Space-based solar power (SBSP) (or historically space solar power- SSP) is a system for the collection of solar power in space, for use on Earth. SBSP differs from the usual method of solar power collection in that the solar panels used to collect the energy would reside on a satellite in orbit, often referred to as a solar power satellite (SPS), rather than on Earth's surface. In space, collection of the Sun's energy is unaffected by the day/night cycle, weather, seasons, or the filtering effect of Earth's atmospheric gases.
The world Radiation Centre's 1985 standard extraterrestrial spectrum for solar irradiance is 1367 W/m2. The integrated total terrestrial solar irradiance is 950 W/m2. Therefore, extraterrestrial solar irradiance is 144% of the maximum terrestrial irradiance. A major interest in SBSP stems from the length of time the solar collection panels can be exposed to a consistently high amount of solar radiation. For most of the year, a satellite-based solar panel can collect power 24 hours per day, whereas a terrestrial station can collect for at most 12 hours per day, and only if weather permits, yielding lower power collection rates around the sunrise and sunset hours.
Collection of solar energy in space for use on Earth introduces new problems of transmitting energy from the collection point, to a receiving antenna on the Earth's surface, ideally near a place of high use, potentially saving transmission line construction costs, and their associated risks of failure, such as the blackouts of 1965 and 2003. Since wires extending from Earth's surface to an orbiting satellite are neither practical or currently possible, many SBSP designs have proposed the use of microwave beams for wireless power transmission. The collecting satellite would convert solar energy into electrical energy, powering a microwave emitter oriented toward a collector on the Earth's surface. Dynamic solar thermal power systems are also being investigated.
Some problems normally associated with terrestrial solar power collection would be entirely avoided by such a design, eg, dependence on weather conditions, contamination or corrosion, damage by wildlife or plant encroachment, etc. Other problems will likely be encountered, such as more rapid radiation damage or micrometeoroid impacts.
The SBSP concept, originally known as Satellite Solar Power System ("SSPS") was first described in November 1968 . In 1973 Peter Glaser was granted U.S. patent number 3,781,647 for his method of transmitting power over long distances (eg, from an SPS to the Earth's surface) using microwaves from a very large (up to one square kilometer) antenna on the satellite to a much larger one on the ground, now known as a rectenna.
Glaser then worked at Arthur D. Little, Inc., as a vice-president. NASA signed a contract with ADL to lead four other companies in a broader study in 1974. They found that, while the concept had several major problems—chiefly the expense of putting the required materials in orbit and the lack of experience on projects of this scale in space, it showed enough promise to merit further investigation and research .
Between 1978 and 1981 the US Congress authorized DOE and NASA to jointly investigate. They organized the Satellite Power System Concept Development and Evaluation Program . The study remains the most extensive performed to date. Several reports were published investigating possible problems with such an engineering project. They include:
The project was not continued with the change in Administrations after the 1980 US Federal elections.
The Office of Technology Assessment concluded
Too little is currently known about the technical, economic, and environmental aspects of SPS to make a sound decision whether to proceed with its development and deployment. In addition, without further research an SPS demonstration or systems-engineering verification program would be a high-risk venture.
More recently, the SBSP concept has again become interesting, due to increased energy demand, increased energy costs, and emission implications, starting in 1997 with the NASA "Fresh Look". In assessing "What has changed" since the DOE study, this study asserts that
Another important change has occurred at the US national policy level. US National Space Policy now calls for NASA to make significant investments in technology (not a particular vehicle) to drive the costs of ETO [Earth to Orbit] transportation down dramatically. This is, of course, an absolute requirement of space solar power.
One might take the NASA "Fresh Look" study as encouraging because the main difficulty identified is driving down Earth to Orbit costs. However, Dr. Pete Worden claimed that space-based solar is about five orders of magnitude more expensive than solar power from the Arizona desert. A major factor in this five orders of magnitude is the cost of transporting materials to orbit. Dr. Worden referred to possible solutions as speculative solutions that would not be available for decades at the best, leaving space-based solar power with no business case for the foreseeable future.
In 1999 NASA's Space Solar Power Exploratory Research and Technology program (SERT) was initiated for the following purpose:
It was to develop a solar power satellite (SPS) concept for a future gigawatt space power systems to provide electrical power by converting the Sun’s energy and beaming it to the Earth's surface. It was also to provide a developmental path to solutions for current space power architectures. Subject to further study, it proposed an inflatable photovoltaic gossamer structure with concentrator lenses or solar heat engines to convert sunlight into electricity. The program looked at both systems in sun-synchronous orbit and geosynchronous orbit.
