No, you could still do this with current-ish technology. The solar sail
would just be a large reflecting mirror, about 1 um thick (which implies
in-space film processing - perhaps take a heavy Mylar and then chemically
remove the polymer). The solar cells would be hanging below the sail on the
sunward side. This would allow you to concentrate the sunlight for higher
efficiency, and be more picky about the direction you emit waste heat. As
long as the waste heat does not go up, it will (on average) push outward
not inward.
Actually, a simplification - just make the entire sphere from 1um think
aluminum, with small solar cells spaced throughout. From the small solar
cell's point of view, it sees a sky entirely made from either direct
sunlight or reflected sunlight. So as long as the mirrors are reasonably
efficient I'd think that would be enough.
Thanks!
-David Summers
On Wed, Mar 4, 2020 at 1:09 PM Henry Spencer <hspencer@xxxxxxxxxxxxx> wrote:
On Tue, 3 Mar 2020, I wrote:
...So it is fairly obvious that you could make a bunch of "fixed" solar
sail like structures in a sphere around the sun...
If you can get total mass down to about 1.5g/m^2, which is what sunlight
will support at 1AU. That's challenging -- the Team Encounter sail
design, circa 20 years ago, was aiming for 3.63g/m^2 (including
structure and a minimal payload), and that was a very aggressive
design... The fundamental limits are much lower, but...
At the cost of having only half the thrust, your sails can be black
instead of reflective... Good solar cells are actually pretty close to
black, although it's not clear that you can make them thin enough...
Fast answer: barring breakthroughs, I suspect not.
Looking at some (admittedly rather old) references on thin-film solar
cells, the thinnest of the technologies can give you working cells where
the active layer plus the contact layers on top and bottom total circa
2.5um thick. Given what materials they're made of, a first guess is maybe
10g/m^2 for that. Such cells are normally made on *much* thicker
substrates, but if you can make them on L'Garde's 0.9um aluminized Mylar,
or make them on something thicker and then transfer them to that, you just
might be able to make working cells at 12g/m^2. Budget a bit for things
like seams and first-level wiring, and 15g/m^2 "membrane" mass would be
plausible; add the ultra-light structure L'Garde was proposing for the
Team Encounter design, and then a modest payload fraction, and you might
optimistically be at 20g/m^2.
Taking that down by a factor of ~25 will be tough. In particular, the
active layer, which is circa 5g/m^2 all by itself, is already about as
thin as it can be made and still be opaque! You'd have to start by
finding a much better cell material. I can't say for sure it's impossible
but I have my doubts.
Actually, what you want on the sunward side is something that absorbs
most of the visible but doesn't emit much thermal IR (with an IR-emitter
surface on the starward side)...
Thinking about this led me to notice another issue: the momentum of a
beam of photons carrying power P is just P/c, and this is independent of
wavelength. If you are absorbing all the sunlight in solar cells, then
re-radiating that energy outward as heat (some of it directly as waste
heat from the cells themselves, some of it from cooling radiators for the
electrical equipment the cells are powering), the photon thrust from the
waste heat has to be subtracted from the absorption thrust on the cells!
(Yes, radiators do produce thrust. The Topex/Poseidon spacecraft, with a
couple of radar altimeters using a lot of power and producing a lot of
heat, exploited its radiator thrust to help with orbit maintenance...)
Since the sunlight is arriving as a unidirectional (well, more or less)
beam, while the waste heat is radiated diffusely, reducing its thrust due
to divergence loss, I think you still end up with net thrust, but it's
even less than the simplistic calculation would indicate.
Henry