[ExI] Physical limits of electromagnetic launchers

[Rafal Smigrodzki]

On Mon, Jun 4, 2012 at 4:36 AM, Anders Sandberg <anders at aleph.se> wrote:

>
> Generally, I wonder about scaling down rockets. While Freitas has shown that
> one can power nanosmall systems using radioactive decay, I seem to recall
> that fission and fusion reactors do not scale down well (once you get below
> the mean free path of particles containment becomes hopeless). So it would
> be interesting to consider the smallest possible rocket. I imagine it would
> be a pencil-sized Orion rocket using antimatter pellets - antimatter at
> least scales perfectly, at least when it comes to exploding. But the shield
> still needs to be dense and thick enough to catch the kinetic energy of the
> photons, pions and muons.

### What about a lightsail probe that eats its huge sail, with a mass
orders of magnitude larger than the payload, to build a fusion-powered
ion drive that uses >99.9 of the initially launched mass to brake at
target location? If you are using a large laser array, the initial
mass of the spacecraft is not strongly constrained, at least much less
than the mass of something that needs to fit into an accelerator. An
alternative would be to accelerator-launch a stream of lighter
spacecraft at high frequency, at slightly varied launch speeds that
would cause them to aggregate a few months or years worth of travel
time before target location, again producing a large craft consisting
of >99.9 reaction mass for an ion drive. The second approach would not
be limited by the physical characteristics of laser propulsion but
achieving large masses at target would lengthen the total launch
duration, from the first to the last sub-spacecraft. And of course,
having a good supply of antimatter would presumably let you keep the
total reaction mass requirement down.

An interesting option would be to have multiple swarms of probes
converging in space and time to allow staged deceleration. At first
let's say 1000 probes would each eat 10 reaction mass pellets, and
slow down while consuming 9/10 of their reaction mass, then 9 of each
group of 10 send their remaining reaction mass to 1, producing 100
fully-fueled but much slower probes, which repeat the cycle, producing
10 even slower ones, which finally produce the final fully fueled
probe capable of braking to the target stellar orbit velocity. I have
a feeling this would improve the payload/reaction mass ratio over what
is doable with a single-stage deceleration, ceteris paribus, but I
don't have the math to prove it. But definitely it would improve
mission reliability (assuming the swarm would successfully stay
together and coalesce efficiently), since the success of the final
payload would be achieved from multiple independently braking objects,
rather than a single huge one.

Rafal