Some of SERT's conclusions include the following:
The SBSP concept is attractive because space has several major advantages over the Earth's surface for the collection of solar power. There is no air in space, so the collecting surfaces would receive much more intense sunlight, unaffected by weather. In geostationary orbit, an SPS would be illuminated over 99% of the time; such an SPS would be in Earth's shadow on only a few days at the spring and fall equinoxes; and even then for a maximum of 75 minutes late at night when power demands are at their lowest. This characteristic of SBSP avoids the expense of storage facilities (dams, oil storage tanks, coal dumps) necessary in many Earth-based power generation systems. Additionally, SBSP would have fewer or none of the ecological (or political) consequences of fossil fuel systems.
SBSP would also be applicable on a global scale. Nuclear power raises questions of proliferation and waste disposal, which pose problems everywhere, but especially in undeveloped areas which are less capable of coping with them. SBSP poses no such known potential threat.
Space-based solar power essentially consists of three parts:
The space-based portion will be in a freefall, vacuum environment and will not need to support itself against gravity other than relatively weak tidal stresses. It needs no protection from terrestrial wind or weather, but will have to cope with space-based hazards such as micrometeors and solar storms.
Most analyses of solar power satellites have focused on photovoltaic conversion (commonly known as “solar cells”). Photovoltaic conversion uses semiconductor cells (e.g., silicon or gallium arsenide) to directly convert photons into electrical power via a quantum mechanical mechanism. Photovoltaic cells are not perfect in practice, as material purity and processing issues during production affect performance; each has been progressively improved for some decades. Some new, thin-film approaches are less efficient (about 20% vs 41% for best in class in each case as of late 2009), but are much less expensive and generally lighter. A March 2010 demonstration by a group at Caltech may have changed this: they claim 85% efficiency for sunlight and 95% efficiency at particular wavelengths, as well as near perfect quantum efficiency for their design. Production costing is unavailable since only experimental cells have been produced.
In an SPS implementation, photovoltaic cells will likely be rather different from the glass-pane protected solar cell panels familiar to many in current terrestrial use, since they will be optimized for weight, and will be designed to be tolerant of the space radiation environment (some thin film silicon solar panels are highly insensitive to ionising radiation), but will not need to be encapsulated against corrosion from environmental exposure or biological deterioration. They do not require the structural support required for terrestrial use, where the considerable gravity and wind loading imposes structural requirements on terrestrial implementations.
Wireless power transmission was proposed early on as a means to transfer energy from collection to the Earth's surface. The power could be transmitted as either microwave or laser radiation at a variety of frequencies depending on system design. Whichever choice is made, the transmitting radiation would have to be non-ionizing to avoid potential disturbances either ecologically or biologically. This established an upper limit for the frequency used, as energy per photon (and consequently the ability to cause ionization) increases with frequency. Ionization of biological materials doesn't begin until ultraviolet or higher frequencies, so most radio frequencies would be feasible.
William C. Brown demonstrated in 1964, during Walter Cronkite's CBS News program, a microwave-powered model helicopter that received all the power it needed for flight from a microwave beam. Between 1969 and 1975, Bill Brown was technical director of a JPL Raytheon program that beamed 30 kW of power over a distance of 1 mile at 84% efficiency. 
A large-scale demonstration of power beaming is a necessary step to the development of solar power satellites. Laser power beaming was envisioned by some at NASA as a stepping stone to further industrialization of space.
In the 1980s researchers at NASA worked on the potential use of lasers for space-to-space power beaming, focusing primarily on the development of a solar-powered laser. In 1989 it was suggested that power could also be usefully beamed by laser from Earth to space. In 1991 the SELENE project (SpacE Laser ENErgy) was begun, which included the study of laser power beaming for supplying power to a lunar base.
In 1988 the use of an Earth-based laser to power an electric thruster for space propulsion was proposed by Grant Logan, with technical details worked out in 1989. He proposed using diamond solar cells operating at six hundred degrees to convert ultraviolet laser light, a technology that has yet to be demonstrated even in the laboratory. His ideas were adapted to be more practical.
The SELENE program was a two-year research effort, but the cost of taking the concept to operational status was too high, and the official project was ended in 1993, before reaching a space-based demonstration.
The size of a solar power satellite would be dominated by two factors: the size of the collecting apparatus (eg. panels and mirrors), and the size of the transmitting antenna. The distance from Earth to geostationary orbit (22,300 miles, 35,700 km), the chosen wavelength of the microwaves, and certain laws of physics (specifically the Rayleigh Criterion or diffraction limit) will all be factors.
It has been suggested that, for best efficiency, the satellite antenna should be circular and about 1 kilometer in diameter or larger; the ground antenna (rectenna) should be elliptical, 10 km wide, and a length that makes the rectenna appear circular from GEO (Geostationary Orbit). (Typically, 14 km at some North American latitudes.) Smaller antennas would result in increased losses to diffraction/sidelobes. For the desired (23 mW/cm²) microwave intensity  these antennas could transfer between 5 and 10 gigawatts of power.
