Space-based solar power
DraftSpace-based solar power has been proposed repeatedly since the 1960s and never built. The physics is sound, the engineering is understood, and the economics have never closed. Whether they now do is a live question rather than a settled one.
The energy argument
A solar collector in orbit has three advantages over one on the ground.
| Factor | Ground | Orbit |
|---|---|---|
| Incident flux | ~1000 W/m² at best, at noon, clear sky | 1361 W/m² always |
| Availability | Day only, weather-dependent | Continuous in a suitable orbit |
| Capacity factor | 15 to 25% typical | Above 99% in geostationary orbit, apart from brief equinox eclipses |
| Atmospheric loss | Absorption and scattering | None |
The combined effect is roughly a factor of five to ten in annual energy per unit collector area. That is a genuine advantage and it is the whole basis of the proposition.
The counterargument is equally simple: the collector must be launched, and ground solar has become extremely cheap. The comparison is not between orbital and ground solar per unit area but between orbital solar and ground solar plus storage per unit delivered energy, and ground solar’s cost has fallen faster than launch cost over the same period.
Wireless power transmission
The power must reach the ground without wires, which is the technically distinctive part.
| Method | Frequency or wavelength | Character |
|---|---|---|
| Microwave | 2.45 or 5.8 GHz | Passes through cloud and rain almost unattenuated; requires a very large aperture |
| Millimetre wave | 35 GHz and above | Smaller apertures; significant atmospheric absorption |
| Laser | ~1 µm | Smallest aperture, highest beam density; blocked entirely by cloud |
Microwave at 2.45 GHz is the classical choice, and the reason is availability rather than efficiency. At that frequency the atmosphere is essentially transparent regardless of weather, so the system delivers power continuously, which is the entire premise. A laser system that stops working under cloud has forfeited the advantage it launched for.
The cost of low frequency is aperture. Diffraction sets the beam spread, and for a transmitter at geostationary distance the numbers are severe:
At 2.45 GHz, a wavelength of 12 cm, a transmitting antenna one kilometre across produces a ground spot several kilometres in diameter. Classic study designs therefore assume a transmitter of about 1 km and a rectenna of 5 to 10 km across. Both are enormous, and the space segment is a kilometre-scale structure that must be assembled in orbit. See In-space manufacturing and assembly.
Beam safety
The safety question is asked constantly and the answer is more reassuring than intuition suggests, though not unqualified.
Beam intensity at the rectenna centre in standard designs is around 230 W/m², roughly a quarter of peak sunlight, falling toward the edges. This is not a weapon and cannot be made into one by adjusting the design, because the same diffraction that forces a large rectenna prevents concentrating the beam into a small spot from that distance.
- Intensity
- Comparable to or below sunlight at the centre of the receiving area, which is why rectennas are usually drawn as transparent structures with agriculture underneath.
- Retrodirective control
- The beam is phase-locked to a pilot signal from the rectenna. If that signal is lost, the phase reference is lost and the beam defocuses rather than wandering.
- Birds and aircraft
- Transient exposure at these intensities is not considered hazardous, but the analysis depends on the design and is a licensing question.
- Spectrum
- A gigawatt-class transmitter is an enormous radio source. Interference coordination with existing users of the band is a genuine regulatory obstacle.
The retrodirective property is the important safety feature and it is structural rather than procedural: the beam cannot be steered off target, because it only forms coherently when locked to the pilot beam from the intended receiver.
The economics
The analysis reduces to a mass and launch cost calculation.
A gigawatt-class system requires kilometre-scale collectors, transmitters and structure. Study estimates for total mass vary widely with assumptions about photovoltaic specific power and structural efficiency, but they are consistently in the thousands to tens of thousands of tonnes.
At historical launch prices of thousands of dollars per kilogram, the launch cost alone exceeds what the delivered electricity could ever be worth. That is why every study from the 1970s onward concluded that the concept was technically feasible and economically hopeless.
Two things have changed and one has not.
Launch cost has fallen substantially, and would fall further with a fully reusable vehicle. This is the change that revived interest. See Launch and mission economics.
Photovoltaic and power electronics performance has improved, reducing mass per delivered watt.
Terrestrial renewables have become much cheaper at the same time, and this is the factor most often omitted from advocacy. The competitor is not fossil generation but ground solar plus batteries, whose cost has fallen at least as fast as launch cost. Space-based solar power must beat a moving target that is moving in the same direction.
Current activity
Interest has revived and the activity is demonstration-scale rather than deployment-scale.
Caltech’s Space Solar Power Demonstrator flew in 2023 and demonstrated wireless power transmission in orbit and detection of the transmitted power on Earth, at milliwatt levels. That is a genuine first and is many orders of magnitude below useful.
The European Space Agency’s SOLARIS programme is a preparatory study assessing feasibility and cost. Japan, China and the United Kingdom have active study programmes, and several private ventures have announced plans.
The honest summary is that the physics has never been in doubt, the engineering is understood at study level, no scaled demonstration exists, and the economic case depends on launch cost falling faster than terrestrial storage cost. That is a defensible bet and not an established result.
Where it might make sense first
The general grid case is the hardest one to win. Narrower applications are more plausible because they compare against a worse alternative rather than against cheap ground solar.
| Application | Why it is easier |
|---|---|
| Power to spacecraft | No atmosphere, small receiver, and the alternative is carrying your own array |
| Lunar polar operations | Beaming power into permanently shadowed craters where solar is impossible |
| Remote or disaster-response sites | Compared against diesel generation and logistics, not against grid solar |
| Military forward operating locations | Value is in avoiding a fuel convoy, not in cost per kilowatt hour |
Beaming power to the lunar surface is the most technically natural of these, because there is no atmosphere, distances are short, and the 14-day night problem described in Lunar surface systems has no cheap alternative.
References
- Glaser, P. E. “Power from the Sun: Its Future”, Science, 162:857, 1968. The original proposal.
- Mankins, J. C. The Case for Space Solar Power, Virginia Edition, 2014.
- NASA, Satellite Power System Concept Development and Evaluation Program, DOE/ER-0023, 1980. The definitive early study.
- Rodenbeck, C. T. et al. “Microwave and Millimeter Wave Power Beaming”, IEEE Journal of Microwaves, 1:229, 2021.
- Hajimiri, A. et al. “MAPLE: Microwave Array for Power-transfer Low-orbit Experiment”, Caltech Space Solar Power Project, 2023.
- ESA, SOLARIS preparatory programme documentation, 2022 onward.