[Robert A Cook PE]

I'd go with solid CO2 since there's no weight lost for tanks, coolant required and refrigerators for the liquid CO2, and the handling of solid CO2 would seem to be easier.

Little bit of thermal stress as cold solid CO2 at one end of the reactor changes into hot gasses in the middle and far end of the reactor. That change would make calculations interesting, shall we say.

[NicknamedBob]

Solid storage would be good for long duration shuttle tugs, asteroid miners (and rock-chuckers), but I think for the atmosphere ships I'd want a tank to pressurize.

I wonder what the trade-off would be for a Star-Launch Tug using an expendable Nuke engine with massive dry ice propellant?