According to some research, to collect and convert the target volume of power, the satellite would require between 50 and 100 square kilometers of collector area (if readily available ~14% efficient monocrystalline silicon solar cells were deployed). State of the art triple-junction gallium arsenide solar cells with a maximum efficiency of 40.7%  could reduce the necessary collector area by two thirds, and the recent Caltech design would halve that if practical, but neither may turn out to be feasible. In any case, an SPS's structure will necessarily be large (perhaps kilometers across), making it larger than most man-made structures on Earth, and building structures of such size in orbit has never been attempted.
A collection of LEO (Low Earth Orbit) space power stations has been proposed as a precursor to GEO (Geostationary Orbit) space-based solar power. There would be both advantages (shorter energy transmission path, lower cost) and disadvantages (frequent changes in antenna geometries, increased debris collisions, more power stations needed to receive power continuously). It might be possible to deploy LEO systems sooner than GEO because the antenna development would take less time, but it may take longer to prepare and launch the number of required satellites.
The Earth-based receiver antenna (or rectenna) is a critical part of the original SPS concept. It would probably consist of many short dipole antennas, connected via diodes. Microwaves broadcast from the SPS will be received in the dipoles with about 85% efficiency. With a conventional microwave antenna, the reception efficiency is still better, but the cost and complexity is also considerably greater, almost certainly prohibitively so. Rectennas would be multiple kilometers across. Crops and farm animals may be raised underneath a rectenna, as the thin wires used for support and for the dipoles will only slightly reduce sunlight, or non arable land could be used, so such a rectenna would not be as expensive in terms of land use as might be supposed.
One problem for the SBSP concept is the cost of space launches and the amount of material that would need to be launched.
Much of the material launched need not be delivered to its eventual orbit immediately, which raises the possibility that high efficiency (but slower) engines could move SPS material from LEO to GEO at an acceptable cost. Examples include ion thrusters or nuclear propulsion.
Power beaming from geostationary orbit by microwaves carries the difficulty that the required 'optical aperture' sizes are very large. For example, the 1978 NASA SPS study required a 1-km diameter transmitting antenna, and a 10 km diameter receiving rectenna, for a microwave beam at 2.45 GHz. These sizes can be somewhat decreased by using shorter wavelengths, although they have increased atmospheric absorption and even potential beam blockage by rain or water droplets. Because of the thinned array curse, it is not possible to make a narrower beam by combining the beams of several smaller satellites. The large size of the transmitting and receiving antennas means that the minimum practical power level for an SPS will necessarily be high; small SPS systems will be possible, but uneconomic.
To give an idea of the scale of the problem, assuming a solar panel mass of 20 kg per kilowatt (without considering the mass of the supporting structure, antenna, or any significant mass reduction of any focusing mirrors) a 4 GW power station would weigh about 80,000 metric tons, all of which would, in current circumstances, be launched from the Earth. Very lightweight designs could likely achieve 1 kg/kW,, meaning 4,000 metric tons for the solar panels for the same 4 GW capacity station. This would be the equivalent of between 40 and 150 heavy-lift launch vehicle (HLLV) launches to send the material to low earth orbit, where it would likely be converted into subassembly solar arrays, which then could use high-efficiency ion-engine style rockets to (slowly) reach GEO (Geostationary orbit). With an estimated serial launch cost for shuttle-based HLLVs of $500 million to $800 million, and launch costs for alternative HLLVs at $78 million, total launch costs would range between $11 billion (low cost HLLV, low weight panels) and $320 billion ('expensive' HLLV, heavier panels).
Gerard O'Neill, noting the problem of high launch costs in the early 1970s, proposed building the SPS's in orbit with materials from the Moon. Launch costs from the Moon are potentially much lower than from Earth, due to the lower gravity. This 1970s proposal assumed the then-advertised future launch costing of NASA's space shuttle. This approach would require substantial up front capital investment to establish mass drivers on the Moon.
Nevertheless, on 30 April 1979, the Final Report ("Lunar Resources Utilization for Space Construction") by General Dynamics' Convair Division, under NASA contract NAS9-15560, concluded that use of lunar resources would be cheaper than Earth-based materials for a system of as few as thirty Solar Power Satellites of 10GW capacity each.
In 1980, when it became obvious NASA's launch cost estimates for the space shuttle were grossly optimistic, O'Neill et al. published another route to manufacturing using lunar materials with much lower startup costs. This 1980s SPS concept relied less on human presence in space and more on partially self-replicating systems on the lunar surface under remote control of workers stationed on Earth. This proposal suffers from the current lack of such automated systems. The design and construction of these automated systems and their use to produce a mass driver launching system on the moon from lunar materials is expected to take more than twenty years. The partially self replicating systems would include locally produced power generation, perhaps solar cells or heat engine produced electrical power.
