risk from (higher) radioactivity or the greater effectiveness of safety measures’. He also notes that dead reactors will add to the growing problem of space debris, and that a collision involving one could scatter highly radioactive debris into orbits occupied by missions carrying people as well as cargo.
The Federation of American ¿ìè¶ÌÊÓƵs and the Committee of Soviet ¿ìè¶ÌÊÓƵs for Peace and Against the Nuclear Threat have proposed a joint ban on nuclear reactors orbiting the Earth. The two groups say it would be relatively straightforward to police the ban, because a reactor would emit detectable infrared, gamma and neutron radiation. Advocates of the ban say that it would affect only the most provocative military space systems. These would include, among others, satellite battle stations, the Soviet Union’s reconnaissance satellites and elements of the SDI.
There is less opposition to nuclear power for spacecraft that go beyond Earth orbit. ‘We recognise that deep space and future lunar and Mars missions . . . can be done only with space nuclear power,’ says Aftergood. In addition, deep-space systems are ‘relatively benign’, that is they have no immediate military purpose, and they will not return to pollute Earth or nearby space.
Some observers question whether anyone needs such powerful reactors in space. A report published in 1987 by the General Accounting Office notes that neither NASA nor the Department of Defense has yet committed itself to using the SP-100 ‘for any specific space applications’, although both say they want the technology to be available. On the other hand, the National Academy of Sciences, in its report this January, said that a space reactor ‘would be a step toward meeting SDI requirements and would be applicable to other civil and military space missions’.
With power-hungry missions such as a lunar base far in the future, NASA seems content with a leisurely pace of development. The people planning star wars convey more sense of urgency, but their requirements are not well defined, and they are also looking at alternatives. Many of the weapons they are thinking of mounting on satellites would have other sources of burst, or short-term, power, including high-energy lasers powered by chemical fuels and small high-speed rockets that would destroy targets on impact. Free-electron lasers, which require large amounts of power, would be based on the ground and direct their beams to orbiting ‘battle mirrors’. Some cynics suggest that the impetus for developing nuclear power for space comes more from the DOE and the nuclear industry than from the agencies that are supposed to want it.
* * *
The orbiting reactors at arm’s length on satellites
THE SP-100 power generation module has a nuclear reactor at one end and a boom, linking it to the payload, at the other. The boom supports power cables and isolates the rest of the satellite from the reactor’s radiation and heat. This avoids the need for massive shielding of the entire reactor as would be needed on the ground.
Liquid lithium carries heat from the reactor core to the thermoelectric system that generates electricity, and to the secondary cooling system that radiates waste heat – more than two million watts at full power – into space. The cooling system has both fixed and movable radiator panels.
The fuel in the reactor is uranium nitride, enriched to contain between 76 and 97 per cent fissionable uranium-235. The percentage of uranium-235 in nature is only 0.7 per cent. The reactor is designed to work at full power for seven years, and not to sustain a chain reaction in an accident even if water replaced all the lithium coolant.
Because the reactor is on a spacecraft it is small and light compared with a reactor on Earth. To generate the high power levels, it therefore needs to operate at higher temperatures than a larger reactor to achieve a comparable power output. The lithium will be heated to more than 1000 Degree C, far above the temperatures of liquid-metal cooled reactors on the ground, which normally operate at less than 500 Degree C, and well above the 700 Degree C of liquid-metal cooled space reactors developed in the 1960s.
According to the US Department of Energy, an important safety feature of the SP-100 system is that, when launched, it contains no radioactive fission products. These are produced only after the reactor has reached its planned operational orbit where it would remain indefinitely.
