story
Here's the summary from NASA:
"Here are four examples [large graphic][1] of what it would take to send a canister about the size of a Shuttle payload (or a school bus) past our nearest neighboring star...and allowing 900 years for it to make this journey.
Well....If you use chemical engines like those that are on the Shuttle, well..., sorry, there isn’t enough mass in the universe to supply the rocket propellant you’d need.
So let’s step up to next possibilities, nuclear rockets with a predicted performance that’s 10 to 20 times better!
Well...it’s still not looking all that good. For a fission rocket you would need a BILLION SUPERTANKER size propellant tanks to get you there, and even with fusion rockets you would still need a THOUSAND SUPERTANKERS!
Even if we look at the best conceivable performance that we could engineer based on today’s knowledge, say an Ion engine or an antimatter rocket whose performance was 100 times better that the shuttle engines, we would need about ten railway tanker sized propellant tanks.
That doesn’t sound too bad, until you consider that we didn’t bring along any propellant to let us stop when we get to the other star system...or if we want to get there quicker than 9 centuries.
Once you add the desire to actually stop at your destination, or if you want to get there sooner, you’re back at the incredible supertanker situation again, even for our best conceivable rockets.
In conclusion, we’d really like to have a form of propulsion that doesn’t need any propellant! This implies the need to find some way to modify gravitational or inertial forces or to find some means to push against the very structure of spacetime itself."
The rest of the website[2] has some good information, even if the organization is a bit disjoint.
[1]: http://www.nasa.gov/centers/glenn/images/content/84509main_w...
[2]: http://www.nasa.gov/centers/glenn/technology/warp/ipspaper.h...