This is very interesting, but I think Lithium might prove less than ideal if Starship ever becomes practical (and cost per kilogram to LEO falls significantly). At that point, it might be cheaper to use other propellant, even if specific impulse suffers, you can just carry more propellant. The nuclear powerplant will be the same in both cases.

> A human mission to Mars might need 2 to 4 megawatts of power

That’s 15x to 30x more than the ISS’s solar panels produce at their peak [1]. It’s 100x to 200x more power than the most-powerful reactors we plan to put in space before 2030 [2]. And that’s coming out of the nozzles—you have to put even more in.

This reminds me of Zubrin’s takedown of the VASIMIR hoax: “electrical propulsion…is limited by electrical power, which comes at a premium in space” [3].

Electric is great for deep space. Until we have big power plants in space, it’s a distraction for manned spaceflight.

[1] https://www.dsneg.com/info/a-comparison-between-the-photovol...

[2] https://www.nextbigfuture.com/2026/04/four-times-the-ion-dri...

[3] https://m.youtube.com/watch?v=myYs4DCCZts&pp=ygULVmFzaW1yIGh...