Current solar panels (plus structure) come in about 10kg per installed kilowatt. Cost of putting something in geosynchronous orbit is currently well in excess of $10,000/kg. So that would be $100M per megawatt before worrying about assembly, pointing, transmission/reception, radiation hardening or even the cost of the panels. It's possible that weight could be reduced by a factor of 100 with thin films and magic, but a similar investment in batteries and other storage would seem likely to pay off sooner.

To reduce launch costs, you could put multiple arrays in lower orbits for what would likely be a smaller total system cost and greater global benefit, but astronomers would likely not be best pleased.

I can see this going down really well with the NO 5G NO Microwave ovens brigade.

Worst idea ever! Extrapolate this idea until all countries have these units up and running. The earth can barely cope with the solar energy falling on it directly. We don't need to redirect more heat that would have otherwise missed. We would end up with low temp/high entropy heat that we would be unable to reject, like a power station with no cooling tower.
Why don't we get a giant space magnifying glass and kill ourselves like ants?

@paul314 - https://theconversation.com/how-spacex-lowered-costs-and-reduced-barriers-to-space-112586 ... says it's nearer US$3k/kg (and that was a year ago). You wouldn't need a lot of the hardware associated with conventional ground-based modules (no gravity) and besides, there's no reason why - in time - the modules could not be manufactured in space, too.

My first thought was how to avoid conflicts with the microwave beam which would have to be pretty seriously powerful but I don't know enough about the tech. What would happen if an aircraft accidentally flew through it?

@Matt - I can't help feeling that the current use of fossil fuels is having a *massively* more serious negative effect on our atmosphere - along with all the pollution, to boot.

A microwave beam with commercial power is also a weapon. I'm guessing the next Falklands war will be very different. If the Argentinians are comfortable with this idea, I won't complain.

@martinwinlow -- that $3k number, from all I could find, is for low earth orbit. Geosynchronous orbit costs a few multiples of that as long as we don't have space elevators. (And from a mass point of view manufacturing in space costs even more, because you have to bring up the manufacturing equipment as well as the material itself. You might have a win if you were using thin-film techniques, because you wouldn't have to make a finished device that could withstand the stress of liftoff.)

Nicola Tesla conceived delivering electricity wirelessly way before Asimov, just so you know. As for setting up a system of solar panels in space to deliver the amounts needed terrestrially, this sci-fi concept is out of this world. Instead of looking skyward for energy, one needs to look inward, where Tesla believed unlimited energy exists, and lightning is proof of that. The answer is right in front of us, not some pie-in-the-sky fantasy.

paul314: Completely agree. Even SpaceX's promise to reduce launch costs by 90% with reusable boosters will still mean eye-wateringly expensive to put such a system into space. It would be FAR less expensive to construct solar farms in the Sahara with long distance power transmission to Europe.

With millions invested and billions made for investors; this is a huge "industry" that maintains its power and profit through manipulating Congress and defeating legislation through millions in advertising against the people who want solar to be "free" from their control.

The comments below do a pretty good job of demonstrating why in a sane Universe somebody reviewing a grant proposal or corporate startup would simply be laughed out of the office or rejected out of hand. There is absolutely zero overlap in the cost-benefit spectrum between the factor of (maybe, possibly) 4 or 5 in mean insolation in orbit vs at points on the surface of the earth and the cost of putting things into orbit. One could build and install 100s of times as much surface collection capacity as one could ever afford to put into orbit, for at most a factor of hell, let's be generous to the point of science fiction, 10 gain relative to those ground collectors, WITHOUT worrying about the technical challenges of large area cells necessarily exposed to the solar wind, atmospheric drag, CMEs, and space junk in orbit. Then there is the simple fact that a GW-scale generator in orbit and transmitting to a kilometer square ground antenna (at some less than 100% efficiency) is receiving a kilowatt per square meter, making it a weapon of mass destruction even BEFORE focusing it down to where it delivers a megawatt per square meter to a circle some 20 meters in radius.

Simply stupid.