Asteroid mining has also been seriously considered. A NASA design study evaluated a 10,000 ton mining vehicle (to be assembled in orbit) that would return a 500,000 ton asteroid fragment to geostationary orbit. Only about 3,000 tons of the mining ship would be traditional aerospace-grade payload. The rest would be reaction mass for the mass-driver engine, which could be arranged to be the spent rocket stages used to launch the payload. Assuming that 100% of the returned asteroid was useful, and that the asteroid miner itself couldn't be reused, that represents nearly a 95% reduction in launch costs. However, the true merits of such a method would depend on a thorough mineral survey of the candidate asteroids; thus far, we have only estimates of their composition.
Having a relatively cheap per pound source of raw materials from space would lessen the concern for low mass designs and result in a different sort of SPS being built. The low cost per pound of lunar materials in O'neill's vision would be supported by using lunar material to manufacture more facilities in orbit than just solar power satellites.
SBSP costs might be reduced if a means of putting the materials into orbit were developed that did not rely on rockets. Some possible technologies include ground launch systems such as mass drivers or Lofstrom loops, which would launch using electrical power, or the geosynchronous orbit space elevator. However, these require technology that is yet to be developed.
Advanced techniques for launching from the moon may reduce the cost of building a solar power satellite from lunar materials. Some proposed techniques include the lunar mass driver and the lunar space elevator, first described by Jerome Pearson. It would require establishing silicon mining and solar cell manufacturing facilities on the Moon.
The use of microwave transmission of power has been the most controversial issue in considering any SPS design.
Power transmission of tens of kilowatts has been well proven by existing tests at Goldstone in California (1975)  and Grand Bassin on Reunion Island (1997). At the Earth's surface, a suggested microwave beam would have a maximum intensity at its center, of 23 mW/cm2 (less than 1/4 the solar irradiation constant), and an intensity of less than 1 mW/cm2 outside of the rectenna fenceline (the receiver's perimeter).
These compare with current United States Occupational Safety and Health Act (OSHA) workplace exposure limits for microwaves, which are 10 mW/cm2, - the limit itself being expressed in voluntary terms and ruled unenforceable for Federal OSHA enforcement purposes. A beam of this intensity is therefore at its center, of a similar magnitude to current safe workplace levels, even for long term or indefinite exposure. Outside the receiver, it is far less than the OSHA long-term levels Over 95% of the beam energy will fall on the rectenna. The remaining microwave energy will be absorbed and dispersed well within standards currently imposed upon microwave emissions around the world. It is important for system efficiency that as much of the microwave radiation as possible be focused on the rectenna. Outside of the rectenna, microwave intensities rapidly decrease, so nearby towns or other human activity should be completely unaffected.
Exposure to the beam is able to be minimized in other ways. On the ground, physical access is controllable (eg, via fencing), and typical aircraft flying through the beam provide passengers with a protective metal shell (ie, a Faraday Cage), which will intercept the microwaves. Other aircraft (balloons, ultralight, etc) can avoid exposure by observing airflight control spaces, as is currently done for military and other controlled airspace.
The microwave beam intensity at ground level in the center of the beam would be designed and physically built into the system; simply, the transmitter would be too far away and too small to be able to increase the intensity to unsafe levels, even in principle.
In addition, a design constraint is that the microwave beam must not be so intense as to injure wildlife, particularly birds. Experiments with deliberate microwave irradiation at reasonable levels have failed to show negative effects even over multiple generations.
A commonly proposed approach to ensuring fail-safe beam targeting is to use a retrodirective phased array antenna/rectenna. A "pilot" microwave beam emitted from the center of the rectenna on the ground establishes a phase front at the transmitting antenna. There, circuits in each of the antenna's subarrays compare the pilot beam's phase front with an internal clock phase to control the phase of the outgoing signal. This forces the transmitted beam to be centered precisely on the rectenna and to have a high degree of phase uniformity; if the pilot beam is lost for any reason (if the transmitting antenna is turned away from the rectenna, for example) the phase control value fails and the microwave power beam is automatically defocused. Such a system would be physically incapable of focusing its power beam anywhere that did not have a pilot beam transmitter.
The long-term effects of beaming power through the ionosphere in the form of microwaves has yet to be studied, but nothing has been suggested which might lead to any significant effect.
When hot rocket exhaust reacts with atmospheric nitrogen, it can form nitrogen compounds. In particular these nitrogen compounds are problematic when they form in the stratosphere, as they can damage the ozone layer. However, the environmental effect of rocket launches is negligible compared to higher volume polluters, such as airplanes and automobiles.