The only real advantages of water over xenon in such applications is that water doubles as a source of fuel for the fusion plant, and oxygen for the LS system, and its ubiquity.
I'm assuming that the power plant is a closed system D-D or D-T fusion reactor at the moment, and that PP fuel isn't a consideration - maybe top it off once a year.
I'm considering a kind of 'universal' engine capable of working in turbojet, ramjet, scramjet and rocket mode.
It would work inatmo like those crazy nuclear-powered jets proposed back in the 50s, with heat transference through liquid metal loops. No propellant would be required. Exatmo, the ducts would be sealed off and the ship would switch to rocket mode - this is where the water would come in.
It strikes me that, economically, the sheer plentifulness of water would make this viable BUT I have no idea how water behaves as a plasma: it would dissociate into OH-, H20+, O-, free electrons and protons. Would this cocktail perform ok?
Being 14 times as dense as L-Hyd, water wouldn't suffer from annoying relativistic or superluminal exhaust velocities (like HEPlaR) in order to get a decent Isp. For a variety of reasons, I want the Isp of rockets to be in the 500,000s - 1,000,000s range.
Does this sound even remotely plausible?