Near term space industries for a solar syncronous industrial estate

[Fat Rooster]

With Starship on the horizon of slashing launch costs, it has occurred to me that we are on the verge of electricity being cheaper in space than on the ground. The idea is that a station in solar synchronous orbit will see perpetual high noon with no losses to angle, atmosphere, or night. The lack of any weather means that moderate concentration is possible, while there being no intermittency means very little storage will be required. Solar panels that would have an average output of 100W (peaking at 1kW) on earth could be pushed to having up to around 2kW continuous output. Cooling would end up being more of a challenge than electrical power.

Spoiler: The numbers

Show

Lets assume a 5 year return on investment cost, and we are aiming to hit a $0.05 per kWh. At ~8,000 hours a year that means we around $2000 per kW. This is viable if we can manage to get a full system down below 10kg per kW, and launch costs of $200/kg. That would translate to Starship getting 50 tons to SSO at a unit launch cost of $10million. Those numbers are realistic. A government subsidised habitation module supporting a few AI training graphics cards could be commercially viable at quite a small scale, and should be viable without subsidies once there is enough work to justify full time astronauts.

For the radiators; at 20C a radiator will expel around 400W per m², so we need around 2.5m² per kW. They would have to be lightweight, but not absurdly so.

As a side note, these numbers also imply we are really not very far from microwave power satellites being viable too, but they have a minimum size to be viable. The solar synchronous industrial estate is the stepping stone to orbital energy harvesting solving our energy crisis by being cheaper than hydrocarbons.

The most realistic scenario for large scale habitation of space is built around industrial stations (possibly one giant linked one so that astronauts can work wherever needed), but I don't think these will be purely dependent on zero G industry. I think they will be build around energy intensive low material industries. The first one that comes to mind is AI training datacentres. The second one is silicon purification. That could even use solar concentrators rather than solar panels.

What other industries exist that have very high energy throughput but low capital and material throughput? If they might benefit from zero G or vacuum conditions even better!