Nor did the panel find that NASA's nuclear programme would support its planned human missions. The NRC acknowledged that fission reactors would be useful for both space travel and long-term human bases on the Moon or Mars. But it said it is not clear whether the nuclear electric propulsion NASA is pursuing is "adequate for either application".

Another reactor technology that uses fission to heat hydrogen so it can be forcefully expelled to provide rocket thrust might get astronauts to Mars more quickly, the panel writes.

The NRC did identify several robotic missions where nuclear electric propulsion could be beneficial, including one to send probes and landers to Neptune and its largest moon, Triton.

But the NRC cautioned that significant hurdles remain for the technology to actually be practical. These include the ability to operate continuously - without repairs - for the decade or so that it would take to reach Neptune. In fact, earlier in 2005, NASA admitted that its first plan to use the technology - in a huge spacecraft called the Jupiter Icy Moons Orbiter - was too ambitious. It is now considering testing the reactor around Earth's Moon instead.

Trunk full of batteries

"There's an awful lot of technological development that's going to take a very long time," says Louise Prockter, a member of the NRC's solar system panel at Johns Hopkins University in Laurel, Maryland, US.

"There's still a lot we can do with the technology we have," she adds, referring to the radioisotope thermoelectric generators (RTGs) that "passively" produce electricity using heat released by the radioactive decay of plutonium-238. RTGs have been used to power dozens of space science missions, from the Voyager probes now at the edge of the solar system to the Cassini spacecraft around Saturn.

But Anthony Hyder, a physicist at the University of Notre Dame University in Indiana, US, points out that RTGs can only generate a few hundred watts of electricity. He says any missions requiring more than that would have to use many more RTGs - the equivalent of using a "trunk full" of flashlight batteries to start a car. "It's much easier at that point to graduate from radioisotopes to fission reactors," he says.

Death knell

However one scientist, who wished to remain anonymous, told New Scientist that public concern over the safety of nuclear reactors resulted in a Catch-22 situation: "You're not going to develop it until someone says they need it and no one is going to say they need it because they know it's a death knell for their programme."

Space researchers generally believe spacecraft reactors can be used safely, for example by launching the reactor in pieces before assembling and starting it in space.>

While reactors would definitely boost a mission's power level, the technology does come at a heavy financial cost. NASA projects Prometheus will cost $3 billion between now and 2010. In the agency's 2006 budget request, the money was scheduled to come from "exploration systems" - and not the science budget.