Well I can't speak for the author of the original article, but I can speak for myself. And what I know of rocket engineering (which is quite a bit for an ameteur, but I am not an engineer. Only a wannabe!) suggests that the performance of the nuclear engines in the Liberty Ship idea were too optimistic.

Having attempted at one time to actually design an open cycle gas-core nuclear engine using hydrogen in direct contact with the fuel, achieving 5000 sec of specific impulse and using centrifugal seperators to reclaim almost all of the fuel, operating at a nominal thermal power of about 25 GW: even without running neutron transport codes on a computer (a very, nontrivial task,) or attempting to simultaneously model the fluid dynamics of such a gasseous system, operating temperature of 250,000 K alone presented almost insurmountable problems with radiation (here thermal radiation) coupling to the walls.

Such an engine will have a low thrust to weight ratio: probably no more than 0.2 or so, which means that such an engine cannot lift itself off the ground--it can only operate in a space environment...

Further, I believe the "Liberty Ship" idea also used transparent quartz pressure vessels (i.e., light bulb reactor) to seperate the fissioning fuel and the hydrogen reaction mass. However, at the temperatures necessarily achieved by the fissioning gas, the 'color' of the UV emitted may be higher than the quartz can transmit: in otherwords, the quartz might become 'opaque' to the spectrum of UV emitted. Also, there is the issue of direct impingement of neutron radiation on the quartz vessel which will demand additional cooling for it, and will also shortly damage or destroy the optical qualities of the fused quartz vessels which will also result in increasing opacity and will also create embrittlement concerns.

For these reasons I believe that a vessel with the properties of the Liberty Ship are not possible. However, that is not to say that a nuclear powered SSTO isn't possible. But SSTO and back are technically very difficult, and this has little to do with nuclear power. The mechanics of rocket engineering is extremely demanding and Tsiokolvsky's Rocket Equation: deltaV=Ce*ln(Mi/Mf) is literally the letter of the law. Any single stage to orbit rocket is demanding whether chemically or nuclearly powered. When one factors in the demands of shielding in addition to structure, reusable surface insulation, etc. etc., then one quickly finds that such a rocket is theoretically possible, but the payload is going to be a miniscule fraction of the whole rocket. Because of this, most SSTO's require rediculously huge rockets for a reasonable payload.

One exception to this is the Nuclear Pulse Rocket such as the vehicle conceived of by Project: Orion which propels itself by exploding nuclear bombs beneath a thick steel plate. For various reasons this technological path also does not seem too attractive. Possible but not desirable, fallout issues besides. The problems of boost attitude control (yaw, pitch and roll) followed by terminal 'conventional rocket powered' RCS (reaction control system for translation and roll control) will eat up almost every kilogram of the supposed high payload capacity of the vehicle, resulting in a very modest 'real' payload that would be far cheaper to launch on a conventional launch vehicle.

Launching payloads into orbit is a non-trivial task: it is technically very hard. And will remain technically hard as long as we must rely on reaction (rocket) propulsion systems.

-- Edited by GoogleNaut at 02:45, 2007-09-04